US Pat. No. 9,471,344

HARDWARE SUPPORT FOR PROCESSING VIRTUAL MACHINE INSTRUCTIONS

Marvell International Ltd...

1. A system on an integrated circuit chip (SOC) for a storage device, the SOC comprising:
an interface for the SOC;
a memory device that is separate from a central processing unit (CPU);
a cache memory device that is separate from the memory device;
the CPU coupled with the interface and the memory device, the CPU configured to process instructions defined by a first instruction
set architecture and configured to write, to the memory device, bytecode representing instructions defined by a second instruction
set architecture that is different from the first instruction set architecture, wherein the instructions defined by the second
instruction set architecture conform to a predefined virtual machine environment;

an interpreter, executed by the CPU, the interpreter configured to convert a scripting code, which conforms to the predefined
virtual machine environment, to generate the bytecode representing the instructions defined by the second instruction set
architecture;

a cache memory manager, which is tightly coupled to the cache memory device, the cache memory manager configured to:
after the CPU has written the bytecode to the memory device and based on an instruction from the CPU, retrieves, at each memory
cycle, stored bytecode from the memory device;

writes, at each memory cycle, the stored bytecode to the cache memory device; and
processing hardware coupled with the interface and the cache memory device, the processing hardware being separate from the
CPU and configured to, after the cache memory manager has written the stored bytecode to the cache memory device, retrieve
the stored bytecode from the cache memory device, and process the stored bytecode to execute the instructions in the predefined
virtual machine environment to cause the processing hardware to perform a self-test of the storage device.

US Pat. No. 9,338,812

METHOD AND APPARATUS FOR SUPPORTING MULTIPLE CONNECTIONS OVER DIFFERENT TYPES OF ACCESS IN 3GPP SYSTEMS

Marvell International Ltd...

1. A method for routing packets from a home agent to a user equipment through (i) a first access network and (ii) a second
access network, the method comprising:
sending a first binding update message from the user equipment over the first access network to create a first binding cache
entry at the home agent, wherein the first binding update message includes an identification of a first flow of packets that
is to be transmitted, via the first access network, between (i) the user equipment and (ii) the home agent;

subsequent to sending the first binding update message, sending a second binding update message from the user equipment over
the first access network to create a second binding cache entry at the home agent, wherein the second binding update message
includes an identification of a second flow of packets that is to be transmitted, via the second access network, between (i)
the user equipment and (ii) the home agent, and wherein (i) the first access network is a foreign link for the user equipment
and (ii) the second access network is a home link for the user equipment; and

based on the first binding cache entry and the second binding cache entry, respectively receiving, by the user equipment from
the home agent, (i) a first plurality of packets through the first access network and (ii) a second plurality of packets through
the second access network.

US Pat. No. 9,473,871

SYSTEMS AND METHODS FOR AUDIO MANAGEMENT

MARVELL INTERNATIONAL LTD...

1. A method for audio management, the method comprising:
determining initial head-related transfer function (HRTF) parameters indicating an initial virtual configuration of a plurality
of audio sources;

detecting a first user operation, through a user interface, to change the initial virtual configuration;
generating target HRTF parameters in response to the first user operation; and
determining a target virtual configuration of the plurality of audio sources based at least in part on the target HRTF parameters.

US Pat. No. 9,363,111

DECISION FEEDBACK EQUALIZER (DFE) UTILIZING ERROR PROPAGATION INFORMATION

Marvell International Ltd...

11. A system comprising:
a first decision feedback equalizer;
a second decision feedback equalizer, wherein each of the first decision feedback equalizer and the second decision feedback
equalizer is configured to receive a signal comprising a plurality of symbols, wherein the first decision feedback equalizer
is further configured to (i) estimate a first value of a first symbol of the plurality of symbols, (ii) based on the first
value of the first symbol, cancel, at least in part, inter symbol interference that the first symbol has on a second symbol
of the plurality of symbols, and (iii) determine a first error associated with the estimation of the first value of the first
symbol; and

an error detector module configured to detect whether the first error is higher than a threshold value,
wherein the second decision feedback equalizer is further configured to, in response to the first error being higher than
the threshold value, (i) receive a second value of the first symbol that is different from the first value of the first symbol,
and (ii) based on the second value of the first symbol, cancel, at least in part, inter symbol interference that the first
symbol has on the second symbol.

US Pat. No. 9,247,243

INTERPOLATED VIDEO ERROR CONCEALMENT

Marvell International Ltd...

1. A method comprising:
receiving a first stream of video frames at a first frame rate;
generating a second stream of video frames at a second frame rate, wherein the second stream comprises (i) video frames of
the first stream of video frames and (ii) sequences of intermediate video frames inserted between video frames of the first
stream of video frames;

interpolating between a pair of adjacent video frames of the first stream of video frames to produce a first number of interpolated
intermediate video frames in a particular sequence of the intermediate video frames;

duplicating at least one of the adjacent video frames of the pair of adjacent video frames to produce a second number of duplicated
intermediate video frames in the particular sequence of the intermediate video frames;

receiving a plurality of interpolation quality values corresponding to one or more interpolated intermediate video frames;
temporally filtering the plurality of interpolation quality values to obtain observed interpolation quality; and
in response to a change in the observed interpolation quality, dynamically changing the first number of interpolated intermediate
video frames and the second number of duplicated intermediate video frames in the second stream of video frames.

US Pat. No. 9,247,621

METHOD AND APPARATUS FOR SELECTIVELY LOADING A DIMMER CIRCUIT

Marvell International Ltd...

1. An apparatus comprising:
a magnetic component arrangement; and
a switch coupled to the magnetic component arrangement, wherein the switch is configured to operate
during a first time period, at a first frequency that results in transfer of a first amount of power to a lighting element
via the magnetic component arrangement, and

during a second time period, at a second frequency,
wherein while the switch operates at the second frequency, zero or nearly zero power is transferred to the lighting element
via the magnetic component arrangement, and

wherein the second frequency is higher than the first frequency.

US Pat. No. 9,183,078

PROVIDING ERROR CHECKING AND CORRECTING (ECC) CAPABILITY FOR MEMORY

Marvell International Ltd...

1. An apparatus, comprising:
a memory interface configured to communicate with a memory, the memory including a plurality of concurrently accessible memory
devices; and

a controller coupled with the memory interface, the controller configured to:
receive a data word to be stored in the memory,
generate error checking and correcting (ECC) information for the data word, the data word and ECC information forming an encoded
word, and

distribute bits of the encoded word across the plurality of concurrently accessible memory devices in accordance with one
or more indications specifying a number of the bits of the encoded word to store in a wordline of each of the plurality of
concurrently accessible memory devices, wherein bits of the data word are to be stored in two or more memory devices of the
plurality of memory devices, and bits of the ECC information are to be stored in two or more memory devices of the plurality
of memory devices,

wherein the controller is configured to determine a distribution of the bits of the encoded word across the plurality of concurrently
accessible memory devices and identify the two or more memory devices of the plurality of memory devices in which the bits
of the data word are to be stored and the two or more memory devices of the plurality of memory devices in which the bits
of the ECC information are to be stored.

US Pat. No. 9,369,914

ADAPTIVELY DETERMINING A DATA RATE OF PACKETIZED INFORMATION TRANSMISSION OVER A WIRELESS CHANNEL

Marvell International Ltd...

1. An apparatus comprising:
circuitry configured to derive two or more measures from a received signal having subcarriers, the two or more measures comprising
a first measure that is based on channel gain values that are determined respectively for the subcarriers and a second measure
that is based on an indication of frequency selectivity in a wireless channel across the subcarriers;

a rate controller configured to determine a data rate of information transmission over the wireless channel based on the two
or more measures, wherein the first measure serves as a primary determinant of the data rate and the second measure serves
as a secondary determinant of the data rate; and

circuitry configured to cause a transmission of information over the wireless channel in accordance with the data rate,
wherein the rate controller is configured to determine a first data rate based on the first measure, and determine a second
data rate based on the first data rate and the second measure, wherein the transmission of information is in accordance with
the second data rate,

wherein the first data rate is determined based on the first measure and one or more signal quality thresholds, and wherein
the second data rate is determined based on an adjustment of the first data rate if the first measure falls within a tolerance
range of a signal quality threshold of the one or more signal quality thresholds and if the second measure indicates an adjustment
is needed.

US Pat. No. 9,232,499

COMMON CHANNELS FOR WIRELESS NETWORK ACCESS

MARVELL INTERNATIONAL LTD...

1. An apparatus comprising:
a wireless receiver configured to scan for a first beacon in one or more of a first plurality of predetermined wireless channels,
wherein each of the first plurality of predetermined wireless channels has the same carrier frequency in a plurality of different
geographical regions; and

a controller configured to
(i) determine whether any of the first plurality of predetermined wireless channels are available,
(ii) responsive to determining that any of the first plurality of predetermined wireless channels are available, establish
a wireless network in one of the first plurality of predetermined wireless channels, and

(iii) responsive to determining that none of the first plurality of predetermined wireless channels are available,
identify, based on the first beacon scanned by the wireless receiver in the one or more of the first plurality of predetermined
wireless channels, a first geographical region of the plurality of different geographical regions,

select a second plurality of wireless channels associated with the first geographical region,
scan for a second beacon in one or more of the second plurality of wireless channels selected by the controller,
determine whether any of the second plurality of wireless channels are available, and
establish the wireless network in one of the second plurality of wireless channels responsive to determining that any of the
second plurality of wireless channels are available.

US Pat. No. 9,167,614

TUNNELED DIRECT LINK SETUP SYSTEMS AND METHODS WITH CONSISTENT LINK INFORMATION MAINTENANCE

Marvell International Ltd...

1. A method comprising:
directly receiving, at a first station, a signal transmitted from a second station, wherein each of the first station and
the second station respectively includes a link status indicator that indicates whether a direct link exists between the first
station and the second station, and wherein the signal directly received at the first station was transmitted by the second
station based on the link status indicator of the second station indicating that a direct link does exist between the first
station and the second station; and

responsive to the link status indicator of the first station indicating that a direct link does not exist between the first
station and the second station and while the link status indicator of the second station indicates that the direct link does
exist and conflicts with the link status indicator of the first station,

dropping the signal received at the first station, and
subsequent to dropping the signal received at the first station, transmitting a teardown signal from the first station to
the second station, wherein responsive to the teardown signal being transmitted from the first station to the second station,
the link status indicator of the second station is to be changed, by the second station, to indicate that a direct link does
not exist between the first station and the second station.

US Pat. No. 9,143,210

SOFT SPHERE DECODER FOR MIMO MAXIMUM LIKELIHOOD DEMODULATION

Marvell International Ltd...

1. A method for decoding a data vector by a processor, the method comprising:
partitioning candidate signal values of the data vector into at least a first partition and a second partition, wherein candidate
signal values in the first partition have a same, first symbol value at a particular symbol position, and candidate signal
values in the second partition have a same, second symbol value at the particular symbol position;

determining a maximum radius value of a first radius associated with the first partition and a second radius associated with
the second partition;

generating a representation of the plurality of candidate signal values, wherein the representation comprises a first portion
representing the first partition and a second portion representing the second partition; and

performing a search based on the generated representation of the candidate signal values and the maximum radius value to compute
decoding information for the data vector.

US Pat. No. 9,224,419

DATA DETECTOR FOR DETECTING DATA IN PRESENCE OF CONTROL SIGNALS

Marvell International Ltd...

1. A detector for detecting a data signal from a storage device, comprising:
a signal estimator configured to generate an estimate of a control signal by determining characteristics of the control signal
from a read signal, wherein the read signal includes the data signal and the control signal embedded together, wherein the
signal estimator is configured to generate the estimate of the control signal as a function of the characteristics, wherein
the control signal is a continuous servo signal, and

wherein the signal estimator is configured to generate the estimate of the control signal by filtering the read signal and
sampling values of the filtered read signal to produce the characteristics of the control signal from which the estimate is
generated; and

a cancellation unit configured to produce the data signal by cancelling the estimate of the control signal from the read signal.

US Pat. No. 9,166,505

PROTECTION CIRCUITS FOR MOTORS

MARVELL INTERNATIONAL LTD...

1. A protection system for a direct current (DC) motor, comprising:
a motor drive component configured to drive a DC motor;
a feedback component configured to generate a feedback signal related to motion of the DC motor; and
a watchdog circuit configured to output a fault signal to the motor drive component to stop the DC motor in response to the
feedback signal indicating abnormal motion of the DC motor, wherein the watchdog circuit includes:

a comparator configured to generate a comparison signal indicating a comparison of (i) a modulation signal output by a motor
controller and (ii) a threshold signal;

a first AND gate configured to receive the comparison signal and an enable signal and generate an first signal;
a hardware watchdog timer configured to receive the first signal and the feedback signal and generate a watchdog signal; and
a second AND gate configured to receive the modulation signal and the watchdog signal and generate a second signal to affect
the motion of the DC motor.

US Pat. No. 9,402,265

SYNCHRONIZED INTERFERENCE MITIGATION SCHEME FOR HETEROGENEOUS WIRELESS NETWORKS

MARVELL INTERNATIONAL LTD...

1. A method of synchronizing time domain multiplexing interference mitigation schemes of at least a first wireless communication
link within a first network of a heterogeneous wireless network, a second wireless communication link within a second network
of the heterogeneous wireless network, and a third wireless communication link within a third network, the method comprising:
determining, by a network controller of the heterogeneous wireless network, a first set of one or more parameters defining
a first set of time intervals, wherein

the first set of time intervals restrict access to the first wireless communication link to reduce interference with the second
wireless communication link;

determining, by the network controller, a second set of one or more parameters defining a second set of time intervals based
on the first set of time intervals, wherein

the second set of time intervals restrict access to the first wireless communication link to reduce interference with the
third wireless communication link; and

causing, by the network controller, the first set of one or more parameters and the second set of one or more parameters to
be transmitted to a first user device of the first wireless communication link and the third wireless communication link so
that the first user device schedules i) data transmissions of the first wireless communication link to reduce interference
with the second wireless communication link, and ii) data transmissions of the third wireless communication link to reduce
interference with the first wireless communication link.

US Pat. No. 9,225,658

PACKET AGGREGATION

Marvell International Ltd...

1. A networking device comprising:
a transmitting device configured to store packets in a transmit queue, wherein the transmitting device is further configured
to transmit the packets from the transmit queue in accordance with a local area network protocol, and wherein the transmitting
device is further configured to selectively generate a signal indicating an available capacity of the transmit queue; and

a host circuit configured to receive input packets for transmission by the transmitting device,
wherein the host circuit is further configured to, in response to an aggregation mode being enabled, (i) generate an aggregate
packet using ones of the received input packets, and (ii) in response to the signal indicating the available capacity of the
transmit queue, provide the aggregate packet to the transmitting device for transmission in accordance with the local area
network protocol, and

wherein the host circuit is further configured to, in response to the aggregation mode being disabled, individually provide
the received input packets to the transmitting device for transmission in accordance with the local area network protocol.

US Pat. No. 9,185,052

SWITCH DEVICE HAVING A PLURALITY OF PROCESSING CORES

Marvell International Ltd...

1. A switching device, comprising:
a plurality of ports; and
a plurality of switch cores disposed on one or more integrated circuits;
wherein:
a first switch core in the plurality of switch cores is configured to:
receive network data frames only via a first subset of the plurality of ports, the first subset corresponding to the first
switch core,

perform an ingress processing operation on a first network data frame received at a first port in the first subset, including
determining that the first network data frame is to be transmitted via a second port in a second subset of the plurality of
ports, the second subset corresponding to a second switch core, and

direct, after ingress processing of the first network data frame, at least i) first header information corresponding to the
first network data frame and ii) an identifier of the second port to the second switch core in the plurality of switch cores;

the second switch core being configured to:
perform an egress processing operation on the first network data frame,
direct the first network data frame to the second port in the second subset of the plurality of ports using the identifier
of the second port, wherein none of the ports in the second subset are in the first subset, and

transmit the first network data frame via the second port.

US Pat. No. 9,160,489

ORIGIN DODGING FOR TRANSMITTED SIGNAL

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
origin dodging logic configured to:
receive a transmission sequence for transmission to a receiver, wherein the transmission sequence includes a sequence of transmit
symbols, wherein each transmit symbol corresponds to a coordinate pair positioning the transmit symbol's phase on an I-Q plane;

determine whether a difference in phase between consecutive transmit symbols is 180 degrees;
in response to determining that the difference in phase is 180 degrees:
sample and filter the transmission sequence to create transition samples;
determine whether any transition sample is within a predetermined radius of an origin of the I-Q plane;
when the difference in phase is 180 degrees and a transition sample is positioned within the radius:
generate a dodging symbol; and
insert the dodging symbol into the transmission sequence to create a modified transmission sequence; and
sampling and filtering the modified transmission sequence to produce filtered samples; and
provide a filtered signal that includes the filtered samples for transmission to a receiver.

US Pat. No. 9,313,789

SYSTEMS AND METHODS FOR TRANSMITTING USING PRE-CODING WITH MULTIPLE CODEBOOKS

Marvell International LTD...

1. A system comprising:
a transmitter configured to
transmit data on a first set of frequencies of a plurality of frequencies,
wherein the data transmitted on the first set of frequencies is respectively transmitted using a first set of pre-coding matrices
from a plurality of pre-coding matrices, and

wherein data on each frequency of the first set of frequencies is transmitted using a different pre-coding matrix from the
first set of pre-coding matrices; and

transmit data on a second set of frequencies of the plurality of frequencies,
wherein the data transmitted on the second set of frequencies is respectively transmitted using the first set of pre-coding
matrices,

wherein data on each frequency of the second set of frequencies is transmitted using a different pre-coding matrix from the
first set of pre-coding matrices,

wherein transmission of data on the second set of frequencies is subsequent to transmission of data the first set of frequencies,
wherein the first set of pre-coding matrices is arranged in a predetermined order, and
wherein the predetermined order includes an ascending order, a descending order, or an alternating order.

US Pat. No. 9,325,346

SYSTEMS AND METHODS FOR HANDLING PARITY AND FORWARDED ERROR IN BUS WIDTH CONVERSION

Marvell International Ltd...

1. A method of encoding forwarded error for data bus width conversion, the method comprising:
receiving, over a first number of input buses having a first total width, input data and input forwarded error information,
wherein a first input bus of the first number of input buses has a first width;

computing system input parity information based on the input data and the input forwarded error information;
providing, over a second number of output buses having the first total width, output data based on the input data and system
output parity information based on the system input parity information, wherein a first output bus of the second number of
output buses has a second width different from the first width;

determining output parity data based on the output data;
comparing the output parity data and the system output parity information to determine whether there is an error in the output
data; and

determining whether to indicate a forwarded error output based on the comparing.

US Pat. No. 9,244,762

METHOD AND APPARATUS FOR FLEXIBLE BUFFERS IN AN XOR ENGINE

Marvell International Ltd...

1. An apparatus comprising:
a hardware engine that is configured to generate a first result, a second result, and a third result from the received user
data, the first result comprising any of a first type of checksum, a second type of checksum, or an unmodified copy of the
received user data, the second result comprising any of the first type of checksum, the second type of checksum, or the unmodified
copy of the received user data, and the third result comprising any of the first type of checksum, the second type of checksum,
or the unmodified copy of the received user data;

a first local result buffer configured to store the first result;
a second local result buffer configured to store the second result; and
a third local result buffer configured to store the third result.

US Pat. No. 9,106,295

REUSE OF A MATRIX EQUALIZER FOR THE PURPOSE OF TRANSMIT BEAMFORMING IN A WIRELESS MIMO COMMUNICATION SYSTEM

MARVELL INTERNATIONAL LTD...

1. A method, comprising:
receiving channel state information (CSI);
selecting for an input of a matrix equalizer computational device of a matrix equalizer either i) a signal to be equalized
or ii) the CSI, including selecting the CSI at a time when matrix computation circuitry of the matrix equalizer computational
device is not needed for matrix equalization;

when the signal is selected for the input of the matrix equalizer computational device,
providing the signal to the input of the matrix equalizer computational device, and
applying an equalizer to the signal, wherein the signal is i) to be transmitted via at least one multiple input, multiple
output (MIMO) channel or ii) received via the at least one MIMO channel, and wherein the equalizer is generated using the
matrix computation circuitry of the matrix equalizer computational device; and

when the CSI is selected for the input of the matrix equalizer computational device,
providing the CSI to the input of the matrix equalizer computational device,
performing a beamsteering calculation by reusing the matrix computation circuitry of the matrix equalizer computational device,
wherein the beamsteering calculation is based on the CSI, and

selecting one or more transmit beamsteering codewords from a transmit beamsteering codebook based on an output of the beamsteering
calculation performed by the matrix equalizer computational device in response to the CSI provided to the matrix equalizer
computational device.

US Pat. No. 9,082,418

METHODS FOR READING DATA FROM A STORAGE MEDIUM USING A READER AND STORAGE DEVICES

MARVELL INTERNATIONAL LTD...

1. A method for reading data from a storage medium using a reader, the method comprising:
reading a first signal with a reader head of the reader at a first position of the storage medium, wherein the first signal
includes a first portion corresponding to a first track and a second portion corresponding to a second track;

reading a second signal with the reader head of the reader at a second position of the storage medium, wherein the second
signal includes a third portion corresponding to the first track and a fourth portion corresponding to the second track;

estimating, based on the first portion and the second portion, (i) a first ratio of an amplitude of the first signal contributed
by the first track with the reader head of the reader at the first position and (ii) a second ratio of the amplitude of the
first signal contributed by the second track with the reader head of the reader at the first position;

estimating, based on the third portion and the fourth portion, (i) a third ratio of the amplitude of the second signal contributed
by the first track with the reader head of the reader at the second position and (ii) a fourth ratio of the amplitude of the
second signal contributed by the second track with the reader head of the reader at the second position; and

determining data stored at a pre-determined position of the storage medium based on the estimated first ratio, the estimated
second ratio, the estimated third ratio, and the estimated fourth ratio,

wherein the storage medium comprises a plurality of tracks including the first track and the second track, each track having
a width smaller than a width of the reader head of the reader.

US Pat. No. 9,497,789

METHODS, APPARATUS, AND DEVICE FOR WIRELESSLY CONNECTING A PLURALITY OF DEVICES FOR COMMUNICATION IN A NETWORK

Marvell International Ltd...

1. A method implemented in a first wireless transceiver device, the method comprising:
simultaneously using each of a first technique and a second technique in attempting to wirelessly connect the first wireless
transceiver device with one or more wireless transceiver devices, the first technique comprising a paging based technique
usable to establish an initial wireless connection between the first wireless transceiver device and a second wireless transceiver
device and the second technique comprising a low latency reconnect technique usable to reestablish a wireless connection between
the first wireless transceiver device and a third wireless transceiver device;

listening, on a particular one of multiple channels, for a first particular type of packet from the second wireless transceiver
device to establish the initial wireless connection with the second wireless transceiver device using the first technique;

listening, on the particular one channel, concurrently with listening for the first particular type of packet, for a second
particular type of packet from the third wireless transceiver device to reestablish the wireless connection with the third
wireless transceiver device using the second technique; and

in response to receiving the second particular type of packet, interrupting the first technique and performing the second
technique to wirelessly connect the first wireless transceiver device to the third wireless transceiver device.

US Pat. No. 9,402,238

METHOD AND APPARATUS FOR CONTROLLING A TRANSMIT POWER LEVEL OF A DEVICE IN A WIRELESS COMMUNICATIONS NETWORK

Marvell International Ltd...

1. A wireless communication system comprising:
a first wireless access point; and
a first set of first stations in wireless communication with the first wireless access point;
wherein the first set of first stations is divided into groups, each group including at least one of the first stations;
wherein each group of first stations is assigned to a respective one of a plurality of power levels;
wherein the first wireless access point controls power for communications with at least one of the groups of first stations
by announcing a power level on a channel;

wherein a station in the at least one of the groups of first stations accesses the channel when the announced power level
is a respective one of the plurality of power levels to which the at least one group of first stations is assigned; and

wherein the first wireless access point implicitly assigns first stations within the first set of at least one first station
to respective ones of the plurality of power levels by announcing the plurality of power levels, allowing a respective one
of the first stations to select a respective one of the plurality of power levels according to power needs of the respective
one of the first stations.

US Pat. No. 9,210,531

SERVER FOR WIRELESS APPLICATION SERVICE SYSTEM

Marvell International Ltd...

1. A wireless client device comprising:
a receiver configured to wirelessly receive a beacon from a wireless access point, wherein the beacon includes an information
element indicating whether the wireless access point is capable of providing a wireless application service;

a controller configured to, in response to the information element indicating that the wireless access point is capable of
providing a wireless application service, generate a wireless application service request, wherein the wireless application
service request

identifies a requested wireless application service, and
identifies a source of an application to be obtained, installed, and executed by the wireless access point, wherein execution
of the application by the wireless access point enables the wireless access point to provide the requested wireless application
service to the wireless client device, and wherein the source of the application is separate from the wireless access point;
and

a transmitter configured to wirelessly transmit a packet to the wireless access point, wherein the packet includes the wireless
application service request.

US Pat. No. 9,110,674

SYSTEMS AND METHODS FOR DYNAMIC POWER MANAGEMENT AND PERFORMANCE ADJUSTMENT OF MEDIA PLAYING

MARVELL INTERNATIONAL LTD...

1. A method for dynamically adjusting a processing speed of a pipeline, wherein the pipeline includes components that are
synchronized for processing media, the method comprising:
determining a speed at which the pipeline processes the media, wherein the speed is a measure of frames per unit time;
determining if the speed exceeds or is lower than a predetermined threshold set for a target pipeline performance level;
selecting, from among the components of the pipeline, one or more of the components that is the most power consuming or that
is causing a bottleneck in the speed at which the pipeline processes the media;

if the speed at which the pipeline processes the media exceeds the predetermined threshold, then decreasing a component performance
level of the one or more of the selected components to decrease the speed at which the pipeline processes the media; and

if the speed at which the pipeline processes the media is lower than the predetermined threshold but the component performance
level of the one or more of the selected components is high enough to support the target pipeline performance level, then
increasing the component performance level of the one or more of the selected components to increase the speed at which the
pipeline processes the media.

US Pat. No. 9,408,090

SIGNALING GUARD INTERVAL CAPABILITY IN A COMMUNICATION SYSTEM

MARVELL INTERNATIONAL LTD...

1. A method for determining capabilities of a first communication device in a wireless network, wherein the wireless network
is configured to use a first guard interval between symbols or a second guard interval between symbols, wherein the first
guard interval has a length shorter than a length of the second guard interval, the method comprising:
analyzing, at a second communication device, a field in a data unit received from the first communication device to determine
a set of two or more modulation and coding schemes (MCSs) supported by the first communication device and to determine whether
one or more MCSs in the set of two or more MCSs are supported by the first communication device when using the first guard
interval, including,

analyzing a first subfield to determine the set of two or more MCSs supported by the first communication device, and
analyzing a second subfield to determine a highest supported data rate,
wherein the highest supported data rate indicates whether each of the two or more MCSs is supported by the first communication
device when using the first guard interval; and

utilizing, at the second communication device, i) one MCS in the set of two or more MCSs and ii) the first guard interval
a) when communicating with the first communication device and b) when it is determined that the one MCS is supported by the
first communication device when using the first guard interval.

US Pat. No. 9,262,249

ADJUSTING A DEFECT THRESHOLD

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
defect detection logic configured to determine, using a defect threshold, when a read channel has read data from a defective
portion of a storage medium; and

threshold adjustment logic, implemented in at least hardware, configured to adjust the defect threshold based, at least in
part, on a gain value and a delayed gain value of a gain loop of the read channel.

US Pat. No. 9,124,454

METHOD AND APPARATUS FOR ADAPTIVELY DETERMINING SETTINGS OF A TRANSMIT EQUALIZER

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
a coefficient adapter configured to remove a signal component from a signal received from a transmitter to produce a residual
signal, wherein the signal component is added to the signal by a transmit equalizer of the transmitter that functions according
to tap coefficients, and

wherein the coefficient adapter is configured to generate settings for modifying the tap coefficients at the transmitter as
a function of the residual signal; and

an adjuster configured to adaptively adjust the tap coefficients of the transmit equalizer by providing the settings to the
transmitter, wherein the transmitter provides the signal across a channel to the apparatus,

wherein the adjuster is configured to accumulate a gradient of revised tap coefficients, wherein the adjuster periodically
transmits the settings that are based, at least in part, on the gradient of revised tap coefficients to the transmitter, and
wherein the transmitter is on a separate side of the channel from the apparatus.

US Pat. No. 9,402,279

MULTI-LEVEL ARBITRATION FOR WIRELESS DEVICE HAVING MULTIPLE RADIO RESOURCES

Marvell International Ltd...

1. A wireless communication device, comprising:
two or more radio resources configured to be collectively used by multiple different wireless radio technologies, wherein
the multiple different wireless technologies include a first technology, a second technology, a third technology, and a fourth
technology; and

an arbitrator configured to perform a multi-level arbitration among the first technology, the second technology, the third
technology, and the fourth technology of the multiple different wireless radio technologies to produce an arbitration result,

wherein the multi-level arbitration comprises performing at least one arbitration based on a criterion comprising potential
inter-technology frequency interference among a subset of the multiple different wireless radio technologies, wherein the
multi-level arbitration comprises performing at least one arbitration based on a criterion comprising temperature of the device,
a collective current draw of the two or more radio resources, or both,

wherein the multi-level arbitration comprises a first level arbitration to produce a first level result, a second level arbitration
to produce a second level result when required by the first level result, and a third level arbitration to produce a third
level result when required by the second level result, the arbitration result being based on (i) the first level result, (ii)
the second level result, and (iii) the third level result, if any, and

wherein the two or more radio resources are operated in accordance with the arbitration result.

US Pat. No. 9,122,590

FLASH MEMORY READ PERFORMANCE

Marvell International Ltd...

1. A flash memory controller, configured to:
issue a sequence of requests to a memory cell, each request in the sequence of requests having a different reference voltage
associated with the request;

responsive to issuing the sequence of requests, receive a plurality of reference bits from the memory cell;
process a first reference bit of the plurality of reference bits to obtain a quantized value;
translate the quantized value to a Log Likelihood Ratio (LLR);
decode the first reference bit using the LLR to obtain a data bit that corresponds to a most-significant bit of a data value
stored in the memory cell; and

store the first reference bit of the plurality of references bits in a first page memory.

US Pat. No. 9,300,328

METHODOLOGY FOR IMPROVED BIT-FLIPPING DECODER IN 1-READ AND 2-READ SCENARIOS

MARVELL INTERNATIONAL LTD...

1. A decoder comprising:
a memory; and
decoding circuitry configured to:
retrieve data related to a symbol from the memory; and
process the data to determine whether to toggle a value of the symbol based at least in part on whether the symbol was previously
toggled.

US Pat. No. 9,253,175

AUTHENTICATION OF COMPUTING DEVICES USING AUGMENTED CREDENTIALS TO ENABLE ACTIONS-PER-GROUP

Marvell International Ltd...

1. An authentication system comprising:
an authenticator;
an authentication server; and
an association table associated with the authentication server, the authenticator, and one or more computing devices capable
of a network connection with the authenticator, the association table:

mapping ports of the authenticator and one or more credentials identifying the one or more computing devices, and
configured to be used by the authentication server to authenticate the one or more computing devices, the authentication being
based on the one or more credentials mapped in the association table to the ports that connect the one or more computing devices
to the authenticator, the authentication being effective to permit an enabling action or a disabling action of at least one
of the one or more computing devices.

US Pat. No. 9,141,543

SYSTEMS AND METHODS FOR WRITING DATA FROM A CACHING AGENT TO MAIN MEMORY ACCORDING TO A PRE-CLEAN CRITERION

MARVELL INTERNATIONAL LTD...

1. A method for writing data from a caching agent to a main memory in a computer system, the method comprising:
receiving notice that a cache line of the caching agent has been evicted, the eviction of the cache line including:
opening a page of the main memory,
writing data of the cache line to the page, and
removing the data of the cache line from the caching agent; and
in response to the receipt of the notice and prior to a closing of the page:
identifying one or more other cache lines of the caching agent based on a proximity of the one or more other cache lines to
the evicted cache line, each of the one or more other cache lines being a dirty cache line that is different than the evicted
cache line, wherein the one or more other cache lines of the caching agent are identified based on a pre-clean criterion that
includes a requirement that each of the one or more other cache lines be associated with a memory address that differs from
a memory address of the evicted cache line by a threshold number of memory address locations, and

writing all data of the one or more other cache lines to the main memory without performing an eviction of the one or more
other cache lines.

US Pat. No. 9,119,082

REFERENCE SIGNAL DESIGN FOR INTERFERENCE MEASUREMENT

MARVELL INTERNATIONAL LTD...

1. A method, comprising:
in a mobile communication terminal, receiving signals from a group of cells in multiple time-frequency Resource Elements (REs),
wherein the signals comprise a desired signal component and an interfering signal component;

identifying in the terminal a subset of the REs, in which the cells are known deterministically to transmit at least part
of the interfering signal component and to not transmit the desired signal component; and

estimating in the terminal an interference caused by the interfering signal component to the desired signal component, by
measuring the signals in the REs in the subset.

US Pat. No. 9,185,653

MEMORY EFFICIENT FILTERING

MARVELL INTERNATIONAL LTD...

1. A method for controlling operation of a wireless network device that includes (i) a host assembly, (ii) a wireless device
assembly, and (iii) an interface assembly having one or more device interfaces connecting the host assembly and the wireless
device assembly, the method comprising:
receiving a filter expression at the wireless device assembly from the host assembly;
while the host assembly is in a sleep mode,
receiving a broadcast signal at the wireless device assembly, the broadcast signal including packet data,
determining, at the wireless device assembly, based on the filter expression, whether the packet data contains a particular
bit pattern indicated by the filter expression,

if the packet data contains the particular bit pattern indicated by the filter expression, the wireless device assembly communicating
a wakeup signal to the host assembly for changing the host assembly from the sleep mode to an active mode, and

if the packet data does not contain the particular bit pattern indicated by the filter expression, not communicating the wakeup
signal to the host assembly.

US Pat. No. 9,143,168

METHOD AND SYSTEM FOR REMOVING INTERFERENCE CAUSED BY SERVO DATA FROM USER DATA

Marvell International Ltd...

1. A method comprising:
receiving encoded data, wherein the encoded data comprises (i) first data, (ii) servo data, and (iii) second data that is
affected by a contribution of inter-symbol-interference due to the servo data;

removing, from the second data, the contribution of inter-symbol-interference due to the servo data;
adding, to the second data, a contribution of inter-symbol-interference due to the first data; and
subsequent to (i) removing, from the second data, the contribution of inter-symbol-interference due to the servo data and
(ii) adding, to the second data, the contribution of inter-symbol-interference due to the first data,

generating user data from the first data and the second data.

US Pat. No. 9,197,268

METHOD AND APPARATUS FOR COMPENSATING FOR IQ MISMATCH IN RADIO FREQUENCY RECEIVERS

Marvell International Ltd...

1. A receiver comprising:
a physical layer module configured to (i) receive a signal, and (ii) downconvert the signal to generate a downconverted signal,
wherein the downconverted signal has in-phase and quadrature phase (IQ) mismatch;

a function module configured to determine (i) a first value based on a first function and a resultant signal, and (ii) a second
value based on a second function;

a residual module configured to estimate a residual IQ mismatch based on the first value and the second value;
an IQ mismatch module configured to estimate an IQ mismatch coefficient based on the estimate of the residual IQ mismatch;
and

a compensation module configured to compensate for the IQ mismatch of the downconverted signal including generating the resultant
signal based on the estimate of the IQ mismatch.

US Pat. No. 9,413,356

CHIP OR SOC INCLUDING FUSIBLE LOGIC ARRAY AND FUNCTIONS TO PROTECT LOGIC AGAINST REVERSE ENGINEERING

Marvell International Ltd...

1. A security chip, comprising:
an input configured to receive, from a verification module external to the security chip, a seed value, wherein the seed value
corresponds to one of a predetermined value and a generated value;

a fusible logic array configured to generate a logic result using the received seed value, wherein
the fusible logic array includes a logic gate having structure corresponding to both i) a first type of logic gate configured
to perform a first logic operation and ii) a second type of logic gate configured to perform a second logic operation different
from the first logic operation,

the logic gate includes a fusible link located within the logic gate, wherein the logic gate is configured to operate, based
on a state of the fusible link, as both the first type of logic gate and the second type of logic gate, and wherein the fusible
link is arranged to conceal which of the first type of logic gate and the second type of logic gate the logic gate is configured
as, and

the fusible logic array is configured to generate the logic result based on the state of the fusible link; and
an output configured to provide a key value, representative of the logic result, to the verification module external to the
security chip.

US Pat. No. 9,152,580

METHOD AND APPARATUS FOR TRANSFERRING DATA BETWEEN A HOST AND AN EMBEDDED DEVICE

Marvell International Ltd...

1. A method comprising:
receiving, by interface hardware from a host, a single write command that (i) specifies two or more ports from among multiple
ports of the interface hardware, and (ii) includes two or more data packets to be respectively written to the specified ports,
wherein the multiple ports of the interface hardware are mapped to corresponding locations of memory associated with the interface
hardware;

saving, by the interface hardware in response to the single write command, the two or more data packets at two or more memory
locations to which the specified ports are mapped; and

upon completion of said saving the two or more data packets, sending a single notification to the host indicating that the
interface hardware is ready to receive another write command.

US Pat. No. 9,071,280

METHOD AND APPARATUS FOR CONTROLLING A NUMBER OF DECODING ITERATIONS OF A DECODER BASED ON A TEMPERATURE OF AN INTEGRATED CIRCUIT THAT INCLUDES THE DECODER

Marvell International LTD...

6. A method comprising:
using a first decoder, (i) performing a first number of decoding iterations, (ii) generating first data in response to performing
each of the first number of decoding iterations, and (iii) generating second data after performing all of the first number
of decoding iterations;

using a second decoder, (i) performing a second number of decoding iterations, wherein each of the second number of decoding
iterations is performed in response to the first decoder performing all of the first number of decoding iterations, and (ii)
generating soft information based on one or more of input samples and the second data generated by the first decoder,

using the first decoder, generating the first data based on one or more of (i) the soft information generated by the second
decoder and (ii) the first data from one of the first number of decoding iterations;

sensing a temperature of an integrated circuit comprising the first decoder and the second decoder; and
controlling, based on the temperature of the integrated circuit, at least one of the first number of decoding iterations or
the second number of decoding iterations.

US Pat. No. 9,237,538

SIMULTANEOUS UPLINK TRANSMISSION IN A WIRELESS NETWORK

Marvell International Ltd...

1. A method for use in a communication system in which a plurality of communication devices simultaneously transmit respective
signals to an access point via respective sets of spatial streams, the method comprising:
obtaining a selection of a first sub-matrix of a channel estimation matrix that corresponds to an estimate of a MIMO communication
channel between the plurality of communication devices and the access point; and

generating, at a first communication device in the plurality of transmitters, a physical layer (PHY) preamble of a first data
unit, including generating a plurality of training fields;

generating, at the first communication device, a data portion of the first data unit; and
causing the first communication device to transmit the first data unit to the access point simultaneously with a second communication
device in the plurality of communication devices transmitting a second data unit to the access point,

wherein causing the first communication device to transmit the first data unit includes applying, at the first communication
device, a spatial mapping to the plurality of training fields in accordance with the selected first sub-matrix.

US Pat. No. 9,184,877

METHOD AND APPARATUS FOR DECODING INDEPENDENTLY ENCODED SIGNALS

Marvell International Ltd...

1. A method comprising:
demodulating a received signal that includes a plurality of user signals, the plurality of user signals including at least
(i) a first user signal corresponding to a first user and (ii) a second user signal corresponding to a second user, wherein

the first user signal corresponds to first user data that has been modulated independently of second user data corresponding
to the second user signal,

the first user signal is encoded with a first finite state machine encoder independently of the second user signal,
the first finite state machine encoder has a number of states S1,

the second user signal is encoded with a second finite state machine encoder independently of the first user signal,
the second finite state machine encoder has a number of states S2, and

demodulating the received signal includes calculating distances between (i) each transmit symbol in the received signal and
(ii) expected symbol values, wherein each expected symbol value corresponds to user data for multiple users including the
first user and the second user; and

decoding the first user data in the received signal based on the calculated distances utilizing a finite state machine decoder
having S1 states, wherein decoding the first user data comprises:

calculating a branch metric corresponding to the first user data based on a minimum distance in a plurality of distances for
all possible values corresponding to the second user data, and

decoding the first user data based on the calculated branch metric.

US Pat. No. 9,319,179

METHODS AND APPARATUS FOR GENERATING EXPANDED CODE WORDS TO SUPPORT ERROR CORRECTION IN A DATA COMMUNICATION SYSTEM

Marvell International Ltd...

1. A method comprising:
obtaining data to be encoded;
obtaining a first subset of bits of a first bit sequence;
combining the data and the first subset of bits to generate a first combined bit sequence;
applying an error correcting code to the first combined bit sequence to generate a first code word having a greater length
than the first combined bit sequence;

combining the first code word with the first bit sequence, resulting in first encoded data that includes both first parity
information and the data; and

repeating the applying and each combining with a second subset of bits of a second bit sequence, the second subset of bits
being derived from the first subset of bits by:

including a lowest single bit that has a same value as a lowest single bit of the first subset of bits; and
including remaining higher bits each having a value that is the complement of a corresponding bit in the first subset of bits,
the repeating resulting in second encoded data that includes both second parity information and the data, the second encoded
data being based on a second code word generated by the error correcting code rather than the first code word.

US Pat. No. 9,313,804

METHOD AND SYSTEM FOR PROVIDING ARBITRATION OF COMMUNICATIONS FOR COLLOCATED WIRELESS TRANSCEIVER MODULES OPERATING BASED ON DIFFERENT WIRELESS COMMUNICATION STANDARDS

Marvell International Ltd...

1. A coexistence system comprising:
a first transceiver module, in a first network device, configured to (i) generate an indication signal, and (ii) subsequent
to a first predetermined period after having transmitted the indication signal, transmit first signals from the first network
device to a second network device during a first portion of an uplink period, wherein the first portion of the uplink period
is allocated to the first transceiver module, wherein the indication signal indicates the first signals are to be transmitted
subsequent to the first predetermined period, wherein the uplink period includes the first predetermined period, and wherein
the first signals conform to a first wireless communication standard; and

a second transceiver module, in the first network device, configured to (i) receive the indication signal from the first transceiver
module, and (ii) based on the indication signal, transmit second signals to or receive the second signals from the second
network device, wherein the second signals conform to a second wireless communication standard,

wherein the second transceiver module is configured to, based on the indication signal, (i) while operating in a non-access
point mode, transmit the second signals during a second portion of the uplink period unused by the first transceiver module,
and (ii) while operating in an access point mode and during the first portion of the uplink period, transmit a signal indicating
a second predetermined period, and wherein the second predetermined period indicates to the second network device not to transmit
the second signals to the second transceiver module during the second predetermined period.

US Pat. No. 9,301,269

ACQUISITION OF PERIODIC SYNCHRONIZATION SIGNALS

Marvell International Ltd...

1. An apparatus comprising:
first circuitry configured to receive wireless communication transmissions having periodic synchronization signals transmitted
at a fixed interval, wherein the periodic synchronization signals convey information used to establish communication between
wireless communication devices; and

second circuitry configured to acquire the periodic synchronization signals of the wireless communication transmissions by,
at least in part, (i) calculating objective function values for hypotheses of parameter sets for a current period, (ii) combining
only a portion of the calculated objective function values with stored objective function value data from a prior period to
form combined data used in sync signal acquisition, and (iii) storing the combined data or the portion of the calculated objective
function values for use in a next period, wherein the acquisition of the sync signal is used to establish a wireless communication
with a source of the wireless communication transmissions;

wherein the combining comprises combining a value of an objective function with a same hypothesis of parameter set across
more than two periods of the fixed interval.

US Pat. No. 9,257,140

SYSTEMS AND METHODS FOR DETERMINING A POSITION ERROR OF A READ/WRITE HEAD

Marvell International Ltd...

1. A system for tracking a position error of a read/write head, the system comprising:
the read/write head, the read/write head configured to (i) read data recorded to a data layer of a recording medium and (ii)
read servo information recorded to a dedicated servo layer of the recording medium, wherein the dedicated servo layer includes
a first track and a second track adjacent to the first track,

wherein, to read the servo information, the read/write head is configured to
read, from the first track, a first servo signal including first servo bursts of a predetermined frequency,
read, from the second track adjacent to the first track, a second servo signal including second servo bursts of the same predetermined
frequency as the first servo bursts, and

read the servo information from the first servo bursts and/or the second servo bursts, and
wherein the first servo bursts are orthogonal to the second servo bursts; and
a position error signal determination circuit configured to determine the position error using (i) at least one of the first
servo signal and the second servo signal and (ii) a waveform having the same predetermined frequency as the first servo bursts
and the second servo bursts.

US Pat. No. 9,252,860

METHOD AND APPARATUS FOR GENERATING A SECOND STEERING MATRIX BASED ON A FIRST STEERING MATRIX FOR USE IN TRANSMIT BEAMFORMING OF A PACKET IN WIRELESS COMMUNICATION

MARVELL INTERNATIONAL LTD...

1. An apparatus configured to perform transmit beamforming of a packet, the apparatus comprising:
a steering matrix cache unit configured to store a first steering matrix, wherein the first steering matrix includes weights
set forth in n rows and c columns that are applicable to be applied to the packet in transmit beamforming of the packet; and

a read control unit configured to
i) generate a second steering matrix having weights set forth in less than n rows or less than c columns, wherein the weights
of the second steering matrix are selected from the weights of the first steering matrix, and

ii) provide the weights from the second steering matrix for the transmit beamforming of the packet.

US Pat. No. 9,055,008

DEVICE AND PROCESS FOR EFFICIENT MULTICASTING

MARVELL INTERNATIONAL LTD...

1. A network switch configured to manage a multicast transmission of a frame, the network switch comprising:
a header memory;
a plurality of output queues, each of the plurality of output queues being associated with a respective egress module of the
network switch;

a controller configured to
i) store an egress variable in the header memory, wherein the egress variable is associated with the frame, and wherein the
egress variable is indicative of a number of transmissions in the multicast transmission, and

ii) provide the egress variable from the header memory to the plurality of output queues; and
at least one egress module configured to i) transmit the frame from the network switch, and ii) after transmission of the
frame, return the egress variable to the controller,

wherein the controller is configured to update the egress variable stored in the header memory in response to the egress variable
being returned from the at least one egress module.

US Pat. No. 9,294,418

64B/66B CODEC FOR ETHERNET APPLICATIONS

Marvell International Ltd...

1. A codec comprising:
an input circuit configured to input data; and
an encoding circuit configured to
transform the input data into a codeword, the codeword including a header portion and a payload portion, the payload portion
including (i) a type control block, and (ii) a plurality of blocks that are either control blocks or data blocks;

determine a location of a first control block of the plurality of blocks in the payload portion of the codeword based on the
type control block;

modify the first control block to create a reference block by modifying a most significant bit of the first control block,
the reference block being eight bits long and including the most significant bit of the first control block and seven unused
bits; and

insert data blocks in the payload portion of the codeword subsequent to the location now corresponding to the reference block.

US Pat. No. 9,240,372

SEMICONDUCTOR DIE HAVING LEAD WIRES FORMED OVER A CIRCUIT IN A SHIELDED AREA

Marvell International Ltd...

1. A semiconductor die, comprising:
a first bond pad, arranged in an outer periphery of the semiconductor die, connected to a first lead wire, wherein the first
bond pad is configured to receive a reference voltage potential through the first lead wire;

a second bond pad arranged in the outer periphery of the semiconductor die, between the first bond pad and an interior portion
of the semiconductor die, such that the second bond pad is arranged inside of the first bond pad relative to the interior
portion of the semiconductor die;

a first via arranged in a substrate of the semiconductor die, wherein the first via is connected between the first bond pad
and an interconnecting layer in the substrate of the semiconductor die;

a second via arranged in the substrate of the semiconductor die, wherein the second via is connected between the interconnecting
layer and the second bond pad such that the reference voltage potential, received by the first bond pad, is provided to the
second bond pad through the first bond pad, the first via, and the second via;

a third bond pad arranged in the interior portion of the semiconductor die, wherein the third bond pad is connected to the
second bond pad using a second lead wire, and wherein the third bond pad is configured to receive the reference voltage potential
through the second lead wire; and

a circuit arranged in the interior portion of the semiconductor die, the circuit arranged between the second bond pad and
the third bond pad below the second lead wire.

US Pat. No. 9,413,567

SYSTEMS AND METHODS FOR FINITE IMPULSE RESPONSE ADAPTATION FOR GAIN AND PHASE CONTROL

MARVELL INTERNATIONAL LTD...

1. A method for signal processing by an equalizer having a plurality of stage weights, the method comprising:
receiving an input signal;
generating a plurality of sample values from the input signal;
respectively multiplying each sample value of the plurality of sample values by a corresponding stage weight of the plurality
of stage weights of the equalizer to yield a plurality of weighted sample values;

summing the plurality of weighted sample values to yield an output signal;
designating, based on a phase change of the equalizer, which of the plurality of stage weights are to be adjustable;
determining an adjustment value for each of one or more of the adjustable stage weights;
adjusting the one or more adjustable stage weights by the respective adjustment values; and
repeating the multiplying and the summing to yield an adjusted output signal.

US Pat. No. 9,301,325

METHOD AND APPARATUS FOR LOW-ENERGY SENSOR WIRELESS CONNECTION

Marvell International Ltd...

1. A method comprising:
activating a wireless interface of a sensor to initiate a data transfer session with an access point of a wireless network,
the sensor being configured to generate data corresponding to an environmental condition surrounding the sensor and store
the data in a sensor data storage that is associated with the sensor;

associating the wireless interface with the access point to establish a wireless network connection with the access point,
the associating being performed:

(i) in accordance with a setting that is usable to establish the wireless network connection with the access point, and
(ii) without having to scan for the access point to discover the setting that is usable to establish the wireless network
connection;

after associating the wireless interface with the access point, configuring the wireless interface to communicate with the
access point over the wireless network connection using an Internet Protocol (IP) address, the configuring being performed
without obtaining the IP address during the data transfer session;

transmitting, via the wireless interface, the data to the access point, the data being transmitted over the wireless network
connection during the data transfer session;

responsive to transmitting the data to the access point, clearing the data from the sensor data storage; and
subsequent to the data being transmitted from the sensor to the access point, deactivating the wireless interface of the sensor
to terminate the data transfer session with the access point.

US Pat. No. 9,313,595

LOW-POWER METHOD AND CIRCUITRY OF DETERMINING HEADPHONE TYPE AND MONITORING ACTIVITY

MARVELL INTERNATIONAL LTD...

1. A method comprising:
detecting insertion of a headphone into an electronic device;
comparing a voltage level detected by a detection circuit of the electronic device to a set of threshold voltages to determine
a type of the headphone, the detection circuit including a first path, a second path, and a bias capacitor;

determining the headphone to be a first type when a first signal travels through the first path to ground; and
determining the headphone to be a second type when a second signal travels through the second path to charge the bias capacitor
to a reference voltage.

US Pat. No. 9,246,379

CURRENT LIMIT PROTECTION CIRCUITS FOR USE WITH DEPLETION MOSFET

Marvell International Ltd...

1. A circuit comprising:
a depletion mode transistor coupled to a power supply; and
a voltage clamp component including a diode and a switch coupled in series, between gate terminal of the depletion mode transistor
and a capacitor, the voltage clamp component configured to restrict a gate voltage of the depletion mode transistor to the
capacitor voltage during a first stage, and restrict the gate voltage to a source terminal voltage of the depletion mode transistor
during a second stage when the capacitor is charged to a predetermined voltage level.

US Pat. No. 9,082,187

METHOD AND APPARATUS FOR CORRECTING DISTORTION IN AN IMAGE DUE TO ROTATIONAL MOTION OF AN IMAGE CAPTURE DEVICE OCCURRING WHILE THE IMAGE IS BEING CAPTURED

Marvell International Ltd...

1. A system operable to correct for distortion in a captured image, the distortion in the captured image having been caused
by a rotation of an image capture device while the captured image was being captured by the image capture device, the system
comprising:
an orientation sensor configured to perform a first measurement concurrently with the image capture device capturing the captured
image;

a rotation module configured to generate, based on the first measurement, a rotation matrix;
a correction module configured to, based on the rotation matrix, produce a corrected image by correcting the distortion in
the captured image caused by the rotation of the image capture device; and

a restoration module configured to selectively revert the corrected image back to the captured image based on (i) the corrected
image and (ii) the rotation matrix.

US Pat. No. 9,299,589

BALL GRID ARRAY PACKAGE WITH LASER VIAS AND METHODS FOR MAKING THE SAME

Marvell International Ltd...

1. A method comprising:
forming a substrate, wherein forming the substrate comprises
forming a plurality of layers in the substrate, wherein the plurality of layers in the substrate include at least (i) an uppermost
layer, (ii) a lowermost layer, and (iii) a middle layer disposed between the uppermost layer and the lowermost layer; and

forming, in the substrate, (i) a plurality of mechanical vias and (ii) a plurality of laser vias, wherein each of the plurality
of mechanical vias is formed through all of the plurality of layers of the substrate,

wherein each of the plurality of laser vias is formed in the uppermost layer of the substrate, and
wherein none of the plurality of laser vias is formed in the lowermost layer of the substrate.

US Pat. No. 9,292,429

OFF-CHIP NON-VOLATILE MEMORY ACCESS

Marvell International Ltd...

1. An apparatus, comprising:
a memory interface configured to communicate with a non-volatile memory device, the non-volatile memory device being configured
to store device information at a predetermined address;

circuitry configured to iteratively issue read command sequences to the non-volatile memory device based on the predetermined
address, using different probing parameters, until a read result includes an indication of a successful read of the non-volatile
memory device, wherein the different probing parameters include different address lengths, different command lengths, or both;
and

circuitry configured to access the non-volatile memory device, using one or more access parameters included in the read result,
to retrieve additional data.

US Pat. No. 9,445,475

BUCK REGULATOR FOR LED LIGHTING COLOR MIXING AND/OR CURRENT COMPENSATION

Marvell International Ltd...

1. A light emitting diode (LED) lighting system, comprising:
a first string of first LEDs emitting light having a first color;
a second string of second LEDs emitting light having a second color and connected in series with the first string of first
LEDs;

a first switch;
a second switch connected in series with the first switch;
an inductor having (i) a first end connected between the first switch and the second switch, and (ii) a second end connected
between the first string of first LEDs and the second string of second LEDs;

a first current sense resistor connected in series with the first string of first LEDs and the second string of second LEDs,
wherein a first current flows through the second string of second LEDs and the first current sense resistor;

a second current sense resistor connected in series with the first switch and the second switch, wherein a second current
flows through the first string of first LEDs and the second current sense resistor; and

a regulator module configured to
communicate with the first current sense resistor to receive a first current sensing input indicating the first current,
communicate with the second current sense resistor to receive a second current sensing input indicating the second current,
and

modulate the first switch and the second switch, based on the received first current sensing input, the received second current
sensing input, and a desired current ratio, to provide the desired current ratio,

wherein the desired current ratio corresponds to a ratio of (i) the second current flowing through the first string of first
LEDs and the second current sense resistor to (ii) the first current flowing through the second string of second LEDs and
the first current sense resistor.

US Pat. No. 9,294,997

WAKEUP BEACONS FOR MESH NETWORKS

Marvell International Ltd...

1. A method comprising:
receiving, from a transmitting node of a mesh network, data intended for a receiving node currently in a power-saving mode;
transmitting, according to a previously determined wakeup beacon schedule and by an intermediate node between the transmitting
node and the receiving node, a wakeup beacon to the receiving node, the wakeup beacon indicating that the data is available
but not including the data within the wakeup beacon, the previously determined wakeup beacon schedule including a start time
and a broadcast interval, the start time offset from non-wakeup beacon broadcasts and the broadcast interval about equal to
or about a multiple of a non-wakeup beacon broadcast interval;

receiving, at the intermediate node and from the receiving node, an indication that the receiving node is able to receive
the data; and

responsive to the indication from the receiving node, transmitting the data to the receiving node.

US Pat. No. 9,226,337

METHOD AND APPARATUS FOR ESTABLISHING CONNECTIONS IN A WIRELESS NETWORK

MARVELL INTERNATIONAL LTD...

1. A first communication device, comprising:
a wireless communication module including:
a transceiver, and
a hardware controller device configured to:
cause the wireless communication module to listen, during advertise states of the first communication device, for probe requests
from one or more second communication devices on a first channel included in a plurality of channels, wherein at least some
respective lengths of the advertise states of the first communication device are different,

when one or more first probe requests are received from the one or more second communication devices during the advertise
states of the first communication device, cause the wireless communication module to transmit one or more first probe responses
responsive to the one or more first probe requests,

cause the wireless communication module to select second channels from the plurality of channels for use during search states
of the first communication device, wherein the second channels are different than the first channel,

cause the wireless communication module to transmit second probe requests on respective second channels during respective
search states of the first communication device, and

cause the wireless communication module to listen, during the search states, for one or more second probe responses on the
second channels.

US Pat. No. 9,281,879

METHODS AND APPARATUS FOR INDIVIDUALLY CONTROLLING POWER SPECTRAL DENSITY (PSD) PER TRANSMISSION PORT IN MIMO TRANSMITTERS

Marvell International Ltd...

1. A method for controlling power spectral density (PSD) of a multiple-input-multiple-output (MIMO) transmitter, the method
comprising:
receiving, with precoding circuitry, data corresponding to multiple transmission streams of the MIMO transmitter from a constellation
encoder;

precoding the received data using the precoding circuitry;
sending the precoded data to a plurality of transmission ports of the MIMO transmitter; and
adjusting power allocation of the precoded data for a first of the plurality of transmission ports independently of adjusting
power allocation of the precoded data for a second of the plurality of transmission ports, wherein adjusting power allocation
of the precoded data comprises modifying a first power value stored in a memory associated with first power allocation circuitry
of the first transmission port and modifying a second power value stored in a memory associated with second power allocation
circuitry of the second transmission port.

US Pat. No. 9,128,859

METHOD AND APPARATUS FOR DYNAMICALLY SELECTING AN ERROR CORRECTION CODE FOR ENCODING AND DECODING DATA IN A COMMUNICATION SYSTEM

Marvell International Ltd...

1. A device comprising:
a memory;
an encoder coupled to the memory, wherein the encoder is configured to (i) receive data, (ii) encode the data with a first
error correction code of a plurality of error correction codes to generate encoded data, and (iii) write the encoded data
to the memory; and

a decoder coupled to the memory, wherein the decoder is configured to
(i) read the encoded data from the memory,
(ii) select a second error correction code of the plurality of error correction codes,
(iii) attempt to decode the encoded data using the second error correction code of the plurality of error correction codes,
and

(iv) in response to a failure in the attempt to decode the encoded data using the second error correction code of the plurality
of error correction codes,

select a third error correction code of the plurality of error correction codes, and
attempt to decode the encoded data using the third error correction code of the plurality of error correction codes.

US Pat. No. 9,117,472

PROCESSING A DATA SIGNAL WITH AN EMBEDDED POSITION SIGNAL

MARVELL INTERNATIONAL LTD...

1. An apparatus associated with processing a read signal produced from reading a storage medium of a storage device, the apparatus
comprising:
read logic configured to control the storage device to generate the read signal by reading a first layer and a second layer
of the storage medium, wherein the first layer defines data and the second layer defines embedded position information,

wherein the read signal is comprised of two separate signals that include a first signal that is the data and a second signal
that is the embedded position information;

data detection logic configured to process the read signal to recover the data by removing an estimate of the second signal
from the read signal to produce a reproduction of the first signal; and

servo logic configured to reproduce the embedded position information by subtracting the reproduction of the first signal
that includes the data from the read signal, wherein the read logic is configured to control the storage device based, at
least in part, on the embedded position information.

US Pat. No. 9,420,594

METHODS AND APPARATUS FOR ADAPTIVELY SELECTING A COMMUNICATIONS MODE IN HIGH FREQUENCY SYSTEMS

Marvell International Ltd...

11. A first wireless device comprising:
receiver circuitry configured to receive a signal through a high-frequency communications channel in the range of 57-67 GHz;
and

control circuitry configured to:
determine a length of the high-frequency communications channel based on the received signal;
transmit a data signal comprising an indication of the determined length of the high-frequency communications channel from
the first wireless device to a second wireless device, wherein the data signal is transmitted using a low band communications
channel in the range of 2-5 GHz;

cause the length of the high-frequency communications channel to be compared to a threshold; and
cause a selection between a first modulation scheme and a second modulation scheme, wherein the first modulation scheme is
selected when the length of the high-frequency communications channel exceeds the threshold, and the second modulation scheme
is selected when the length of the high-frequency communications channel is below the threshold.

US Pat. No. 9,407,481

SYSTEMS AND METHODS FOR PERFORMING EFFICIENT BLIND DECODING AT A WIRELESS RECEIVER

Marvell International Ltd...

1. A method for decoding data, the method comprising:
receiving a first decision metric associated with first timing information;
accumulating, using a storage device, a first portion of the first decision metric with an adjacent second portion of the
first decision metric;

receiving a second decision metric different from the first decision metric associated with second timing information different
from the first timing information;

accumulating, using the storage device, portions of the second decision metric with the accumulated portions of the first
decision metric;

determining how many antennas correspond to data based on information stored in the storage device;
determining an instance of timing information corresponding to a start of a period of periodic decoding information, wherein
the timing information includes the first timing information or the second timing information; and

configuring receiver decoding circuitry based on how many of the antennas are determined to correspond to the data and the
instance of the timing information.

US Pat. No. 9,275,985

RC NETWORKS THAT INCLUDE AN INTEGRATED HIGH RESISTANCE RESISTOR ON TOP OF A MOS CAPACITOR

Marvell International Ltd...

1. An apparatus including a resistor capacitor network, the resistor capacitor network comprising:
a metal oxide semiconductor capacitor portion that includes a high-k gate oxide layer, wherein the value of k is in a range
of 4.0 to 100.0, wherein the high-k gate oxide layer comprises (i) a high-k dielectric layer and (ii) a conductive layer deposited
over the high-k dielectric layer, wherein the high-k gate oxide layer does not include any metal layer;

a high resistance polysilicon gate layer formed over the high-k gate oxide layer, wherein the resistance of the high resistance
polysilicon gate layer is in a range of 100 to 2000 ohms per square, wherein no metal layer is deposited directly over any
portion of an entirety of the high-k gate oxide layer;

a metal gate embedded within the high resistance polysilicon gate layer, wherein the metal gate is configured to reduce a
parasitic resistance of the high resistance polysilicon gate layer, and wherein the metal gate is embedded within the high
resistance polysilicon gate layer subsequent to the formation of the high resistance polysilicon gate layer over the high-k
gate oxide layer;

a substrate;
a well formed on the substrate, wherein the well comprises (i) a first corner region, (ii) a second corner region, and (iii)
a middle region interposed between the first corner region and the second corner region, wherein the high-k gate oxide layer
is formed over the middle region of the well;

a first resistor contact and a second resistor contact formed over the high resistance polysilicon gate layer, wherein the
first resistor contact and the second resistor contact are used as contacts for a resistor of the resistor capacitor network;

a first capacitor contact formed on the first corner region; and
a second capacitor contact formed on the second corner region,
wherein the first capacitor contact, the second capacitor contact, the high-k gate oxide layer and the high resistance polysilicon
gate layer, in combination, form a capacitor of the resistor capacitor network, and

wherein the first capacitor contact and the second capacitor contact are used as contacts for the capacitor of the resistor
capacitor network.

US Pat. No. 9,179,365

METHOD AND APPARATUS FOR PROVIDING QUALITY OF SERVICE (QOS) IN A WIRELESS LOCAL AREA NETWORK

Marvell International Ltd...

1. A method comprising:
switching a mode of operation of a wireless communication device from a first transmission mode to a second transmission mode,
wherein the first transmission mode supports transmission of packets comprising header information in accordance with a communication
protocol specified for the first transmission mode, and the second transmission mode supports transmission of real-time data
packets that do not contain the header information in accordance with the communication protocol specified for the first transmission
mode; and

configuring a transmission parameter associated with the wireless communication device for operation in the second transmission
mode for transmission of a real-time data packet containing real-time data.

US Pat. No. 9,356,638

SYSTEM AND METHOD FOR DETECTING RATED MAXIMUM SYSTEM DEVIATION (RMSD) AND ALLEVIATING RMSD MISMATCH

Marvell International Ltd...

1. A method of receiving, comprising:
demodulating a modulated carrier bearing a first signal by a demodulator configured to operate with a plurality of modes including
at least a first mode to demodulate the modulated carrier bearing the first signal based on a first rated maximum system deviation
(RMSD) value of the modulated carrier, the demodulator operating in the first mode;

extracting a second signal from the first signal;
generating a third signal based on the second signal, the third signal being fed back to the demodulator; and
instructing, by the third signal, the demodulator to change from the first mode to a second mode to demodulate the modulated
carrier bearing the first signal based on a second rated maximum system deviation (RMSD) value of the modulated carrier.

US Pat. No. 9,304,777

METHOD AND APPARATUS FOR DETERMINING RELATIVE AGES OF ENTRIES IN A QUEUE

Marvell International Ltd...

1. A system comprising:
a queue configured to store a plurality of instructions, wherein the queue comprises a plurality of entries, wherein the queue
is associated with (i) a first pointer pointing to an entry of the queue that is configured to store an instruction that newly
arrives in the queue and (ii) a second pointer pointing to an entry of the queue that stores an oldest instruction of the
plurality of instructions, wherein each entry of the plurality of entries of the queue is associated with a corresponding
identification, wherein each identification associated with each entry of the queue comprises (i) a wrap bit that is based
on a position of the corresponding entry relative to one of the first pointer or the second pointer and (ii) position bits
that are based on an absolute position of the corresponding entry relative to the plurality of entries of the queue; and

a processing unit configured to
receive, from the queue, (i) a first identification associated with a first entry of the plurality of entries, wherein the
first entry stores a first instruction of the plurality of instructions, and (ii) a second identification associated with
a second entry of the plurality of entries, wherein the second entry stores a second instruction of the plurality of instructions,

compare (i) a first wrap bit of the first identification and (ii) a second wrap bit of the second identification, and
based on comparing (i) the first wrap bit of the first identification and (ii) the second wrap bit of the second identification,
determine a relative age of the first instruction with respect to the second instruction.

US Pat. No. 9,271,192

RATE ADAPTATION IN WIRELESS SYSTEMS

Marvell International Ltd...

1. A method comprising:
generating confidence measures for a plurality of data rates based at least on a first indicator and a second indicator, wherein
the first indicator corresponds to packet loss associated with packets transmitted by a device at a first data rate that is
selected from the plurality of data rates, and the second indicator corresponds to one or more signal quality values associated
with one or more packets received by the device;

selecting, from the plurality of data rates, a second data rate based at least on the confidence measures; and
transmitting one or more additional packets at the second data rate.

US Pat. No. 9,750,071

METHOD AND APPARATUS FOR SUPPORTING MULTIPLE CONNECTIONS OVER DIFFERENT TYPES OF ACCESS IN 3GPP SYSTEMS

Marvell International Ltd...

5. The method of claim 3, wherein the second plurality of packets is included in the second flow of packets.

US Pat. No. 9,872,305

OPTIMIZED CELLULAR TRANSMISSION SCHEDULING BASED ON KNOWLEDGE OF INTERFERENCE-MITIGATION CAPABILITIES IN NEIGHBOR CELL

MARVELL INTERNATIONAL LTD...

1. A method, comprising:
in a cellular communication network that comprises a first base station serving a first user terminal and a second base station
serving a second user terminal, identifying, by the first base station, (i) capabilities of the first base station to reduce
interference caused to the first base station by transmissions of the second base station or by transmissions of the second
user terminal, or (ii) capabilities of the first user terminal to reduce interference caused to the first user terminal by
transmissions of the second base station or by transmissions of the second user terminal;

reporting by the first base station, to the second base station, the capabilities to reduce the interference, which were identified
by the first base station; and

based on the reported capabilities of the first base station or of the first user terminal to reduce the interference, scheduling,
by the second base station, transmissions (i) of the second user terminal to the second base station, or (ii) of the second
base station to the second user terminal,

wherein scheduling the transmissions comprises checking, based on the reported capabilities, whether or not the first user
terminal is capable of directional transmission,

wherein if the checking indicates that the first user terminal is capable of directional transmission, reusing time/frequency
resources that are allocated for uplink transmissions of the first user terminal in uplink transmissions of the second user
terminal, and if the checking indicates that the first user terminal is incapable of directional transmission, refraining
from reusing time/frequency resources that are allocated for uplink transmissions of the first user terminal in uplink transmissions
of the second user terminal.

US Pat. No. 9,444,964

METHOD AND APPARATUS FOR SCANNING AN IMAGE

Marvell International Ltd...

1. A method of reducing electromagnetic emissions generated by an image scanner, the method comprising:
cyclically dithering a frequency of a clock signal of the image scanner by repeated dither cycles;
upon commencement of a scan pass of a page of a document, determining a phase of the dither cycle;
suspending the dithering for a duration that is based, at least in part, on the determined phase of the dither cycle; and
after the duration, commencing all subsequent scan passes of the page of the document at a particular phase shift from the
determined phase of the dither cycle.

US Pat. No. 9,118,347

METHOD AND APPARATUS FOR OFDM ENCODING AND DECODING

MARVELL INTERNATIONAL LTD...

1. A system for communicating on a cellular telecommunications network, comprising:
a receiver interface configured to receive an orthogonal frequency division multiplexing (OFDM) signal that includes transmitted
data encoded according to a particular coding matrix;

a processor configured to receive the OFDM signal, the processor being configured to decode data from the OFDM signal by solving
a matrix optimization, wherein the processor is configured to decode the data from the OFDM signal without error when an error
rate, ?, of the OFDM signal is less than

where n/m is a code-rate of the OFDM signal; and
a computer-readable memory configured to store the decoded data.

US Pat. No. 9,262,374

MODEL DRIVEN COMMAND LANGUAGE TRANSLATOR

MARVELL INTERNATIONAL LTD...

1. A method, comprising:
identifying, by at least a processor, functional configurations into which a computer networking device can be placed based,
at least in part, on a first model;

identifying, by at least the processor, a subset of functional configurations in the first model into which the computer networking
device can be placed by executing commands from a source command language;

accessing, by at least the processor, a target command language model, wherein the target command language model identifies
the subset of functional configurations into which the computer networking device can be placed by executing commands from
a target command language;

translating, by at least the processor, a plurality of source commands written in the source command language to produce an
output comprising a plurality of target commands written in the target command language, wherein the target commands are identified
and selected as being commands that produce a target functional configuration when executed on the computer networking device
and wherein the target functional configuration is equivalent to a source functional configuration that would be produced
by executing the plurality of source commands on the computer networking device; and

executing, by at least the processor, the plurality of target commands to produce the target functional configuration in the
computer networking device.

US Pat. No. 9,137,030

MULTICAST QUEUEING IN A NETWORK SWITCH

Marvell International Ltd...

1. A method for transmitting multicast network traffic, the method comprising:
storing, in a first queue of a network switching device, first data units corresponding to a first multicast group (MCG) and
second data units corresponding to a second MCG;

aggregating into a first frame (i) at least one first data unit retrieved from the first queue and (ii) at least one second
data unit retrieved from the first queue;

transmitting, with the network switching device, the first frame to a superset MCG that includes at least the first MCG and
the second MCG;

storing, in a second queue of the network switching device, only third data units corresponding to a third MCG; and
transmitting, with the network switching device, third data units retrieved from the second queue to the third MCG.

US Pat. No. 9,285,421

SERIALIZER/DESERIALIZER AND METHOD FOR TRANSFERRING DATA BETWEEN AN INTEGRATED CIRCUIT AND A TEST INTERFACE

Marvell International Ltd...

1. A serializer/deserializer for communicating with an integrated circuit, the serializer/deserializer comprising:
a serializer configured to serialize, from a parallel format into a serial format, first data to be transferred from an external
device to the integrated circuit, wherein the external device is configured to perform testing on the integrated circuit;

a deserializer configured to deserialize, from the serial format into the parallel format, second data to be transferred from
the integrated circuit to the external device; and

a test access port module configured to (i) receive, from a test interface arranged between the serializer/deserializer and
the external device, third data for controlling the serializer and (ii) provide, to the test interface, fourth data associated
with control of the test interface, wherein the fourth data corresponds to the deserialization of the second data to be transferred
from the integrated circuit to the external device.

US Pat. No. 9,355,066

ACCELERATED CALCULATION OF ARRAY STATISTICS

Marvell International Ltd...

1. A device comprising:
a memory configured to store a data array and a counter array, wherein the data array comprises multiple values, and each
of the multiple values is encoded in a respective row of the data array; and

processor electronics configured to add up a number bits found in respective columns of the data array by treating respective
columns of the counter array as respective ones of multiple counters with rows of the counter array corresponding to bit significance
positions spanning the multiple counters; and

wherein the processor electronics operate using n bit wide words, and the processor electronics are configured to (i) retrieve
data from the data array in the n bit wide words, (ii) AND the retrieved n bit wide words with an n bit wide portion of the
multiple counters to generate an n bit wide carry array, (iii) XOR the retrieved n bit wide words with the n bit wide portion
of the multiple counters, and while the n bit wide carry array remains non-zero, (iv) AND the n bit wide carry array with
a next n bit wide portion of the multiple counters to update the n bit wide carry array, and XOR the n bit wide carry array
with the next n bit wide portion of the multiple counters.

US Pat. No. 9,268,941

METHOD FOR SECURE SOFTWARE RESUME FROM LOW POWER MODE

MARVELL INTERNATIONAL LTD...

1. A device, comprising:
secure flash memory configured to store data in a manner that can be validated by a secure boot ROM;
a low power data retentive memory (LPDRM);
a package logic configured to:
receive a request from the device to enter a low power mode;
in response to the request to enter a low power mode,
store a data section in the secure flash memory;
store the data section in the LDPRM;
construct a resume package that includes i) a location of the data section in the LPDRM, ii) an image identifier for an image
to be loaded from the secure flash memory in a subsequent boot operation, and iii) a validation value that results from performing
a validation function on the data section, and

store the resume package in the LPDRM for use by the device in resuming operation when exiting the low power mode; and
a resume logic configured to, in response to a boot signal from a device operating in low power mode:
access the resume package;
perform the validation function on the data section stored in the LPDRM to compute a resulting validation value;
compare the resulting validation value with the validation value in the resume package;
when the resulting validation value agrees with the validation value in the resume package:
identify, in the resume package, the image identifier;
load the image from the secure flash memory; and
transfer control of the device to an operating system without loading an image corresponding to the data section to the LPDRM.

US Pat. No. 9,264,005

TECHNIQUES FOR EFFICIENT RADIO FREQUENCY POWER AMPLIFIER

MARVELL INTERNATIONAL LTD...

7. A transmitter comprising:
a first circuit configured to adjust an amplitude of a power amplifier output voltage based on an amplitude feedback signal;
a second circuit configured to adjust a phase of the power amplifier output voltage based on a phase feedback signal; and
a third circuit configured to control a power efficiency of the power amplifier output voltage,
wherein the third circuit includes a switch control circuit configured to compare an amplitude of an envelope of a modulated
signal with a threshold voltage that has a constant level and switch a bias voltage of a power amplifier from a higher voltage
to a lower voltage when the envelope of the modulated signal has the amplitude below the threshold voltage,

wherein the switch control circuit is configured to switch the bias voltage of the power amplifier from the lower voltage
to the higher voltage, in response to the envelope of the modulated signal having the amplitude above the threshold voltage,
and

wherein the switch control circuit is configured to compare the amplitude of the envelope of the modulated signal with the
threshold voltage at a first time, determine to switch the bias voltage of the power amplifier based on the comparison result,
detect a local minimum amplitude of an output power of the power amplifier, and switch the bias voltage of the power amplifier
at a second time when the local minimum amplitude of the output power of the power amplifier is detected, the second time
being subsequent to the first time.

US Pat. No. 9,231,633

METHOD AND APPARATUS FOR PERFORMING RUN-TIME COMPENSATION OF I/Q MISMATCH IN A BASEBAND SIGNAL

Marvell International LTD...

1. A receiver comprising:
a downconversion module configured to (i) receive a signal and (ii) downconvert the signal to generate a downconverted signal,
wherein the downconverted signal includes in-phase/quadrature (IQ) mismatch;

a compensation module configured to generate a compensated signal based on (i) the downconverted signal and (ii) an IQ mismatch
compensation value;

a first mixer configured to mix the compensated signal with a first oscillating signal generated by a local oscillator to
shift a frequency of the compensated signal;

a second mixer configured to mix the compensated signal with a second oscillating signal to shift a frequency of the compensated
signal;

a first Fast Fourier Transform (FFT) module configured to perform a FFT on an output of the first mixer to generate a first
FFT signal;

a second FFT module configured to perform a FFT on an output of the second mixer to generate a second FFT signal; and
an IQ mismatch estimation module configured to
capture a first portion of the first FFT signal and a second portion of the first FFT signal generated in response to the
received signal including a first predetermined sequence and a second predetermined sequence, respectively,

capture a first portion of the second FFT signal and a second portion of the second FFT signal generated in response to the
received signal including the first predetermined sequence and the second predetermined sequence, respectively, and

generate the IQ mismatch compensation value based on: (i) the first portions of the first and second FFT signals; (ii) the
second portions of the first and second FFT signals; (iii) the first predetermined sequence; and (iv) the second predetermined
sequence.

US Pat. No. 9,223,327

UNIVERSAL ADAPTIVE VOLTAGE SCALING SYSTEM

Marvell International Ltd...

1. A system, comprising:
a universal adaptive voltage scaling monitor (UAVSM) configured to
receive a first signal generated by a signal path,
delay the first signal by an adjustable time period,
receive a second signal associated with the signal path,
compare the delayed first signal and the second signal,
provide an error signal indicating a result of the comparison, wherein the error signal is asserted when the result of the
comparison indicates that the delayed first signal is different from the second signal, and

maintain the asserted error signal until an asserted reset signal is received;
a transition detector configured to
detect a transition in the second signal, and
provide a transition signal, where the transition signal is asserted in response to detecting the transition in the second
signal; and

an adaptive voltage scaling (AVS) controller connected with the UAVSM and the transition detector, the AVS controller configured
to

determine a voltage being provided to the signal path,
receive the error signal from the UAVSM,
receive the transition signal from the transition detector,
determine whether the signal path is active when the received transition signal is an unasserted transition signal,
provide a first control signal indicating that the voltage is to be increased, wherein the first control signal is asserted
when

the received error signal is an asserted error signal,
the received transition signal is an asserted transition signal, and
the voltage is less than a maximum voltage, and
provide a second control signal indicating that the voltage is to be decreased, wherein the second control signal is asserted
when

the received error signal is an unasserted error signal,
the signal path is active, and
the voltage is greater than a minimum voltage.

US Pat. No. 9,202,512

DATA STORAGE DEVICE

Marvell International LTD...

1. A data storage device comprising a motor having a stator, the stator comprising:
a substrate having a first surface and a second surface opposite to the first surface; and
a n phase winding arrangement having n phase windings;
wherein each of the n phase windings comprises a first set of m flat fractional-pitch coils arranged on the first surface
of the substrate such that the m flat fractional-pitch coils are spaced apart uniformly along a closed loop and connected
in series;

wherein each one of the m flat fractional-pitch coils, together with an angular section of the substrate between the one of
the m flat fractional-pitch coils and an adjacent one of the m flat fractional-pitch coils of one of the n phase windings
defines a stator pole-pair;

wherein m is an integer larger than 1; and
wherein each of the n phase windings further comprises a second set of m flat fractional-pitch coils arranged on the second
surface of the substrate such that m flat fractional-pitch coils of the first set of m flat fractional-pitch coils are at
least substantially in alignment with the m flat fractional-pitch coils of the second set of m flat fractional-pitch coils.

US Pat. No. 9,253,122

PREAMBLE FOR COMMUNICATION IN POWER LINE CHANNEL

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
a transmitter configured to transmit a communication signal on a communication channel for communicating with a remote device,
wherein the communication channel is a power line channel that carries electric power transmissions; and

a transmission logic configured to control the transmitter to generate at least two polarity transitions in a preamble of
the communication signal,

wherein the transmission logic is configured to control the transmitter to generate the at least two polarity transitions
in the preamble by generating the preamble using a first sequence of symbols, a second sequence of symbols that occur after
the first sequence of symbols, and a final sequence of symbols that occur after the second sequence of symbols, wherein symbols
in the second sequence of symbols have a different polarity than symbols of i) the first sequence of symbols and ii) the final
sequence of symbols.

US Pat. No. 9,099,182

METHOD AND APPARATUS FOR SELECTIVE USE OF WEAR-LEVELING TO STORE DATA IN A MEMORY

Marvell International Ltd...

1. A method comprising:
receiving data to be stored to a memory;
based on a time period for which the data is expected to be stored in the memory, characterizing the data;
based on characterizing the data, determining whether to use wear-leveling to store the data in the memory;
in response to determining to use wear-leveling to store the data in the memory,
(i) using wear-leveling to select a first memory block of the memory, and
(ii) storing the data in the selected first memory block of the memory; and
in response to determining to not use wear-leveling to store the data in the memory, refraining from using wear-leveling while
storing the data in the memory.

US Pat. No. 9,094,080

METHOD AND APPARATUS FOR ESTIMATING STATISTICS IN WIRELESS SYSTEMS

Marvell International Ltd...

1. A method for processing a signal using simplified channel statistics, the method comprising:
estimating a channel profile for a multipath channel, wherein estimating the channel profile for the multipath channel comprises
detecting, within a given range, a plurality of channel taps associated with a channel impulse response, and
assigning each of the plurality of channel taps a uniform amplitude value corresponding to a constant signal strength;
receiving the signal through the multipath channel; and
filtering the received signal based on the estimated channel profile.

US Pat. No. 9,986,521

SYSTEMS AND METHODS FOR DETECTING A PRIMARY SYNCHRONIZATION SIGNAL IN A WIRELESS COMMUNICATION SYSTEM

Marvell International Ltd...

1. A method for detecting a received synchronization signal, the method comprising:receiving, at a receiver, a signal from a transmitter, wherein the signal includes the received synchronization signal;
processing, in frequency domain, the received signal and a plurality of candidate synchronization signals to obtain a frequency transform of at least one of a plurality of correlation signals, wherein the processing comprises multiplying a frequency transform of the received signal with a frequency transform of at least one of the plurality of candidate synchronization signals;
selecting, based at least in part on the at least one of the plurality of correlation signals, one of the plurality of candidate synchronization signals, wherein the selected candidate synchronization signal is correlated with the received synchronization signal; and
detecting the received synchronization signal based at least in part on (i) the received signal, and (ii) a characteristic obtained from the at least one of the plurality of correlation signals that is associated with the selected candidate synchronization signal.

US Pat. No. 9,356,534

METHOD AND APPARATUS FOR TURNING ON A LIGHTING DEVICE

Marvell International Ltd...

1. A method, comprising:
regulating a time for turning on a switch to transfer energy via a transformer in a first control mode;
determining a first turn-on time for the switch in the first control mode based on the regulated time in the first control
mode;

generating a count per a cycle, based on a clock signal in response to pulses in a pulse width modulation (PWM) signal that
controls the switch;

searching a minimum counted value among counts generated over a plurality of cycles;
determining whether a dimmer is coupled with the switch or not;
determining a second turn-on time for the switch in a second control mode based on the searched minimum counted value; and
controlling the switch based on the determined second turn-on time in the second control mode to transfer energy via the transformer,
wherein the second turn-on time in the second control mode is a constant turn-on time for the switch only when the dimmer
is decoupled with the switch.

US Pat. No. 9,355,001

METHOD AND APPARATUS FOR SELECTING AN OPERATING FREQUENCY OF A CENTRAL PROCESSING UNIT, BASED ON DETERMINING A MILLIONS OF INSTRUCTION PER SECOND (MIPS) VALUE ASSOCIATED WITH THE CENTRAL PROCESSING UNIT

Marvell International Ltd...

1. A method comprising:
monitoring an operation of a processing unit within a first time window, wherein within the first time window, the processing
unit operates in an active power mode at (i) a first frequency for a first time period, and (ii) a second frequency for a
second time period;

determining, (i) based on the first frequency and the first time period, a first factor, and (ii) based on the second frequency
and the second time period, a second factor;

based on the first factor and the second factor, determining a millions of instructions per second (MIPS) value for the first
time window, wherein the MIPS value for the first time window is based on a ratio of (i) a sum value that is based on a sum
of the first factor and the second factor, and (ii) a duration of the first time window; and

based on the MIPS value for the first time window, selecting an operating frequency of the processing unit for a second time
window that occurs subsequent to the first time window.

US Pat. No. 9,143,219

EQUAL POWER OUTPUT SPATIAL SPREADING MATRIX FOR USE IN A WIRELESS MIMO COMMUNICATION SYSTEM

Marvell International Ltd...

1. An apparatus, comprising:
a symbol encoder unit configured to produce two or more encoded spatial data streams including at least first and second encoded
spatial data streams, wherein a number, NSS, of the encoded spatial data streams is less than a number, NTX, of antennas to be used to transmit the encoded spatial data streams; and

a spatial spreading unit configured to utilize a spatial spreading matrix Q to distribute the two or more encoded spatial
data streams to the antennas,

wherein:
at least a first portion of the first encoded spatial data stream is correlated with a corresponding portion of the second
encoded spatial data stream,

the spatial spreading matrix Q has NTX rows and NSS columns,

a ratio of squared norms of a sum of components of a row, for different rows of the spatial spreading matrix Q, is equal to
a first constant sequence, and

a ratio of squared norms of a sum of a symbol S1 to be transmitted, when the symbol S1 is equal to 1 or ?1, multiplied by each of the components of a row, for different rows of the spatial spreading matrix Q,
is equal to a second constant sequence.

US Pat. No. 9,407,562

SYSTEMS AND METHODS FOR OPERATING ACCORDING TO AN AUTO-NEGOTIATED DATA RATE IN A NETWORK SWITCH

Marvell International Ltd...

1. A network switch, comprising:
a physical layer device including
an auto-negotiation circuit configured to negotiate a first data rate for transmission of data between the physical layer
device and a network client external to the network switch, and

a first serializer interface configured to receive the negotiated first data rate from the auto-negotiation circuit; and
a media access controller including a second serializer interface,
wherein the first serializer interface of the physical layer device is configured to (i) based on the negotiated first data
rate received from the auto-negotiation circuit, selectively replicate portions of data received from the network client,
and (ii) transmit the data to the second serializer interface of the media access controller at a second data rate regardless
of the first data rate negotiated between the auto-negotiation circuit and the network client, wherein the first data rate
and the second data rate are different.

US Pat. No. 9,323,269

VOLTAGE REGULATOR WITH POSITIVE AND NEGATIVE POWER SUPPLY SPIKE REJECTION

Marvell International Ltd...

1. A voltage regulator comprising:
a supply filter configured to operate from a supply voltage and generate a filtered supply voltage at a first node, wherein
the supply filter includes a transistor and a capacitor,
a first terminal of the transistor receives the supply voltage,
a second terminal of the transistor is connected to the first node,
a control terminal of the transistor receives the supply voltage, and
a first terminal of the capacitor is connected to the first node;
a first circuit configured to operate from both the supply voltage and the filtered supply voltage, receive an input reference
voltage, and generate a second reference voltage based on the input reference voltage;

a bias filter configured to receive the second reference voltage and generate a filtered second reference voltage based on
the second reference voltage; and

a second circuit configured to operate from the filtered supply voltage, receive the filtered second reference voltage, and
generate a regulated voltage based on the filtered second reference voltage.

US Pat. No. 9,319,904

WIRELESS DEVICE CALIBRATION FOR IMPLICIT TRANSMIT BEAMFORMING

MARVELL INTERNATIONAL LTD...

1. A method of calibrating a wireless communication device, the method comprising:
generating, at a calibration station, a first channel descriptor based on a first sounding packet received, at the calibration
station, from the wireless communication device via a first radio frequency (RF) communication channel from the wireless communication
device to the calibration station, the first channel descriptor corresponding to the first RF communication channel;

transmitting, with the calibration station, a second sounding packet via a second RF communication channel from the calibration
station to the wireless communication device;

obtaining, at a processor device, the first channel descriptor;
obtaining, at the processor device, a second channel descriptor corresponding to the second RF communication channel, wherein
the second channel descriptor was generated, at the wireless communication device, based on the second sounding packet, and
the second channel descriptor was obtained, from the wireless communication device, via a communication link separate from
the second RF communication channel;

generating, using the processor device and based on the first channel descriptor and the second channel descriptor, calibration
coefficients indicative of one or both of (i) phase imbalance between a receive RF chain and a transmit RF chain at the wireless
communication device, and (ii) amplitude imbalance between the receive RF chain and the transmit RF chain at the wireless
communication device; and

sending the calibration coefficients from the processor device to the wireless communication device so that the wireless communication
device can use the calibration coefficients to compensate for the one or both of (i) the phase imbalance between the receive
RF chain and the transmit RF chain, and (ii) the amplitude imbalance between the receive RF chain and the transmit RF chain
at the wireless communication device.

US Pat. No. 9,308,455

SYSTEM AND METHOD FOR GAMING IN AN AD-HOC NETWORK

Marvell International Ltd...

1. A method of mobile gaming for a mobile device, the mobile device configured to include a plurality of gaming parameters
representative of a game to be played, the method comprising:
communicating a gaming request from a first mobile device to a second mobile device in order to create an ad-hoc wireless
network, wherein the gaming request includes a game identifier;

detecting a gaming request response from the second mobile device, wherein the gaming request response is associated with
the game identifier;

determining a difference of a display of a user indication for a game play based on a first wireless beacon signal of the
first mobile device and a second wireless beacon signal of the second mobile device, wherein the difference of the display
of the user indication for the game play corresponds to a difference in receiving the first wireless beacon signal of the
first mobile device and the second wireless beacon signal of the second mobile device; and

modifying a game playing sequence to include at least a delay of the display of the user indication on at least one of the
first mobile device or second mobile device based on the determined difference, the delay reducing the determined difference
of the display of the user indication for the game play.

US Pat. No. 9,270,960

SYSTEM AND METHOD FOR GREEN IMBALANCE COMPENSATION IN IMAGES

MARVELL INTERNATIONAL LTD...

1. A method comprising:
obtaining a pixel output signal for each green pixel in a group of pixels of an image captured by a pixel array device, the
group of pixels including a plurality of non-overlapping pixel arrays, wherein each pixel array of the plurality of pixel
arrays includes (i) a first green pixel in a row that includes a red pixel, and (ii) a second green pixel in a row that includes
a blue pixel;

determining a green imbalance value for the group of pixels based on the obtained pixel output signals, the determining of
the green imbalance value including:

for each pixel array of the plurality of pixel arrays, calculating a difference between the pixel output signal for the first
green pixel and the pixel output signal for the second green pixel, and

for each of the calculated differences, determining if the calculated difference is a result of image edge pixels in the group
of pixels; and

adjusting the pixel output signal of at least one of the green pixels in the group of pixels based on the green imbalance
value.

US Pat. No. 9,596,715

LONG WIRELESS LOCAL AREA NETWORK (WLAN) PACKETS WITH MIDAMBLES

Marvell International Ltd...

1. A method of providing channel coherence during transmission of a single data unit from a single communication device to
a network interface via a communication channel, the method comprising:
receiving, by the network interface, a preamble of the single data unit;
updating, by the network interface, a first characteristic based on the preamble;
receiving, by the network interface, a first portion of a data payload of the single data unit;
processing, by the network interface, the single data unit based on the updated first characteristic:
receiving, by the network interface, a midamble of the single data unit, the midamble includes a plurality of Orthogonal Frequency
Division Multiplexing (OFDM) symbols that include calibration information, wherein

the plurality of OFDM symbols of the midamble is generated based on one or more channel training matrices corresponding to
a maximum number of space-time streams of the communication channel,

a first format of a channel training matrix of the one or more channel training matrices utilized to generate the midamble
corresponds to a subset of a second format of a channel training matrix utilized to generate the preamble, and

tones of each OFDM symbol of the midamble are interleaved across multiple space-time streams;
at least one of (i) updating, by the network interface, the first characteristic based on the midamble, or (ii) updating,
by the network interface, a second characteristic based on the midamble;

receiving, by the network interface, a second portion of the data payload; and
processing, by the network interface, the single data unit based on i) the updated first characteristic based on the midamble,
or ii) the updated second characteristic based on the midamble, to provide channel coherence during the transmission of the
single data unit.

US Pat. No. 9,450,660

METHODS AND APPARATUS FOR DIRECTING A BEAM TOWARDS A DEVICE IN THE PRESENCE OF INTERFERENCE

MARVELL INTERNATIONAL LTD...

1. A method for providing channel state feedback about a transmission source and an interfering source, the method comprising:
receiving a first signal from a transmission source and a second signal from an interfering source;
determining a first channel matrix based on the first signal and a second channel matrix based on the second signal;
processing the first channel matrix to obtain an essential channel matrix; and
transmitting a feedback signal to the transmission source that comprises the essential channel matrix and the second channel
matrix.

US Pat. No. 9,230,960

COMBINED TAP CELL AND SPARE CELL FOR LOGIC CIRCUIT

Marvell International Ltd...

1. An integrated circuit, comprising:
a plurality of logic cells arranged in the integrated circuit, wherein at least a subset of the plurality of logic cells includes
one or more transistors; and

a first cell, wherein the first cell includes
(i) a logic portion including one or more transistors, and
(ii) a tap portion, wherein the tap portion provides tap connectivity to (i) the one or more transistors of the subset of
the plurality of logic cells and (ii) the one or more transistors of the logic portion.

US Pat. No. 9,231,737

METHOD AND APPARATUS FOR REMOTELY INCREASING AVAILABLE PROCESSING TIME AT A WIRELESS DEVICE WHILE MAINTAINING RESPONSE DELAY TIME

Marvell International Ltd...

1. A first wireless device comprising:
a coding module configured to encode a first packet and a second packet, wherein the coding module is configured to add an
amount of dummy data to the second packet;

a transmitter configured to transmit the first packet and the second packet to a second wireless device; and
a first receiver configured to (i) in response to the transmission of the first packet to the second wireless device, receive
a first acknowledgement signal from the second wireless device, and (ii) in response to the transmission of the second packet
to the second wireless device, receive a second acknowledgement signal from the second wireless device,

wherein
reception of the first acknowledgement signal at the receiver is delayed a first delay period from an end of the transmission
of the first packet,

reception of the second acknowledgement signal at the receiver is delayed a second delay period from an end of the transmission
of the dummy data, and

the coding module is configured to, prior to the transmission of the second packet to the second wireless device and based
on (i) a predetermined time to process the first packet at the second wireless device, and (ii) a predetermined time to process
the second packet at the second wireless device, determine the amount of dummy data to add to the second packet such that
a length of the first delay period is a same length as the second delay period.

US Pat. No. 9,094,843

METHOD AND APPARATUS FOR ADJUSTING A GAIN OF A RECEIVER IN A WIRELESS DEVICE

Marvell International LTD...

1. A system comprising:
a first power measuring module configured to
generate a first power measurement of a signal received by a receiver of a wireless device, the first power measurement of
the signal being generated based on a plurality of reference signals associated with the signal received by the receiver of
the wireless device,

wherein each of the plurality of reference signals is transmitted at a predetermined power,
wherein the first power measurement is generated in frequency domain, and
wherein the first power measurement is generated during a first frame of the signal received by the receiver of the wireless
device;

a second power measuring module configured to generate a second power measurement of the signal in time domain;
a gain control module configured to adjust a gain of the receiver based on (i) the first power measurement and (ii) the second
power measurement; and

a signal processing module configured to process a second frame in the signal, wherein the second frame is subsequent to the
first frame, and wherein the signal processing module processes the second frame at the gain as adjusted by the gain control
module.

US Pat. No. 9,066,246

EFFICIENT CHANGE RECOGNITION ALGORITHM

Marvell International Ltd...

1. A first device comprising:
a processor that is configured to:
generate a first hash value based on a first data stream;
generate a second hash value based on a second data stream;
detect a change in recognition data that describes at least one of a first access point or a second access point, the detection
of the change being based on a comparison of the first hash value with the second hash value; and

establish an association with the first access point or the second access point based on the detected change; and
a memory in communication with the processor, the memory configured to store the first hash value, and
wherein the processor is further configured to retrieve the first hash value from memory to perform the comparison of the
first hash value with the second hash value.

US Pat. No. 9,058,272

METHOD AND APPARATUS HAVING A SNOOP FILTER DECOUPLED FROM AN ASSOCIATED CACHE AND A BUFFER FOR REPLACEMENT LINE ADDRESSES

MARVELL INTERNATIONAL LTD...

10. An apparatus comprising:
a snoop filter decoupled from an associated cache of the apparatus such that the associated cache changes states of lines
in the cache from a first state that is a clean state to a second state that is not a clean state without the snoop filter's
knowledge, wherein the snoop filter includes a buffer configured to buffer addresses of replaced lines in the associated cache
that are unknown to be clean, the snoop filter further includes a directory configured such that an organization of the directory
matches an organization of the associated cache, the snoop filter being configured to receive an identification of a cache
way to be filled at a time the snoop filter observes a read transaction of the associated cache.

US Pat. No. 9,190,973

HEADPHONE AMPLIFIER

Marvell International Ltd...

1. An amplifier circuit, comprising:
an input stage configured to receive an electrical signal for amplification;
an intermediate stage configured to amplify the electrical signal with an adjustable gain;
an output stage configured to drive an audio output device in response to the amplified electrical signal; and
a detecting and controlling circuit configured to detect an output current for driving the audio output device, and adjust
the gain of the intermediate stage based on the current to compensate for a pole change of the amplifier circuit due to a
change of the current.

US Pat. No. 9,277,421

SYSTEM AND METHOD FOR ESTIMATING NOISE IN A WIRELESS SIGNAL USING ORDER STATISTICS IN THE TIME DOMAIN

MARVELL INTERNATIONAL LTD...

1. A receiver device comprising:
radio frequency logic configured to receive radio frequency signals and to identify a training signal from the radio frequency
signals;

impulse response logic configured to generate a set of sample values representing an impulse response of a system, wherein
the system includes the receiver device and the set of sample values are generated based at least on the training signal;
and

noise estimation logic configured to generate a noise power estimate by:
(i) generating an initial set of magnitude squared values from the set of sample values,
(ii) calculating a mean value of the initial set of magnitude squared values,
(iii) generating a modified set of magnitude squared values from the initial set of magnitude squared values by eliminating
magnitude squared values from the initial set of magnitude squared values based on the mean value,

(iv) generating an initial noise power estimate based on the modified set of magnitude squared values,
(v) generating a threshold value based on the initial noise power estimate,
(vi) averaging values of the initial set of magnitude squared values that are less than the threshold value to generate a
biased noise power estimate, and

(vii) removing a thresholding bias from the biased noise power estimate to generate a final noise power estimate.

US Pat. No. 9,214,964

SYSTEMS AND METHODS FOR CONFIGURING PRODUCT CODES FOR ERROR CORRECTION IN A HARD DISK DRIVE

Marvell International Ltd...

1. A method for encoding data to be stored on a storage device using a product code having a first dimension and a second
dimension, comprising:
receiving a portion of user data to be written in the first dimension, wherein the first dimension is a row dimension or a
column dimension, and the portion of user data is a portion of a row or a portion of a column;

computing first parity symbols for the portion of user data, wherein the first parity symbols are with respect to the first
dimension;

computing partial parity symbols for the portion of user data, wherein the partial parity symbols are with respect to the
second dimension and are used to obtain second parity symbols for the portion of user data; and

storing the user data, the first parity symbols, and the second parity symbols in the storage device.

US Pat. No. 9,106,371

LINK ADAPTATION FOR OFDM SYSTEMS

Marvell International Ltd...

1. A method implemented in a communication device, the method comprising:
determining, in the communication device, an error rate associated with the communication link;
determining, in the communication device, an interference metric associated with a communication link, wherein determining
the interference metric includes

obtaining a plurality of measurements indicative of quality of the communication link, and
calculating the interference metric using (i) a mean of the plurality of measurements and (ii) a standard deviation of the
plurality of measurements,

wherein the interference metric is determined without using the error rate;
selecting, in the communication device, a modulation and coding scheme based on at least (i) the interference metric, and
(ii) the error rate; and

causing at least one packet to be modulated and coded according to the selected modulation and coding scheme prior to transmission
over the communication link.

US Pat. No. 9,860,862

METHODS AND APPARATUS FOR SELECTING A DEVICE TO PERFORM SHARED FUNCTIONALITY IN A DETERMINISTIC AND FAIR MANNER IN A WIRELESS DATA COMMUNICATION SYSTEM

Marvell International Ltd...

1. A method implemented in a first transceiver device of multiple transceiver devices in a data communication system configured
to distribute functionality among the multiple transceiver devices, the method comprising:
sending an identifier of the first transceiver device to other transceiver devices of the multiple transceiver devices;
receiving identifiers of the other transceiver devices;
updating a synchronization value for the data communication system, the updated synchronization value comprising a value used
to facilitate keeping the multiple transceiver devices synchronized;

generating for a given time, for each transceiver device of the multiple transceiver devices, a selection measure from the
updated synchronization value and the respective identifier of each of the transceiver devices;

ranking for the given time each of the multiple transceiver devices based on the generated selection measures; and
selecting, based on the ranking, one of the multiple transceiver devices as a transceiver to transmit a beacon at the given
time.

US Pat. No. 9,276,618

SYSTEMS AND METHODS FOR SIDELOBE CANCELLATION

Marvell International Ltd...

1. A system for performing noise filtering of a received signal, comprising:
a first receiver configured to capture a first signal;
a second receiver configured to capture a second signal, wherein a desired signal is received at the first receiver and the
second receiver substantially simultaneously, wherein noise received from directions other than an expected signal direction
is received at the first antenna and the second antenna at different times;

a signal processor configured to:
generate a combined signal based on the first signal and the second signal;
subtract the combined signal from the first signal to generate a noise indicating signal;
perform a phase adjustment on the combined signal based on the noise indicating signal;
subtract the phase adjusted signal from the first signal to recreate the noise indicating signal;
wherein the noise indicating signal is subtracted from the combined signal to generate a filtered output signal.

US Pat. No. 9,235,712

SECURING EXTERNAL MEMORY DATA

Marvell International Ltd...

1. A method performed by a system on a chip (SOC), the method comprising:
obtaining a key and a flag;
identifying a proper subset of the key;
selectively modifying the proper subset of the key responsive to the flag;
receiving data to be output to a memory external to the SOC;
scrambling the received data using the proper subset of the key; and
outputting the scrambled data to the memory external to the SOC.

US Pat. No. 9,203,429

SYSTEMS AND METHODS FOR USING MARKOV DISTRIBUTION CODES IN DATA STORAGE SYSTEMS

MARVELL INTERNATIONAL LTD...

1. A method for encoding information using a code specified by a target Markov distribution, the method comprising:
selecting a set of parameters comprising a block length, a plurality of weight metrics, and a threshold;
estimating a Markov distribution associated with the selected set of parameters from a plurality of data blocks defined by
the selected parameters;

modifying the set of parameters based on the estimated Markov distribution; and
encoding the information using the modified set of parameters.

US Pat. No. 9,160,664

DETERMINING A NETWORK NODE WHETHER A RECEIVED PACKET FRAGMENT SHOULD BE TRANSMITTED

Marvell International Ltd...

1. A method of controlling transmission of data units in a network node, the method comprising:
receiving a current fragment of a data unit at the network node, the data unit having been fragmented into an ordered sequence
of fragments prior to the current fragment being received at the network node;

determining, at the network node, whether the current fragment is expected, wherein determining whether the current fragment
is expected includes determining a position of the current fragment within the ordered sequence of fragments; and

in response to determining that the current fragment is expected based on the determined position of the current fragment
within the ordered sequence of fragments, transmitting the current fragment via a network link.

US Pat. No. 9,088,553

TRANSMITTING MESSAGE PRIOR TO TRANSMITTING ENCAPSULATED PACKETS TO ASSIST LINK PARTNER IN DECAPSULATING PACKETS

Marvell International LTD...

1. A network device comprising:
a first circuit configured to receive one or more of a plurality of first packets from a host unit or a processor of the network
device;

a second circuit configured to
generate a plurality of second packets based on the plurality of first packets, and insert, in each of the plurality of second
packets, a different value for a count;

a third circuit configured to
generate a plurality of seeds,
wherein each of the plurality of seeds is based on (i) a first predetermined key, (ii) an address of the network device, and
(iii) a predetermined value for the count;

a fourth circuit configured to encapsulate each of the plurality of second packets using one of the plurality of seeds generated
based on the value for the count in the respective one of the plurality of second packets; and

a fifth circuit configured to
send to an antenna of the network device a message comprising (i) the address of the network device and (ii) the predetermined
value for the count; and

send to the antenna, subsequent to sending the message, the plurality of encapsulated second packets.

US Pat. No. 9,070,410

RECORDING MEDIA, DATA STORAGE DEVICES, AND METHODS FOR DETERMINING A POSITION ERROR SIGNAL IN A RECORDING MEDIUM

MARVELL INTERNATIONAL LTD...

9. A data storage device, comprising:
a recording medium comprising
a data layer configured to record data therein, and
a dedicated servo layer configured to provide servo information, wherein the dedicated servo layer comprises a plurality of
tracks, wherein a first track comprises a first servo signal, the first servo signal including first servo bursts of a predetermined
frequency, wherein a second track adjacent to the first track comprises a second servo signal, the second servo signal including
second servo bursts of the predetermined frequency, and wherein the first servo bursts are orthogonal to the second servo
bursts, and wherein the data in the data layer and the servo information in the dedicated servo layer are provided in a same
direction; and

a reader head configured to read a signal from the recording medium.

US Pat. No. 9,426,745

MEMORY EFFICIENT FILTERING

MARVELL INTERNATIONAL LTD...

1. A method for controlling operation of a wireless network device that includes (i) a host assembly, and a (ii) a wireless
device assembly, the method comprising:
receiving a filter expression at the wireless device assembly;
while the host assembly is in a sleep mode,
receiving, at the wireless device assembly, a signal having packet data,
determining, at the wireless device assembly, whether the packet data contains a data pattern indicated by the filter expression,
if the wireless device assembly determines that the packet data contains the data pattern indicated by the filter expression,
the wireless device assembly communicating a wakeup signal to the host assembly for changing the host assembly from the sleep
mode to an active mode, and

if the wireless device assembly determines that the packet data does not contain the data pattern indicated by the filter
expression, not communicating the wakeup signal to the host assembly.

US Pat. No. 9,354,898

DETECTION OF A USB OS DESCRIPTOR REQUEST TO FACILITATE INSTALLATION OF A DEVICE DRIVER

Marvell International Ltd...

1. A method comprising:
coupling a first device to a client device using a universal serial bus (USB) link;
in response to coupling the first device to the client device using the USB link, detecting, by the first device, a USB operating
system (OS) descriptor request made by the client device;

in response to detecting the USB OS descriptor request, reporting, by the first device, a mass storage interface to the client
device;

using the mass storage interface to install, in the client device, a device driver associated with the first device;
subsequent to the installation of the device driver in the client device, operatively de-coupling the first device from the
client device;

subsequent to operatively de-coupling the first device from the client device, operatively re-coupling the first device to
the client device; and

in response to a lack of a USB OS descriptor request from the client device to the first device subsequent to operatively
re-coupling the first device to the client device, reporting, by the first device, all interfaces of the first device other
than the mass storage interface to the client device.

US Pat. No. 9,322,717

TEMPERATURE SENSOR ARCHITECTURE

MARVELL INTERNATIONAL LTD...

1. A device comprising: a signal generator to receive an input voltage indicative of a temperature and generate a pulse signal
having a period determined from the input voltage; and a counter to output a temperature code based on the pulse signal, the
temperature code being indicative of the temperature, wherein the signal generator comprises a comparator to receive the input
voltage and an oscillator output signal, to compare the input voltage with the oscillator output signal, and to output a comparison
signal that causes a change in a value of the oscillator output signal, wherein the signal generator further comprises: a
capacitor to be charged or discharged according to the comparison signal output from the comparator, the capacitor being coupled
to an output node through which the oscillator output signal is output; and a waveform converter coupled to the capacitor
and to convert the oscillator output signal having a triangular waveform into the pulse signal having a rectangular waveform,
the oscillator output signal having a period that is substantially the same as the period of the pulse signal.

US Pat. No. 9,306,851

APPARATUS AND METHODS TO STORE DATA IN A NETWORK DEVICE AND PERFORM LONGEST PREFIX MATCH (LPM) PROCESSING

Marvell International Ltd...

1. A network device comprising:
a random access memory (RAM);
a buffer coupled with the RAM;
circuitry configured to copy data from a location in the RAM to the buffer responsive to a received identifier corresponding
to a search key corresponding to a received packet; and

circuitry configured to compare the data copied to the buffer with the search key to provide a result for use in forwarding
of the packet, wherein don't care bits for the comparison are determined from a count of don't care bits encoded in a portion
of the location in the RAM indicated by the identifier, and wherein the circuitry configured to compare comprises (i) circuitry
configured to generate a vector mask from the count, (ii) an XOR gate coupled to compare a predetermined bit of the search
key with a corresponding predetermined bit of the data copied to the buffer, and (iii) an AND gate coupled to compare a predetermined
bit of the vector mask with an output of the XOR gate.

US Pat. No. 9,100,910

PACKET CLASSIFICATION BASED POWER SAVING RECEIVER

Marvell International LTD...

1. A method, comprising:
detecting an incoming frame at a receive unit operating in a receive frame state;
determining that the incoming frame is not intended for the receive unit;
calculating expected power savings for entering an ignore state based, at least in part, on a switching time and frame duration
of the incoming frame; and

determining to remain in the receive frame state rather than the ignore frame state when the expected power savings for entering
the ignore state is insufficient, wherein in the receive frame state the incoming frame is received by the receive unit, and
wherein when the receive unit is in the ignore frame state the incoming frame is (i) not received or (ii) not decoded by the
receive unit.

US Pat. No. 9,055,039

SYSTEM AND METHOD FOR PIPELINED ENCRYPTION IN WIRELESS NETWORK DEVICES

Marvell International LTD...

1. An apparatus for transmitting N packets of data, wherein N?1, the apparatus comprising:
a key mixing circuit to generate N groups of encryption seeds, wherein each encryption seed is generated based upon a predetermined
key, a transmitter address, and a predetermined start value for a packet number;

an input circuit to receive the N packets of data;
a packet number circuit to insert a different one of N values for the packet number into each of the N packets of data, wherein
each of the N values for the packet number is greater than or equal to the predetermined start value for the packet number;

an encryption circuit to encrypt each of the N packets of data using the one of the N groups of encryption seeds that was
generated based on the value for the packet number in the respective one of the N packets of data; and

an output circuit to send the N packets of data,
wherein the encryption circuit generates each of the N groups of encryption seeds before the input circuit receives the respective
one of the N packets of data, and

wherein the predetermined start value is initialized when the apparatus is initialized and incremented when each of the N
groups of encryption seeds is generated.

US Pat. No. 9,489,023

SECURE WAKE ON LAN WITH WHITE LIST

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
memory configured to store a white list, wherein the white list includes credentials for entities that are authorized to wake
a host device from a power-saving sleep mode;

sleep manager logic configured to:
receive, from a requesting entity, a request to wake the host device;
determine if credentials for the requesting entity are on the white list; and
wake the host device when the credentials for the requesting entity are on the white list
wherein the sleep manager logic is configured to receive the white list from authorization manager logic that uses the white
list to manage access to the host device when the host device is not in the power-saving sleep mode.

US Pat. No. 9,288,764

DISCOVERY-PHASE POWER CONSERVATION

Marvell International Ltd...

1. A method comprising:
conserving power during a discovery phase of a wireless local area network (WLAN) by:
reducing power at an access point of the WLAN for a reduced-power-usage period, the access point operating in a geographical
region and at a frequency in which client stations operating in the geographical region and at the frequency are required
to passively search for radar transmissions prior to transmitting at the frequency, the reduced power being insufficient to
enable the access point to receive communication from a client station within a communication region of the access point,
the reduced-power-usage period at the access point being configured to vary according to information received by the access
point from one of the client stations indicating a radar search period for the client station; and

increasing power at the access point subsequent to expiration of the reduced-power-usage period, the power being increased
for a reception period in which communication from the client station or another client station within the communication region
of the access point can be received.

US Pat. No. 9,246,827

METHOD AND APPARATUS FOR CONTROLLING THE FLOW OF PACKETS IN A DATA NETWORK

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
a transceiver; and
a flow logic configured to control the transceiver to transmit a frame to a downstream device by
determining whether the frame matches a predetermined pattern, wherein the predetermined pattern identifies a source of frames,
and wherein the predetermined pattern specifies an action to perform for frames received from the source that are to be transmitted
to the downstream device, and

in response to determining that the frame matches the predetermined pattern, controlling transmission of the frame to the
downstream device according to the action specified by the predetermined pattern,

wherein the predetermined pattern is received from the downstream device, and wherein the flow logic is configured to control
transmission of the frame by selecting an action according to the predetermined pattern, wherein the action includes changing
a class of service for the frame by modifying the class of service in a header of the frame, pausing transmission of the frame,
logging information about the frame, redirecting the frame to a different destination, permitting transmission of the frame,
and dropping the frame,

wherein changing the class of service includes promoting the frame to a higher class of service and demoting the frame to
a lower class of service, wherein controlling transmission of the frame according to the action manages the flow of the frame
through the network, wherein the class of service defines a priority level associated transmitting the frame, and wherein
the priority level is selected from a range of priority levels that include at least eight different priority levels.

US Pat. No. 9,100,917

POWER SAVE MODES FOR A SYSTEM-ON-CHIP AND A HOST PROCESSOR OF A WIRELESS DEVICE

Marvell International LTD...

1. A system-on-chip comprising:
a host interface module configured to
interface the system-on-chip to a host processor of a wireless device, and
communicate with the host processor of the wireless device via a bus,
wherein the bus uses an application programming interface of the system-on-chip, and
wherein the host processor uses the application programming interface to configure a power save mode of the system-on-chip;
and

a power management module configured to, in response to the host processor stopping communication with the system-on-chip,
operate the system-on-chip in the power save mode without performing a handshake with the host processor via the bus,

wherein the handshake includes (i) sending a request to enter the power save mode to the host processor via the bus, and (ii)
receiving an acknowledgement of the request from the host processor via the bus.

US Pat. No. 9,059,160

SEMICONDUCTOR PACKAGE ASSEMBLY

Marvell International Ltd...

1. An assembly comprising:
a first semiconductor package comprising:
a first substrate, and
a first semiconductor die mounted on the first substrate in a wirebonding configuration using a bonding wire to electrically
couple the first semiconductor die to the first substrate;

a second semiconductor package comprising:
a second substrate, and
a second semiconductor die (i) mounted on the second substrate and (ii) electrically coupled to the first semiconductor die,
wherein power and/or ground signals are routed between the first semiconductor die and the second semiconductor die using
electrical connections of (i) the second substrate and (ii) not the first substrate;

a molding compound substantially encapsulating the first semiconductor die and the bonding wire; and
a through-hole via formed through the molding compound and coupled to the first semiconductor die, wherein the through-hole
via is electrically connected between the first semiconductor die and the second semiconductor die to route the power and/or
ground signals between the first semiconductor die and the second semiconductor die.

US Pat. No. 9,398,603

APPARATUS TO FACILITATE CO-EXISTENCE OF BLUETOOTH AND WIRELESS LOCAL AREA NETWORKS

Marvell International Ltd...

1. An apparatus for receiving wireless local area network packets and Bluetooth packets, wherein the apparatus comprises:
means for predicting, prior to receiving a wireless local area network packet, (i) a first time at which reception of the
wireless local area network packet is going to end, and (ii) a second time at which a transmission of a first Bluetooth packet
is going to start; and

means for simultaneously receiving both the wireless local area network packet and a second Bluetooth packet in response to
the first time being predicted to occur prior to the second time, wherein the first Bluetooth packet is to be transmitted
subsequent to the second Bluetooth packet being received.

US Pat. No. 9,326,235

METHOD AND APPARATUS FOR WIRELESS ARBITER POWER SAVING

MARVELL INTERNATIONAL LTD...

1. An apparatus comprising:
a first wireless communication protocol component configured to process, according to a first communication protocol, radio
frequency (RF) signals;

a circuit configured to process requests to access an antenna, the requests from a plurality of wireless communication protocols
components including (i) the first wireless communication protocol component and (ii) a second wireless communication protocol
component, wherein the second wireless communication protocol component is configured to process, according to a second communication
protocol, RF signals, and wherein the circuit is further configured to cause a switch to provide the second wireless communication
component with access to the antenna while the circuit is in a power saving mode of operation,

upon the circuit exiting the power saving mode of operation, determine whether the second wireless communication component
is utilizing the antenna,

upon the circuit exiting the power saving mode of operation, apply a first set of arbitration rules for processing requests
to access the antenna when it is determined that the second wireless communication component is utilizing the antenna, wherein
the first set of arbitration rules permits the first wireless communication component access to the antenna under specified
conditions such that communications by the second wireless communication component is disrupted, and

upon the circuit exiting the power saving mode of operation, apply a second set of arbitration rules for processing requests
to access the antenna when it is determined that the second wireless communication component is not utilizing the antenna.

US Pat. No. 9,294,941

PHASE TRANSFORMATION OF REPEATED SIGNALS

MARVELL INTERNATIONAL LTD...

1. An apparatus, comprising:
duplication logic implemented with at least hardware configured to duplicate a first string of data and a second string of
data to form a first duplicate string of data and a second duplicate string of data;

transformation logic configured to convert the first string of data into a string of conjugate values and to rotate symbols
in the second string of data based at least on a constellation diagram to produce a string of rotated symbols, wherein data
values of the string of rotated symbols are not conjugate values of the second string of data; and

signal generation logic configured to generate and output a signal for wireless transmission comprising at least the string
of conjugate values, the first duplicate string of data, the string of rotated symbols, and the second duplicate string of
data,

wherein the signal for wireless transmission is output by the signal generation logic to a device that performs the wireless
transmission of the signal.

US Pat. No. 9,262,312

APPARATUS AND METHODS TO COMPRESS DATA IN A NETWORK DEVICE AND PERFORM CONTENT ADDRESSABLE MEMORY (CAM) PROCESSING

Marvell International Ltd...

1. A method comprising:
receiving, in a network device, records to be stored in a content addressable memory (CAM) of the network device for use in
network packet routing;

comparing the records to identify similar bit values at respective bit positions of at least a portion of the records;
storing in the CAM the similar bit values in a single sample record corresponding to the portion of the records;
storing in the CAM remaining non-similar bit values of the portion of the records, thereby compressing the portion of the
records stored in the CAM;

storing in the CAM one or more remaining records of the received records not included in the portion of the records; and
searching the CAM including the compressed portion of the records and the one or more remaining records to generate an output
search result.

US Pat. No. 9,264,030

ADAPTIVE VOLTAGE SCALING USING A DELAY LINE

MARVELL INTERNATIONAL LTD...

1. A method comprising:
measuring a supply voltage that is powering a series of delay stages;
measuring a number of delay stages corresponding to how many of the delay stages a signal traverses in a predetermined time
period;

converting the measured supply voltage to a derived number of delay stages, based on a predetermined correlation of number
of delay stages versus voltage;

performing a comparison between (i) the measured number of delay stages and (ii) the derived number of delay stages; and
adjusting the supply voltage based on the comparison between (i) the measured number of delay stages and (ii) the derived
number of delay stages.

US Pat. No. 9,256,487

OPTIMIZING ERROR FLOOR PERFORMANCE OF FINITE-PRECISION LAYERED DECODERS OF LOW-DENSITY PARITY-CHECK (LDPC) CODES

Marvell International Ltd...

1. A system comprising:
memory circuitry configured to store check node messages and variable node messages; and
processing circuitry configured to:
update a variable node message of stored variable node messages; and
determine if the variable node message is saturated based on a precision of the variable node message.

US Pat. No. 9,250,992

TEST DATA REPORTING DURING MEMORY TESTING

Marvell International Ltd...

1. A method comprising:
performing, by memory testing circuitry, an execution of a test sequence to test a memory device, wherein performing the execution
comprises generating addresses of the memory device in accordance with the test sequence and advancing a value of a modulo
counter as each of the addresses is generated;

enabling error logging when a generated address and a value of the modulo counter corresponding to the generated address match
an address and a value of the modulo counter stored for a previously detected error;

detecting an error in data read from the memory device after enabling error logging; and
storing information associated with the detected error, wherein storing the information comprises storing an address generated
for reading the data associated with the detected error from a location of the memory device and storing a value of the modulo
counter corresponding to the address generated for reading the data.

US Pat. No. 9,251,831

RECORDING MEDIUM AND METHOD OF FORMING THE SAME

Marvell International Ltd...

1. A recording medium comprising:
a servo layer configured to have a nucleation field value, wherein the servo layer is a magnetic layer;
a data recording layer configured to have a saturation field value, wherein the nucleation field value of the servo layer
is greater than or equal to 80% of the saturation field value of the data recording layer, and wherein the data recording
layer is a magnetic layer;

a first intermediate layer disposed on the servo layer, wherein the data recording layer is disposed on the first intermediate
layer;

a second intermediate layer, wherein the servo layer is disposed on the second intermediate layer;
a crystalline underlayer comprising a first side and a second side, wherein the second side opposes the first side, and wherein
the second intermediate layer is disposed on the first side of the crystalline underlayer; and

an amorphous underlayer disposed on the second side of the crystalline underlayer, wherein the crystalline underlayer or the
amorphous underlayer is a magnetic layer.

US Pat. No. 9,179,523

START-UP VOLTAGE FOR SOLID-STATE LIGHTING CONTROLLER

Marvell International Ltd...

1. A controller, comprising:
a transformer;
a control circuit having a bias voltage input, the control circuit activating when voltage applied to the bias voltage input
reaches a threshold, the transformer having feedback output to the bias voltage input;

a bias voltage circuit applying start-up bias voltage to the bias voltage input, the bias voltage circuit having:
a resonant circuit providing a resonant current,
a bias capacitor charged by the resonant current, wherein voltage on the bias capacitor is the start-up bias voltage applied
to the bias voltage input, and

a cutoff, comprising at least one diode, to shut down the resonant current when the transformer feedback output is above the
threshold; wherein:

the resonant circuit includes a voltage divider that divides a supply voltage to produce a divided voltage for the resonant
current; and

the voltage divider is adjusted so that when the transformer feedback output is above the threshold, voltage on the transformer
feedback output is above the divided voltage, thereby turning off the at least one diode.

US Pat. No. 9,455,644

METHOD AND APPARATUS FOR CHANGING A FREQUENCY OF A SWITCH PRIOR TO A LEVEL OF CURRENT RECEIVED FROM AN INDUCTOR DECREASING TO A PREDETERMINED LEVEL

Marvell International Ltd...

1. A controller comprising:
a switch configured to (i) receive current from an inductor, and (ii) bypass a portion of the current from being received
by a load, wherein the switch is configured to be cycled between a first state and a second state at a frequency;

a first module configured to, for a first cycle of the switch, determine a first amount of time the switch is in the first
state;

a second module configured to, based on the first amount of time, determine a second amount of time for a level of the current
to decrease to a predetermined level, wherein the second amount of time begins during the first cycle and when the switch
transitions from the first state to the second state; and

a control module configured to, based on the second amount of time and prior to the current decreasing to the predetermined
level, change the frequency of the switch.

US Pat. No. 9,423,821

METHOD AND APPARATUS FOR SETTING A FREQUENCY OF A CLOCK SIGNAL FOR A BUS BASED ON A NUMBER OF REQUESTS ASSERTED TO TRANSFER DATA ON THE BUS

Marvell International Ltd...

1. An integrated circuit comprising:
a generator configured to, based on a summation signal, generate a clock signal having a frequency;
a plurality of devices configured to generate respective requests, wherein each of the requests requests transfer of data
on a bus, and wherein each of the plurality of devices is configured to, based on the frequency of the clock signal, transfer
the data for the corresponding request on the bus;

a summer configured to (i) receive the requests generated by the plurality of devices, and (ii) generate the summation signal
based on a number of the requests being in an asserted state during a first period of time; and

a first module configured to, based on the summation signal, increase a second period of time that a first request of one
of the plurality of devices is in an asserted state, wherein the second period of time is increased to include or overlap
the first period of time, and

wherein the summer is configured to, as a result of the increase in the second period of time, generate the summation signal
further based on the first request.

US Pat. No. 9,319,122

BEAMSTEERING IN A SPATIAL DIVISION MULTIPLE ACCESS (SDMA) SYSTEM

MARVELL INTERNATIONAL LTD...

1. A method in a communication network, the method comprising:
obtaining, at a first communication device, respective channel estimate matrices of respective communication channels between
i) the first communication device and ii) respective second communication devices;

generating, at the first communication device, respective steering matrices for use in communicating with the respective second
communication devices, wherein generating each respective steering matrix for use in communicating with each respective second
communication device comprises

generating, at the first communication device, a respective aggregate channel matrix that i) includes channel estimate matrices
for others of the second communication devices, and ii) omits the channel estimate matrix for the respective second communication
device, and

using, at the first communication device, the respective aggregate channel matrix to generate the respective steering matrix
to project to a null-space of a space spanned by channel estimate matrices corresponding to others of the second communication
devices; and

utilizing, at the first communication device, the respective steering matrices to simultaneously transmit respective signals
to the respective second communication devices.

US Pat. No. 9,275,654

METHOD AND APPARATUS FOR WRITING SERVO INFORMATION ON A RECORDING MEDIUM

Marvell International LTD...

1. A servo system for writing servo information on a recording medium of a hard disk drive and for determining head position
based on the servo information with a tracking accuracy, the servo system comprising:
a first magnetic head; and
a servo writer configured to write the servo information on the recording medium of the hard disk drive by:
forming, using the first magnetic head, a first servo pattern comprising a servo burst pattern on a segment of a first track
of a first layer of the recording medium of the hard disk drive; and

forming a second servo pattern comprising addressing information on a segment of a second track of a second layer of the recording
medium of the hard disk drive,

wherein the second track is arranged half a width of the first track from the first track; and
wherein arranging the second track half the width of the first track from the first track increases the tracking accuracy
of the servo system.

US Pat. No. 9,258,219

MULTI-UNIT SWITCH EMPLOYING VIRTUAL PORT FORWARDING

MARVELL ISRAEL (M.I.S.L.)...

1. A multi-unit switch, comprising:
a plurality of network devices; and
a central switch communicatively coupled to the plurality of network devices, the central switch including:
a plurality of physical ports, wherein at least some of the physical ports are communicatively coupled to the plurality of
network devices, and

a forwarding database for making virtual port forwarding decisions, the forwarding database being populated with virtual port
entries corresponding to virtual ports of the central switch, at least some virtual ports of the central switch being associated
with physical ports of a unit in the multi-unit switch other than the central switch, ones of the virtual ports mapped to
corresponding physical ports of the central switch and/or to respective ones of physical ports of the plurality of network
devices;

wherein the central switch is configured to:
make forwarding decisions, using the forwarding database, to determine virtual ports to which packets are to be forwarded,
and

forward packets to appropriate physical ports of the central switch based on the virtual port forwarding decisions based on
the mapping of virtual ports to physical ports; and

wherein network devices among the plurality of network devices are configured to:
receive packets from the central switch, and
forward the received packets to appropriate physical ports of the network devices responsively to indications of the virtual
port forwarding decisions made by the central switch, ones of the respective indications being separate from a virtual local
area network (VLAN) tag.

US Pat. No. 9,147,497

METHOD AND APPARATUS FOR SAMPLING

Marvell International Ltd...

1. A sampling circuit, comprising:
a switch having a control terminal and a transistor;
a first driving module configured to output a first signal to the switch in response to a received analog input signal, the
first signal being substantially equivalent to the received analog input signal; and

a second driving module configured to output a second signal in response to the received analog input signal such that a control
voltage on the control terminal is varied based on the received analog input signal, a substantially constant voltage difference
being applied between the control terminal and the second driving module, the second signal being substantially equivalent
to the received analog input signal, and a first error difference between the first signal of the first driving module and
the received analog input signal being smaller than a second error difference between the second signal of the second driving
module and the received analog input signal, wherein the first driving module is further configured to drive the first signal
substantially equivalent to the received analog input signal with a higher power than the second driving module driving the
second signal.

US Pat. No. 9,070,454

FLASH MEMORY

Marvell International Ltd...

1. A system comprising:
flash memory cells; and
a flash controller, the flash controller configured to:
receive a threshold voltage value, the threshold voltage value calculated based on a cycle count of a flash memory cell of
the flash memory cells; and

write a data value to the flash memory cell using the threshold voltage value as a target threshold voltage.

US Pat. No. 9,509,512

MESSAGE DIGEST GENERATOR

Marvell International Ltd...

1. A network device, comprising:
an append bits module to generate a padded message by appending padding bits to a first message;
an append length module to generate a second message by appending length data to the padded message, the length data representing
an amount of data in the first message;

an initialization module to store predetermined values in buffers;
an array to store constants; and
a looping message-digest calculation module to generate a first message digest in response to (i) the first message, (ii)
the padding bits, (iii) the length data, (iv) the predetermined values in the buffers, and (v) the constants, wherein:

the first message digest corresponds to the first message,
the looping message-digest calculation module generates the first message digest by executing a plurality of iterations of
a first portion of code,

in each iteration of the plurality of iterations of the first portion of code, the looping message-digest calculation module:
selects a different one of a predefined plurality of functions for use in the iteration,
calculates a set of values based on evaluating the selected one of the predefined plurality of functions using data from the
buffers, and

updates the buffers based on the set of values, and
the looping message-digest calculation module outputs the first message digest based on final results in the buffers after
executing the plurality of iterations of the first portion of code.

US Pat. No. 9,490,745

VOLTAGE-CONTROLLED OSCILLATOR

Marvell International Ltd...

1. A voltage-controlled oscillator comprising:
a tank circuit;
a first pair of transistors, wherein each transistor of the first pair of transistors comprises a gate, a drain, and a source,
and wherein the drains of the first pair of transistors are coupled to the tank circuit and the gates of the first pair of
transistors are cross-coupled with the drains of the first pair of transistors;

a second pair of transistors, wherein each transistor of the second pair of transistors comprises a gate, a drain, and a source,
and wherein the drains of the second pair of transistors are coupled to the tank circuit and the gates of the second pair
of transistors are cross-coupled with the drains of the second pair of transistors;

a first resonant filter comprising a first inductor, wherein a first terminal of the first inductor is coupled to the sources
of the first pair of transistors, and wherein the first resonant filter does not include a capacitor coupled in parallel to
the first inductor;

a second resonant filter comprising (i) a second inductor and (ii) a capacitor bank coupled in parallel to the second inductor,
wherein the first inductor of the first resonant filter and the second inductor of the second resonant filter are magnetically
coupled,

wherein the capacitor bank of the second resonant filter is configured to be used to tune each of the first resonant filter
and the second resonant filter, based on the first inductor of the first resonant filter and the second inductor of the second
resonant filter being magnetically coupled,

wherein a first terminal of the second inductor is coupled to the sources of the second pair of transistors; and
a current source having a first terminal coupled to a second terminal of the first inductor.

US Pat. No. 9,461,671

METHOD AND APPARATUS FOR POWER CONSERVATION IN LDPC DECODING

Marvell International Ltd...

1. A method of decoding codewords in conjunction with a low-density parity-check (LDPC) code that defines variable nodes and
check nodes, the method comprising:
receiving a codeword over a data channel;
evaluating quality of the data channel;
iteratively updating values of the variable nodes to decode the codeword, wherein the values of the variable nodes are updated
at different levels of numeric precision depending on the evaluated quality of the data channel, wherein a first one or more
registers are associated with processing high-order bits of the values of the variable nodes, and wherein a second one or
more registers are associated with processing low-order bits of the values of the variable nodes;

storing the values of the variable nodes in a memory, wherein the memory is partitioned in a first memory and a second memory,
wherein the memory is different and separate from the each of (i) the first one or more registers and (ii) the second one
or more registers, wherein the first memory is configured to store the high-order bits of the values of the variable nodes
and the second memory is configured to store the low-order bits of the values of the variable nodes; and

in response to the evaluated quality of the data channel being greater than a predetermined threshold, disabling (i) the second
one or more registers and (ii) the second memory.

US Pat. No. 9,451,057

COMMUNICATION SYSTEM AND ENCODING METHOD HAVING LOW OVERHEAD

MARVELL INTERNATIONAL LTD...

1. A method for coding a packet of information words into a frame for transmission, the method comprising:
receiving blocks of input data, the input data including control words and the packet of information words, the packet having
(a) a start preceded by ones of the control words and (b) an ending followed by others of the control words, the blocks being
smaller than the packet;

determining whether a block consists exclusively of information words and:
when the block consists exclusively of information words, appending to the block a one bit block header having a first sense
to form the frame,

when the block does not consist exclusively of information words:
condensing the block to accommodate a TYPE word, generating the TYPE word to have a value that indicates one of the following
structural properties of the block:

(a) a position of the start of the packet in the block,
(b) a position of the end of the packet in the block, and
(c) the block being composed exclusively of control words,
inserting the TYPE word into the block,
appending to the block a one bit block header having a second sense, opposite to the first sense, to form the frame, and
performing forward error correction (FEC) on multiple frames to generate FEC data.

US Pat. No. 9,368,235

DEFECT DETECTION USING PATTERN MATCHING ON DETECTED DATA

Marvell International Ltd...

1. A method for detecting defects comprising:
receiving, with defect detector circuitry, data from a storage device;
retrieving a symbol comprising a relational expression including a relational operator and a value from memory, wherein the
memory is separate from the storage device;

determining whether the data satisfies the relational expression of the symbol to identify a defect in the storage device;
and

indicating a location of the defect on the storage device.

US Pat. No. 9,343,083

METHOD AND APPARATUS FOR DETERMINING RELATIVE HEAD-TO-DISK SPEED

Marvell International LTD...

1. A system comprising:
a write head configured to write a pattern on a track of a medium of a hard disk drive, wherein the pattern has a first frequency;
a read head configured to read the pattern written on the track of the medium of the hard disk drive;
a mixer configured to mix a first signal generated by reading the pattern with a second signal to obtain a mixed signal, wherein
the second signal has a second frequency, and wherein the second frequency is different than the first frequency; and

a filter configured to filter the mixed signal to determine a relative head-to-disk speed,
wherein the filter has a parameter selected based on a difference between the first frequency and the second frequency.

US Pat. No. 9,299,377

STORAGE MEDIUM FOR STORING DATA AND SERVO INFORMATION

Marvell International LTD...

1. A storage medium comprising:
a servo layer configured to store servo information on the storage medium, wherein the servo information is used to position
a read/write head over the storage medium during a read/write operation performed by the read/write head on the storage medium;

a data recording layer configured to store data on the storage medium during the read/write operation performed by the read/write
head on the storage medium;

a heat sink layer disposed between the servo layer and the data recording layer;
a diamond-like carbon layer with a lubricant covering disposed above the data recording layer;
a thermal isolating layer disposed above the servo layer;
an intermediate layer; and
a thermal conditioning layer,
wherein the intermediate layer, the heat sink layer, and the thermal conditioning layer are disposed between the data recording
layer and the thermal isolating layer.

US Pat. No. 9,256,492

ITERATIVE DECODING SYSTEMS USING NOISE-BIASING

MARVELL INTERNATIONAL LTD...

1. A method comprising:
combining channel samples with a set of noise samples to obtain a first set of biased channel samples;
processing the first set of biased channel samples using an iterative decoder to obtain output samples of the iterative decoder;
determining if iterative decoding failed based on the output samples; and
in response to a determination that the iterative decoding failed:
modifying the set of noise samples and combining the channel samples with the modified set of noise samples to obtain a second
set of biased channel samples; and

re-running the iterative decoder using the second set of biased channel samples.

US Pat. No. 9,160,373

SYSTEMS AND METHODS FOR JOINT DECODING OF SECTOR AND TRACK ERROR CORRECTION CODES

Marvell International Ltd...

1. A method for decoding data stored on a storage device, comprising:
retrieving data from the storage device, wherein the retrieved data are encoded using a product code having a first dimension
and a second dimension;

processing at least one codeword from the first dimension to form detector soft information;
decoding the at least one codeword from the first dimension based on the detector soft information, to form a first decoder
soft information;

decoding at least one codeword from the second dimension based on the first decoder soft information, to form a second decoder
soft information; and

updating the detector soft information based on the second decoder soft information in response to determining that a target
syndrome weight of one of the at least one codeword from the second dimension exceeds a predetermined threshold.

US Pat. No. 9,161,145

SYSTEM AND METHOD FOR FASTER RESPONSE TIME IN STEREO/MONO DETECTION

Marvell International Ltd...

1. A receiver comprising:
an extractor configured to extract a pilot energy signal from an input signal;
a hardware filter having a variable leakage factor that is being varied to filter the pilot energy signal and outputting a
filtered pilot energy signal;

a detector configured to cause the receiver to output one of a monophonic signal and a stereophonic signal based on a residual
signal that is indicative of a difference between the pilot energy signal and the filtered pilot energy signal.

US Pat. No. 9,148,200

DETERMINING POWER OVER ETHERNET IMPAIRMENT

Marvell International Ltd...

11. A device for use in a Power over Ethernet application, the device comprising:
an echo canceller to remove echo energy from a signal received at a receiver; and
a tap monitor configured to:
monitor operation of said echo canceller;
compute a value of the echo energy over time; and
responsive to a determination that the computed value of the echo energy over time is above a threshold, which indicates a
saturation condition, trigger an alert to indicate unbalanced current as a cause of the saturation condition.

US Pat. No. 9,106,323

METHOD AND APPARATUS FOR DETERMINING WHETHER A CHANNEL IS BUSY

Marvell International Ltd...

1. A network device comprising:
a receiver configured to receive a first signal on a channel;
a demodulator configured to output demodulated data based on the first signal;
a correlator configured to, based on the first signal, perform a correlation to generate correlated data;
a first gain device configured to (i) monitor a change in gain of the first signal, and (ii) based on the change in the gain
of the first signal, generate a second signal;

a validating device configured to (i) determine whether the first signal is a valid direct sequence spread spectrum signal,
and (ii) based on whether the first signal is a valid direct sequence spread spectrum signal, generate a third signal, wherein
the validating device is configured to, in determining whether the first signal is a valid direct sequence spread spectrum
signal, determine whether the correlated data includes valid information by determining a distance between consecutive peaks
of the correlated data;

an assessment device configured to
determine whether the demodulated data includes a predetermined header, wherein the predetermined header includes a predetermined
sequence,

determine whether the channel is busy based on (i) the second signal, (ii) the third signal, and (iii) whether the demodulated
data includes the predetermined header with the predetermined sequence, and

generate a channel signal indicating whether the channel is busy; and
a transmitter configured to, based on the channel signal, transmit a fourth signal on the channel.

US Pat. No. 9,596,649

POWER SAVING TECHNIQUE FOR A WIRELESS DEVICE

MARVELL INTERNATIONAL LTD...

1. A method for controlling a data transmitter to transmit data from a buffer, the method comprising:
calculating an activation time interval and deactivation time interval for activating and deactivating the data transmitter,
respectively, based on (i) a known fixed data rate at which the data is received from a data source, (ii) a transmit rate
at which the data is transmitted, and (iii) a capacity of a buffer;

after calculating the activation time interval and deactivation time interval and during the deactivation time interval, (i)
deactivating, by a processor, the data transmitter; (ii) receiving, at the processor, the data from the data source at the
known fixed data rate and (iii) storing the data to the buffer, wherein the activation time interval and deactivation time
interval are alternating, periodic time intervals;

after calculating the activation time interval and deactivation time interval and during the activation time interval, activating
the data transmitter to thereby transmit, by the activated data transmitter, the data from the buffer to a data receiver;

during the activation time interval and before a next deactivation time interval, receiving a message from the data receiver
to stop transmitting the data;

deactivating the data transmitter during the activation interval and before the next deactivation time interval based on the
message; and

adjusting the activation time interval and deactivation time interval for activating and deactivating the data transmitter
based on the message.

US Pat. No. 9,485,728

POWER SAVING WITH WNM-SLEEP MODE

Marvell International Ltd...

1. A method in a communication device, the method comprising:
receiving from a station a request to enter a power save mode wherein the request to enter the power save mode includes one
or more traffic parameters for filtering data intended for the station;

in response to receiving the request to enter the power save mode, enabling a set of services, wherein the set of services
includes (i) buffering of data intended for the station and (ii) filtering of data intended for the station, according to
the one or more traffic parameters, to determine which data intended for the station should be buffered and which data destined
for the station should be discarded;

after enabling the set of services, receiving from the station an indicator that the station is awake;
in response to receiving the indicator that the station is awake, disabling the set of services;
after disabling the set of services, receiving from the station a communication frame having an indicator that the station
is no longer awake, wherein the communication frame does not include the one or more traffic parameters for filtering data
intended for the station;

in response to receiving the indicator that the station is no longer awake, again enabling the set of services;
after again enabling the set of services, receiving from the station a request to exit the power save mode; and
in response to receiving the request to exit the power save mode, disabling the set of services.

US Pat. No. 9,419,587

METHOD AND APPARATUS TO RECONFIGURE A FILTER

Marvell International Ltd...

1. A circuit, comprising:
a filter circuit comprising a transconductance amplifier having a first output circuit coupled with a first resistor and a
second output circuit coupled with a second resistor, the filter circuit driving a load circuit having different input impedances
under different operation conditions of the load circuit; and

a controller coupled to the load circuit and configured to generate control signals to selectively couple one of the first
output circuit and the second output circuit to the load circuit based on an operation condition of the load circuit to provide
a constant gain for the filter circuit.

US Pat. No. 9,324,621

PROVIDING SHALLOW TRENCH ISOLATION STRUCTURES THROUGH A BACKSIDE OF A METAL-OXIDE SEMICONDUCTOR DEVICE

Marvell International Ltd...

1. A method of making a metal-oxide semiconductor (MOS) device, the method comprising:
providing an apparatus comprising
a common source and drain well disposed within a substrate, and
a gate disposed on the substrate, wherein the gate is substantially encapsulated within layers of the apparatus;
removing a portion of the substrate;
creating a shallow trench isolation (STI) structure through the substrate such that the STI structure partially penetrates
the common source and drain well but does not extend all the way through the common source and drain well;

prior to removing the portion of the substrate, coupling the apparatus to a carrier wafer by coupling a surface of the layers
to the carrier wafer; and

subsequent to creating the STI structure, removing the carrier wafer.

US Pat. No. 9,294,397

APPARATUS AND METHOD FOR FORWARDING PACKETS BASED ON APPROVED ASSOCIATIONS BETWEEN PORTS AND ADDRESSES OF RECEIVED PACKETS

Marvell Israel (M.I.S.L) ...

1. A switch comprising:
a plurality of ports comprising a first port and a second port;
a processor configured to approve associations between (i) the plurality of ports and (ii) addresses of packets;
a memory configured to store entries, wherein each of the entries has a respective indicator indicating approval or disapproval
of an association between (i) one of the plurality of ports and (ii) one of the addresses of the packets,

wherein the first port is configured to receive a first packet sent to the switch; and
a controller configured to
determine whether one of the entries includes an address of the first packet,
if none of the entries includes the address of the first packet, send a first message to the processor requesting approval
of the first packet,

if one of the entries includes the address of the first packet, (i) determine whether the respective indicator of the one
of the entries indicates approval, and (ii) based on whether the respective indicator of the one of the entries indicates
approval, send the first message to the processor,

receive a response from the processor based on the first message, and
based on the response from the processor, forward the first packet to the second port.

US Pat. No. 9,159,355

DVD-RAM HEADER OR LAND/GROOVE DETECTION

Marvell International Ltd...

1. A method comprising:
adjusting, based on a previously-integrated portion of a wobble signal, a minimum or a maximum signal threshold configured
to enable integration of subsequently received portions of the wobble signal;

receiving, subsequent to adjusting the minimum or the maximum signal threshold, another portion of the wobble signal from
an optical head;

integrating, when the adjusted minimum or the adjusted maximum signal threshold is exceeded, the other portion of the wobble
signal to produce another integrated portion of the wobble signal;

filtering, via a low-pass filter, the other integrated portion of the wobble signal to determine an offset for the other integrated
portion of the wobble signal;

removing, subsequent to integrating the other portion of the wobble signal, the offset from the other integrated portion of
the wobble signal to produce a non-offset integrated portion of the wobble signal;

detecting a header within the non-offset integrated portion of the wobble signal based on the non-offset integrated portion
of the wobble signal and a header-detection threshold; and

determining that data in a sector associated with the header is stored within a land track or groove track based on a direction
of a transition of the header.

US Pat. No. 9,148,317

METHOD AND APPARATUS FOR CORRECTING A MISMATCH BETWEEN AN IN-PHASE COMPONENT AND A QUADRATURE COMPONENT OF A SIGNAL

Marvell International Ltd...

1. A method comprising:
receiving an input signal, wherein the input signal has (i) an in-phase component and (ii) a quadrature component;
determining whether a mismatch exists between the in-phase component and the quadrature component of the input signal;
estimating a symbol of the input signal;
encoding the estimated symbol of the input signal to generate an encoded signal; and
subsequent to encoding the estimated symbol of the input signal to generate the encoded signal and responsive to a mismatch
existing between the in-phase component and the quadrature component of the input signal, processing the encoded signal to
reduce the mismatch between the in-phase component and the quadrature component, wherein processing the encoded signal to
reduce the mismatch between the in-phase component and the quadrature component comprises:

multiplying the encoded signal by a correction matrix to reduce the mismatch between the in-phase component and the quadrature
component.

US Pat. No. 9,128,634

SYSTEMS AND METHODS OF PACKED COMMAND MANAGEMENT FOR NON-VOLATILE STORAGE DEVICES

Marvell International Ltd...

1. A system comprising:
a host memory to hold a scatter/gather list containing in respective portions two or more interrupt bits, two or more done
bits, and pointers to memory blocks associated with a packed command specifying two or more commands;

a host controller coupled with the host memory to access the pointers to access the memory blocks associated with the packed
command; and

a storage device coupled with the host controller to send or receive data associated with the packed command provided by the
host controller;

wherein the host controller is configured to check each respective one of the two or more interrupt bits in the host memory
that correspond to the two or more commands specified by the packed command and assert at least one interrupt to host software,
for at least one command of the two or more commands specified by the packed command in response to a value of at least one
of the two or more interrupt bits, after data transfer resulting from processing the at least one command is completed, but
before data transfer resulting from processing all of the commands of the packed command is completed;

wherein the host software sets the value of the at least one of the two or more interrupt bits in the scatter/gather list
for the at least one command to ask the host controller to assert the interrupt; and

wherein the host controller is configured to set one of the two or more done bits in the scatter/gather list for the at least
one command before the interrupt is asserted.

US Pat. No. 9,112,698

CRYPTOGRAPHIC DEVICE AND METHOD FOR DATA ENCRYPTION WITH PER-ROUND COMBINED OPERATIONS

Marvell International Ltd...

1. A cryptographic device comprising:
a key addition module configured to generate an input block based on (i) a cipher key and (ii) a plaintext block;
a first module configured to generate an output block by performing a plurality of rounds of processing on the input block;
and

a key module configured to, for each of the rounds, provide a round key based on the cipher key,
wherein the first module comprises
an inversion module configured to, for each of the rounds, perform a matrix inversion operation on first intermediate data
to generate second intermediate data, wherein, in a first round of the rounds, the first intermediate data is set equal to
the input block, and

a combined operation module configured to, for each of the rounds, update the first intermediate data by performing an affine
transformation operation and a mix columns operation on the second intermediate data.

US Pat. No. 9,100,325

CONTROLLING A NETWORK CONNECTION STATUS INDICATOR

Marvell International Ltd...

1. A method comprising:
receiving a data unit from a network;
determining whether the received data unit is a protocol control unit;
disabling activity of a status indicator if the received data unit is determined to be the protocol control unit; and
allowing activity of the status indicator if the received data unit is determined to be data other than the protocol control
unit.

US Pat. No. 9,596,615

METHOD AND APPARATUS FOR ASSESSING A CONDITION OF A WIRELESS CHANNEL IN VIEW OF FEEDBACK UNDER STRONG AND SPORADIC INTERFERENCE RECEIVED OVER THE WIRELESS CHANNEL

Marvell International Ltd...

1. A wireless communication device, comprising:
an interference module configured to (i) detect interference on a communication channel, (ii) identify which of a first type
of interference and a second type of interference the detected interference corresponds to, and (iii) provide information
about which of the first type of interference and the second type of interference the detected interference corresponds to,
wherein the second type of interference occurs less frequently than the first type of interference; and

a feedback generation module configured to (i) receive a request, from a transmitter, for feedback about a condition of the
communication channel, and (ii) based on the request and the information about which of the first type of interference and
the second type of interference the detected interference corresponds to,

(a) determine a measurement region of the communication channel,
(b) generate the feedback about the condition of the communication channel using measurements taken, by a receiver, in the
measurement region as determined by the feedback generation module, and

(c) provide the feedback to the transmitter.

US Pat. No. 9,531,360

HIGH SPEED COMPARATOR

MARVELL INTERNATIONAL LTD...

1. An apparatus comprising:
an input difference determination circuit configured to determine a digital value based on a difference between first and
second input signals, the input difference determination circuit including first, second, third, and fourth transistors, the
first transistor and the second transistor cross-coupled to each other and the third and fourth transistors coupled to a first
supply voltage, wherein:

a gate of the third transistor is coupled to a gate of the first transistor,
a gate of the fourth transistor is coupled to a gate of the second transistor,
a drain of the fourth transistor is coupled to a first output node, and
a drain of the third transistor is coupled to a second output node; and
a charge pump configured to provide a second supply voltage to the first and second transistors to determine the digital value;
and

a reset circuit configured to receive a reset signal, to reset the first and second output nodes when the reset signal has
a first value, and to provide the first supply voltage to the charge pump when the reset signal has the first value,

wherein the first and second transistors are configured to have transconductance values that are greater than those of the
third and fourth transistors, respectively.

US Pat. No. 9,430,339

METHOD AND APPARATUS FOR USING WEAR-OUT BLOCKS IN NONVOLATILE MEMORY

Marvell International Ltd...

1. A method comprising:
designating a portion of a nonvolatile memory as a wear-out portion of the nonvolatile memory, in response to the portion
of the nonvolatile memory being error-prone beyond a particular threshold;

storing volatile data in the wear-out portion of the nonvolatile memory;
over a time interval, periodically refreshing the volatile data in the wear-out portion of the nonvolatile memory by (i) reading
the volatile data from the wear-out portion of the nonvolatile memory and (ii) rewriting the volatile data to the wear-out
portion of the nonvolatile memory;

determining a number of errors in the volatile data that is read during the periodic refresh of the volatile data; and
based, at least in part, on the number of errors that is determined, modifying the time interval.

US Pat. No. 9,275,731

SYSTEMS AND METHODS FOR INCREASING THE READ SENSITIVITY OF A RESISTIVE RANDOM ACCESS MEMORY (RRAM)

Marvell International Ltd...

1. A resistive random access memory system, comprising:
a plurality of bitlines comprising first bitlines and a second bitline;
a plurality of wordlines; and
an array of resistive random access memory cells,
wherein
each of the resistive random access memory cells in the array includes a transistor and a resistive random access memory element
connected in a common gate configuration,

each of the transistors includes a first terminal and a second terminal,
a drain-source voltage of each of the transistors is greater than a difference between (i) a gate-source voltage of the corresponding
transistor and (ii) a threshold voltage of a gate of the corresponding transistor,

a drain current of each of the transistors is equal to a square of a function of a difference between (i) the gate-source
voltage of the corresponding transistor and (ii) the threshold voltage of the gate of the corresponding transistor,

the first terminals of the transistors are output terminals of the array of resistive random access memory cells,
each of the first terminals is connected to a corresponding one of the first bitlines,
the second terminals are connected respectively to the resistive random access memory elements, and
the resistive random access memory elements are connected to the second bitline.

US Pat. No. 9,160,368

SYSTEMS AND METHODS FOR NEAR-CODEWORD DETECTION AND CORRECTION ON THE FLY

Marvell International Ltd...

1. A method for correcting near-codewords, the method comprising:
receiving a near-codeword;
determining an error pattern associated with the near-codeword;
comparing the determined error pattern to a known error pattern to determine whether a shift relationship exists between the
determined error pattern and the known error pattern; and

adjusting a value of a symbol in a symbol location in the near-codeword, wherein the symbol location is determined based on
the shift relationship between the determined error pattern and the known error pattern when the shift relationship exists.

US Pat. No. 9,106,456

ADAPTIVE LOW-COMPLEXITY CHANNEL ESTIMATION

MARVELL INTERNATIONAL LTD...

1. A device comprising:
a controller configured to control a switch to select between at least two processing paths when performing channel estimation
processing of an orthogonal frequency-division multiplexed (OFDM) signal, wherein the at least two processing paths include:

a first processing path configured to process the OFDM signal according to a first order of processing operations; and
a second processing path configured to process the OFDM signal according to a second order of processing operations, wherein
the first order of processing operations is different from the second order of processing operations;

wherein the controller is further configured to:
select the first processing path if a frequency correlation for the OFDM signal is within a predetermined frequency threshold;
and

select the second processing path if i) a time correlation for the OFDM signal is within a predetermined time threshold, and
ii) the frequency correlation is not within the predetermined frequency threshold.

US Pat. No. 9,088,483

PACKET IDENTIFICATION TRACKER

MARVELL INTERNATIONAL LTD...

13. A device, comprising:
a receiver configured to receive data from a data source, wherein the data is transmitted in packets;
a decoder configured to decode a packet identifier from a received packet to produce a decoded packet identifier, wherein
the decoder performs error correction if an error exists in a received packet identifier; and

a packet identification tracker configured to:
estimate a reliability of the decoded packet identifier;
determine a packet identifier of the received packet based, at least in part, on the estimated reliability of the decoded
packet identifier; and

maintain a first counter and a second counter, wherein, when a new decoded packet identifier is received, i) the first counter
stores a value corresponding to a most recent packet identifier that was deemed reliable incremented by an increment value
and ii) the second counter stores a value corresponding to a most recent packet identifier that was deemed to be either reliable
or correctable incremented by an increment value.

US Pat. No. 9,064,135

HARDWARE IMPLEMENTED KEY MANAGEMENT SYSTEM AND METHOD

Marvell International Ltd...

1. A method of managing keys by a controller implemented in a single integrated circuit package, said method comprising:
controlling, by the controller, data transfer between a host system and a device, wherein the controller is configured as
a host bus adaptor and implemented as an integrated circuit;

connecting the controller to a communication bus to the host system via a first port, and connecting the controller to the
device via a second port;

receiving, by the integrated circuit of the controller, a data encryption key in encrypted form, the data encryption key associated
with a command directed to the device connected to the controller;

employing, by the integrated circuit, a key encryption key to unwrap the data encryption key;
storing, by the integrated circuit of the controller, the unwrapped data encryption key in a protected memory, wherein the
protected memory is connected to the integrated circuit; and

selectively controlling access to the protected memory, wherein the integrated circuit protects the protected memory from
being accessed by the host system.

US Pat. No. 9,288,836

ELECTRONIC BRACELET

Marvell International LTD...

1. A bracelet configured to assist a second device in performing a near-field communication (NFC) transaction with a near-field
communication (NFC) device, the bracelet comprising:
a near-field communication (NFC) circuit configured to, in response to the bracelet being placed proximate to the near-field
communication (NFC) device, exchange data related to the near-field communication (NFC) transaction with the near-field communication
(NFC) device;

a wireless communication circuit configured to wirelessly communicate the data related to the near-field communication (NFC)
transaction to the second device;

a power supply circuit configured to supply power to one or more of the near-field communication (NFC) circuit and the wireless
communication circuit,

wherein the near-field communication (NFC) circuit is configured to transmit, to the near-field communication (NFC) device,
authentication information to authenticate the near-field communication (NFC) transaction, wherein the authentication information
includes unique information included in a subscriber identity circuit of the second device, and

wherein the bracelet is configured to (i) operate as a companion device of the second device, and (ii) use the unique information
included in the subscriber identity circuit of the second device to communicate with the near-field communication (NFC) device;

a plurality of display devices arranged along an outer surface of the bracelet; and
a display circuit configured to display data on the plurality of display devices, wherein each of the plurality of display
devices comprises a capacitive sensor configured to generate an output in response to touch;

wherein the display circuit is configured to display, on the plurality of display devices, i) a first set of characters at
a first time, and ii) a second set of characters at a second time;

wherein the first set of characters includes at least one of letters, numerals, symbols, shapes, and colors;
wherein the second set of characters includes a rearrangement of the first set of characters; and
wherein the second time corresponds to one or more of the capacitive sensors on the plurality of display devices having sensed
touch within a predetermined time.

US Pat. No. 9,252,999

METHOD AND APPARATUS FOR PERFORMING CHANNEL ESTIMATION AND EQUALIZATION BASED ON A SIGNAL FIELD OF A RECEIVED PACKET

Marvell International Ltd...

1. A network device comprising:
a receiver module configured to receive a packet wirelessly transmitted over a channel to the network device, wherein the
packet comprises a signal field, and wherein the signal field (i) comprises a plurality of bits, or (ii) has a corresponding
plurality of sub-carriers;

a first channel estimate module configured to, (i) based on the signal field, perform a plurality of determinations, wherein
each of the plurality of determinations indicates whether a respective one of the plurality of bits or a respective one of
the plurality of sub-carriers has been modulated with a first predetermined value or a second predetermined value, and (ii)
based on the plurality of determinations, output a first channel estimate of the channel; and

a frequency equalizer configured to (i) receive a first signal comprising the packet, and (ii) equalize the first signal based
on the first channel estimate.

US Pat. No. 9,264,054

DLL LOCK DETECTOR

MARVELL INTERNATIONAL LTD...

1. An apparatus comprising:
a lock detect circuit configured to receive a phase detect signal and generate a lock signal according to the phase detect
signal,

wherein the lock detect circuit is configured to receive a clock signal and generate the lock signal according to a count
of a number of clock cycles of the clock signal since a most recent change in a value of the phase detect signal.

US Pat. No. 9,220,087

DYNAMIC POINT SELECTION WITH COMBINED PUCCH/PUSCH FEEDBACK

MARVELL INTERNATIONAL LTD...

1. A method for providing channel feedback within a system comprising (i) a mobile communication terminal and (ii) three or
more cells, the method comprising:
receiving, in the mobile communication terminal, signals from multiple cells of the three or more cells that coordinate transmission
of the signals with one another in a Dynamic Point Selection (DPS) Cooperative Multipoint (CoMP) scheme, wherein, in accordance
with the DPS CoMP scheme, a single cell of the multiple cells is selected to transmit to the mobile communication terminal
in a given time-frequency resource;

calculating in the terminal, based on the received signals, (i) a respective channel feedback for each of the multiple cells,
and (ii) a single rich channel feedback, wherein the single rich channel feedback comprises, per spectral sub-band, only a
single-bit indication that identifies the single cell selected to transmit to the mobile communication terminal, and wherein
the single-bit indication is sufficient to identify the multiple cells of the three or more cells that cooperate in the DPS
CoMP scheme; and

transmitting the channel feedback for the multiple cells on a periodic uplink control channel, and transmitting the single
rich channel feedback on an a-periodic uplink channel.