Abstract: One example discloses an OFDM wireless communications device, including: a memory configured to support processing of OFDM tones; a controller, coupled to the memory, and configured to set the wireless communication device to a first mode and a second mode; wherein the first mode is configured to transmit or receive a first wireless communication signal having a first set of OFDM tones contained within an OFDM channel bandwidth; wherein the second mode is configured to transmit or receive a second wireless communication signal having a second set of OFDM tones contained within the OFDM channel bandwidth; and wherein the memory used for processing the first set of OFDM tones is same as the memory used for processing the second set of OFDM tones.

Abstract: A first portion of a physical layer (PHY) preamble of a PHY data unit is generated for transmission via a communication channel that comprises a plurality of sub-channels. A first portion of the PHY preamble is generated to include a legacy portion and a plurality of first non-legacy signal fields spanning the respective sub-channels. A second portion of a PHY preamble that immediately follows the first portion of the PHY preamble of the PHY data unit is generated to include: a non-legacy short training field spanning all sub-channels in the plurality of sub-channels, a plurality of non-legacy long training fields immediately following the non-legacy short training field, each non-legacy training field spanning all sub-channels in the plurality of sub-channels, and a second non-legacy signal field immediately following the plurality of non-legacy long training fields, the second-legacy signal field spanning all sub-channels in the plurality of sub-channels.

Abstract: An adjustable high resolution timer (100) for synchronizing a local clock to an external reference clock includes frequency offset acquisition and compensation unit (110) configured to acquire a frequency offset difference between the local and external reference clock and to generate frequency adjustment signals based on the frequency offset difference; a time drift tracking and adjustment unit (120) configured to continuously monitor the local and external reference clocks for phase offset differences therebetween and to generate timing adjustment signals based on the phase offset difference; a nanosecond timer core unit (140) configured to generate a frequency and phase adjusted nanosecond timer output signal in response to the frequency adjustment signals and timing adjustment signals; and a pulse generation unit (130) for generating a plurality of output pulse signals that are synchronized with the external reference clock in response to the frequency and phase adjusted nanosecond timer output signal.

Abstract: A training device and a training method for a neural network model are provided. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.

Abstract: A preamble of a physical layer (PHY) data unit that conforms to a first communication protocol is generated. The preamble includes a legacy field that is formatted according to a second communication protocol, and a signal field having a first orthogonal frequency division multiplexing (OFDM) symbol and a second OFDM symbol. The first OFDM symbol (i) immediately follows the legacy field and (ii) is modulated using binary phase shift keying (BPSK) modulation, whereas a third communication protocol specifies that an OFDM symbol, defined by the third communication protocol, that immediately follows the legacy field is modulated using BPSK modulation rotated by 90 degrees (Q-BPSK). The second OFDM symbol (i) immediately follows the first OFDM symbol and (ii) is modulated using Q-BPSK to indicate to a receiver device that conforms to the first communication protocol that the data unit conforms to the first communication protocol.

Abstract: Flexible structured-pipelined CORDIC techniques efficiently perform various CORDIC operations and support different parameters for MIMO MEQ processing. The structured-pipelined CORDIC techniques simplify signal processing flow, unify input requirements and output delay, and simplify integration. Look-up table techniques provide quick generation of control signals, reduce design and verification efforts, and facilitate design automation. In addition, the structured-pipelined CORDIC techniques are conducive to hardware sharing and reuse. The structured-pipelined CORDIC techniques reduce integrated circuit area and power consumption.

Abstract: The present invention discloses a method for a source driver, for preventing image sticking in a display panel coupled to the source driver. The source driver includes a first output driver and a second output driver having different polarities. The first output driver is configured with a first driving capability and the second output driver is configured with a second driving capability. The method includes the steps of: obtaining an effective voltage of a pixel in the display panel; and adjusting the second driving capability of the second output driver to be identical to the first driving capability of the first output driver, to allow the adjusted second driving capability to drive the effective voltage to reach a level having the same magnitude as a level of the effective voltage driven by the first source driver with the same variation of display data.

Abstract: A fusion-based classifier, classification method, and classification system, wherein the classification method includes: generating a plurality of probability vectors according to input data, wherein each of the plurality of probability vectors includes a plurality of elements corresponding to a plurality of class respectively; selecting, from the plurality of probability vectors, a first probability vector having an extremum value corresponding to a first class-of-interest according to the first class-of-interest; and determining a class of the input data according to the first probability vector.

Abstract: An adjustable high resolution timer (100) for synchronizing a local clock to an external reference clock includes frequency offset acquisition and compensation unit (110) configured to acquire a frequency offset difference between the local and external reference clock and to generate frequency adjustment signals based on the frequency offset difference; a time drift tracking and adjustment unit (120) configured to continuously monitor the local and external reference clocks for phase offset differences therebetween and to generate timing adjustment signals based on the phase offset difference; a nanosecond timer core unit (140) configured to generate a frequency and phase adjusted nanosecond timer output signal in response to the frequency adjustment signals and timing adjustment signals; and a pulse generation unit (130) for generating a plurality of output pulse signals that are synchronized with the external reference clock in response to the frequency and phase adjusted nanosecond timer output signal.

Abstract: A neural network-based classification method, including: obtaining a neural network and a first classifier; inputting input data to the neural network to generate a feature map; cropping the feature map to generate a first cropped part and a second cropped part of the feature map; inputting the first cropped part to the first classifier to generate a first probability vector; inputting the second cropped part to a second classifier to generate a second probability vector, wherein weights of the first classifier are shared with the second classifier; and performing a probability fusion on the first probability vector and the second probability vector to generate an estimated probability vector for determining a class of the input data.

Abstract: A method for deinterleaving a plurality of resource units (RUs) where each RU includes data and parameters. The method includes assigning each RU into one of a first deinterleaver and a second deinterleaver and storing the parameters of each respective RU to a respective buffer in an order. The method also includes processing the data of the respective RU in the one of the first deinterleaver and the second deinterleaver and upon completion of the processing of the data of a respective RU, outputting the data of the respective RU from the one of the first deinterleaver and the second deinterleaver into which the respective RU was assigned, outputting the parameters of the respective RU corresponding to one of the first deinterleaver and the second deinterleaver into which the respective RU was assigned, and aligning the parameters of the respective RU with the data of the respective RU based on the order of storage of the parameters in the respective buffer.

Abstract: Flexible structured-pipelined CORDIC techniques efficiently perform various CORDIC operations and support different parameters for MIMO MEQ processing. The structured-pipelined CORDIC techniques simplify signal processing flow, unify input requirements and output delay, and simplify integration. Look-up table techniques provide quick generation of control signals, reduce design and verification efforts, and facilitate design automation. In addition, the structured-pipelined CORDIC techniques are conducive to hardware sharing and reuse. The structured-pipelined CORDIC techniques reduce integrated circuit area and power consumption.

Abstract: A communication device determines that a data rate of a data packet exceeds a threshold, and in response, lowers a processing speed of an equalizer device to prevent a buffer from overflowing. The buffer stores outputs generated by the equalizer device. A forward error correction code decoder device of the communication device processes outputs of the equalizer device corresponding to the data packet to generate decoded information corresponding to the data packet. The communication device transmits an acknowledgment packet to acknowledge the data packet such that transmission of the acknowledgment packet begins within required time period, defined by a communication protocol, after an end of the data packet.

Abstract: A radiofrequency system includes a plurality of transceivers providing a plurality of channels, and circuitry configured to perform off-channel functions. A first subset of channels performs radiofrequency communication functions, and a second subset of channels provides signals for the off-channel functions. The first subset of channels may be provided by a first subset of transceivers, and the second subset of channels may be provided by a second subset of transceivers. A spectrum sensing unit may select between the first subset of transceivers and the second subset of transceivers. The circuitry configured to perform the off-channel functions may use the second subset of the plurality of channels to perform neighborhood discovery.

Abstract: A first field of a preamble includes a plurality of indications of respective numbers of spatial or space-time streams for respective receivers to enable each receiver among the multiple receivers to determine a respective set of one or more training sequences, in a plurality of training sequences, that corresponds to the receiver. A second field of the preamble includes respective modulation and coding scheme information for the respective receivers. The preamble is generated such that i) the first field is transmitted prior to the plurality of training sequences, and ii) the second field of the preamble is transmitted after the plurality of training sequences are transmitted. A data portion of the multi-user data unit is generated using respective modulation and coding schemes for the respective receivers. Thee multi-user data unit is transmitted such that data for the respective receivers are transmitted via respective sets of one or more spatial or space-time streams.

Abstract: The present invention discloses a method for a source driver, for preventing image sticking in a display panel coupled to the source driver. The source driver includes a first output driver and a second output driver having different polarities. The first output driver is configured with a first driving capability and the second output driver is configured with a second driving capability. The method includes the steps of: obtaining an effective voltage of a pixel in the display panel; and adjusting the second driving capability of the second output driver to be identical to the first driving capability of the first output driver, to allow the adjusted second driving capability to drive the effective voltage to reach a level having the same magnitude as a level of the effective voltage driven by the first source driver with the same variation of display data.

Abstract: A method for deinterleaving a plurality of resource units (RUs) where each RU includes data and parameters. The method includes assigning each RU into one of a first deinterleaver and a second deinterleaver and storing the parameters of each respective RU to a respective buffer in an order. The method also includes processing the data of the respective RU in the one of the first deinterleaver and the second deinterleaver and upon completion of the processing of the data of a respective RU, outputting the data of the respective RU from the one of the first deinterleaver and the second deinterleaver into which the respective RU was assigned, outputting the parameters of the respective RU corresponding to one of the first deinterleaver and the second deinterleaver into which the respective RU was assigned, and aligning the parameters of the respective RU with the data of the respective RU based on the order of storage of the parameters in the respective buffer.

Abstract: A first field of a preamble includes a plurality of indications of respective numbers of spatial or space-time streams for respective receivers to enable each receiver among the multiple receivers to determine a respective set of one or more training sequences, in a plurality of training sequences, that corresponds to the receiver. A second field of the preamble includes respective modulation and coding scheme information for the respective receivers. The preamble is generated such that i) the first field is transmitted prior to the plurality of training sequences, and ii) the second field of the preamble is transmitted after the plurality of training sequences are transmitted. A data portion of the multi-user data unit is generated using respective modulation and coding schemes for the respective receivers. The multi-user data unit is transmitted such that data for the respective receivers are transmitted via respective sets of one or more spatial or space-time streams.

Abstract: Systems and methods are provided that securely authenticate a user of a web application. For example, the user may utilize a bot from within a first application, such as a chat application. The user may request the bot to access a second application (e.g., a social-networking application) that is remote from the first application. If the bot does not have authorization, the bot may redirect the user to a webpage for the second application, where the user may enter login credentials. Upon verification, the second application may provide an access token to a webpage associated with the bot. To authenticate the bot user, the bot webpage may generate and cache a nonce that is transmitted back to the first application, which then transmits it to the bot. The bot may then compare the received nonce with the cached nonce. If the nonces match, the user may be securely authenticated.

Abstract: Provided is a neural-network-based classification method, including: generating, by a neural network, one or more score vectors corresponding to one or more samples respectively; determining a first subset of the one or more samples according to the one or more score vectors and a first decision threshold, wherein the first subset is associated with a first class; and selecting samples to be re-examined from the one or more samples according to the first subset.