During a training procedure for communicating via a full-duplex communication link, a first communication device receives training information from a second communication device. The training information corresponds to first signal processing parameters developed at the second communication device for use by the second communication device to process signals received by the second communication device via the full-duplex communication link. After receiving the training information from the second communication device, the first communication device develops second signal processing parameters to be used by the first communication device to process signals received by the first communication device via the full-duplex communication link. The second signal processing parameters are developed using the training information received from the second communication device.

During a training procedure for communicating via a full-duplex communication link, a first communication device receives training information from a second communication device. The training information corresponds to first signal processing parameters developed at the second communication device for use by the second communication device to process signals received by the second communication device via the full-duplex communication link. After receiving the training information from the second communication device, the first communication device develops second signal processing parameters to be used by the first communication device to process signals received by the first communication device via the full-duplex communication link. The second signal processing parameters are developed using the training information received from the second communication device.

Various embodiments include methods performed in receiver circuitry of a wireless communication device for demodulating wireless transmission waveforms to reconstruct data tones, which may include receiving, from a transmitter, wireless transmission waveforms that includes peak reduction tones (PRTs) that were inserted by a PRT neural network in the transmitter, and demodulating the received wireless transmission waveforms using a decoder neural network that has been trained based on outputs of the transmitter to output a reconstruction of the data tones. Further embodiments include exchanging information between the transmitter and receiver circuitry to coordinate the PRT neural network used for inserting PRTs in the transmitting wireless communication device and the decoder neural network used in the receiving wireless communication device for demodulating transmission waveforms received from the transmitting wireless communication device.

Certain aspects of the present disclosure provide techniques for generating and decoding orthogonal frequency division (OFDM) waveforms with peak reduction tones (PRTs) designed to reduce PAPR. By generating PRT tones with a machine learning (e.g., neural network) based encoder and mapping some of the PRT tones to subcarriers used for physical channels or signals, PAPR may be reduced while efficiently using system resources.

Facilitating echo cancellation within communication networks is contemplated, such as but not necessarily limited to facilitating echo cancellation within full-duplex (FDX) communication networks. The echo cancellation may optionally be performed with an echo canceller included as part of or otherwise associated with an FDX node used to facilitate interfacing signaling between a digital domain and an analog domain of a FDX or other communication network.

In some embodiments, a method receives a first signal that is sent in a first direction in a network. Communications in the network are full duplex communications in a same frequency band. The first signal is amplified in the first direction. The method trains a first echo canceller to cancel a first echo signal from the first signal where the first echo signal is received in a second direction. After training the first echo canceller, the trained first echo canceller is enabled. The method receives a second signal in the second direction that is sent in the second direction in the network. The second signal is amplified in the second direction. The method trains a second echo canceller to cancel a second echo signal received in the first direction from the second signal where the first echo canceller cancels the first echo signal that is received in the second direction.

Echo cancellation to alleviate timing varying channels may be provided. First, a feedback signal corresponding to one of a plurality of downstream paths and a combination upstream signal comprising a combination of upstream signals from a plurality of upstream paths may be received. Next, a plurality of echo corrected signals may be created using the feedback signal, the combination upstream signal, and a plurality of echo cancelation coefficients that each respectively correspond to each one of the plurality of echo corrected signals and that are different from each other. Then a one of the plurality of echo cancelation coefficients that corresponds to a one of the plurality of echo corrected signals that provides a best echo cancelation performance as compared to other ones of the plurality of echo corrected signals may be selected to use.

A signal transmission method, a signal transmission apparatus and a display device are provided which belonging to the field of display technology. The signal transmission method is applied to a signal sender, the signal sender includes a first interface, a point-to-point connection is established between the first interface and a second interface of a signal receiver, the signal transmission method includes: sending a non-periodic training signal to the second interface via the first interface in an interval of transmission of pixel data. EMI generated during data transmission is reduced by using a technical solution of the present disclosure.

Methods and devices for aggregating hardware loopback streams of a plurality of display devices in communication with a computer device may include a plurality of hardware loopback streams with rendered audio data from the plurality of display devices in communication with the computer device. The methods and devices may include combining the rendered audio data from the plurality of hardware loopback streams into a loopback buffer to create aggregated loopback audio data. The methods and devices may include providing the loopback buffer with the aggregated loopback audio data to one or more applications executing on the computer device.

Proactive Echo Cancellation (EC) training may be provided. First, a plurality of Echo Cancellation Training Opportunities (ECTOs) may be identified in an upstream bandwidth allocation. Identifying the ECTOs may comprise identifying a corresponding plurality of mini-slots in a two dimensional time frequency space designated as not to be used for Upstream (US) traffic. Then Echo Cancellation Training (ECT) may be conducted for each of the plurality of ECTOs.