A radio receiver includes a channel estimator processing circuit including: a feature extractor configured to extract one or more features from a received signal, the features including a channel correlation estimated based on a reference signal in a current slot, the estimated channel correlation indicating a rate of change of a wireless channel over time; and a Doppler spread estimator configured to estimate a Doppler spread of the wireless channel by supplying the features to one or more Doppler shift predictors trained on training data across a training signal-to-noise ratio (SNR) range and across a training Doppler shift range, each Doppler shift predictor being trained on a portion of the training data corresponding to a different portion of the training data.

A method performed by a communication device includes generating an initial channel estimate of a channel for a current time step with a Kalman filter based on a first signal received at the communication device. The method also includes inferring, with a neural network, a residual of the initial channel estimate of the current time step. The method further includes updating the initial channel estimate of the current time step based on the residual.

To provide demapping at a receiving side, a trained model for a demapper is used to output log-likelihood ratios of received signals representing data in a multi-user transmission. Inputs for the trained model for the demapper comprise a resource grid of equalized received signals.

A method for reporting, by a terminal, a measurement result for location determination according to an embodiment of the present disclosure comprises the steps of: determining whether a channel characteristic between the terminal and each base station included in a plurality of base stations configured for location determination of the terminal corresponds to a visible ray (line of sight: LoS); calculating a reference signal time difference (RSTD) by configuring, as a reference cell, one of the base stations, the channel characteristics of which correspond to a visible ray (LoS); and reporting a measurement result including the RSTD.

Provided are a method and device for communication. The method comprises: a network device transmits first information to a terminal device, the first information being used for indicating whether the network device acquires downlink channel information via an artificial intelligence algorithm model, and the artificial intelligence algorithm model being constructed by training with past uplink channel information and past downlink channel information serving as samples. A terminal device measures a channel state on the basis of the first information to acquire channel state information.

A channel information processing method and apparatus. The method includes: receiving first information and second information from a terminal, where the first information includes a first parameter without uplink/downlink channel reciprocity, the first parameter is determined based on downlink channel estimation, the second information is used to indicate a deviation between a second parameter and a third parameter that have uplink/downlink channel reciprocity, the second parameter is determined based on uplink channel estimation, and the third parameter is determined based on downlink channel estimation; and determining channel information of a downlink channel based on the first information, the second information, and the second parameter.

A machine learning (ML) agent operates at a transmitter to optimize signals transmitted across a communications channel. A physical signal modifier modifies a physical layer signal prior to transmission as a function of a set of signal modification parameters to produce a modified physical layer signal. The ML agent parses a feedback signal from a receiver across the communications channel, and determines a present tuning status as a function of the signal modification parameters and the feedback signal. The ML agent generates subsequent signal modification parameters based on the present tuning status and a set of stored tuning statuses, thereby updating the physical signal modifier to generate a subsequent modified physical layer signal to be transmitted across the communications channel.

Examples described herein include methods, devices, and systems which compensates input data for I/Q imbalance or noise related thereto to generate compensated input data. In doing such the above compensation, during an uplink transmission time interval (TTI), a switch path is activated to provide converted input data to a receiver stage including a recurrent neural network (RNN). The RNN calculates an error representative of the noise based partly on the input signal to be transmitted and a feedback signal to generate filter coefficient data associated with the I/Q imbalance. The feedback signal is provided, after processing through the receiver, to the RNN. During an uplink TTI, the converted input data is transmitted as the RF wireless transmission via an RF antenna. During a downlink TTI, the switch path is deactivated and the receiver stage receives an additional RF wireless transmission to be processed in the receiver stage.

Method, comprising receiving a terminal location information or a location-like information from a terminal; selecting one or more first pairs of prior channel information among one or more stored first pairs of prior channel information based on the terminal location information or the location-like information, respectively; inputting the terminal location information or the location-like information, respectively, and the selected one or more first pairs of prior channel information into a trained interpolation neural network to obtain a first estimation of a channel between the terminal and a base station as an output from the interpolation neural network; providing the weights of the trained neural network to the terminal; wherein each of the one or more first pairs of prior channel information comprises a location information related to a respective prior channel and a first representation of the respective prior channel.

A system, apparatus, and method are provided for performing fast re-training of fully fused neural networks configured to implement at least a portion of a transceiver. At least one of a demapping module, an equalization module, or a channel estimation module can be implemented, at least in part, using a fully fused neural network. The neural network can be trained online during operation by acquiring training data sets using a number of received frames of data. Re-training of the neural network is performed periodically to adapt the neural network to changing channel characteristics. In various embodiments, a neural demapper, a neural channel estimator, and a neural receiver are disclosed to replace or augment one or more components of the transceiver. In another embodiment, an auto-encoder can be implemented across a transmitter and receiver to replace most of the components of the transceiver, the auto-encoder being trained via an end-to-end learning algorithm.