Samsung Electronics Co., Ltd.

Ziming He

Fei Tong

Jacob Christopher Sharpe

A communication device, including a receiver configured to receive a signal comprising a plurality of training symbols and a data symbol; and at least one processor configured to: select, from among the plurality of training symbols, a first training symbol and a second training symbol based on a first parameter and a second parameter; obtain a first symbol difference corresponding to the first training symbol, and a second symbol difference corresponding to the second training symbol; estimate a carrier frequency offset based on the first symbol difference and the second symbol difference; obtain a compensated data symbol corresponding to the data symbol based on the estimated carrier frequency offset; and demodulate and decode the compensated data symbol to obtain decoded data.

Low complexity carrier frequency offset estimation algorithm using known symbols

A method for detecting a modification, due to an interferer, of a wireless channel between a transmitting device and a receiving device. First and second samples of channel frequency response (CFR) of the wireless channel are measured within a time interval less than or equal to a pre-established static time interval throughout which the wireless channel is expected to exhibit a static characteristic in an environment without any interferer. An interferer is detected to have modified the wireless channel when a similarity condition reflecting a degree of similarity between the first and second CFR samples is not satisfied.

Detection of spoofing or jamming attacks in wireless communication system

One or more aspects of the techniques and models described herein provide for bidirectional recurrent neural network (BiRNN)-based digital pre-distortion techniques for radio frequency (RF) power amplifiers (PAs). As an example, a digital pre-distorter (DPD) system may implement residual learning and long short-term memory (LSTM) projection layer features to reduce computational complexity and memory requirements. Implementing the described unconventional techniques of applying residual learning in RNN (e.g., in BiLSTM), using LSTM projection to develop a DPD structure, or both, may provide several advantages over preexisting techniques. For instance, the complexity in training and pre-distortion may be reduced and significantly less memory may be required to store the DPD neural network coefficients (e.g., while achieving similar or better linearization performance compared to other LSTM models). Further, faster training convergence speed may be achieved (e.g., compared to other LSTM models).

Residual neural network models for digital pre-distortion of radio frequency power amplifiers

The present disclosure describes systems and methods for time-of-arrival (TOA) estimation techniques. Some embodiments of the disclosure provide for estimating radio propagation path parameters based on a training signal received over a set of active frequencies. The radio propagation path parameters (e.g., fading coefficients for each path) are used to reconstruct a channel frequency response on null frequencies (e.g., frequencies that did not include or carry the received training signal). A time-of-arrival parameter can then be estimated based on the estimated channel frequency response and the reconstructed channel frequency response (e.g., the channel frequency response estimated using both active frequencies and null frequencies).

Growing community of inventors

Cambridge, United Kingdom

Ziming He