Signal detection method based on model-driven deep learning

Abstract

The invention relates to a signal detection method based on model-driven deep learning. According to the invention, a channel estimation and signal detection model is established based on an OFDM system. Channel estimation adopts a combined neural network model taking an MMSE estimator based on DFT and FC-DNN as sub-networks, pilot frequency data distributed in a self-adaptive mode are preprocessed through the MMSE estimator, DNN network initialization information is extracted, and a more accurate channel estimation model is obtained according to a training learning network ChannelEstNe. The SignalDetNet adopts a ZF equalization detection preprocessor, and the LSTM and the DNN form a combined network, so that final signal detection is realized, and an original signal is recovered. According to the structure, the mode that the OFDM system processes signals block by block is kept, sending data with linear and nonlinear distortion in the OFDM system can be recovered, the initialization training speed is higher by combining a traditional algorithm, and therefore the deployment efficiency is improved.

Description

technical field [0001] The invention belongs to the field of intelligent communication, and relates to a signal detection method based on model-driven deep learning, in particular to a signal detection method of an OFDM wireless communication receiver based on model-driven and deep learning. Background technique [0002] The OFDM receiver scheme mainly includes two functional modules of channel estimation and signal detection, that is, firstly obtain accurate channel state information (CSI) through channel estimation, and then use the estimated CSI to restore the transmitted signal. Most of the traditional channel estimation and signal detection technologies use complex algorithms to improve the receiving performance of the communication system. However, for the current 5G wireless communication that requires high dimensions, high speed, and high density, the high complexity calculation greatly affects the communication performance. effectiveness. Intelligent communication ...

AI Technical Summary

Problems solved by technology

[0004] The present invention solves the problem of low signal reliability caused by traditional channel estimation and signal detection algorithms in existing Orthogonal Frequency Division Multiplexing (OFDM) receivers, and proposes a method that uses ChannelEstNet and SignalDetNet neural network models to realize receiving signals Algorithms for Channel Estimation and Signal Detection

Images