5G Massive MIMO Signal Detection Algorithm Based on Deep Learning

Abstract

Aiming at the problems of poor signal detection effect caused by many interference factors in large-scale MIMO technology scene, this paper proposes a 5G massive MIMO signal detection algorithm based on deep learning. Firstly, the MIMO system model based on neural network is constructed, and Deep Neural Network (DNN) detection is introduced into the receiver of the traditional MIMO system to obtain the information bits or codewords and channel state information transmitted by transmitters. Then, the end-to-end training method is adopted to make neural network learn the mapping relationship of information bits or codewords transmitted by system transceivers. Furthermore, DNN detector is improved based on Simplified Message Passing Detection (sMPD) algorithm, and the correction factor is updated continuously to optimize network parameters to realize the accurate detection and decoding of the MIMO system. Finally, the proposed algorithm is experimentally analyzed based on the TensorFlow deep learning framework. Experimental results show that when signal-to-noise ratio is 10 dB, the bit error rate and mean square error are lower than 0.005 and 0.1, respectively.

An approach to analyse the error performance of linear receivers for Multi-User Massive MIMO communication systems under Imperfect CSI

“…An interesting future work will be to consider developing the BER expressions for a 5G MIMO system that uses machine learning based detection. For example, the work in [57] employed 5G massive MIMO based detection to optimize the network parameters of a Deep Neural Network (DNN). The simplified Message Passing Detector (sMPD) was used to optimize the network parameters iteratively in order to reduce the computational complexity of the DNN.…”

“…Significant gains in terms of BER and MSE were obtained when the DNN detector was used in conjunction with the sMPD algorithm. However, the work in [57] did not consider the communication error rate performance that combines it with channel estimation, channel coding, and also assumed an ideal state channel environment with no interference. Moreover, closed form expressions for the BER when DNN and sMPD are used, have not been derived for realistic channel conditions in [57].…”

“…However, the work in [57] did not consider the communication error rate performance that combines it with channel estimation, channel coding, and also assumed an ideal state channel environment with no interference. Moreover, closed form expressions for the BER when DNN and sMPD are used, have not been derived for realistic channel conditions in [57]. It will therefore be a major step to modify the equations derived in this paper so that they can be used for predicting the bit error rate distributions of user terminals that employ DNN based detection with sMPD as described in [57].…”

“…Moreover, closed form expressions for the BER when DNN and sMPD are used, have not been derived for realistic channel conditions in [57]. It will therefore be a major step to modify the equations derived in this paper so that they can be used for predicting the bit error rate distributions of user terminals that employ DNN based detection with sMPD as described in [57].…”

Implementation of the deep learning method for signal detection in massive-MIMO-NOMA systems

“…However, the PSD and complexity of the system are not analyzed and discussed. Using channel estimation as the detector [ 14 ], provided a cooperative MIMO channel approximation and signal recognition. The channel estimate is enhanced by the found data and incorporates detection errors, while also accounting for channel statistics.…”