Large-scale multi-antenna channel estimation method based on deep convolutional neural network

Abstract

The invention discloses a large-scale multi-antenna channel estimation method based on a deep convolutional neural network, and belongs to the technical field of wireless communication. The method comprises: a transmitting antenna transmitting an OFDM data block to the receiving antenna, receiving a corresponding normalized pilot signal after wireless multipath channel propagation, and stacking the normalized pilot signal in rows to obtain a large-scale multi-antenna channel estimation model; constructing a deep convolutional neural network and carrying out weight training, and then estimatingthe stacked channel impact response to obtain an estimated stacked channel impact response; selecting sub-channel vectors corresponding to a transmitting antenna in the estimated stacked channel impact response to form an estimated sparse support set; and optimizing the estimation sparse support set corresponding to each transmitting antenna to obtain a joint estimation sparse support set, and further obtaining large-scale multi-antenna refined channel estimation. Under the condition of high noise intensity, the large-scale multi-antenna channel can be accurately estimated, the estimation precision of the large-scale multi-antenna channel is effectively improved, and the channel estimation time delay is effectively reduced.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to a large-scale multi-antenna channel estimation method based on a deep convolutional neural network. Background technique [0002] Since it can significantly improve the data transmission rate and spectral efficiency, massive multi-antenna technology has been widely used in 5G communication systems. However, with the increasing number of antennas, the difficulty of large-scale multi-antenna channel estimation is also increasing. At the same time, in low-latency communication scenarios, the complex time-varying characteristics of the channel are not conducive to real-time channel estimation using traditional complex methods or iterative channel estimation algorithms. Therefore, it is necessary to design a method that can accurately estimate large-scale multi-antenna channels with low estimation delay to meet the growing communication needs in the future....

AI Technical Summary

Problems solved by technology

The existing channel estimation method based on compressed sensing utilizes the sparse characteristics of the wireless channel, and uses compressed sensing to restore the channel, which can reduce the overhead of the required training sequence and improve the estimation accuracy to a certain extent, such as the Chinese patent application publication number CN104052691A A MIMO-OFDM system channel estimation method based on compressed sensing is proposed. Chinese patent application publication number CN105681232A proposes a MIMO channel estimation method based on shared channel and compressed sensing; however, in the case of strong background noise, lack of sampling data, and In severe and complex scenarios such as high sparsity, the accuracy of the channel estimation method based on compressed sensing is limited, and the high computational complexity of compressed sensing and a large number of iterative processes cause large delays, making it difficult to meet the needs of low-latency scenarios

Images