PC-SCMA joint iterative detection decoding method based on deep learning

Abstract

The invention belongs to the technical field of computers and communication, and particularly relates to a joint iterative detection decoding method based on deep learning. The method comprises the following steps: adding a learnable weight factor in a transmission path between a resource node and a user node in a factor graph of a message passing algorithm of the SCMA; adding a learnable offset which can be learned off line in information iteration in the factor graph of the confidence propagation algorithm; forming a joint factor graph; using a learnable weight factor and a learnable offset in the joint factor graph as hidden layer parameters of the deep neural network, and inputting a scrambled information bit value to perform iterative updating in the deep neural network; iteratively updating the resource node information, the prior node information and the user node information in sequence; after updating the resource node information, the priori information of the user node and the user node information, outputting an estimated transmission symbol; and according to the method, the detection decoding BER performance can be effectively improved when the operation complexity is increased within an acceptable range.

Description

technical field [0001] The present invention is a deep learning (Deep Learning, DL) based PC-SCMA (Polar Coded-Sparse Code Multiple Access) joint iterative detection and decoding (Joint Iterative Detection and Decoding Algorithm, JIDD) method. It has the characteristics of high code and detection accuracy, and can be applied to scenarios with high spectrum utilization and mass user access in the 5th generation mobile communication system, and belongs to the field of computer and communication technology. Background technique [0002] With the continuous development of mobile communications, diversified application scenarios have higher and higher performance requirements for communication technologies. The 5G mobile communication network is expected to achieve high spectral efficiency, which puts forward a demand for the improvement of channel coding and decoding technology and multiple access technology. Channel coding and decoding technology is directly related to the perf...

AI Technical Summary

Problems solved by technology

[0004] However, with the increase of the polar code length and the number of iterations of the decoding algorithm, a large number of learnable parameters introduced after the combination of deep learning technology increase a large number of multiplication operations and storage space occupation, so how to improve the decoding BER performance while Not increasing the computational complexity too much has become a current hot research direction

On the other hand, the SCMA system multi-user detection algorithm combined with deep learning also has problems such as high computational complexity or unsatisfactory detection performance that need to be solved urgently

Images