A user activity and multi-user joint detection method

Abstract

The invention discloses a user activity and multi-user joint detection method. According to the Bayesian theorem, the matrix formed by the signals received by the base station on all sub-carriers of all time slots is obtained under the condition that all users at all times are known. The probability of the matrix formed by the symbols transmitted on the slot is introduced into the factor graph, and the factor graph model is obtained according to the relationship between the variables and the factors; on the basis of the factor graph model, the user activity and multi-user are jointly detected. During the joint detection process Approximate message passing algorithm and variational Bayesian inference algorithm are adopted; the advantage is that it can perform user activity and multi-user joint detection without knowing user sparsity; it combines approximate message passing algorithm and variational Bayesian inference algorithm The advantage of using Bayesian thinking is to introduce structured prior knowledge in the form of prior probability, which improves the performance accuracy of joint detection.

Description

technical field [0001] The invention relates to a joint detection technology, in particular to a user activity and multi-user joint detection method applied in an uplink scheduling-free non-orthogonal multiple access system. Background technique [0002] With the development of mobile communication, multiple access technology, as the key technology of each generation of mobile communication system, has undoubtedly become one of the research hotspots of many scholars. With the increase in the number of users, in order to achieve the goal of being able to communicate in any form at any time, any place, mobile communication has undergone technological innovations again and again. To sum up, from the first generation mobile communication system to the fourth generation mobile communication system all use orthogonal multiple access technology. However, on the one hand, theoretical studies have shown that the orthogonal multiple access technology cannot always achieve the maximum...

AI Technical Summary

Problems solved by technology

Existing user activity and multi-user joint detection methods, such as based on the Orthogonal Matching Pursuit (OMP) algorithm, propose an iterative order recursive least squares algorithm (IORLS) based on prior knowledge of user sparsity, which requires In fact, user sparsity should usually be unknown; then, a structured iterative support detection (SISD) algorithm for user activity and multi-user joint detection using frame joint sparsity is proposed. Exhibits better performance than the Iterative Support Detection (ISD) algorithm with separate sparse multiuser detection, and does not require the sparsity of the users to be known, however, the SISD algorithm does not take advantage of the prior information of the discrete symbols sent, making Combined test results are not accurate enough

Images