An improved matched filter message passing detection method based on probability clipping in communication systems

Abstract

The present invention provides an improved matched filter message delivery detection method based on probability cut in a communication system. The present invention is applied to the symbol detection and judgment part of the receiver of the wireless communication system. design. The beneficial effects of the present invention are as follows: 1. The present invention utilizes matched filtering and real number operation so that the message passing detection algorithm can be applied to the communication symbol detection not limited to the mMIMO system under the condition of lower complexity. 2. After the symbol probability distribution of the matched filtering message passing detection algorithm is updated, the present invention introduces a new "probability clipping" operation, so that the Gaussian property required by the algorithm can be maintained in the algorithm iteration process, thereby greatly improving the algorithm Symbol detection performance at high SNR comes at the expense of additional sorting algorithm complexity in the iterative process and simulation overhead for clipping threshold selection prior to system application.

Description

technical field [0001] The invention relates to the field of wireless communication technology and probability graph theory, can be applied to communication symbol detection of massive MIMO, OFDM, SC, OTFS, and particularly relates to an improved matched filter message delivery detection method based on probability clipping in a communication system. Background technique [0002] In wireless communication, Orthogonal Frequency Division Multiplexing (OFDM) signals have better anti-multipath performance and spectral efficiency through OFDM, and are obtained in modern wireless communication systems. a wide range of applications. Since OFDM is a multi-carrier system, there is inevitably the problem that the Peak to Average Power Ratio (PAPR) is too high, so based on Cyclic Prefix (CP) and Frequency Domain Equalization (FDE) The single carrier (Single Carrier, SC) signal was proposed, the signal is a constant envelope signal, and at the same time it can achieve better communicat...

AI Technical Summary

Problems solved by technology

However, this method uses Gaussian approximation in the mathematical derivation of probability. When the received signal-to-noise ratio is high, the Gaussian process gradually no longer holds true as the algorithm iterates, so that CHEMP increases with the increase of the received signal-to-noise ratio. The bit error rate (Bit Error Rate, BER) tends to a stable value, that is, the detection algorithm has a BER limit for the receiving signal-to-noise ratio

At the same time, this symbol detection method has only been proposed in the mMIMO system so far

Images