MIMO system signal detection method and system

Abstract

The invention discloses a MIMO system signal detection method and system. The invention relates to the field of communication. The method comprises the steps of: performing real number processing on areceived signal and a channel matrix, and inputting the real number processed received signal and channel matrix into a trained detection network which is an approximate message passing algorithm network containing the damping mechanism to obtain an estimated signal. The embodiment of the invention provides an approximate message passing algorithm detection network. Compared with a full-connection neural network, a convolutional neural network and a recurrent neural network, the MIMO system signal detection method and system have the advantages of being low in data dependence, interpretable in network structure, low in overfitting risk, high in accuracy, low in complexity and the like, the convergence speed of the detection algorithm is further increased by adding a damping mechanism intothe approximate message passing algorithm and setting learnable damping parameters, the detection network complexity is reduced, and meanwhile the detection performance is improved.

Description

technical field [0001] The present invention relates to the communication field, in particular to a MIMO system signal detection method and system. Background technique [0002] With the rapid development of wireless communication technology, people's communication demands for high communication rate and reliable service quality are constantly increasing. Existing frequency band resources are increasingly difficult to meet the needs of wireless communication. MIMO technology is equipped with multiple antennas at the sending end and receiving end to generate multiple transmission paths to combat wireless channel fading, and utilizes the spatial freedom provided by the transmission channel to improve system capacity and link reliability. Its main features are : Deploy a large number of low-cost, low-power amplifiers and hundreds of antennas at the base station to increase the spatial freedom of the system. The base station can communicate with many users on the same time-freq...

AI Technical Summary

Problems solved by technology

The linear detection algorithm headed by the minimum mean square error will introduce the channel matrix inversion operation in the detection process. For the massive MIMO system, the detection technology complexity is very high, and the detection performance is average.

Compared with the linear detection algorithm, the nonlinear detection algorithm and suboptimal detection have better detection performance and avoid the process of inverting the channel matrix, but the detection process still requires a lot of calculations, and the complexity is still high. It is suitable for application in large-scale MIMO systems, but the training process of signal detection networks based on fully connected neural networks, convolutional neural networks, and recurrent neural networks requires a large amount of labeled data. The process of accumulating and labeling a large amount of information is not only time-consuming but also expensive. Moreover, the generalization and adaptability of the above-mentioned network are weak, and when the channel changes, the detection accuracy of the network will be greatly reduced.

Images