A design method of beamforming matrix for mimo system based on deep learning

Abstract

A method for designing a beamforming matrix of a MIMO system based on deep learning. The steps are as follows. First, use a known algorithm to obtain the training sample set required by the deep learning network; then construct a deep learning neural network model, initialize model-related parameters and use the training sample set Carry out training; then use the pilot to obtain the channel and send it to the neural network to predict the beamforming matrix coefficients, and finally combine the channel and the beamforming matrix coefficients to form a beamforming matrix. This method uses the beamforming matrix obtained by the deep learning neural network to take into account both performance and algorithm complexity, and can reduce delay under the premise of ensuring performance, so that the MIMO system can provide real-time services.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and in particular relates to a design method of a MIMO system beamforming matrix based on deep learning. Background technique [0002] Multiple-input multiple-output (MIMO) systems can effectively increase the capacity of communication networks, and beamforming technology, as a key technology of MIMO systems, has received extensive attention. Traditional optimization algorithms rely on an iterative process. Although good performance can be achieved, the complexity of the algorithm is high, and the calculation delay is large, which cannot meet the needs of real-time services. However, some heuristic methods, such as zero-forcing method and canonical zero-forcing method, although the complexity is low, the performance is greatly reduced. As a way to realize artificial intelligence, deep learning greatly reduces the online service delay by transferring online computing complexity to of...

AI Technical Summary

Problems solved by technology

Traditional optimization algorithms rely on an iterative process. Although good performance can be achieved, the algorithm complexity is high and the calculation delay is large, which cannot meet the needs of real-time services.

However, some heuristic methods, such as zero-forcing method and regular zero-forcing method, although the complexity is low, the performance is greatly reduced

Images