LLL aided MIMO detection algorithm based on BP neural network optimization

In this paper, we propose a Lenstra-Lenstra-Lovasz (LLL) lattice-reduction aided multiple input and multiple output (MIMO) detection algorithm optimized by Back Propagation (BP) neural network to uplink the large-scale MIMO system. Unlike the conventional case, our algorithm uses the forward propagation of signals and the backward propagation of errors. Through the training of sample data, it constantly modifies the network weights and thresholds to make the error function drop along the negative gradient direction, thus approaching the expected output. Simulation results show that BP neural network has an obvious improvement effect on the MMSE criterion of LLL lattice-reduction processing and has low computational complexity.