Channel Parameters Extraction Based on Back Propagation Neural Network

Accurately mastering the information of wireless channel characteristics is of great significance for improving the spectrum utilization rate and channel capacity. This paper studies the application of back propagation neural network (BPNN) in channel parameters extraction based on QuaDriGa platform. In this paper, the QuaDriGa platform is used to generate the Channel Impulse Response (CIR) in urban scenes, and SAGE algorithm is used to extract channel parameters such as delay spread, azimuth angle (AOA, AOD) in horizontal dimension and elevation angle (EOA, EOD) in vertical dimension. Then BPNN is trained with sample data to extract different channel parameters. The results show that there is little difference between the prediction results of BPNN model and SAGE algorithm, so BPNN model can replace SAGE algorithm to extract channel parameters for MIMO channel simulation. In addition, the time complexity of the two methods is also compared. The results show that BPNN has higher time complexity than SAGE algorithm. Besides, the simulation results of three common error back propagation algorithms are compared. The results show that the L-M algorithm has the lowest mean square error and the best effect in training BPNN model.