Channel Parameter Extraction Based on BP-Neural Network in NLOS Environment

Zhexin Hu, Fanliang Bu, Dandan Ding

Abstract: The extraction of wireless channel parameters is important for channel modeling and wireless resource scheduling in complex environments. Based on QuaDriGa studies the application of back-propagation neural network in MIMO channel parameters extraction, and compares with the effect of traditional space-alternating generalized expectation-maximization algorithm. The results show that the back-propagation neural network model algorithm works well and can be used for channel parameter extraction in future complex scenarios.