基于神经网络无线通信中时变信道建模应用

摘要

The modeling of time varying channel in mobile communication was presented in the paper.It took BP neural network Levenberg-Marquardt algorithm for time-varying multi-path channel model,and according to the generalization of the complex exponential basis expansion model,fitting the training data, and finally through the simulation to simulate the signal-to-noise ratio and bit error rate,it was proved that the model has a good forecast effect.%对移动通信中时变信道进行建模，采取 BP 神经网络 Levenberg－Marquardt 算法对多径时变信道进行建模，同时根据泛化复指数基扩展模型，拟合出训练数据，通过仿真来模拟信噪比与误比特率之间的关系，最后验证了该模型具有很好的预测效果。