Channel Estimation of MIMO Ofdm based system using pilot based neural networks

 ABSTRACT In this paper Artificial Neural Network (ANN) is used for channel estimation in MIMO OFDM systems over AWGN fading channels. Neural Network is trained on the basis of pilots over the different layers using back propagation algorithm. In OFDM system, pilots are used here in order to estimate the channel. Pattern of pilots must be able to match the channel behavior both in time and frequency domains. In this arrangement, the performance of the channel estimation is done through estimators based on obtaining weights and biases of ANN are carried out through MATLAB Simulation and coding. The performance of OFDM is evaluated on the basis of Bit Error Rate (BER). This technique is useful in achieving different parameters the high data rate, transmission capability with high bandwidth, efficiency and its compatibility to noise.