Semi blind neural network based channel estimation technique for OFDM receivers

Orthogonal frequency division multiplexing has high data rate capacity and lowest ISI, among the different present technologies and hence considered as the modulation technique for next generation wireless communication. Channel estimation is one of the crucial challenges in designing high performance receivers based on OFDM. In this paper, different operation mode for Levenberg Maquardt algorithm powered back propagation feed forward neural networks are examined for channel estimation of OFDM receivers. A novel, semi blind, optimized channel estimation technique is presented based on the outcomes of different experiments. The tests were conducted in the background of varying channel characteristics, subcarriers, pilot positioning etc. The performance of the different techniques were analyzed based on bit error rate, which is also used for the comparison between the different techniques. The simulations of the tests are included for illustrating the results.