Nonlinear distortion cancellation based on neural network technique for MIMO-OFDM systems over Rayleigh fading channel

In this paper, we investigate the effect of high power amplifier (HPA) nonlinear distortions on vertical Bell Laboratories Layered Space Time multiple input multiple output-orthogonal frequency division multiplexing (V-BLAST MIMO-OFDM) systems when the HPA is modeled as polynomial one. Then, we introduce an iterative technique for the detection of the nonlinear received signal. This technique is based on neural network (NN) techniques and aims to estimate and compensate effectively the nonlinear distortions due to HPA. The numerical results evaluation shows that the proposed method of compensation reduces significantly bit error rate (BER) of MIMO OFDM systems.