Nonlinear channel identification and equalization for OFDM systems

Orthogonal frequency division multiplexing (OFDM) has become increasingly popular due to its potential applications in digital audio broadcasting, digital terrestrial TV broadcasting, and satellite communication. A notable drawback of OFDM systems is their sensitivity to nonlinear distortion. For maximum power efficiency, amplifiers and transmitters of modern communication systems often operate near their saturation regions which leads to nonlinear distortion. We use the special property that the transmitted OFDM symbols are asymptotically white Gaussian to derive an algorithm that identifies the nonlinear channel. A nonlinear equalizer is built to compensate for the undesired nonlinearities. Simulation results show that the nonlinear equalizer outperforms its linear counterpart when nonlinear distortion is present.