Adaptive lattice equalizer to compensate nonlinearities of digital radio channels with MPSK modulation

In this paper we analyze the performance of lattice structure adaptive equalization to compensate for the nonlinearities of digital radio channels with MPSK modulation (satellite channels) and AWGN (additive white Gaussian noise). The nonlinearities are outperformed by TWTA AM/AM and AM/PM characteristics. We simulate a polynomial approximation optimized under the minimax criterion (Karam and Sari). In the receiver we simulate: (a) a DFE (decision feedback equalizer) with a FSE T/2 feedforward part with 64 tap adjustments and symbol spaced tap adjustments; (b) an equalizer proposed by Benedetto and Biglieri (1983) with 22 linear and 81 nonlinear tap adjustments (based on the Volterra series for nonlinear systems with memories); (c) an adaptive lattice equalizer (linear structure) with 11 tap adjustments. The simulation results are outperformed with by 8PSK and 32PSK modulations for a digital radio system. We found that the adaptive lattice equalizer is capable of compensating for nonlinearities of the digital radio system very successfully (“naturally” without any additional tap adjustments as a consequence of solution of the Yule-Walker equation). The adaptive lattice equalizer gives better results (in the sense of mean square error and convergence rate) than the equalizers simulated in (a), (b), but with greatly reduced complexity