CSA-MWF based nonlinear equalization for data relay satellite channel

A low complexity nonlinear equalizer for data relay satellite channel is presented. The nonlinear equalizer is base on the truncated third order Volterra filter, which is approximately implemented in low complexity form using the recently developed Correlation Subtraction Architecture of the Multistage Wiener Filter (CSA-MWF). CSA-MWF offers significantly faster convergence, as well as low sample support. After the Volterra kernels are estimated through CSA-MWF in training mode, a pipeline structure of the nonlinear equalizer is described, more than half of multipliers are saved. The proposed algorithm is verified by computer simulation. Results show that the nonlinear equalization outperforms linear equalization with more than 1dB gain for 16QAM, and only one to fifth of samples are needed comparing to Least Mean Square algorithm based.