Abstract:

To improve the performance of satellite communication systems, one of the key problems is to realize the equalization of the nonlinearity generated by the high-power amplifier of the channel. This paper firstly establishes a satellite channel model based on volterra series and analyses the nonlinear effects of two-dimensional modulation signals generated by the channel. For the slow convergence of the volterra equalizer based on the minimum mean square error, a partially decoupled equalization algorithm based on the volterra model of complex baseband is proposed. To achieve the decoupling between the each weight of equalizer, the linear weights of equalizer are forced to converge to the optimal solution of the linear filter and the new cost function is constructed using the Lagrange multiplier method for the every order coefficients of the equalizer to satisfy the constraints. Simulation results show that, compared with the whole coupling minimum mean square error algorithm, the new algorithm ensures the steady-state error and the number of iteration required for the convergence of algorithm is reduced from 16 000 to 1 000 times.