New quasi-analytic BER estimation technique on the nonlinear satellite communication channels

A novel quasi-analytic (QA) simulation technique to estimate bit error rate (BER) over nonlinear satellite communication channels is introduced. Generally, QA simulation is performed by assuming that the noise probability density function (PDF) at the receiver is known, and that, in most cases, the noise entering the receiver is considered to be additive white Gaussian noise (AWGN). However, because of the nonlinear characteristics of a high power amplifier such as a travelling wave tube amplifier (TWTA) in the satellite, the noise PDF present in the down-link cannot be assumed to be a Gaussian PDF any longer. An efficient QA technique which finds a non-Gaussian noise PDF at the output of the TWTA and hence estimates BER accurately, is introduced. Simulation has been performed using a QPSK satellite system, and BERs are estimated using various methods including the Monte Carlo (MC) method, classical QA methods, and the proposed method. Simulation results showed that the proposed QA technique produces BER values approximating to those of the MC technique