Statistical performance estimation of digital radio over fading channels

Since the error rate of digital radio is one of the primary measures of performance and since the computation of the error rate as a function of multipath parameters, tap weight coefficients, filter characteristics, etc. is extremely complex, the authors use a simulation method to determine this performance. The average error rate is determined by using a Monte Carlo simulation for a set of channel parameters. Since the estimation of the performance using simulation can be computationally intensive, especially for low error rates and high order communication systems, the authors use a quasi-analytic approach for determination of equalizer taps and error rate. Equalizer taps are expressed directly in terms of the parameters of the channel and modulation characteristics of the signal. Error rate is estimated by using the moment method. Simulation techniques are used to evaluate the residual intersymbol interference after using the equalizers, and moments of this intersymbol interference are calculated using a very fast and accurate recursive method of computation. The standard techniques of Gaussian quadrature are then used to obtain the error rate of the system