Filter Parameter Optimization for Linear and Nonlinear BPSK Transmission

Abstract

The parameter optimization of practical filters for baseband equivalents of linear and nonlinear BPSK transmission systems is investigated. An estimate of system error rate is defined and this is minimized. Here the received signal is detected in the presence of intersymbol interference and white Gaussian noise. A closed form expression for receiver BT product minimizing mean-square distortion on a simple channel is developed. The numerical optimization scheme introduced in this paper minimizes the error rate estimate. This method only requires the transfer functions of the various system components and the data spectrum. The optimization is applied to linear systems which include four and six pole Butterworth and four pole Chebyshev receive filters for a channel with a fixed four pole Chebyshev filter as a channel filter. A joint transmit/receiver filter optimization is considered with practical constraints introduced on the transmit filter. The numerical optimization is also applied to a bandlimited nonlinear satellite channel with high uplink SNR.