Analysis of channel performance for binary signaling

In this thesis, the performance of binary signaling through a stationary Gaussian noise channel and a nonstationary channel is analyzed. The impulse response of the nonstationary channel is comprised of a direct component that arises from the line of sight propagation and a reflected component. The reflected component is a function of Gaussian noise weighted by an exponential function of time. The exponential function gives rise to the time-dependent statistics of the nonstationary channel model. The optimal detector obtained for the AWGN model is used to detect the transmitted symbols. The statistical properties of the nonstationary channel model are evaluated and the performance of both the channels is characterized using the bit error rate (BER). The performance of the nonstationary channel model is shown to be comparable to that of the AWGN channel as the envelope of decay approaches zero.