THE IMPACT OF WHITE GAUSSIAN-NOISE MISMATCH ON LINEAR AND DECISION-FEEDBACK EQUALIZERS

In this paper, we analyze the sensitivity of equalizers to white gauss ian noise mismatch, which exists whenever the noise power observed dur ing training differs from that observed during detection, A closed-for m solution for the mean-square error is obtained for linear and decisi on feedback equalizers as a function of the channel spectrum and degre e of noise mismatch. Our results show that increasing equalizer comple xity may actually degrade performance in the presence of channel misma tch.