Diversity with Pilot Symbol Assisted Channel Estimation

摘要

In this paper, we present a framework for evaluatingrnthe bit error probability of N_d-branch diversity combiningrnin the presence of non-ideal channel estimates. The estimatorrnstructure is based on the maximum likelihood (ML) estimaternand arises naturally as the sample mean of N_p pilot symbols.rnThe framework presented requires only the evaluation of a singlernintegral involving the moment generating function of the normrnsquare of the channel-gain vector, and is applicable to channelsrnwith arbitrary distribution, including correlated fading. Ourrnresults show that the practical ML channel estimator preservesrnthe diversity order of an N_d-branch diversity system, contraryrnto conclusions in the literature based upon a model that assumesrna fixed correlation between the channel and its estimate.