Superimposed periodic pilots for blind channel estimation

Multipath is a major impairment in a wireless communications environment. Blind channel estimation methods are of interest because they avoid training and thus make efficient use of the available bandwidth. In the past few years, we have seen an evolution from higher-order statistics-based, to cyclic second-order statistics based approaches for blind channel estimation. We propose here a simple, alternative scheme which employs only a first-order statistic. Pilot symbols are periodically added to the intended symbol stream prior to transmission. There is no loss of information rate in our approach, and neither is there any restriction on the FIR channel to be identified. We derive the variance expression of our linear channel estimate and compare with the Cramer-Rao Bound (CRB). Numerical examples indicate that the variance approaches the CRB at high SNR.