Iterative Channel Estimation and Equalization Techniques for Multipath Channels with Multiple Frequency Offsets

Abstract

We provide an iterative joint channel estimation and data detection algorithm for multipath channels with multiple frequency offsets (FOs). Pilot symbols are generally not long enough to provide enough resolution to estimate frequency offsets. Hence we used initial pilot based channel estimation to generate soft decision of the transmitted signal using log likelihood ratios, and the frequency offsets were estimated using these soft estimates of the transmitted signal together with pilots. Using extensive simulation study, we demonstrated the convergence of this joint channel and frequency offsets estimation and data detection. We also showed the variances of the iterative channel and the frequency offset estimates using 26 pilot symbols attain the Cramer-Rao lower bound (CRLB) derived assuming all 142 symbols in the burst are pilot symbols.