Channel Estimation and Equalization Techniques in Downsampled ARQ Systems

In this paper, we study channel estimation and equalization for a class of bandwidth efficient automatic repeat request (ARQ) systems. In these systems, retransmissions achieve bandwidth efficiency through symbolwise downsampling. Unlike a standard ARQ scheme, in which an erroneous packet is fully repeated, more sophisticated hybrid ARQ systems retransmit a recoded data packet. However, many of these ARQ schemes do not consider the integration of ARQ protocols with receiver signal processing such as channel estimation and signal detection. In this paper, we present a joint multichannel estimation algorithm that exploits the structure of a symbolwise downsampled ARQ equipped network. We show that this new method can improve the identifiability conditions for channel estimation as well as the overall system performance