Multistage decoding-aided channel estimation and equalization for DVB-H in single-frequency networks

In single-frequency network (SFN) broadcasting systems, the presence of several transmitters translates into an equivalent channel with relevant sparsity in the delay domain, which may significantly impair the error performance. In DVB-H systems this channel sparsity, coupled with the time variation due to mobile users, renders the channel estimation task very challenging, especially when only a moderate complexity is tolerated at the receiver side. To solve this problem, this work proposes a new low-complexity multistage channel estimation technique that exploits the data at the output of the channel decoder, as well as a simple parallel intercarrier interference (ICI) canceller. The proposed scheme iteratively estimates data and channel in a joint fashion, and also includes a regularization-based preprocessing stage to specially deal with severe channel sparsity. Simulation results show that our low-complexity scheme is adequate for practical implementations of fast-mobility DVB-H receivers in all SFN scenarios.