Capacity-approaching block-based transceivers with reduced redundancy

We present a transceiver structure for a frequency selective channel that allows the introduction of reduced redundancy. At the same time we optimize the transmitter and receiver in this structure to maximize the information rate. We show that we can decouple the problem of optimizing the transmitter and the receiver in two different problems by using to the data processing inequality of information theory. The problem of finding the transmitters is a difficult non-linear optimization problem. For that reason we present two simple algorithmic procedures in order to obtain suboptimal solutions. We present simulation results that show that the proposed design outperforms other existing ones in the literature. We also show that the proposed scheme is asymptotically optimal in the sense that it approaches the channel capacity.