A new method of joint filterbank precoders and decision feedback equalizers optimization over dispersive channels

Abstract

Joint transceiver optimization has received considerabIe attention very recently. Optimal FIR transmit filterbank precoders and nonlinear decision feedback receivers are derived for block-based transmissions over frequency-selective AWGN channels. The joint optimization of precoders and equalizers is based on the maximum mutual information criteria subject to fixed transmission power constraint. By adding one FIR feedback filter after the filterbank linear equalizers, we get a new result from which the precoder can also be interpreted as converting the frequency fading channel into a set of independent flat fading subchannels, and there is a direct path in each subchannel because of the feedback filter, thus the SNR of each subchannel and the maximum information rate are increased and the system has better overall BER performance while maintaining all the advantages of that in Scaglione et al., (1999). Simulation illustrates the merits of our design.