摘要

The capacity of broadband wireless communication is severely limited by intersymbol interference (ISI) and co-channel interference (CCI). Although the maximum likelihood sequence estimator (MLSE) can be used to combat ISI efficiently when implemented with the Viterbi algorithm, CCI is still the bottleneck of the radio link performance. We herein propose an adaptive detection (AD) that can suppress the ISI and CCI simultaneously. By carefully choosing the length of a beamformer, our AD can actually be much simpler than the conventional MLSE. In the case of imperfect channel estimation, a robust AD (RAD) is proposed to maintain the good performance. We also show that both AD and RAD can be further improved by using an iterative scheme, referred to as iterative AD (IAD) and iterative RAD (IRAD), respectively. Numerical examples show that all the proposed methods outperform the conventional MLSE significantly in the presence of CCI, and that they are also computationally more efficient.