Adaptive MIMO channel shortening with post-FEQ diversity combining

Abstract

Cyclic prefixed communication systems are generally called Discrete MultiTone (DMT) for wireline systems, and Orthogonal Frequency-Division Multiplexing (OFDM) for wireless systems. A cyclic prefix should be as long as the channel impulse response (channel delay spread) to combat inter symbol interference (ISI). If this requirement is not met, a channel shortening equalizer is used to shorten the channel. In time varying channels, it is more sensible to use an adaptive channel shortener. In literature, adaptive channel shorteners and frequency domain equalizers (FEQ) have been well studied. But on the other hand, in wireless communication systems the performance metric is the bit error rate (BER) and minimizing the BER is the primary concern of the designing a system. In this paper we reviewed the adaptive channel shortener which uses recursive least squares algorithm (RLS-MERRY), and proposed an adaptive FEQ with different post combining techniques in order to get a lower BER than existing adaptive TEQ-FEQ architectures.

Blind Channel Shortening for Space-Time-Frequency Block Coded MIMO-OFDM Systems

“…A viable alternative is to shorten the channel in blind mode, by deriving the parameters of the channel-shortening prefilter directly from the received data, without performing any explicit channel-estimation procedure. Blind FIR channel-shortening techniques for SISO-OFDM systems have been developed in [20]- [23], some of which have been generalized [24], [25] to MIMO-OFDM systems. These techniques are based on optimization of multimodal heuristic cost functions, which do not admit closedform solutions in the presence of noise; hence, they must be solved by means of iterative procedures, whose local and/or global convergence is not always ensured.…”