Bayesian decision feedback techniques for blind equalization

Abstract

In this paper we propose a family of Bayesian conditional decision feedback estimators (BCDFE) suitable for blind equalization. The BCDFEs are indexed by two parameters: a "chip" length and an estimation lag. These algorithms can be used with estimation lags greater than the equivalent channel length, and have a complexity which is exponential in the chip length but only linear in the estimation lag. Recursive channel estimation is combined with the BCDFE to produce high performance in unknown channel equalization. Extensive simulations characterize the performance of the BCDFE for uncoded linear modulations over unknown channels. Also, a simple adaptive complexity reduction scheme can be combined with the BCDFE resulting in further substantial reductions in complexity, especially for large constellations.