Computationally-efficient methods for blind adaptive equalization

This paper proposes a new computationally-efficient method for blind equalization and presents a comprehensive review of existing computationally-efficient methods. The proposed method selectively updates the equalizer taps based on the equalizer output radius for QAM signal constellations. Signed-error, dithered signed-error, quantized-error, block, and update-decimated methods are discussed for computationally-efficient blind equalization. The proposed and discussed methods are applied to the constant modulus algorithm (CMA) and multi-modulus algorithm (MMA), and comparisons are provided.