A New Blind Equalization Algorithm for Improving the Channel Equalized Performance

Abstract

Adaptive channel equalization accomplished without a training sequence is known as blind equalization. The decision directed algorithm (DDA), Godard algorithm (GA), Sato algorithm (SA), Benveniste and Goursat algorithm (BGA), and the stop-and-go algorithm (SNGA), are examples of Blind equalization techniques. These algorithms exhibit very slow convergence rates when compared to algorithms employed in conventional data equalization schemes. In order to speed up their convergence process, a modified algorithm (MA) can be employed which uses a combination of DDA and GA. The modified algorithm has demonstrated their effectiveness compared to other conventional techniques especially in the noisy environment. Keyword: LE, DFE, blind equalization, adaptive equalization algorithms. IntroductionAn equalizer is necessary when the inter-symbol interference (ISI) component of a channel output becomes large relative to the signal component and causes a high symbol error rate (SER). Some equalizers filter the channel outputs directly in order to present the decision slicer with a reduced ISI estimated symbol sequence. Other synthesize an output sequence using a channel estimate and the knowledge of the symbol alphabet, and search for the closest match to the actual channel output sequence.