Adaptive equalization for mobile communication systems based on constant modulus algorithm

In this paper the performance of the modified-and improved modified constant modulus algorithm (MCMA-IMCMA)-for adaptive equalization is investigated. The receiver uses a multistage constant modulus (CM) array for multiuser detection. It consists of several identical stages connected in a cascade structure. Each stage is composed of a CM array that captures one of the source signals, and an adaptive signal canceller that removes the source from the array input. Since the captured signal is corrupted by fading, Doppler shift and noise after each stage, it needs to be equalized in order to recover the original signal. Since CMA is phase blind, the output constellation of the algorithm has a phase error. A modified CMA that accomplishes blind equalization and carrier offset is proposed. In this paper the performance of MCMA-IMCMA for recovering the phase and carrier offset of the captured signals is evaluated. In order to speed up the steady state performance, once the error is in a so-called confidence zone the algorithm can switch between MCMA-IMCMA and a decision directed (DD) algorithm. Computer simulations for signals with the QPSK constellation in a Rayleigh fading channel corrupted by additive white gaussian noise (AWGN) and in the presence of Doppler shift are presented to illustrate the performance of CMA for equalization.