A blind variable step-size method to enhance the blind multi-user detector performance for DS-CDMA systems

In this paper we will propose a new blind variable step-size method which is mainly based upon the exponentially weighted step-size method to enhance the performance of the recursive versions of the CMA and the CC-CMA algorithms in multi-user applications. The proposed method presents a new method which detects blindly the channel impulse response variation according to the ever-changing environmental conditions, and then uses this variation to adjust the step-size parameter that attenuates this variation by the exponential ratio at each iteration of the adaptive blind equalization algorithm. As a result we will gain a step-size parameter sufficient for updating the tap weight coefficient vector of the blind equalizer iteratively, and that will speed up the convergence rate, minimize the mean square error (MSE) and enhance the performance of the blind detectors for DS-CDMA systems.