The Use of Transform Domain LMS Algorithm to Improve The Performance of Blind Multi-user Detection

The convergence speed of LMS algorithm is determined by the spread of the eigenvalues of the input signal´s auto-correlation matrix, which is too slow to meet the need in many conditions. Transform domain LMS algorithm turns the auto-correlation matrix into unit matrix in transform domain, so it can speed up the convergence effectively. In this paper, transform domain LMS algorithm was applied in multi-user detection. The simulation result showed that the algorithm can improve the performance of multi-user detector.