This paper represents a comparative Study of filter algorithms Least Mean Square (LMS), Normalized Least Mean Square (NLMS) and Recursive Least Square (RLS) by considering a Quasi Orthogonal Space Time Block Code (QO-STBC) encoded Multiple Input Multiple output (MIMO) Code Division Multiple Access (CDMA) system. MIMO-CDMA system has been currently acknowledged as one of the most competitive technology. Here, the adaptive behaviors of the algorithm are studied. Implementation aspects of these algorithms are their computational complexity and Signal to Noise ratio which are also examined. Recently adaptive filtering algorithms have a nice tradeoff between the complexity and the convergence rate. In this system, by comparative study of three adaptive filter algorithms, the RLS algorithm has faster convergence rate than LMS and NLMS algorithms with better robustness to unpredictable situation and better tracking capability.