Efficient parallel form of multiuser detection in CDMA receivpers

In this paper, we present an alternative approach with reduced complexity for blind adaptive multiuser detection based on subband decomposition. Subband decomposition leads to considerable reduction of computational cost, which is a fundamental requirement in practical systems. Gradient algorithms can be then applied to iteratively calculate the weights in every subband. However, to difference from conventional blind adaptive algorithms, the adaptation process is performed over filters with shorter lengths, which leads to reduce the complexity overall system. Simulations indicate that the performance of the proposed method is acceptable while conserve a reasonable computational cost.