Computationally efficient real-time blind multiuser detection for cellular DS/CDMA based on inverse QRD-RLS algorithm and subspace tracking

A new computationally efficient real-time blind multiuser receiver based on constrained optimization approach with multiple constraints, inverse QR decomposition (IQRD), and a new fast subspace tracking algorithm is developed in this paper. The IQRD-based recursive least-squares (RLS) algorithm (IQRD-RLS) is very popular because, it has good numerical stability and can be mapped onto coordinate rotation digital computer (CORDIC) processor-based systolic arrays, which is suitable for very large scale integrated circuits (VLSI) architecture and real time applications. Robustness will be added to our receiver by implementing a new fast and simple subspace tracking algorithm for tracking channel vector based also on IQRD algorithm. In IQRD updating method the weights can be calculated without back substitution which gives numerical stability to such algorithms. In this paper we develop a novel multiuser detection (MUD) receiver based on the minimum output energy (MOE) detector and linearly constraint IQRD-RLS algorithm with optimized constraints via subspace tracking algorithm for tracking channel vector. Simulations are presented in a rich multipath environment with near-far effect to demonstrate the potential and robustness of our approach