Fast convergence of distributed joint spreading sequence and power control algorithms

In this paper a distributed iterative algorithm is developed for joint spreading sequence design and power control in the uplink of CDMA systems. Simultaneous iterations are implemented using gradient approaches for convex cost functions with nonlinear QoS constraints. Performing parallel iterations instead of round robin iterations, the algorithm convergence speed is significantly increased. Different from conventional approaches the algorithm convergence speed will not explicitly depend on the number of active users. Moreover, the efficiency of our method increases by increasing the number of active users.