A Novel Scheme for Sum Power Minimization in MIMO MAC Multicarrier Systems

The sum power minimization (SPmin) subject to per user rate constraint problem in the multiple access channel (MAC) of a multiple-input multiple-output (MIMO) multicarrier system is investigated. The SPmin can be transformed into a series convex problems and solved iteratively. The conventional solutions requires a large number of iterations. Furthermore, though in each iteration the optimality can be achieved by the efficient interior-point method, the slow convergence speed is still a burden that makes the calculation with large numbers of users prohibitively complex. We propose an efficient algorithm to find the optimal solution of the SPmin problem. The algorithm utilizes the convexity of the rate region to reduce the number of iterations. The convergence behavior of the algorithm is compared to the existing methods and shown to be much faster.