A Modified Iterative Water-filling Algorithm with Per-iteration Power Normalization in Multiuser MMSE-Precoded MIMO Systems

In this paper, we deal with an optimum per-user power allocation problem in terms of the sum capacity in multiuser multiple input multiple output (MU-MIMO) downlink systems when the minimum mean square error (MMSE) scheme has been used for multiuser preceding. However, MMSE preceding allows some inter-user interference. In this paper, we propose a modified iterative water-filling (MIWF) algorithm with per-iteration power normalization (PIPN) in multiuser MMSE-precoded MIMO systems. In the proposed algorithm, power levels for respectacle users are re-adjusted prior to the consecutive iteration, so that the total transmit power constraint should be satisfied. This normalization can reduce greatly the number of iterations for the iterative water-filling process since the inter-user interference terms at every iteration can be computed more accurately without performance degradation. From computer simulations and complexity analyses, we see that the proposed algorithm has lower complexity but with the same capacity as compared with the original MIWF algorithm.