A windowing-based joint user pairing and resource allocation algorithm for V-MIMO systems

Abstract

In this paper, we investigate the user pairing and resource allocation (UP-RA) problem which is a critical step in uplink virtual multiple input multiple output (V-MIMO) systems. It becomes a joint problem under the framework of Long Term Evolution (LTE) where single carrier frequency division multiple access (SC-FDMA) is adopted in uplink due to the intrinsic consecutive resource allocation constraint imposed therein. Traditional two-step algorithms tackle this joint problem by separating the UP step and the RA step and turning them into two independent and successive procedures. Different from them, in this paper a windowing-based UP-RA algorithm is developed in which the user pairs are selected by maximizing their performance on a series of contiguous RBs rather than a single one as discussed in those traditional algorithms. With the windowing technique introduced in the new algorithm heuristically, improved performance in uplink V-MIMO systems can be achieved. In addition, we provide the relationship between the performance and the window size, and present that the window size can be selected adaptively to further exploit the diversity gain. Furthermore, to corroborate the universal validity of the proposed windowing technique, another algorithm named pairing and extension (P&E) is also modified accordingly by it. Corresponding simulation results manifest the great potential of the windowing technique to be applied to other general two-step algorithms in V-MIMO systems and improve their performances.