On the Performance of Joint Processing Schemes over the Cluster Area

In this paper, three joint processing schemes for the downlink are characterized and compared within a cluster of base stations. The motivation of this study is to analyze the performance of these schemes over the cluster area, as a first step towards designing an adaptive joint processing scheme supporting dynamic usage scenarios. Each one of the analyzed schemes, the centralized, partial and distributed joint processing approaches, requires a different amount of available channel knowledge at the transmitter side, inter-base information exchange and feedback from the users. In addition, these schemes show varying capabilities to serve the users depending on their location in the cluster area. Therefore, in a real scenario, an adaptive joint processing scheme encompassing the three schemes could be used by the cluster of base stations. Simulation results show that, assuming coherent transmission, the centralized joint processing scheme outperforms with 25% the partial joint processing scheme and with 50% the distributed joint processing approach in the cell edge when a backhaul-load weighted average sum-rate per cell metric is taken into account.