Abstract: Cooperative user selection is studied for downlink coordinated multi-point transmission systems. We first argue the necessity of full spatial multiplexing multi-user transmission for such systems, and then study the cooperative user selection algorithm to support full multiplexing, where the users selected as a group are served together on the same time and frequency resources by multiple cooperative base stations. The average received signal to interference and noise ratio (SINR) is analyzed for each user in the concerned systems. It shows that each user’s average received SINR is an increasing function of the large scale signal to interference ratio (SIR) of the users in neighboring cooperative cells. Based on this observation, a novel cooperative user selection algorithm is proposed based on the large scale SIR. Simulation results show that comparing with existing large scale channel information based cooperative user selection algorithms, the proposed algorithm significantly improves cell edge throughput with only minor cell average throughput reduction.