Overlapped clustering for comp transmissions in massively dense wireless networks

Coordinated multipoint (CoMP) transmission is widely recognized as a promising technique for inter-cell interference (ICI) mitigation. Because both the signal processing complexity and the inter-cell data exchanging overhead required by CoMP are limited in practical applications, cooperative clusters which consist of a limited number of base stations (BSs) need to be formed so as to mitigate the interference locally. The clustering scheme significantly affects the average sum-rate of a massively dense cellular network. In this paper, a novel dynamic clustering scheme with overlapped clusters is proposed for multicell cooperative processing (MCP) in the uplink transmission. The clusters formed by this scheme may overlap with each other, thus leading to the desired coupling between clusters. Accordingly the BS which belongs to multiple clusters chooses the detection output with the highest signal to interference plus noise ratio (SINR) as the final detection result. This scheme is compared with the traditional disjoint clustering scheme in terms of network average sum-rate and the fairness of service. Simulation results show that the overlapped clustering obtains higher average sum-rate with much smaller cluster size and is fairer than the scheme proposed in the existing work.