An Improved Convolution Based User Clustering Scheme in Ultra Dense Network

Abstract

In ultra-dense networks (UDN), one of the most important technologies is clustering algorithm which can provide extra benefits for system performance through increasing system throughput, as well as raising data rates of cell-edge users. In this paper, a user cluster scheme based on improved convolution is presented to maximize system Spectral Efficiency(SE) and cell-edge users' throughput. We proposed once-and twice- convolution algorithms, which means the convolution implemented once or twice. Simulation results show that the recommended algorithm improves the SE of cell-edge users and the system throughput obviously. Moreover, the once-convolution algorithm has better improvement compared with the existing clustering scheme than twice-convolution algorithm.