A clustering-based downlink resource allocation algorithm for small cell networks

Abstract

This paper proposes a clustering-based downlink resource allocation (CDRA) algorithm for allocating downlink spectrum resources in small cell networks. To improve the system capacity, the proposed CDRA algorithm introduces clustering before resource allocation, and proposes a new clustering method to divide all small cells in the system into a set of clusters. Based on three clustering principles, those small cells that are adjacent to each other are grouped into one cluster. After clustering, resource allocation is performed for the macro cell and each small cell based on the priority of the macro cell and each small cell. The macro cell is given a higher priority than all small cells while the priority of a small cell depends on the size of the largest cluster it belongs to and the number of clusters it belongs to. Moreover, the users in a particular cell or cluster can only be allocated different physical resource blocks (PRBs). Only those users in different clusters or between the macro cell and small cells can share the same PRBs. Simulation results show that the proposed CDRA algorithm can significantly improve the system capacity as compared to two existing clustering-based resource allocation algorithms for small cell networks.