Clustering Strategy Based on Graph Method and Power Control for Frequency Resource Management in Femtocell and Macrocell Overlaid System

Hongjia Li,Xiaodong Xu,Dan Hu,Xiaofeng Tao,Ping Zhang,Song Ci,Hui Tang
DOI: https://doi.org/10.1109/jcn.2011.6157483
2011-01-01
Journal of Communications and Networks
Abstract:In order to control interference and improve spectrum efficiency in the femtocell and macrocell overlaid system (FMOS), we propose a joint frequency bandwidth dynamic division, clustering and power control algorithm (JFCPA) for orthogonal-frequency-division-multiple access-based downlink FMOS. The overall system bandwidth is divided into three bands, and the macro-cellular coverage is divided into two areas according to the intensity of the interference from the macro base station to the femtocells, which are dynamically determined by using the JFCPA. A cluster is taken as the unit for frequency reuse among femtocells. We map the problem of clustering to the MAX k-CUT problem with the aim of eliminating the inter-femtocell collision interference, which is solved by a graph-based heuristic algorithm. Frequency bandwidth sharing or splitting between the femtocell tier and the macrocell tier is determined by a step-migration-algorithm-based power control. Simulations conducted to demonstrate the effectiveness of our proposed algorithm showed the frequency-reuse probability of the FMOS reuse band above 97.6% and at least 70% of the frequency bandwidth available for the macrocell tier, which means that the co-tier and the cross-tier interference were effectively controlled. Thus, high spectrum efficiency was achieved. The simulation results also clarified that the planning of frequency resource allocation in FMOS should take into account both the spatial density of femtocells and the interference suffered by them. Statistical results from our simulations also provide guidelines for actual FMOS planning.
What problem does this paper attempt to address?