Message-Passing Strategy for Joint User Association and Resource Blanking in HetNets

Abstract

This paper develops a self-organizing approach to joint user association and resource blanking between network-tiers in highly dense heterogeneous networks (HetNets). HetNets that populate many small cells within each macrocell have been studied extensively as a promising solution to exponentially increasing traffic demands. To benefit from small cells, traffic loads are efficiently spread over the network through user association, and strong interference from macrocells are carefully controlled. A simple but very effective solution is resource blanking that partitions time slots into two orthogonal groups and dedicates them to macrocells and small cells exclusively. The underlying challenge is the distributed management of joint user association and resource blanking because a centralized coordination becomes intractable for highly dense HetNets. Unlike existing approaches, this paper aims at developing a distributed solution for this joint optimization task without relaxing nonlinear constraints or decoupling the joint optimization. To this target, a novel message-passing algorithm is developed that enables a fully distributed solution for the joint optimization. The presented approach includes consensus mechanism that determines the optimal macrocell blanking via the negotiation among base stations. Simulation results show that the developed algorithm provides very efficient solutions and outperforms existing techniques consistently.