Dynamic clustering and sleep mode strategies for small cell networks

In this paper, a novel cluster-based approach for optimizing the energy\nefficiency of wireless small cell networks is proposed. A dynamic mechanism\nbased on the spectral clustering technique is proposed to dynamically form\nclusters of small cell base stations. Such clustering enables intra-cluster\ncoordination among the base stations for optimizing the downlink performance\nthrough load balancing, while satisfying users' quality-of-service\nrequirements. In the proposed approach, the clusters use an opportunistic base\nstation sleep-wake switching mechanism to strike a balance between delay and\nenergy consumption. The inter-cluster interference affects the performance of\nthe clusters and their choices of active or sleep state. Due to the lack of\ninter-cluster communications, the clusters have to compete with each other to\nmake decisions on improving the energy efficiency. This competition is\nformulated as a noncooperative game among the clusters that seek to minimize a\ncost function which captures the tradeoff between energy expenditure and load.\nTo solve this game, a distributed learning algorithm is proposed using which\nthe clusters autonomously choose their optimal transmission strategies.\nSimulation results show that the proposed approach yields significant\nperformance gains in terms of reduced energy expenditures up to 40% and reduced\nload up to 23% compared to conventional approaches.\n