Dynamic small cell on/off control for green ultra-dense networks

摘要

In next generation networks, ultra-dense small cell networks are emerging to deal with exponential data traffic increasing. With the deployment of small cells (SCs) becoming increasingly dense, these cells may be under-utilized during quite a few period of time. Depending on this circumstance, we aim to maximize the network energy efficiency by deactivating such under-utilized cells, subjecting to the rate fairness constraints among users. To overcome this issue, we propose an energy efficient SCs on/off approach that allows SCs to activate or deactivate based on estimated traffic load and UEs' preference requirements. Small cells are coupled into clustered based on distance firstly, and then UE-preference aware network selection scheme is performed, after which the SCs on/off pattern is derived iteratively in each formed cluster. For the network selection scheme, the optimal user association is obtained according to the SINR, estimated traffic load, traffic type and delay. Due to the complexity of finding the optimal SCs on/off pattern, we also give a practical suboptimal greedy SCs on/off algorithm with polynomial computational complexity. Simulation results demonstrate that energy expenditure can be reduced by 72% at most. Comparing to optimal search and traditional algorithms, the greedy algorithm achieves a comparable performance with much lower complexity.