Research On Energy-saving Technology Of Base Station In 5G Ultra Dense Network

Recently,the number of user terminal(UE)is greatly increased due to the rapid development of wireless communication system.Traditional network architecture,which is typically overlapped with macro cells,will not meet the increasing demand for system capacity and higher data rate for massive connections.Ultra dense network,which is comprised of a massive number of low power cell node,has been viewed as one pioneering technology to resolve these issues and attracted much attention lately.Although this architecture provides a better service for users,the problem of energy consumption is considerable.As known,all cell nodes are working with maximum power all the time,even under a low traffic load.Therefore,UDN can make great benefits to maintain network coverage and satisfy the performance requirements of UEs with the cost of non-negligible energy consumption.In this context,the network energy saving technology of UDN has been widely studied to maintain the quality of service(QoS).In this paper,we investigate the technology of energy saving for UDN network.In this paper,we formalize the problem of minimizing the energy consumption which caused by the imbalanced distribution of UEs in space and time by optimally controlling the power mode of small cell base station(SBS),i.e.,switching between awake mode and sleep mode.We focus on optimal SBS selection with the object of energy efficiency,while considering the data rate of UEs,the coverage of UEs and the capacity of BSs.Based on the analysis of the principle and performance of distributed algorithm and centralized algorithm,a dynamic clustering sleep strategy is proposed with the purpose of reducing the high computation complexity with the increasing of network size.Firstly,we formalize a multi-objective problem to divide the sleep period by equilibrium optimization algorithm.Then,the base stations are clustered reasonably in time.The clustering process is regarded as a benefit forest generating process based on interest degree,and the exclusion factor is introduced to make the cluster scale balanced.Finally,we transform the problem into a 0-1 integer programming problem within the promise of certain QoS for UEs,and use genetic algorithm(GA)to get the sleeping combination in every cluster.To validate the effectiveness of the proposed scheme,we have conducted simulation campaigns with considering of UE density,distribution,as well as SBS deployment.The results demonstrate that the new scheme can effectively reduce the number of active base stations,decrease the network energy consumption and improve the energy efficiency.Meanwhile,the complexity of clustering algorithm is greatly reduced.As a conclusion,our proposed framework provides an essential solution on the deployment of future green cellular network.