Resource Management Strategy For Energy-Efficient In Dense Networks

With the popularity of 4G,the research about 5G is also on the way.The emergence of 5G will satisfy the diverse service needs of people in various areas such as residence,work,leisure,transportation and so on.However,it will face the challenges of ultra-high traffic density and high density of connections.One of the effective ways to solve this problem is to deploy heterogeneous networks.However,with a large number of low-power nodes deployed in a dense network,it is important to ensure network performance while minimizing energy consumption.In view of the above problems,this paper focuses on the energy-efficient network resource management strategy in 5G,and mainly focuses on the base station sleep strategy in the resource management strategy.The main working of this paper are as follows:1.Aiming at the problem that the base station is difficult to determine the sleeping cycle due to the uncertainty of traffic loads that reaching the small cells in dense heterogeneous network,this paper proposes a strategy of sleeping strategy which can dynamic adjusting the length of deeply sleeping cycle based on the traffic prediction.First of all,this strategy includes a base station sleeping mechanism which has a long or short sleeping cycle,and the sleeping states are divided into deep sleeping and light sleeping.Secondly,this strategy uses Partially Observable Markov Decision Process(POMDP)to predict the trend of traffic.Finally,this strategy takes The Length of Sleeping Time Selection Policy(STSP)to dynamic select the optimal deep sleeping cycle which can maximize reduce the energy consumption of base station.The simulation results show that compared with the base station sleeping strategy based on the traffic threshold,this strategy that this paper proposed can not only predict the traffic well,but also get a better energy saving performance.2.As we all know,the total power consumption of the system is too high during the low traffic periods when deployed a large number of low power nodes in a dense network.To solve this problem,this paper proposes a strategy which could dynamic turn off small base stations(SBSs)based on distance and load.Firstly,this strategy will rank the SBSs' number from near and far according to the distance between the SBSs and macro base station(MBS),and it has a high probability to turn off the SBSs that closer to the MBS.What's more,this paper analysis and establish an optimal model that can minimize the total power consumption under the circumstance of turning the SBSs.Finally,this strategy use the sleeping base station selection based on distance and load algorithm(SSDLA)to get the minimum energy consumption of all SBSs' operator state.Simulation demonstrate that the strategy that this paper proposes can dynamic adaption the SBSs' operation modes(on or off)to save the total power consumption compared with traditional random sleeping strategy and distance-based sleeping strategy,and the performance is better when the traffic is low.