Energy-saving Algorithm Based On Load Transfer For Base Station In Heterogeneous Wireless Networks

With the booming development of mobile communication technology in recent years, avariety of new technologies and services are emerging. Wireless networks have beenexpanding in order to meet the demand and number for the growth of users, lead to the rapidgrowth of wireless network energy consumption, which attract the attention of experts andscholars. In order to provide a variety of services to user, the wireless communication systemhas been coexistence of multiple access networks. The future of the wireless communicationsystem is bound to be an all-IP heterogeneous wireless network. Heterogeneous wirelessnetwork base station usually have mutual overlapping coverage, due to the presence of thewireless network traffic uneven distribution in time and space, a large number of denselydeployed base stations cause a serious problem of waste energy consumption when the loadof networks is low. Therefore the energy-saving technologies for heterogeneous wirelessnetwork base station become a hot research area.By the summary of research on the base station energy-saving technologies, mostenergy-saving algorithms required to carry out the calculation of the reserve resources whenneed to shut down network resources, generally assumed the traffic arrival rate is Poissondistribution. But traditional traffic model lack the ability to describe self-similarcharacteristics of the traffic in all-IP network, making existing energy-saving algorithmsinaccurate estimates the performance of network. To solve this problem, this articleintroduces a traffic model which has the ability of describe self-similarity characteristics tothe base station energy-saving algorithm. The traffic generated by this model is closer to theactual situation of the network traffic, which has a guiding significance for the base stationenergy-saving algorithms design and optimization.Based on a hierarchical semi-centralized network structure which suitable forheterogeneous wireless networks collaborative energy-saving algorithms, an energy-savingalgorithm based on load transfer for base station in heterogeneous wireless networks isproposed. This algorithm set the threshold load transfer strategies to cell dormancy processesand cell wake up process. The proposed algorithm can take advantage of the mutualcooperation between the heterogeneous wireless network cells to absorb the low load celltraffic to go to sleep state to reduce the energy consumption of the networks. Simulationresults prove the feasibility of the proposed algorithm with the energy-saving effect,reflecting the traffic model based on Poisson distribution inaccurate estimate the networkqueuing performance and the energy-saving effect of algorithm. Future research shouldconsider the impact of traffic self-similar characteristics in the design and analysis for basestation energy saving algorithm.To avoid the frequent switching of dormant cell, this paper introduces a state durationthreshold and adjusts it by fuzzy neural network dynamically. Improved algorithm based onfuzzy neural network which can dynamic adjusts the threshold according to the network loadsituation. Simulation results show the improved algorithm has a better performance. This work has a reference value for base station energy-saving algorithm inheterogeneous wireless network.