Optimization Of Energy Efficiency In Self-Organized Network

With the rapid development of global information technology, and the vigorous of the new technologies such as big data, cloud computing and Internet+, mobile Internet is becoming a basic platform which could carry a variety of new business in future. More and more services which are provided by the fixed internet service through wireless access to the Internet, resulting in exponential mobile data traffic growth, and new network struture such as intensive Small Cell Network (SCN) and heterogeneous networks (HetNets) could through the deployment of Low Power Node (LPN) and increased density of base stations in a manner that effectively add Strong network coverage and upgrade system capacity. However, large-scale wireless network can bring good network coverage performance, but it poses a serious problem of energy consumption, while increasing the number of nodes in the network also increase the complexity of network management. In the paper, under the background of 5G wireless technology research projects, the paper researchs energy efficiency optimization techniques based on self-organizing network(SON) are carried out research.Firstly, the paper studies the theories and key characteristics of SON, and gives an architecture of SON in LTE network. It mainly introduces principle of HetNets and challengs, such as interference coordination technologies. On the basis, it introduces the energy saving technniques with respect to SCN and HetNets, emphatically studies the existing base station (BS) energy saving techniques, that LPNs can be dynamically switched into sleeping mode and achieve the optimization of network resoures, while the macrocells are responsible for network coverage.Subsequently, targeting on saving energy in dense SCN, the paper proposes dynamic sector sleeping algorithm based on tilt optimization to periodically configure the sectors’mode. The scheme is based on a distributed SON and divided into two phases to achieve the energy-saving purposes. The first-stage tests to active sectors according to sector utility function. The second-stage readjusts the mode of sectors which is not suitable for sleeping keep active, and takes optimal downtilt adjustment strategy to ensure network coverage and Quality of Service (QoS). The paper poposes a fixed method to distubute downtilt scheme, then propsed two improved schemes which are optimization tilting program and improved optimization tilting program, they realizes of downtilt adjustment without fixed allocation Angle, but rather through adjust index function to find out the optimal combination of Angle, so as to realize the purpose of downtilt adjustment. Simulations demonstrate that these algorithms allow the number of active sectors tracking the network fluctuation effectively, through downtilt optimization ensures network coverage and QoS. However the fixed allocation scheme can effectively reduce complexity, and optimization tilting scheme can keep the blocking probability relatively stable, improved optimization tilting scheme is the best compromise in network energy efficiency and the blocking probability.Finally, on the bases of the sleeping algorithm which bases on tilt optimization in SCN, the paper puts forward the suitable metod for the Pico BS in HetNets. According the HetNets mode and its characteristics, it designed a performance indicator function to measure the performace of the macrocell antenna tilt in order to assist the tilt optimization.on this basis, and proposed two dynamic Pico BS sleeping algorithms (i.e., GPS and GPSTA) to save energy in HetNets by periodically configure the working mode of all pico BSs. The first algorithm relies on the utility function to test every pico BSs, and without considering the network possible coverage of the decline of the covering loopholes and QoS. The second algorithm ensure maximum closed pico BS through tilt adaptive to effectively solve the covering loopholes and QoS. Simulations demonstrate that GPSTA has significantly improved in energy saving and network congestion.