SDWN Energy Efficiency Methodology Based On Multi-path Mechanism

Densely-deployed wireless networks meet the communication traffic demands of mobile users,while causing a serious waste of energy.In traditional single-path based energy-saving researchs,user's traffic allocation is constrained by associated mode.The utilization efficiency of most APs is still not optimal,resulting in incomplete system energy saving.Multi-path software defined wireless network(MP-SDWN)integrates SDN/NFV technology with wireless network to implement concurrent communications between users and multiple APs.Meanwhile,MP-SDWN achieves centralized management and resource optimization for wireless networks,and provides new research ideas for further energy saving optimization.The thesis mainly studies the energy saving methods based on multi-path mechanism in SDWN.This method effectively predicts user traffic requirements and formulates reasonable AP work/sleep,user-AP link traffic allocation strategies to achieve real-time optimization of wireless network systems energy consumption,while ensuring users' quality of experience(QoE)and system stability.Firstly,the network parameters affecting the AP energy consumption are measured and analyzed.The mathematical relationship model between AP traffic load and energy consumption is obtained by fitting the experimental data.At the same time,in order to effectively satisfy the user's traffic demand while saving energy,by analyzing the characteristics of wireless user traffic data,and base on the idea of Matrix Completion,a wireless user traffic prediction model and algorithm are designed.Based on the above two achievements,using AP work/sleep and user-AP traffic allocation strategies as variables,the wireless network energy-saving problem under the multi-path mechanism is analyzed and modeled.Further,an energy-saving algorithm,based on Bender's Decomposition,is proposed to solve the energy consumption optimization model.For handing the high computational complexity of the algorithm in large-scale wireless scenarios,the algorithm is optimized by the perspective of reducing the number of variables and constraints to improve the practicability.Finally,software simulations and SDWN platform experiments validate that compared with single-path energy-saving method,the proposed energy saving method can effectively predict traffic demands of wireless network users,and realize better real-time energy saving of wireless network while guaranteeing user traffic,system coverage and throughput.By studying the energy efficiency methodology based on mutil-path mechanism,5%~10% redundant energy consumption can be reduced,which is of great significance to the construction and development of the future green 5G.