摘要
为了最大化基站的可节能空间,解决全网绿色基站的智能发现问题,通过改进Affinity Propagation聚类算法对基站日负荷曲线进行自适应聚类,并进一步挖掘分析周效应下的日潮汐现象和汐节能时段,智能化识别基站的个性化节能场景,从而周期性地采取差异化节能策略。实验分析验证了该算法的高效性和准确性,预计节能空间可达20%以上,可应用于5G基站能耗的智能化管理,提高5G网络能效。
In order to maximize the energy-saving space of the base station(BS),this paper solves the problem of smart discovery of green BSs in the entire network.Specifically,an improved Affinity Propagation clustering algorithm is used to adaptively cluster the daily load curve of 5G BSs,and to further explore and analyze the daily tidal phenomena and the duration in the night tides under the weekly effect.Then the personalized energy-saving scenarios of BSs are intelligently identified to periodically adopt differentiated energy-saving strategies.Experimental analysis demonstrates the efficiency and accuracy of the algorithm,and the estimated energy-saving pace can reach more than 20%.It can be applied to the intelligent management of 5G BS energy consumption to improve the energy efficiency of 5G networks.
作者
郑佳欢
向勇
ZHENG Jiahuan;XIANG Yong(Research Institute of China Telecom Co.,Ltd.,Guangzhou 510630,China)
出处
《移动通信》
2021年第3期91-96,共6页
Mobile Communications
