摘要
对物联网网络节点能耗进行实时估算,可以提高网络安全性能和网络寿命。进行物联网网络节点能耗检测时,需要获取物联网网络节点能耗序列。对序列进行指数平滑的预处理,消除节点检测过程的干扰,但是传统方法依据节点至基站的距离和近邻节点的总量,对待选节点转发数据能耗实现检测,但是缺少节点能耗序列预处理过程,在进行检测时存在大量的干扰,降低了络节点能耗实时检测的准确性。提出基于人工智能的物联网网络节点能耗实时检测方法。首先融合于灰色理论的思想对原始网络节点能耗序列进行新陈代谢等维处理,组建新的节点能耗序列。对上述序列进行指数平滑,将灰色理论与马尔可夫模型相结合组建网络节点能耗的检测状态转移矩阵,利用人工智能方法对上述矩阵检测结果进行训练和验证,有效地完成了对物联网网络节点能耗实时检测。仿真结果表明,所提方法可以大幅度地提升物联网网络寿命。
In this paper, we proposed an energy consumption real - time detection method of network nodes in the Internet of things based on artificial intelligence. Firstly, we integrated the grey theory to make dimension processing of metabolism with the energy consumption sequence of original network node, and built a new sequence of node energy consumption. The sequence was smoothed exponentially. Then we integrated the grey theory with the Markov model to build the detecting state transition matrix of network node＇s energy consumption. Finally, we used the artificial intelligence method to train and verify the detection result of the matrix and completed the energy consumption real - time detection of network node in the Internet of Things. The simulation results show that the method mentioned above can improve the network lifetime of Internet of things apparently.
出处
《计算机仿真》
CSCD
北大核心
2016年第12期453-456,共4页
Computer Simulation
基金
2011年国家自然科学基金项目(61072087)
关键词
人工智能
物联网网络
节点能耗实时检测
Artificial intelligence
Internet of things network
Node energy consumption real - time detection
