Electric Power ›› 2022, Vol. 55 ›› Issue (5): 158-165.DOI: 10.11930/j.issn.1004-9649.202006211

• Information and Communication • Previous Articles     Next Articles

Information Entropy Based Multi-Source Power IoT Terminal Equipment Trust Degree Evaluation Method

ZHAI Feng1, FENG Yun1, CHENG Kai2, CAI Shaotang3, YU Liying3, YANG Ting3   

  1. 1. China Electric Power Research Institute, Beijing 100192, China;
    2. State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050021, China;
    3. School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
  • Received:2020-06-23 Revised:2021-09-20 Online:2022-05-28 Published:2022-05-18
  • Supported by:
    This work is supported by Science and Technology Project of SGCC(End-to-End Security Threat Analysis and Accurate Protection Technology of Ubiquitous Power Internet of Things, No.5700–201958466A-0-0-00)

Abstract: Power Internet of things terminal (IoT) equipment is vulnerable to identity camouflage, information theft, data tampering and other security threats. Traditional security methods can not resist the network internal attacks from damaged terminals. Trust evaluation system is an effective mechanism to protect power IoT terminals from internal attacks. A trust evaluation method is proposed based on information entropy for power wireless private network communication terminals. Firstly, the direct trust value is estimated by the credibility model based on exponential distribution, and then the sliding window and forgetting factor are used to update the direct trust value. According to the entropy theory, the uncertainty of direct trust value is measured, and the indirect trust value is introduced to make up for the inaccuracy of direct trust judgment, and the judgment accuracy is improved through comprehensive evaluation of both. Simulation results show that the proposed method can effectively resist switch attacks and collusion attacks, and compared to the binomial trust management method and beta distribution based trust evaluation method, it can better evaluate malicious terminals and normal terminals.

Key words: power internet of things, exponential distribution, information entropy, cyber attack, trust evaluation