Scholarlink

<p>Cognitive Internet of Things (CIoT) uses cognitive radio technology in the Internet of Things (IoT). It can effectively solve the problem of spectrum shortage and enable IoT devices to better access the network. But malicious IoT devices (MIDs) in the CIoT may launch Spectrum Sensing Data Forgery (SSDF) attacks to disrupt network performance. Especially when these MIDs collude, the consequences may be more serious. For solving these problems, This paper adopts an association rule mining algorithm to detect collusion SSDF attacks. For a large number of devices in CIoT, the FP-growth detection algorithm based on cluster division is proposed in the paper. We divide IoT devices in a certain area into sever sevices in each sub-cluster only need to send the perception report to the sub-fusion center (S-FC) to which it belongs, and execute the FP-growth detection algorithm based on cluster division in the S-FC to identify colluding malicious IoT devices (C-MIDs) and filter them out. Its sensing report, only the perception data of is finally sent to the fusion center (FC), and the FC performs data fusion on the received sensing report and makes a global decision. Simulation results show that the proposed FP-growth detection algorithm based on cluster division is superior to traditional methods in detection accuracy and execution time, and achieves better detection results for different types of collusion SSDF attacks in CIoT.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cognitive IoT Collusion SSDF Attack Detection Based on FP-Growth Algorithm

  • 期刊论文

  • 发表日期：

    2025年1月27日

查看全文
  • Miao Liu,
  • Longxi Liu,
  • Di Xu,
  • Li Zhang

摘要

Cognitive Internet of Things (CIoT) uses cognitive radio technology in the Internet of Things (IoT). It can effectively solve the problem of spectrum shortage and enable IoT devices to better access the network. But malicious IoT devices (MIDs) in the CIoT may launch Spectrum Sensing Data Forgery (SSDF) attacks to disrupt network performance. Especially when these MIDs collude, the consequences may be more serious. For solving these problems, This paper adopts an association rule mining algorithm to detect collusion SSDF attacks. For a large number of devices in CIoT, the FP-growth detection algorithm based on cluster division is proposed in the paper. We divide IoT devices in a certain area into sever sevices in each sub-cluster only need to send the perception report to the sub-fusion center (S-FC) to which it belongs, and execute the FP-growth detection algorithm based on cluster division in the S-FC to identify colluding malicious IoT devices (C-MIDs) and filter them out. Its sensing report, only the perception data of is finally sent to the fusion center (FC), and the FC performs data fusion on the received sensing report and makes a global decision. Simulation results show that the proposed FP-growth detection algorithm based on cluster division is superior to traditional methods in detection accuracy and execution time, and achieves better detection results for different types of collusion SSDF attacks in CIoT.