北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2017, Vol. 40 ›› Issue (3): 76-84.doi: 10.13190/j.jbupt.2017.03.011

• 论文 • 上一篇    下一篇

时空域信息融合的智能家居入侵检测算法

李民政1,2,3, 蓝剑平1   

  1. 1. 桂林电子科技大学 计算机与信息安全学院, 桂林 541004;
    2. 广西信息科学实验中心, 桂林 541004;
    3. 广西可信软件重点实验室, 桂林 541004
  • 收稿日期:2017-02-12 出版日期:2017-06-28 发布日期:2017-05-25
  • 作者简介:李民政(1972-),男,教授;蓝剑平(1991-),男,硕士研究生,E-mail:jianpinglan@guet.edu.cn.
  • 基金资助:
    国家自然科学基金项目(61362007);广西信息科学实验中心项目(YB1407);广西可信软件重点实验室基金(KX201414)课题

Smart Home Intrusion Detection Algorithm Based on Spatial-Temporal Field Information Fusion

LI Min-zheng1,2,3, LAN Jian-ping1   

  1. 1. School of Computer and Information Security, Guilin University of Electronic Technology, Guilin 541004, China;
    2. Guangxi Information Science Research Centre, Guilin 541004, China;
    3. Guangxi Key Laboratory of Trusted Software, Guilin 541004, China
  • Received:2017-02-12 Online:2017-06-28 Published:2017-05-25

摘要: 针对现有智能家居入侵检测算法收敛速度慢及抗干扰能力差进而导致决策系统实时性和鲁棒性差的缺陷,分析了时空域证据融合的特性,得出时域融合存在冲突融合、空域融合更存在去除冲突融合的结论.基于此,提出了基于证据理论的时域自适应加权算法及空域证据修正的3种证据融合入侵检测算法,并将提出的检测算法应用于智能家居入侵检测系统.检测结果表明,提出的算法能够加快融合结果的收敛速度,增强抗干扰能力,并能提高入侵检测系统决策的实时性和鲁棒性.

关键词: 时空域, 信息融合, 证据理论, 入侵检测

Abstract: The existing smart home intrusion detection algorithm has the defects of slow convergence speed and weak anti-interference ability which can weaken the real-time and robustness of decision-making system. By analyzing the characteristics of time domain data and spatial domain evidence, it is concluded that the temporal fusion is focused more on conflict data but more attention is paid to remove conflict data fusion in the spatial domain. Therefore, an adaptive weighted algorithm for temporal fusion and three evidence fusion intrusion detection solution based on modified evidence were proposed. Further analysis is done by applying the algorithm in intelligent house intrusion detection system. Simulations verify that the fusion result of new proposed algorithms not only accelerate the convergence rate and enhance anti-jamming capability comparing with existing algorithms, but also improve the real-time and robustness of intrusion detection system decision-making.

Key words: spatial-temporal field, information fusion, evidence theory, intrusion detection

中图分类号: