摘要
当无线传感网络受到不同攻击类型的入侵时,当前的无线传感网络入侵检测系统对于入侵节点的检测率和数据传输成功数量过低,因此基于隐马尔可夫模型(Hidden Markov Model, HMM)设计了一种无线传感网络入侵检测系统。首先,基于HMM建立检测流程,分析HMM结构;其次,将观测序列之间匹配的概率作为入侵检测的判断依据,并提取入侵攻击特征,包括节点的工作状态、活动情况、位置等信息;最后,优化入侵检测算法,将传感器节点和数据中心分离进行匹配计算,最大限度化简计算步骤,提升系统检测效率。系统性能测试结果表明,设计的系统在不同的攻击类型下,测试周期内入侵节点检测数量和成功接收的数据量均高于原有系统,验证了系统的高可靠性。
When the wireless sensor network is invaded by different types of attacks,the detection rate and the number of successful data transmission of the current wireless sensor network intrusion detection system for intrusion nodes are too low.For this reason,a wireless sensor network intrusion detection system based on Hidden Markov Model(HMM)is designed.First,the detection process is established based on the HMM,and the structure of HMM is analyzed.Then,the probability of matching between observation sequences is used as the judgment basis for intrusion detection,and the characteristics of intrusion attacks are extracted,including the working status,activity and location of nodes.Finally,the intrusion detection algorithm is optimized,and the sensor nodes and data centers are separated for matching calculation to minimize the calculation steps,improve system detection efficiency.The system performance test results show that under different attack types,the number of intrusion nodes detected and the amount of data successfully received in the test cycle of the designed system are higher than the original system,which verifies the high reliability of the system.
作者
李秋月
晁绪耀
LI Qiuyue;CHAO Xuyao(College of Information Engineering and Big Data,Zhengzhou Technical College,Zhengzhou Henan 451450,China)
出处
《信息与电脑》
2023年第3期204-206,共3页
Information & Computer
作者简介
李秋月(1992-),女,河南周口人,硕士研究生,助教。研究方向:网络信息安全、物联网技术、大数据、人工智能。
