摘要
物联网终端信道连续数据流中潜在的可以被旁路功耗攻击的节点被定义为脆弱点。该类节点一旦被攻击成功,将严重影响物联网终端的通信能力。为此,进行物联网终端信道连续数据流脆弱点快速识别研究。利用简单网络管理协议(Simple Network Management Protocol,SNMP)捕捉物联网终端信道连续数据流并实施离散化处理,提取信息熵、流入流出比、平均流长及单边连接密度四类特征。以提取特征为输入,利用支持向量机识别其是否为脆弱点,通过人工神经网络进一步确定脆弱点的脆弱程度,完成物联网终端信道连续数据流脆弱点快速识别。结果表明,所提方法的连续数据流脆弱点识别质量指数最大、时间复杂度最小,由此说明该方法的识别准确性更高,速度更快。
Potential nodes in the continuous data stream of the Internet of Things terminal channel that can be attacked by bypass power consumption are defined as vulnerabilities.Once such nodes are successfully attacked,the communication capability of the Internet of Things terminals will be seriously affected.Therefore,the research on rapid identification of vulnerability of continuous data flow in the Internet of Things terminal channel is carried out.The Simple Network Management Protocol(SNMP)is used to capture the continuous data flow of the Internet of Things terminal channel and implement the discretization process,and extract four characteristics:information entropy,inflow/outflow ratio,average flow length and unilateral connection density.Taking the extracted feature as input,support vector machine is used to identify whether it is a vulnerability.Further determine the vulnerability of vulnerable points through Artificial neural network,and complete the rapid identification of vulnerability of continuous data flow in the Internet of Things terminal channel.The results show that the proposed method has the largest quality index and the smallest time complexity for vulnerability identification of continuous data flow,which shows that the method has higher accuracy and faster speed.
作者
于刘
YU Liu(Shanghai Aviation Industrial(Group)Co.,Ltd.,Shanghai 201206,China)
出处
《电子设计工程》
2024年第9期161-164,169,共5页
Electronic Design Engineering
关键词
物联网
终端信道
连续数据流
脆弱点
支持向量机
人工神经网络
Internet of Things
terminal channel
continuous data flow
vulnerability
Support Vector Machine
Artificial Neural Network
