基于物联网的异构传感数据入侵风险识别方法

Intrusion risk identification method of heterogeneous sensor data based on Internet of things

摘要:

异构传感数据入侵风险识别过程中未对采集的异构传感数据进行融合处理，导致数据入侵风险识别完整性较差，为此，本文引入物联网技术，提出一种新的异构传感数据入侵风险识别方法。根据物联网的组成层次，构建物联网结构模型。依据不同网络入侵攻击类型，设置风险类型识别标准。在物联网结构模型下，采集异构传感数据，得出初始数据的融合处理结果。从传感数据结构以及数据时域变化两个方面，提取异构传感数据特征，计算入侵风险值。最终输出可视化的异构传感数据入侵风险等级以及类型的识别结果。实验结果表明，设计识别方法的风险值识别误差降低了0.015，风险类型识别正确率提高了1.6%，且风险识别方法的响应时间更短，即优化设计的入侵风险识别方法在精度和时效性两个方面更加具有优势。

Abstract:

In the process of heterogeneous sensor data intrusion risk identification, the collected heterogeneous sensor data is not fused, which leads to the poor integrity of data intrusion risk identification. Therefore, this paper introduces the Internet of things technology and proposes a new method of heterogeneous sensor data intrusion risk identification.According to the composition level of the Internet of things, build the structure model of the Internet of things. According to different types of network intrusion attacks, set up risk type identification standards. Under the structure model of the Internet of things, collect heterogeneous sensor data and get the fusion processing results of the initial data.From the two aspects of sensor data structure and data time-domain changes, extract the characteristics of heterogeneous sensor data, and calculate the intrusion risk value. Finally, the visual recognition results of heterogeneous sensor data intrusion risk level and type are output.The experimental results show that the risk value identification error of the design identification method is reduced by 0.015, the accuracy of risk type identification is increased by 1.6%, and the response time of the risk identification method is shorter, that is, the optimized intrusion risk identification method has more advantages in accuracy and timeliness.