基于深度学习的物联网动态数据完整性检测方法  

Dynamic data integrity detection method of Internet of things based on deep learning

作　　者：周挺[1] ZHOU Ting(Xi’an Aeronautical Polytechnic Institute,Xi’an 710089,China)

机构地区：[1]西安航空职业技术学院,西安710089

出　　处：《自动化与仪器仪表》2021年第5期4-8,共5页Automation & Instrumentation

基　　金：西安航空职业技术学院自然科学类科研项目“语音编程教学软件研究与设计”(No.19XHZK-022)。

摘　　要：受到物联网动态采集数据质量的影响,导致传统的动态数据完整性检测方法存在检测精度低的问题,为此,提出了基于深度学习的物联网动态数据完整性检测方法。首先分析物联网环境下的数据结构,并以此确定数据完整性的参考标准。选择PDP作为数据完整性的验证机制,利用硬件设备实时采集物联网中的动态数据,并利用深度学习理论中的循环神经网络算法实现对初始数据的处理,提升采集数据的质量。将物联网动态数据处理结果作为输入项,导入到完整性验证机制中,并在检测与数据更新协议的约束下,得出数据完整性的检测结果。通过与传统方法的对比实验得出结论:通过深度学习算法的应用,数据完整性的检测误差降低了约0.55%,且在时间开销方面更加具有优势。Due to the influence of the quality of the dynamic data acquisition of the Internet of things, the traditional dynamic data integrity detection methods have the problem of low detection accuracy.Therefore, a dynamic data integrity detection method based on deep learning of the Internet of things is proposed.Firstly, the data structure under the environment of Internet of things is analyzed, and then the reference standard of data integrity is determined.The PDP is chosen as the verification mechanism of data integrity, and the dynamic data in the Internet of things is collected in real time by using hardware equipment.The cyclic neural network algorithm in deep learning theory is used to process the initial data and improve the quality of the collected data.The dynamic data processing results of the Internet of things are imported into the integrity verification mechanism as input items, and the detection results of data integrity are obtained under the constraints of detection and data update protocol.Compared with the traditional methods, the results show that: through the application of deep learning algorithm, the detection error of data integrity is reduced by about 0.55%,and it has more advantages in time cost.

关 键 词：深度学习 物联网数据 动态数据 数据完整性 完整性检测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]