Analysis of anomaly detection method for Internet of things based on deep learning,Transactions on Emerging Telecommunications Technologies

当前位置： X-MOL 学术 › Trans. Emerg. Telecommun. Technol. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Analysis of anomaly detection method for Internet of things based on deep learning
Transactions on Emerging Telecommunications Technologies ( IF 2.5 ) Pub Date : 2020-04-16 , DOI: 10.1002/ett.3893
Wei Ma _{1,

2,

3}

Affiliation

With the rapid development of the Internet of things technology, the connection between things and people is realized in a real sense, and the intelligent perception, recognition, and management of goods and processes are also achieved. Cloud computing, as one of the core technologies of the Internet of Things, has been widely used in online services of various networks, but the generation of abnormal data will affect the service performance of cloud computing systems. Therefore, effective detection of abnormal data is of great significance to improve the efficiency of the system. Because of the large amount of data and nonlinear distribution in the cloud computing system, the accuracy of traditional methods is low. Based on this, this article proposes a deep learning algorithm based on recurrent neural network (RNN) to implement anomaly detection in the cloud computing system. Based on the basic principle of the RNN algorithm, this article analyses the properties and defects of the activation functions commonly used in RNN, and then improves the RNN algorithm, so as to realize the effective detection of abnormal data in cloud computing system. The simulation results show that the optimized RNN deep learning algorithm for anomaly detection in cloud computing system can effectively improve the detection success rate, effectively reduce the detection time and cost, show strong robustness, and effectively improve the online service efficiency of the Internet of things technology.

中文翻译：

基于深度学习的物联网异常检测方法分析

随着物联网技术的飞速发展，物与人之间的联系得以真正实现，并实现了对商品和过程的智能感知，识别和管理。云计算作为物联网的核心技术之一，已经广泛应用于各种网络的在线服务中，但是异常数据的产生会影响云计算系统的服务性能。因此，有效检测异常数据对提高系统效率具有重要意义。由于云计算系统中的大量数据和非线性分布，传统方法的准确性较低。基于此，本文提出了一种基于递归神经网络（RNN）的深度学习算法，以在云计算系统中实现异常检测。基于RNN算法的基本原理，分析了RNN常用的激活函数的性质和缺陷，并对RNN算法进行了改进，以实现对云计算系统中异常数据的有效检测。仿真结果表明，针对云计算系统中异常检测的优化RNN深度学习算法可以有效提高检测成功率，有效减少检测时间和成本，显示出强大的鲁棒性，并有效提高物联网在线服务效率技术。本文分析了RNN常用的激活函数的性质和缺陷，并对RNN算法进行了改进，以实现对云计算系统中异常数据的有效检测。仿真结果表明，针对云计算系统中异常检测的优化RNN深度学习算法可以有效提高检测成功率，有效减少检测时间和成本，显示出强大的鲁棒性，并有效提高物联网在线服务效率技术。本文分析了RNN常用的激活函数的性质和缺陷，并对RNN算法进行了改进，以实现对云计算系统中异常数据的有效检测。仿真结果表明，针对云计算系统中异常检测的优化RNN深度学习算法可以有效提高检测成功率，有效减少检测时间和成本，显示出强大的鲁棒性，并有效提高物联网在线服务效率技术。

更新日期：2020-04-16

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11