权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Security Testbed for Internet of Things

物联网安全测试平台

基本信息

批准号：
18K11287
负责人：
シルバーストントーマス
金额：
$ 2.83万
依托单位：
Shibaura Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2018
资助国家：
日本
起止时间：
2018-04-01 至 2022-03-31
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-18K11287/
关键词：
Internet of Things Network Traffic Network Entropy Cybersecurity Network Model Testbed IoT Traffic Model Traffic Analysis Networking

项目摘要

This project focuses on IoT Traffic and cybersecurity. We set an IoT Testbed and collected traffic; based on our observations, we were able to design a new IoT Traffic generator tool, which is an essential tool to help detecting anomalies for future IoT Traffic.We observed the collected traffic and the generated traffic from IoTTGen through traffic entropy and we showed that our generator is able to capture the essential characteristics of IoT Traffic. We then add IoT traffic anomaly into our data set. We observe the impact on the traffic entropy on both generated and synthetic data set. Our Tool is therefore accurate to model IoT traffic and help detecting IoT anomaly.Based on this observation we propose a new framework based on Machine Learning algorithms to identify devices based on their traffic profile. Our framework allows detecting very precisely the nature of smart home IoT devices and is also able to distinguish the devices with anomalies. Indeed, our results showed that more than 90% of devices were properly identified and also in the case of cybersecurity threats. It is an important step toward the recognition of legitimate IoT traffic pattern with regards to anomalous ones.

该项目侧重于物联网流量和网络安全。我们建立了一个物联网测试床并收集了流量;根据我们的观察，我们能够设计一个新的物联网流量生成器工具，这是一个重要的工具，可以帮助检测未来物联网流量的异常。我们通过流量熵来观察收集的流量和从IoTTGen生成的流量，我们表明我们的生成器能够捕获物联网流量的基本特征。然后，我们将物联网流量异常添加到我们的数据集中。我们观察到对生成的和合成的数据集上的交通熵的影响。因此，我们的工具可以准确地对物联网流量进行建模，并帮助检测物联网异常。基于这一观察，我们提出了一个基于机器学习算法的新框架，以根据流量配置文件识别设备。我们的框架允许非常精确地检测智能家居物联网设备的性质，并且还能够区分具有异常的设备。事实上，我们的研究结果表明，超过90%的设备都得到了正确的识别，在网络安全威胁的情况下也是如此。这是识别合法物联网流量模式与异常流量模式的重要一步。