[文献阅读笔记]Machine Learning DDoS Detection for Consumer Internet of Things Devices 2018 IEEE SPW

1.INTORDUCTION

1. 提出了IOT设备往往具有与其他网络设备不同的网络行为，例如IoT设备发送数据的目标总是固定的而不是大范围的web服务器，IoT设备更倾向于形成重复性的流量模式，例如出于记录需要在固定的时间发送小包进行周期性的网络连接测试。

Fortunately,however,IoT traffic is often distinct from from that of other Internet connected devices(e.g. laptops and smart phone.) For example,IoT devices are also more likely to have repetitive network traffic patterns, such as regular network ping with small packets at fixed time intervals for logging purpose.

2. 文章中的测试环境部署在了家庭网络环境中的网关上，或者ISP控制的交换机上。

Therefore, it is well suited for deployment on consumer home gateway routers or ISP-controlled switches.

2.BACKGROUND AND RELATED WORK

1.IoT设备的正常网络通信种类往往是有限的，例如一个智能灯泡，只有“On”,“Off”,“Connecting to Wi-Fi”,三种状态，每种状态都存在对应的，有明显差别的网络