For both of the use cases, we can generate our own datasets and train and test the models on them. In the following paragraphs, we briefly present how we can create a dataset for device-level host intrusion detection through a DoS attack.
Data collection
CPU utilisation data
For the DoS attack, we need an attack machine and a target machine. We are using a Kali Linux machine as the attacker and a Windows machine as the target (which can be home gateway/Raspberry Pi 3/sensors). In Kali Linux, a DoS attack can be achieved in multiple ways. One way is to use the hping3. hping3 command as a network tool to send custom TCP/IP packets, and allowing for the testing of firewalls, port scanning, address spoofing, and more.
This can...