Most of intrusions detection systems are based on centralized architecture, what makes them vulnerable in case of denial-of-service attacks. Furthermore, the performance and response time of intrusions detection systems could be negatively influenced by network traffic density.
In this project, we proposed an intrusion detection system based on decentralized architecture through a multi agent system. In particular, mobile agents are used to analyze the network traffic, which can improve the performance of the system. To analyze and detect intrusions, we used machine learning classifiers such as decision tree, support vector machine and neuronal networks.
We’ve also developed a mobile application to help network administrator in supervising the network traffic in order to insure quick intervention in case of anomalies.
link to mobile application: https://github.com/chihab4real/MAS_IDS_UI
| Dashboard | Details about classifiers |
|---|---|
 |
 |
| Performance Metrics | Performance Metrics |
|---|---|
 |
 |
| System Logs | Platform Settings |
|---|---|
 |
 |
Real-time Monitoring
| 
| 
|
| 
|