Network-Security-Based-On-ML
NIPS — Machine-Learning-Based Network Intrusion Prevention System | Kitsune/LUCID integrated | Online unsupervised anomaly detection | Kernel-level real-time tr
What is Network-Security-Based-On-ML?
NIPS — Machine-Learning-Based Network Intrusion Prevention System | Kitsune/LUCID integrated | Online unsupervised anomaly detection | Kernel-level real-time traffic interception | FastAPI + CLI management
Overview of Network-Security-Based-On-ML
What it is
NIPS (Network Intrusion Prevention System) is a server-side IPS that intercepts traffic on Linux, scores each packet through a rule engine plus an anomaly detector, and drops malicious packets via iptables. The rule engine is deterministic and always present; learning detectors (Kitsune, LUCID) are plug-ins off by default. The system uses NFQUEUE for live interception and provides a FastAPI + CLI management interface.
Key features
- Rule engine with blacklist, whitelist, protocol filtering, rate limiting, and signature matching - Kitsune (NDSS'18): online unsupervised anomaly detection using AfterImage incremental statistics and KitNET autoencoder ensemble - LUCID (IEEE TNSM 2020): optional 1D CNN-based DDoS detector requiring a trained TensorFlow model - Kernel-level real-time traffic interception via nfqueue and iptables - FastAPI REST API and CLI for management and monitoring - Pluggable detector architecture: third-party detectors can implement a simple contract - Configurable grace periods, threshold percentiles, and blocking escalation policies
Use cases
- Real-time network intrusion prevention on Linux servers - Online unsupervised anomaly detection without labeled training data - DDoS attack mitigation using CNN-based detection (LUCID)
- Traffic filtering and rate limiting based on deterministic rules
Getting started
bash git clone https://github.com/zimingttkx/Network-Security-Based-On-ML.git cd Network-Security-Based-On-ML python -m venv venv source venv/bin/activate pip install -r requirements.txt # Optional: for LUCID detector pip install -e .[lucid]
License and pricing
MIT license. Open source.
Based on the project's public README and repository metadata.