Tracenox: Towards Adaptive Cybersecurity using Hybrid Deep Learning and FSM Models

Tracenox is a real-time, AI-powered Intrusion Detection System (IDS) designed for monitoring, analyzing, and defending network traffic. It integrates deep learning and rule-based heuristics (FSM) for hybrid threat detection, offering high accuracy and adaptive response to evolving network attacks.Using Deeplearning and FSM

---

Features

• Real-time Intrusion Detection using a hybrid Deep Learning + FSM model
• Network Traffic Monitoring with live analysis of packet flows
• Attack Classification covering DDoS, DoS variants, Web attacks, Bot attacks, Infiltration, Port Scans, and more
• High Accuracy with deep learning achieving 96%+ detection
• Configurable Thresholds for alerts, confidence margins, and automatic threat blocking
• File Upload & Testing for CSV, PCAP, and PCAPNG formats (Max 100MB)
• Visual Dashboards for system overview, traffic monitoring, attack timelines, and detailed logs

---

Security Dashboard

Provides a high-level overview of network health and threat landscape.

• Total Packets: 268,038 (+12% from last hour)
• Attacks Detected: 20 (5 critical)
• Detection Accuracy: 96.3%
• System Status: Active, all systems operational
• Attack Types Distribution: DDoS, Web Attacks, Bot Attacks, Infiltration, Port Scan
• Attack Timeline (24h): Hourly detection of attacks

---

Real-time Traffic Monitoring

Monitor ongoing network traffic, identify suspicious flows, and classify attacks in real-time.

• Live Network Traffic Table with:
  • Timestamp
  • Source & Destination IP
  • Protocol
  • Attack Type & Severity
  • Detection Confidence
  • Number of Packets
• Filter traffic by severity: All / Critical
• Refresh data in real-time for continuous monitoring

---

Upload & Test Network Data

Test and analyze custom network data files or manually input flow information.

• File Upload: CSV, PCAP, PCAPNG (Max 100MB)
• Manual Entry: Enter network flow features like Source IP, Destination IP, Protocol, and Packet Count
• Start Analysis: Run the hybrid DL+FSM detection on uploaded or manually entered data

---

System Settings

Configure detection thresholds, FSM parameters, and system preferences.

• Detection Threshold: Minimum confidence to trigger alerts
• Confidence Margin: FSM state transition confidence threshold
• Log Retention: Days of logs to retain
• Max Concurrent Connections: Limit for live monitoring
• Auto-block Threats: Automatically block detected malicious traffic
• Real-time Monitoring: Enable continuous traffic analysis
• Alert Notifications: Send alerts for critical threats
• System Status: CPU usage, memory, uptime, and version information

---