i46

i46 logo white
菜单

Our projects

Our flagship projects are where innovation meets impact.

DARK-5G (2025)

✍️ Project Description


DARK-5G delivers reliable 5G connectivity in underground environments to support emergency response. The solution combines the i46 portable 5G Standalone (SA) network with Rinicom’s leaky cable system, ensuring continuous coverage in tunnels and confined spaces. Validated at Patras5G/PNET using the FIDAL platform, it supports advanced features such as network slicing, mission-critical (MCX) services, and AI-driven operations. Complementary IoT devices enhance detection and extend connectivity, resulting in a resilient, rapidly deployable network for real-world emergency conditions.

 

🎯 Project Goals

  • Develop a portable, rapidly deployable 5G network solution specifically designed for emergency communications in challenging underground environments (e.g., subway tunnels, mining sites, collapsed buildings).
  • Integrate advanced 5G standalone architecture with resilient hardware (network-in-a-box and leaky feeder cables) to overcome signal blockage and infrastructural obstacles where conventional networks fail.
  • Implement IoT-based mobile device detection using WiFi and 5G sensors to enable real-time localization and situational awareness of survivors and responders in low-visibility or hazardous conditions.
  • Leverage AI-driven resource optimization and network slicing to ensure prioritized, ultra-reliable communications for first responders, while maintaining connectivity for trapped individuals.
  • Validate the system through comprehensive laboratory and field trials, demonstrating operational readiness, scalability, and seamless integration with the FIDAL experimental framework.

 

🔍 Project Outcome

Currently in progress.

 

💰 Funded by
FIDAL – Field Trials beyond 5G

SHIELD (Jul, 2025 - Dec, 2025)

✍️ Project Description


The SHIELD project, developed under the CyberSecDome Open Call framework and led by i46 s.r.o., aims to advance cybersecurity defenses by integrating a state-of-the-art Intrusion Prevention System (IPS) into network flow analysis pipelines.
The project focuses on the creation and delivery of high-quality, representative network traffic datasets to train and validate machine learning models for intrusion detection and prevention. These datasets include both realistic attack scenarios (e.g., port scanning, DDoS, ping floods, insider threats) and benign activities (e.g., OS updates), captured in packet capture (PCAP) format to allow replay, labeling, and detailed analysis.

The work is guided by user and technical requirements agreed with CyberSecDome stakeholders, ensuring alignment with real-world use cases. Dissemination and a robust Data Management Plan run in parallel to guarantee ethical and secure handling of all data.


🎯 Project Goals

  • Develop and validate datasets in PCAP format that accurately capture both attack and benign scenarios.
  • Enable training and evaluation of AI-driven IPS models, achieving high detection accuracy while minimizing false positives.
  • Ensure secure, ethical, and compliant data governance aligned with GDPR and EU AI standards.

 


🔍 Project Outcome

Currently in progress.

💰 Funded by
CyberSecDome

✍️ Project Description


The CATT-46 project (CyberSecDome AI Tool Testing with i46) rigorously evaluates CyberSecDome’s AI-driven penetration testing tools on the i46.io platform—an IoT compliance monitoring system supporting the EU’s Cyber Resilience Act (CRA). Leveraging a representative testbed, the project establishes a dedicated testing environment integrating cloud servers, local servers, and geographically distributed IoT devices.

CATT-46 provides a dedicated testbed to support the diverse attack scenarios. The scenarios include website penetration (with and without credentials), server- and client-side intrusions, and API-level exploitation, to assess the tools’ performance under real-world conditions.


🎯 Project Goals

  • Simulate real-world IoT cyber threats (malware, DDoS, data exfiltration, phishing, etc.) in a controlled environment.
  • Evaluate the effectiveness of CyberSecDome’s AI-driven penetration testing tools across multiple scenarios.
  • Achieve ≥90% tool utilization
  • Attain ≥90% vulnerability discovery rate
  • Ensure ≥90% resilience assessment accuracy
  • Deliver actionable recommendations to enhance CyberSecDome’s tools, improving accuracy, usability, and adaptability.



🔍 Project Outcome

Currently in progress.


💰 Funded by
CyberSecDome

5G-SAIIoT (2024 - 2025)

Ready to explore our solutions?

Let's talk
zh_CNChinese
en_USEnglish zh_CNChinese