The CapriTech team enjoys writing software mainly for evaluating its research findings. Therefore, during the last years a number of software have been developed; these play key role in demonstrations required as part of research and innovation projects. CPT proposes the extension of this software in such projects so that a higher Technology Readiness Level (TRL) is achieved. Apart from this, CPT contributes to the integration of its software with other interfacing software modules to achieve the desired innovation outcome, such as: technology demonstrated in relevant environment (TRL 6); system prototype demonstration in operational environment (TRL 7); system complete and qualified (TRL 8). CPT's prominent sofware include:
5G-SAT (5G Security Analysis and visualisation Tool)
5G-SAT is a graphical security analysis tool for 5G systems. It is a fork of the IoT security analysis tool ASTo (Apparatus Software Tool) which is developed to support the Apparatus security framework. The Apparatus Framework provides a modeling language and analysis procedures for a 5G system during the following phases:
- design phase (model the idea of a system) [high-level concepts]
- design phase state (model the idea of a system different states)
Each phase has different concepts and rules on how those concepts interact with each other.
The concepts of each phase are defined via UML class diagrams that in turn define the metamodels of Apparatus.
The metamodels are translated into schemas that ASTo uses to validate modules. The Apparatus Framework uses
a graph-based front-end representation of models. In 5G-SAT, we leverage powerful graph-based algorithms for a variety
of 5G system modeling and security analysis tasks.
Link to GitHub Repository
IDCM4AS (Intrusion Detection and Classification Module for Autonomous Systems)
IDCM4AS is a specialized software for the detection and classification of cyber
and cyber-physical attacks against autonomous systems (ASs). In IDCM4AS, we leverage the AS's cyber features and physical features
for implementing the detection logic and then using lightweight heuristic techniques or advanced deep learning techniques,
we decide whether the autonomous system is in an attack state or not. As an extension, IDCM4AS employs a method based on
Bayesian Networks to determine the domain (cyber or physical) from which the attack originated from.
FINDER (Finding INtruDERs based on advanced machine learning)
FINDER is a cloud-based intrusion detection tool which enables the identification of malicious activities from different viewpoints of the network overcoming the deficiency of classical intrusion detection tools. By understanding the "attack logic" of the advanced and determined adversary harnessing powerful machine-learning techniques and security visualisations, FINDER can detect a wide variety of cyber attacks in private and public clouds.