Model Learning and Model Checking of SSH Implementations

The dataset contains source code and data for the paper 'Model Learning and Model Checking of SSH Implementations'. It includes Java and Python code for an automata learning setup, learned Mealy Machine models for three SSH server implementations, and utilities for model conversion and beautification. The work was authored by Paul Fiterău-Broştean of Radboud University Nijmegen.

Use Cases

• Reusing the learner setup to connect to a different system over sockets based on the provided Java code.
• Converting any Mealy Machine .dot file to an .smv file suitable for the NuSMV model checker using the included converter.
• Analyzing or extending the learned Mealy Machine models for three SSH server implementations (BitVise, DropBear, OpenSSH).
• Applying the beautifying Python program to make complex .dot files more readable.

Strengths

• Includes learned Mealy Machine models for three distinct SSH server implementations (BitVise, DropBear, OpenSSH).
• Provides a complete experimental setup with Java learner code, Python mapper code, and conversion utilities.
• The dataset is associated with a peer-reviewed ACM paper, providing context and validation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Radboud University Nijmegen

Collection Method

Created as part of a research project using automata learning with abstraction to infer models of SSH implementations.