# Structure

This repository is organized into three main directories—`CWEs`, `Firmware`, and `Machine_Learning`—each grouping related artifacts for fault analysis and detection.



## Directory Descriptions

- **CWEs**  
  Contains subfolders grouping specific Common Weakness Enumeration (CWE) mappings for detected fault motifs in ASCON micro-architecture.  
  - `Spinner_fault` and `Tunnel_fault` map detected spinner and tunnel faults to their corresponding CWE identifiers.

- **Firmware**  
  Source code for various hardware targets used in fault-injection and detection experiments:  
  - **IBEX, MSP430, PICO**: each folder houses C implementations and test harnesses tailored to that core.

- **Machine_Learning**  
  Code and notebooks for training and evaluating machine-learning models on the extracted state-transition graphs:  
  - **GNN_analyser_fault_patterns_detection**: Graph Neural Network pipelines for multi-cycle fault-pattern detection.  
  - **Single-Chain_Driven_Fault_Classification(LLM)**: LoRA- and LLM-based classification experiments across six dataset configurations.