Resilient Edge: Architectural Optimization for Real-Time Network Fault Diagnosis on Resource-Constrained Embedded Systems

Experimental Dataset & Codebase: TinyML for Embedded Network Fault Diagnosis

This repository contains the raw experimental datasets, the revised machine learning pipeline (Jupyter Notebook), and the compiled C++ hardware deployment model from a longitudinal experimental campaign. These artifacts evaluate the "Resilient Edge" TinyML framework on severely resource-constrained ESP32 microcontrollers. The codebase and data were generated to validate a novel Decision Tree Distillation technique and physics-based feature extraction pipeline designed for noisy and attenuated 2.4 GHz wireless environments.

Note: The standalone, comprehensive raw telemetry dataset corresponding to this prior work is available at DOI: https://doi.org/10.5281/zenodo.18470802

📄 ASSOCIATED MANUSCRIPT

This dataset and codebase support the findings presented in the research article prepared for The Journal of Supercomputing:

"Resilient Edge: Architectural Optimization for Real-Time Network Fault Diagnosis on Resource-Constrained Embedded Systems"

📂 REPOSITORY STRUCTURE

Rakshit_ResilientEdge_Zenodo_Artifacts/
│
├── esp32EdgeAI_Revision.ipynb (Revised Jupyter Notebook for the ML pipeline)
├── model.h (Compiled C-header for ESP32 TinyML deployment)
│
└── 01_Raw_Data/ (Original Server Logs - ESP32 to Gateway)
├── Baseline_Server_Day-1.csv
├── Baseline_Server_Day-2.csv
├── Baseline_Server_Day-3.csv
├── Distance_Server_Day-1.csv
├── Distance_Server_Day-2.csv
├── Distance_Server_Day-3.csv
├── Noise_Server_Day-1.csv
├── Noise_Server_Day-2.csv
└── Noise_Server_Day-3.csv

📊 KEY DATA / COLUMN DEFINITIONS

R_i: Extracted resistance/congestion feature from the wireless channel.

L_i: Extracted inductance/attenuation feature.

J_i: Extracted jitter/jerk feature from packet dynamics.

latency_us: On-device inference latency measured in microseconds (µs).

fault_class: Ground truth label for the diagnosed network state.

🛠 USAGE

Machine Learning Pipeline: To reproduce the data processing, feature extraction, and model training, run the esp32EdgeAI_Revision.ipynb notebook. This notebook directly ingests the CSV files located in the 01_Raw_Data/ directory.

Hardware Deployment: To replicate the hardware benchmarks (e.g., the 4.16 µs inference latency / 998 CPU cycles), include the provided model.h file within your Arduino IDE project and flash it directly to an ESP32 microcontroller.

Environment: Python 3.x, Jupyter, Scikit-Learn, Pandas, Matplotlib, Seaborn, Arduino IDE (v1.8+ or v2.x with ESP32 Core).