Supplementary Dataset: Physical Current Corroboration for Cross-Layer Benchmarking of Hardware-Accelerated Security in Edge IoT

Experimental Companion Dataset: INA219 Physical Current Corroboration (EdgeSec-Bench) This dataset contains 22,640 high-resolution physical current measurements collected using a Texas Instruments INA219 current sensor in series with the power supply of an ESP32-DevKitC v4 IoT node. The measurements provide independent physical corroboration of the energy model presented in the associated peer-reviewed paper: "Cross-Layer Benchmarking of Hardware-Accelerated Security in Edge IoT Under Signal Degradation and Network Congestion". Key Finding: AES-GCM-128 hardware-accelerated encryption adds a statistically negligible current differential (S4 − S3 = −0.35 mA, Cohen's d = 0.114), supporting the conclusion that Wi-Fi transmission energy dominates cryptographic energy by 14×–42× (analytical model) to 9× (duty-cycle-averaged measurement). 📄 ASSOCIATED MANUSCRIPT This dataset supports the findings presented in the research article: "Cross-Layer Benchmarking of Hardware-Accelerated Security in Edge IoT Under Wi-Fi Signal Degradation and Network Congestion" 📂 DATASET STRUCTURE INA219_Current_Monitoring_Dataset/ │ ├── 01_Raw_Data/ (Original Experimental Output) │ ├── Client_Serial/ (15 Local CSVs: 5 scenarios × 3 runs) │ └── Server_MQTT/ (6 Server CSVs: S3 & S4 × 3 runs) │ ├── 02_Processed_Data/ (Cleaned & Merged Datasets) │ ├── combined_client_energy.csv (All 22,640 merged samples) │ ├── combined_server_mqtt.csv (All server-side MQTT packets merged) │ └── scenario_statistics.csv (Summary statistics: mean, std, CI) │ ├── 03_Analysis_Code/ (Reproducibility) │ ├── INA219_Reproducibility_Analysis.ipynb │ └── requirements.txt │ ├── 04_Firmware/ (Hardware Setup & Collection Scripts) │ ├── S1_Idle_INA219.ino (Idle baseline firmware) │ ├── S2_SensorOnly_INA219.ino (Sensor read only, WiFi disabled) │ ├── S3_Plaintext_INA219.ino (Plaintext MQTT transmission) │ ├── S4_Encrypted_INA219.ino (Hardware-accelerated AES-GCM MQTT) │ ├── S5_CryptoOnly_INA219.ino (Crypto active, transmission disabled) │ └── energy_collector_server.py (Raspberry Pi MQTT gateway logger) │ ├── 05_Results/ (Pre-generated Outputs) │ ├── Datasheet_Validation_Table.md │ ├── Scenario_S1..S5_Report.md (Per-scenario quality reports) │ └── Figures/ (6 × PDF + PNG publication-ready figures) │ ├── LICENSE (MIT License) └── README.md (Detailed documentation) 📊 COLUMN DEFINITIONS Client Serial CSV (01_Raw_Data/Client_Serial/): type: Always 'E' for energy readings. timestamp_us: ESP32 micros() timestamp (microseconds since boot). phase: Processing phase at time of measurement ('I' = Idle baseline, 'S' = Sensor Read, 'C' = AES-GCM Cryptography, 'T' = MQTT Transmit). current_mA: INA219-measured current in milliamperes (0.1 mA resolution). voltage_V: INA219-measured bus voltage in volts. crypto_us: AES-GCM execution time in microseconds (0 for S1/S2/S3). tx_us: MQTT publish time in microseconds (0 for S1/S2/S5). scenario: Scenario identifier (e.g., 'S4_Encrypted'). seq: Per-scenario packet sequence number. Server MQTT CSV (01_Raw_Data/Server_MQTT/): recv_ts_ms: Server receive timestamp (epoch ms). DevID: ESP32 device identifier. MsgSeq: Message sequence number. SendTs: ESP32 send timestamp (epoch ms). Scen: Scenario identifier. CrypT_us: Client-measured crypto time (µs). TxT_us: Client-measured transmit time (µs). RSSI: Wi-Fi signal strength (dBm). Heap: ESP32 free heap memory (bytes). has_crypto: 1 if encrypted, 0 if plaintext. payload_size_bytes: Raw MQTT payload size. ⚡ EXPERIMENTAL SCENARIOS S1_Idle: WiFi connected, no operations — idle baseline (8.80 ± 1.32 mA) S2_SensorOnly: DHT22 sensor read, WiFi disabled — sensor baseline (8.58 ± 0.52 mA) S3_Plaintext: DHT22 + MQTT publish, no encryption — transmission control (10.74 ± 3.14 mA) S4_Encrypted: DHT22 + AES-GCM-128 + MQTT publish — full secure system (10.40 ± 2.94 mA) S5_CryptoOnly: DHT22 + AES-GCM-128, no MQTT — crypto isolation baseline (7.79 ± 1.34 mA) 🛠 USAGE & REPRODUCIBILITY The included Jupyter Notebook (INA219_Reproducibility_Analysis.ipynb) contains the complete Python code to process the raw datasets and reproduce all 6 figures, validation tables, and Cohen's d effect size calculations. Required Environment: Python 3.x (Tested on 3.14.0). Please ensure the following dependencies are installed: pandas>=2.0.0 numpy>=1.24.0 scipy>=1.10.0 seaborn>=0.12.0 matplotlib>=3.7.0 jupyter>=1.0.0 🔗 RELATED DATASET This physical current corroboration dataset is a supplementary companion to the primary EdgeSec Benchmark Dataset v2 (30,631 packets) available at DOI: 10.5281/zenodo.20254881.