Companion Artefacts for Workload-Derived Firmware-Readiness Screening in Agricultural IoT Data Quality

This repository contains the companion artefacts for the manuscript entitled “Workload-Derived Firmware-Readiness Screening for Agricultural IoT Data Quality”. The artefacts support the reproducibility of a workload-derived pre-deployment screening method for Agricultural Internet of Things (Ag-IoT) data-quality processing under ESP32- and STM32-oriented microcontroller unit (MCU) readiness profiles. The study treats firmware readiness as an early screening problem before direct board-level validation. It does not claim measured firmware execution, measured current draw, packet loss, reset behaviour, watchdog behaviour, or long-duration runtime stability. Instead, the artefacts provide the workload summaries, calculation scripts, board-profile parameters, generated tables, and generated figures used to estimate whether Ag-IoT data-quality workloads are safe, borderline, or unsuitable under study-defined readiness budgets. The repository includes workload summaries for seven Primary Field-Validation (PFV) workloads and four public Ag-IoT datasets. The PFV workloads contain field-derived workload conditions related to missingness, repair demand, and integrity-tagged processing. The public datasets are included to support external reuse checking and cross-board-profile screening. For public datasets, the repair ratio is represented by a repair-demand proxy because repaired-output and integrity-tagged states are not available. The MATLAB scripts provided in this repository reproduce the main screening calculations, including missing-cell ratio, repair ratio or repair-demand proxy, workload factor, estimated latency, peak RAM, energy per sample, payload size, bytes per hour, loop jitter, budget margin, and readiness category. The output tables and figures correspond to the manuscript’s workload indicator summaries, estimated resource behaviour, readiness budget settings, deployment-readiness interpretation, cross-board-profile screening, and visual readiness summaries. The repository also includes ESP32, STM32, and NodeMCU Lua starter templates for future MCU-side data acquisition. These files are provided only as optional future validation templates. They are not used as measured firmware evidence in the present study and should not be interpreted as proof of physical ESP32 or STM32 firmware performance. Repository Contents 01_input_workload_summaries/Contains workload-summary files for the PFV workloads and public Ag-IoT datasets, including selected workload indicators and variable dictionaries. 02_matlab_analysis/Contains MATLAB scripts for calculating workload indicators, estimating board-profile resource behaviour, calculating budget margin, and generating tables and figures. 03_output_tables/Contains CSV versions of the main output tables used to support the manuscript. 04_output_figures/Contains generated high-resolution figures used for manuscript preparation. 05_board_profile_parameters/Contains ESP32 and STM32 profile settings, readiness budgets, and baseline parameters used in the screening calculations. 06_optional_mcu_templates/Contains optional starter templates for future MCU-side data acquisition using ESP32, STM32 HAL C, and NodeMCU Lua. These templates are not direct validation results. 07_manuscript_text/Contains manuscript-supporting notes and short text snippets related to reproducibility and study boundaries. Intended Use This repository is intended for researchers who wish to reproduce, inspect, or adapt the workload-derived firmware-readiness screening method for Ag-IoT data-quality workloads. The artefacts may also be used as a starting point for future direct ESP32, STM32, or other MCU board-level validation, provided that board-specific budgets and measured firmware logs are added separately. Reproducibility Statement The artefacts allow users to reproduce the estimated screening outputs reported in the manuscript. Users can inspect the workload indicators, run the MATLAB scripts, regenerate the output tables and figures, and modify the board-profile parameters for other MCU-oriented readiness studies. The results should be interpreted as pre-deployment estimates and not as direct physical firmware measurements. Suggested Keywords Agricultural Internet of Things; Ag-IoT; firmware-readiness screening; ESP32; STM32; microcontroller unit; data quality; missing data; workload screening; edge computing; MATLAB; board-profile screening; sensor data; embedded systems. Suggested Upload Type Dataset / Software companion artefacts Suggested Version Version 1.0.0 Suggested License Creative Commons Attribution 4.0 International (CC BY 4.0) for documentation, tables, and figures.MIT License for reusable scripts and optional MCU templates, if Zenodo allows separate licensing notes in the README. Suggested Related Identifier Use the manuscript title as the related work: Workload-Derived Firmware-Readiness Screening for Agricultural IoT Data Quality After the manuscript is submitted or accepted, add the article DOI as a related identifier in Zenodo.