THERMO-SAFE-RV: A Thermal-Aware Secure RISC-V Architecture for Edge-AI Nodes
DOI:
https://doi.org/10.37936/ecti-eec.2027251.267397Abstract
Edge-AI nodes execute perception, classification and control tasks near sensors under tight thermal, energy and security budgets. RISC-V offers an open and extensible foundation for such nodes, but conventional edge-AI execution stacks treat thermal control, security monitoring and accelerator scheduling as separate modules. This separation adds latency during throttling, raises energy in attack-free operation, and leaves a gap when adversarial code, abnormal memory access or peripheral misuse occurs during sustained inference. This paper proposes THERMO-SAFE-RV, a thermal-aware secure RISC-V edge-AI architecture that jointly coordinates dynamic voltage and frequency scaling (DVFS), lightweight runtime intrusion monitoring, secure boot attestation, accelerator scheduling and thermal guard policies. A composite risk–energy–thermal controller selects operating points from observed temperature, power, task criticality and a runtime anomaly score, and binds every admissible response to a cryptographically signed policy table, so that compromised software cannot relax a thermal or security limit. A constrained optimisation model is solved online by a table-driven search over a small candidate set. Architecture-level simulation at 80% workload intensity lowers energy per inference by 36.8% and peak die temperature by 13.5 °C, raises attack detection from 91.2% to 96.8%, cuts false alarms from 7.8% to 3.1%, and keeps area and power overhead below 0.65% and 0.30%.
Downloads
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Keshav Kumar, Mohit Kumar Srivastava, Vivek Kumar, Utkarsh Pandey, Amandeep Kaur

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
- Creative Commons Copyright License
The journal allows readers to download and share all published articles as long as they properly cite such articles; however, they cannot change them or use them commercially. This is classified as CC BY-NC-ND for the creative commons license.
- Retention of Copyright and Publishing Rights
The journal allows the authors of the published articles to hold copyrights and publishing rights without restrictions.
