Adaptive Estimation of the Thermal Behavior of CPU-GPU SoCs for Prediction and Diagnosis

Oussama Djedidi (1) , Nacer K M'Sirdi (1) , Aziz Naamane (1)

1 LIS - Laboratoire d'Informatique et des Systèmes (LIS) (Marseille, Toulon)

This paper proposes a dynamic behavioral model for temperature variations of systems on chips (SoC) in embedded systems. We use identification techniques (ARMAX modeling) to construct a data-driven online temperature model that estimates the temperature according to the CPU and GPU frequencies, the used RAM and the power consumed by the chip. Furthermore, we used two the Recursive Least Squares (RLS) to estimate the parameters of the ARMAX model. This method allows us to update the parameters of the model online in case of a change in the system or its characteristics. Finally, we validate the temperature model and compare between booth estimation methods.

Mots clés: support vector re- gression; Fault Diagnosis; Systems-on-Ship; Identification; Embedded systems; Control; Remaining Useful Life

Domaines: Systèmes embarqués; Automatique / Robotique; Traitement du signal et de l'image [eess.SP]; Electronique

Oussama Djedidi, Nacer K M'Sirdi, Aziz Naamane. Adaptive Estimation of the Thermal Behavior of CPU-GPU SoCs for Prediction and Diagnosis. Proceedings of the International Conference on Integrated Modeling and Analysis in Applied Control and Automation, 2019–IMAACA 2019, Sep 2019, Lisbon, Portugal. pp.93-98. ⟨hal-02311475v2⟩