Energy–Efficient Configuration of Embedded Data-Processing Systems in Public Transportation
Context
Modern public transportation vehicles, such as trams, buses, trolleybuses and trains, increasingly rely on on-board computing units to process and securely transfer large volumes of data generated by sensors and surveillance cameras. These systems often operate on limited battery power during night-time parking, when vehicles are disconnected from external energy sources. During this time window, the on-board computer must complete several computationally intensive tasks—such as software updates, video decoding, compression, encryption, and data upload—before service resumes.
In collaboration with Supercomputing Systems AG (SCS) and a public transportation company in Romandie, this project addresses the challenge of executing these tasks reliably under strict energy and time constraints. Understanding how to configure the embedded system and how to select optimal communication protocols for data transfer in order to remain both energy-efficient and predictable is essential for dependable fleet operations.
Motivation
Although modern processors provide features such as Dynamic Voltage and Frequency Scaling (DVFS) to control energy consumption, the actual impact of hardware configuration on energy consumption depends strongly on the workload characteristics. Algorithms for tasks such as data processing, compression, encryption, or transfer differ significantly in CPU usage, memory access patterns, and I/O behavior. These choices also have a direct impact on the volume of data to be transferred. The combined influence of hardware settings and algorithmic choices on energy consumption remains an open field to study, particularly in industrial embedded systems as it is the case in public transportation systems.
Goal
The goal of this project is to determine how hardware and software configurations and communication strategies shape energy–performance behavior in embedded systems. The student will experiment with different setups and analyze their impact on efficiency. The outcome of this thesis includes:
- A multi-dimensional model linking algorithmic choices, data transfer and hardware settings to energy and performance outcomes.
- A cleaned dataset, energy–performance models and visualizations.
- A report with practical recommendations.
- A contribution for a potential scientific publication based on anonymized data.
Requirements
- Experience with Linux-based systems
- Programming skills in Python and C/C++
- Interest in systems programming and energy/performance analysis
Pointers
- Garcia, A. M., Serpa, M., Griebler, D., Schepke, C., Fernandes, L. G. L., and Navaux, P. O. A., “The Impact of CPU Frequency Scaling on Power Consumption of Computing Infrastructures,” ICCSA 2020, LNCS 12254, Springer, 2020.
- T. Maseda, J. Enes, R. R. Expósito and J. Touriño, “Automated Approach for Accurate CPU Power Modelling,” 2024 IEEE International Conference on Cluster Computing (CLUSTER), Kobe, Japan, 2024, pp. 97-107, doi: 10.1109/CLUSTER59578.2024.00016.
- Ibraheem, M. K. I., Dvorkovich, A. V., & Al-Khafaji, I. M. A. (2024). A Comprehensive Literature Review on Image and Video Compression: Trends, Algorithms, and Techniques. Ingénierie Des Systèmes D’information, 29(3), 863–876. https://doi.org/10.18280/isi.290307