A Survey of Energy Consumption Measurement in Embedded Systems

• Published in: IEEE access Vol. 9; pp. 60516 - 60530
• Main Authors: Guo, Chen, Ci, Song, Zhou, Yanglin, Yang, Yang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; embedded system; Embedded systems; Energy consumption; Energy efficiency; Energy measurement; Hardware; Instruments; Internet of Things (IoT); Optimization; Performance evaluation; Power consumption; Power measurement; Software
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2021.3074070

The low-power design of software and hardware is crucial for the efficiency and sustainability of embedded systems. Accurate energy consumption measurement plays a vital role in evaluating the energy performance of software and hardware, which also provides design insights to develop new paradigms for algorithm or code optimization in terms of energy efficiency and system cost. A few literature has surveyed this research field, but it lacks multi-faceted comparisons and comprehensive analysis. In this paper, we first show the necessity of accurate energy consumption analysis for embedded systems. Then, we study major methods in literature for measuring energy consumption of embedded systems, which can be summarized with three categories: 1) measurement-based energy profiling, 2) model-based energy estimation, and 3) simulator-based energy estimation. Some subcategories are further made based on characteristics of these approaches. The pros and cons of each category have been reviewed and evaluated through multi-faceted comparisons. Finally, for transparent energy analysis and improving the energy efficiency of embedded systems, we come up with some contributing factors that matter the energy consumption measurements and discuss the challenges and future research directions.