Abstract
Power consumption has become a major design constraint, especially for battery-powered embedded systems. However, the impact of software applications is typically considered in later phases, where both software and hardware parts are close to their finalization. Power-related issues must be detected in early stages to keep the development costs low, satisfy time-to-market, and avoid cost-intensive redesign loops. Moreover, the variety of hardware components, architectures, and communication interfaces make the development of embedded software more challenging. To manage the complexity of software applications, approaches such as model-driven development (MDD) may be used. This article proposes a power-estimation approach in MDD for software application models in early development phases. A unified modeling language (UML) profile is introduced to model power-related properties of hardware components. To determine the impact of software applications, we defined two analysis methods using simulation data and a novel in-the-loop concept. Both methods may be applied at different development stages to determine an energy trace, describing the energy-related behavior of the system. A novel definition of energy bugs is provided to describe power-related misbehavior. We apply our approach to a sensor node example, demonstrate an energy bug detection, and compare the runtime and accuracy of the analysis methods.
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference84 articles.
1. Global IoT Forecast Report, 2020–2030
https://transformainsights.com/research/reports/global-iot-forecast-report-2020-2030
2. Energy efficiency of the Internet of Things;Friedli,2016
3. Commodity Markets Outlook: Causes and Consequences of Metal Price Shocks
https://openknowledge.worldbank.org/handle/10986/35458
4. LoRaWAN in a rural context: Use cases and opportunities for agricultural businesses;Grunwald;Proceedings of the Mobile Communication-Technologies and Applications, 24. ITG-Symposium,2019
5. A Manifesto for Energy-Aware Software
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献