D
2018
Analytical Solution for Long Battery Lifetime Prediction in Nonadaptive Systems
IVANOV, Dmitry, Kim G. LARSEN, Sibylle SCHUPP and Jiří SRBA
Basic information
Original name
Analytical Solution for Long Battery Lifetime Prediction in Nonadaptive Systems
Authors
IVANOV, Dmitry (643 Russian Federation), Kim G. LARSEN (208 Denmark), Sibylle SCHUPP (276 Germany) and Jiří SRBA (203 Czech Republic, guarantor, belonging to the institution)
Edition
Netherlands, Proceedings of the 15th International Conference on Quantitative Evaluation of SysTems (QEST'18), p. 173-189, 17 pp. 2018
Other information
Type of outcome
Stať ve sborníku
Field of Study
10201 Computer sciences, information science, bioinformatics
Country of publisher
Netherlands
Confidentiality degree
není předmětem státního či obchodního tajemství
Publication form
printed version "print"
Impact factor
Impact factor: 0.402 in 2005
RIV identification code
RIV/00216224:14330/18:00106626
Organization unit
Faculty of Informatics
Keywords in English
UPPAAL SMC; battery models; statistical model checking; wireless protocols
V originále
Uppaal SMC is a state-of-the-art tool for modelling and statistical analysis of hybrid systems, allowing the user to directly model the expected battery consumption in battery-operated devices. The tool employs a numerical approach for solving differential equations describing the continuous evolution of a hybrid system, however, the addition of a battery model significantly slows down the simulation and decreases the precision of the analysis. Moreover, Uppaal SMC is not optimized for obtaining simulations with durations of realistic battery lifetimes. We propose an analytical approach to address the performance and precision issues of battery modelling, and a trace extrapolation technique for extending the prediction horizon of Uppaal SMC. Our approach shows a performance gain of up to 80% on two industrial wireless sensor protocol models, while improving the precision with up to 55%. As a proof of concept, we develop a tool prototype where we apply our extrapolation technique for predicting battery lifetimes and show that the expected battery lifetime for several months of device operation can be computed within a reasonable computation time.
Displayed: 10/11/2024 04:32