Automatic energy model generation with MSP430 EnergyTrace

Autor(en): Friesel, D.
Kaiser, L.
Spinczyk, O. 
Stichwörter: Automatic model generation; automation; Benchmarking; Drift compensation; Embedded systems; Energy measurement system; energy measurements; energy models; energytrace; Hardware cost; Internet of things, Automated measurement; Maximum error; Measurement range; Single point, Error compensation
Erscheinungsdatum: 2021
Herausgeber: Association for Computing Machinery, Inc
Journal: CPS-IoTBench 2021 - Proceedings of the 2021 Benchmarking Cyber-Physical Systems and Internet of Things
Startseite: 26
Seitenende: 31
Zusammenfassung: 
Professional energy measurement systems are expensive, especially when it comes to systems usable for automated measurements. In exchange, they provide a well-known measurement range, detailed accuracy guarantees, and a computer interface. DIY solutions from researchers have improved the situation considerably, with hardware cost in the $100 range, but are typically not commercially available. We are interested in even more affordable solutions, and examine the capabilities of the "EnergyTrace"technology embedded on the TI MSP430FR5994 LaunchPad, which is commercially available for less than $20. Out of the box, we observe a maximum error of 210 μA in the 100 μA to 10 mA range, but no support for automatic model generation. With single-point calibration and a custom synchronization and drift compensation algorithm, we are able to further reduce the error to 53 μA and perform entirely automated measurements and energy model generation on 3.3V MCUs and peripherals with a maximum timestamp error of 0.95 ms. © 2021 ACM.
Beschreibung: 
Conference of 4th Benchmarking Cyber-Physical Systems and Internet of Things, CPS-IoTBench 2021 ; Conference Date: 18 May 2021 Through 18 May 2021; Conference Code:168532
ISBN: 9781450384391
DOI: 10.1145/3458473.3458822
Externe URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85105468901&doi=10.1145%2f3458473.3458822&partnerID=40&md5=31167fc07420773616a24458ca9f3ca4

Show full item record

Page view(s)

5
Last Week
0
Last month
0
checked on May 19, 2024

Google ScholarTM

Check

Altmetric