A Framework for Automated Detection of Power-related Software Errors in Industrial Verification Processes

Autor: Giovanni Squillero, Walter Ruzzarin, Ernesto Sanchez, Alberto Tonda, Stefano Gandini
Přispěvatelé: dauin, Dipartimento di Automatica e Informatica [Torino] (DAUIN), Politecnico di Torino = Polytechnic of Turin (Polito)-Politecnico di Torino = Polytechnic of Turin (Polito), DAUIN Dipartimento di Automatica e Informatica, Politecnico di Torino = Polytechnic of Turin (Polito)
Rok vydání: 2010
Předmět:
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR]
Computer science
02 engineering and technology
[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE]
computer.software_genre
Diagnostics
Evolutionary algorithms
Mobile phones
Power consumption
Software testing
Testing tools
diagnostics
0202 electrical engineering
electronic engineering
information engineering
Software quality analyst
Software verification and validation
evolutionary algorithms
Electrical and Electronic Engineering
mobile phones
business.industry
Software development
software testing
power consumption
020207 software engineering
[SPI.TRON]Engineering Sciences [physics]/Electronics
Software framework
Embedded system
Software construction
Personal software process
020201 artificial intelligence & image processing
business
Software engineering
computer
Software quality control
Software verification
testing tools
Zdroj: Journal of Electronic Testing
Journal of Electronic Testing, Springer Verlag, 2010, 26 (6), pp.689-697. ⟨10.1007/s10836-010-5184-5⟩
ISSN: 1573-0727
0923-8174
DOI: 10.1007/s10836-010-5184-5
Popis: International audience; The complexity of cell phones is continually increasing, with regards to both hardware and software parts. As many complex devices, their components are usually designed and verified separately by specialized teams of engineers and programmers. However, even if each isolated part is working flawlessly, it often happens that bugs in one software application arise due to the interaction with other modules. Those software misbehaviors become particularly critical when they affect the residual battery life, causing power dissipation. An automatic approach to detect power-affecting software defects is proposed. The approach is intended to be part of a qualifying verification plan and complete human expertise. Motorola, always at the forefront of researching innovations in the product development chain, experimented the approach on a mobile phone prototype during a partnership with Politecnico di Torino. Software errors unrevealed by all human-designed tests have been detected by the proposed framework, two out of three critical from the power consumption point of view, thus enabling Motorola to further improve its verification plans. Details of the tests and experimental results are presented.
Databáze: OpenAIRE
Externí odkaz: