A selective DVS technique based on battery residual

Sukwon Choi, Hojung Cha, Rhan Ha

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Conventional research on DVS assumes that the battery, which is the power source of most mobile devices, provides constant power; the battery, however, has unique discharge characteristics due to its chemical properties. Based on the residual characteristics of the battery, this paper presents a practical approach towards a battery-aware DVS mechanism that can easily be applied to conventional operating systems. The proposed policy monitors the battery residual in runtime, classifies the current battery state as one of three - the recovery effect region, the rate capacity region and the alarm region - then applies a suitable DVS algorithm according to each region, which takes advantage of the underlying battery characteristics. Experiments on commercial PDAs (iPAQ5550 and Zaurus SL-860 PDA) running MPEG applications showed that our policy significantly extends the battery lifetime, compared to the conventional DVS method, with only a marginal QoS degradation.

Original languageEnglish
Pages (from-to)33-42
Number of pages10
JournalMicroprocessors and Microsystems
Volume30
Issue number1
DOIs
Publication statusPublished - 2006 Feb 1

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'A selective DVS technique based on battery residual'. Together they form a unique fingerprint.

  • Cite this