Hardware-assisted energy monitoring architecture for micro sensor nodes

  • Authors:
  • Sukwon Choi;Hayun Hwang;Byunghun Song;Hojung Cha

  • Affiliations:
  • Department of Computer Science, Yonsei University, Seodaemun-gu, Shinchon-dong 134, Seoul 120-749, Republic of Korea;Department of Computer Science, Yonsei University, Seodaemun-gu, Shinchon-dong 134, Seoul 120-749, Republic of Korea;RFID/USN Convergence Research Center, Korea Electronics and Technology Institute, #68 Yatap-dong, Bundang-gu, Seongnam, Gyeonggi Province 463-816, Republic of Korea;Department of Computer Science, Yonsei University, Seodaemun-gu, Shinchon-dong 134, Seoul 120-749, Republic of Korea

  • Venue:
  • Journal of Systems Architecture: the EUROMICRO Journal
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present a hardware-assisted energy monitoring architecture, HEMA, which provides power information for individual system components in micro sensor nodes. HEMA gives information on how the system components of sensor nodes spend energy while applications are running. To develop a practical runtime monitoring system we used a power monitoring technique based on a battery monitor IC (Integrated Circuit) which is typically used in embedded systems. HEMA uses a software technique to monitor device usage patterns, and combines it with hardware-assisted power information in runtime. However, the battery monitor IC has problems in monitoring individual devices. In this paper, we developed a software technique to supplement its drawbacks. To evaluate the proposed system, we built a micro sensor node with battery monitor IC. We operated real applications on the sensor node and conducted a comparative analysis with a dedicated power monitor. Our experiment results show that HEMA indeed provides a suitable architecture for runtime power analysis with low overhead.