Embedded Systems Power Consumption Measurement Methods Overview

Estimation of embedded systems (ES) power consumption is becoming a standard step in the development cycle of battery powered equipment. The importance of the power consumption is well realized by the vendors of integrated circuits, especially microprocessors and filed programmable gate arrays (FPGA). The issue is also targeted by the designers of ES development tools. It is also covered in the research publications. Power consumption and therefore its estimation is a key issue in designing battery powered devices in the field of wireless, medical, utility metering, etc., applications. The less power is consumed by the electronics the longer it will continue functioning without recharging or changing the battery. This is not only improves convenience of device use but is also important from the ecology point of view. Development of ultra-low power electronics opened a possibility to create absolutely new class of ES scavenging a necessary energy for their operation from the ambience [1]. In the larger energy sector the key issue is the energy dissipation of the integrated circuits that leads to the need of cooling equipment and expensive packages. The key component of today’s ES is microcontroller and/or FPGA. Both are user programmable units and their power consumption depends upon the executed program and data processed. Therefore, so called software related power is widely considered [2−4]. Having in mind that most of modern microcontrollers and FPGAs are designed for run time switching between active and low power modes, the software related power is the part of overall power consumption budget that can be optimized during development of a system. To be able to minimize power consumption methods and tools for its estimation are needed. There exist two main approaches towards estimation of the power consumption of ES containing programmable unit: simulation based and measurement based. The simulation based approach uses models relating power consumption and programming instructions. This type of tools are becoming a standard utility in the FPGA design packages like Altera PowerPlay or Synopsys Power Compiler. The second approach is based on physical measurements of power consumption while running the system under development. While the measurement based approach requires measurement instruments and in most of the cases special setups inside the ES, it does not suffer from the inadequacy or even absence of reliable models. To add more, measurement approach is the only way to verify correctness of simulation based approach. Therefore measurement is important to validate the power consumption models. The goal of this paper is to give an overview of the current status of ESs power consumption measurement problems, methods and instrumentation.

[1]  Sharad Malik,et al.  Power analysis of embedded software: a first step towards software power minimization , 1994, IEEE Trans. Very Large Scale Integr. Syst..

[2]  Rolf Ernst,et al.  Timing and power measurement in static software analysis , 2002 .

[3]  Naehyuck Chang,et al.  Cycle-accurate energy measurement and characterization with a case study of the ARM7TDMI [microprocessors] , 2002, IEEE Trans. Very Large Scale Integr. Syst..

[4]  Spiridon Nikolaidis,et al.  Instruction-level power consumption estimation of embedded processors for low-power applications , 2002, Comput. Stand. Interfaces.

[5]  Theodore Laopoulos,et al.  Measurement of current variations for the estimation of software-related power consumption [embedded processing circuits] , 2003, IEEE Trans. Instrum. Meas..

[6]  Chandra Krintz,et al.  Application-level prediction of battery dissipation , 2004, Proceedings of the 2004 International Symposium on Low Power Electronics and Design (IEEE Cat. No.04TH8758).

[7]  Theodore Laopoulos,et al.  Measurements analysis of the software-related power consumption in microprocessors , 2003, IEEE Transactions on Instrumentation and Measurement.

[8]  Lizy Kurian John,et al.  Runtime identification of microprocessor energy saving opportunities , 2005, ISLPED '05. Proceedings of the 2005 International Symposium on Low Power Electronics and Design, 2005..

[9]  Joseph A. Paradiso,et al.  Energy scavenging for mobile and wireless electronics , 2005, IEEE Pervasive Computing.

[10]  J. Millet,et al.  Adapting power consumption to performance requirements in a MSP430 microcontroller , 2005, Conference on Electron Devices, 2005 Spanish.

[11]  Dushyanth Narayanan Software Power Measurement , 2005 .

[12]  V. Konstantakos,et al.  Measurement of Power Consumption in Digital Systems , 2005, 2005 IEEE Instrumentationand Measurement Technology Conference Proceedings.

[13]  D. Macii,et al.  An Automatic Power Consumption Measurement Procedure for Bluetooth Modules , 2006, 2006 IEEE Instrumentation and Measurement Technology Conference Proceedings.

[14]  V. Konstantakos,et al.  A Power Measuring Technique for Built-in Test Purposes , 2006, 2006 IEEE Instrumentation and Measurement Technology Conference Proceedings.

[15]  Yung-Hsiang Lu,et al.  Power Measurement of Software Programs on Computers With Multiple I/O Components , 2007, IEEE Transactions on Instrumentation and Measurement.

[16]  A. Borovyi,et al.  Analysis of Circuits for Measurement of Energy of Processing Units , 2007, 2007 4th IEEE Workshop on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications.

[17]  Eduardo I. Boemo,et al.  Arithmetic Operations and Their Energy Consumption in the Nios II Embedded Processor , 2008, 2008 International Conference on Reconfigurable Computing and FPGAs.