能耗量測分析工具在MP3上之應用

Energy Consumption Measurement and Analysis Tool on MP3

吳維凱(Wei-Kai Wu)

指導教授 ： 薛智文

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2007年)

https://doi.org/10.6342/NTU.2007.01923

摘要

嵌入式系統的設計開發人員，目前亟需一套精確可靠的能耗量測工具。MP3隨身聽是目前最普遍的掌上型多媒體播放器，其中續航力的問題，已成為一項重要的設計考量。在這篇論文中，我們改進一個在多工環境下，整合系統執行資訊的實體量測能耗分析工具，並透過大量的實驗，驗證這套工具，建立一套MP3能耗量測標準方法。實驗中，考慮到經過各種最佳化的MP3播放器，對不同長度、位元率或曲風的MP3歌曲，利用正規的統計方法，驗證量測結果的誤差。相信這套能耗量測分析工具的實驗方法，同樣能適用於其它應用上。

關鍵字

嵌入式系統 ； 能耗值 ； 實體量測 ； MP3 ； 統計方法

並列摘要

An accurate and reliable energy consumption measurement tool is very important for the designers of modern embedded systems. MP3 Walkman is almost the most popular hand-held multimedia player, and the endurance of a
MP3 player has become an critical issue. In this thesis, we revise a physical measurement-based tool which integrates energy consumption with system execution information in multi-tasking systems. We validate this tool and establish a standard approach of MP3 energy measuring by a large number of experiments. Different optimizations of a MP3 player or variable lengths, bit
rates, and music styles of a MP3 music are considered in the experiments, and we make use of a statistical approach to verify the measurement results. We believe the experiment approach using energy consumption measurement and analysis tool can be adopted to other applications as well.

並列關鍵字

embedded system ； energy consumption ； physical measurement ； MP3 ； statistical approach

參考文獻

[4] J. Flinn and M. Satyanarayanan. Powerscope: A tool for profiling the energy usage of mobile applications. In Proceedings of IEEE Workshop Mobile Comput. Syst. Applicat., pages 2-10, Feb 1999.

[6] Chuling Hu, Danial A. Jimenez, and Ulrich Kremer. Toward an evaluation infrastructure for power and energy optimizations. In proceedings of the 41 19th IEEE Internationl Parallel and Distributed Processing Symposium,

[10] Tajana Simunic, Luca Benini, and Giovanni De Micheli. Energy-efficient design of battery-powered embedded systems. In IEEE Transactions on VLSI Systems, pages 15-28, Feburary 2001.

[11] Phillip Stanley-Marbell and Michael Hsiao. Fast, fexible, cycle-accurate energy estimation. In International Symposium on Low Power Electronics

[12] V. Tiwari, S. Malik, and A. Wolfe. Power analysis of embedded software: A first step toward software power minimization. IEEE Transactions on VLSI Syst., 2:437-445, Dec. 1994.