电子科技大学学报
電子科技大學學報
전자과기대학학보
JOURNAL OF UNIVERSITY OF ELECTRONIC SCIENCE AND TECHNOLOGY OF CHINA
2014年
2期
272-277
,共6页
段林涛%郭兵%沈艳%王毅%张文丽%熊伟
段林濤%郭兵%瀋豔%王毅%張文麗%熊偉
단림도%곽병%침염%왕의%장문려%웅위
应用程序能耗模型%电池电能%能耗模型%移动终端%功耗
應用程序能耗模型%電池電能%能耗模型%移動終耑%功耗
응용정서능모모형%전지전능%능모모형%이동종단%공모
application energy model%battery energy%energy model%mobile devices%power consumption
应用程序能耗分析与建模是智能移动终端能耗优化的重要组成部分。针对智能移动终端丰富的应用程序,提出了一种基于应用程序运行时间的时间能耗模型。与精度高和复杂的应用程序组件能耗模型相比,该模型使用时间变量刻画和包含终端的功耗、性能等多种因素,而且运行时间容易精确测量和获取,能够快速地估算应用程序运行时移动终端产生的能耗。实验结果表明,在GT-I9108、GT-I9308和GT-P3108实验平台下,该模型的能耗估算结果与Android操作系统应用程序框架提供的组件能耗模型测量值相比平均误差分别为0.89%、1.37%和0.29%,能够为移动终端用户便捷地预测应用程序消耗的电池电能提供帮助。
應用程序能耗分析與建模是智能移動終耑能耗優化的重要組成部分。針對智能移動終耑豐富的應用程序,提齣瞭一種基于應用程序運行時間的時間能耗模型。與精度高和複雜的應用程序組件能耗模型相比,該模型使用時間變量刻畫和包含終耑的功耗、性能等多種因素,而且運行時間容易精確測量和穫取,能夠快速地估算應用程序運行時移動終耑產生的能耗。實驗結果錶明,在GT-I9108、GT-I9308和GT-P3108實驗平檯下,該模型的能耗估算結果與Android操作繫統應用程序框架提供的組件能耗模型測量值相比平均誤差分彆為0.89%、1.37%和0.29%,能夠為移動終耑用戶便捷地預測應用程序消耗的電池電能提供幫助。
응용정서능모분석여건모시지능이동종단능모우화적중요조성부분。침대지능이동종단봉부적응용정서,제출료일충기우응용정서운행시간적시간능모모형。여정도고화복잡적응용정서조건능모모형상비,해모형사용시간변량각화화포함종단적공모、성능등다충인소,이차운행시간용역정학측량화획취,능구쾌속지고산응용정서운행시이동종단산생적능모。실험결과표명,재GT-I9108、GT-I9308화GT-P3108실험평태하,해모형적능모고산결과여Android조작계통응용정서광가제공적조건능모모형측량치상비평균오차분별위0.89%、1.37%화0.29%,능구위이동종단용호편첩지예측응용정서소모적전지전능제공방조。
Analysis and modeling of application energy consumption plays a vital role in the optimization of energy consumption of smart mobile devices. An application energy model based on application’s running time is proposed. Compared with application’s component energy models with high accuracy and complexity, the model is characterized by the time variable and contains a variety of mobile devices’ properties, such as power consumption and performance, which can be used to rapidly estimate the mobile devices’ energy consumption during application execution, and the execution time of the application is easy to measure and obtain. The experiment results show that the average error rate of proposed model is 0.89%, 1.37% and 0.29% compared with that measured by component energy model provided by Android application framework on GT-I9108, GT-I9308 and GT-P3108, respectively. The model can be used to help the end users to rapidly and conveniently predict the battery energy consumption of applications.