指導教授：蘇雅韻臺灣大學：資訊網路與多媒體研究所陳政柏Chen, Cheng-PoCheng-PoChen2014-11-292018-07-052014-11-292018-07-052014http://ntur.lib.ntu.edu.tw//handle/246246/263450隨著智慧型手機的普及，其應用軟體也隨之快速增長，然而即時通訊軟體也為其一。本篇論文旨在探討搭載安卓系統的智慧型手機在安裝多個即時通訊軟體下，對其電量帶來過量耗損的情形，並提出一套程式設計架構，以解決此一現象。With the development of mobile devices, more and more applications has been added to its application store by developers and installed by users in their own devices. However, one main concern for mobile users is that whether an application drains a device’s battery. Therefore, improving battery life always is a popular issue in recent years because of limited battery. One popular application on the market is a mobile instant messaging (MIM) application. MIM applications allow mobile users owning devices with different platforms (e.g., Android, iOS) to communicate through text, voice, or images over Wi-Fi or cellular networks. This thesis targets to reduce power consumption on MIM applications and examines the application and energy behavior of users who install multiple MIM applications on their phones. We designed and implemented Consultant, a light-weight platform to optimize energy consumption for MIM applications. Consultant analyzes the data provided by different MIM applications and help applications to better use phone resource. In our evaluation, we replayed the synthetic real trace collected from the users and experimental results showed that Consultant can save 52% MIM energy with 3 installed MIM applications during a day.摘要i Abstract ii 1 Introduction 1 2 Problem 3 2.1 Mobile Instant Messaging . . . . . . . . . . . . . . . 3 2.2 Radio Power in 3G Cellular Network . . . . . . . . . . 4 3 Mobile Instant Messaging Application Behavior 7 3.1 Measurement and Detecting . . . . . . . . . . . . . . 7 3.2 Evaluate Detecting Flow . . . . . . . . . . . . . . . 10 3.3 Measurement Results . . . . . . . . . . . . . . . . . 10 3.4 Instant or Non-instant .. . . . . . . . . . . . . . . 11 4 Power-Saving Strategies 15 4.1 Keep-alive Mechanism . . . . . . . . . . . . . . . . 15 4.2 Message-Delayed Delivery . . . .. . . . . . . . . . . 16 4.3 APIs . . . . . . . . . . . . . . . . . . . . . . . . 17 5 Implementation 20 5.1 Client-side . . . . . . . . . . . . . . . . . . . . . 20 5.2 Server-side . . . . . . . . . . . . . . . . . . . . . 20 6 Evaluation 22 6.1 Methodology . . . . . . . . . . . . . . . . . . . . . 22 6.2 Consultant for Keep-alive . . . . . . . . . . . . . . 23 6.3 Trace Replay . . . . . . . . . . . . . . . . . . . . 24 7 Related Work 29 8 Conclusion 31 Bibliography 321952997 bytesapplication/pdf論文公開時間：2014/07/29論文使用權限：同意無償授權行動計算即時通訊安卓系統耗電量保持活動[SDGs]SDG7thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/263450/1/ntu-103-R01944039-1.pdf