E-APK: Energy pattern detection in decompiled android applications

Energy efficiency is a non-functional requirement that developers must consider, particularly when building software for battery-operated devices like mobile ones: long-lasting battery an essential enjoyable user experience. In previous studies, it has been shown many applications include inefficiencies cause to be drained faster than necessary. Some of these result from patterns have catalogued, and which more energy-efficient alternatives are also known. The existing catalogues, however, assume as fundamental one access the source code application in order able analyse it. This makes independent energy analysis challenging, or even impossible, e.g. or, most significantly, App Store trying provide information on how efficient being submitted publication is. We study viability looking known by decompiling them analysing resulting code. For this, we decompiled analysed 420 open-source extending tool, now capable transparently android applications. With collected data, performed comparative presence four between two types analysis: (i) comparing total number pattern detections; (ii) similarity detections. When detections against code, found 79.29% reported same To test APKs, calculated, each application, score based our implemented detectors. Of all applications, 35.76% achieved perfect 4, 89.40% got 3 out 4. Furthermore, only 0. viewed tandem, results analyses point promising direction. They initial evidence static techniques, typically used can viable method inspect APKs restricted, further research this area worthwhile.