Screening smartphone applications using malware family signatures

摘要: The sharp increase in smartphone malware has become one of the most serious security problems. Since Android platform taken dominant position popularity, number grown correspondingly and represents critical threat to users. This rise is primarily attributable occurrence variants existing malware. A set stem from can be considered as family, families cover more than half population. conventional technique for defeating use signature matching which efficient a time perspective but not very practical because its lack robustness against variants. As counter approach handling issue behavior analysis techniques have been proposed require extensive resources. In this paper, we propose an detection mechanism that uses automated family extraction matching. Key concept extract representative binary patterns evaluated members classify each into via estimation similarity signatures. offers flexible variant does legacy matching, strictly dependent on presence specific string. Furthermore, compared with previous considering detection, higher accuracy without need significant overhead data control flow analysis. Using signature, detect new known efficiently accurately by static We our 5846 real world samples belonging 48 collected April 2014 at anti-virus company; experimental results showed that; achieved greater 97% also demonstrated linear complexity target applications.