Design And Implementation Of Malicious Software Detection Scheme Based On Android

Mobile Internet developing fast,smart phones are no longer just traditional communication tools.Applications on Android can meet the needs of users in social contact,video,game,etc.Among them,smart phones with Android based on Linux kernel develop fastest.However,malwares consider the users' devices as the potential attack targets and cause malicious chargeback,privacy theft,and system damage.Consequencetly,detecting the security of applications,finding and cleaning up threat have important significance to ensure users' interests.Data traffic automatic consumption becomes the main security problems worried by users,therefore,this paper focuses on the detection and analysis of network connection,message sending.At first,this paper designs and realizes a malicious behavior dynamic detection system that simulate user to run the test software and monitor whether suspicious behavior happened in the running process.It describes the framework of the system and the function of each module and demonstrates the performance of the system through the experiments.Secondly,detect malicious behaviors from the static point of view.It designs a method based on feature matching and establishes a API database of triggering malicious behaviors.To achieve the purpose of behavior judgement,we extract the software's information of API call after decompiled and calculate the similarity.Two kinds of detection methods can verify results and reduce the rate of false positives and omissions.