Design And Implementation Of Malwares Detection System On Android

In recent years, with the rapid development of Mobile Internet, smart phones with mobile operating system become the mainstream of mobile device development. Among them, smart phones with Android based on the Linux kernel develop most rapidly. Applications on Android can meet the users to browse the Internet, send and receive e-mail, SMS, and exchange data to other devices needs; but the malwares of them also consider the users’smart phones as the potential attack targets. Among them, malicious chargeback, privacy theft, and system damage are the main hazards of malwares. Consequently, it is an important issue to software manager that how to evaluate and measure the security of software effectively, which uploaded to Internet or the Application Store.This paper points out the advantages and disadvantages of current malicious code detection technology on Android. It puts forward that from the perspective of source code, we use static analysis technology of object-oriented language to analyze the Java source code decompiled from android applications, can effectively get the possible malicious behaviors of input applications by finding the key API calls and analyzing the sensitive data sources. It solves the problems of the reverse analysis of Android applications, semantic understanding of malicious code, data flow tracking of sensitive data, and determination of user submits behavior. This paper gives the design and implementation of malwares static detection system on Android. Through experimental tests to verify the system can effectively detect the common malicious behavior at a certain rate of false positives, false negative range. On the other hand, we design a malicious behavior dynamic detection system on Android based on Android dynamic automated testing and end-monitor technologies. Dynamic analysis is a good addition to static analysis, it monitors software whether to have network connection and message sending behavior in the actual operation process by simulating to execute the software in a real mobile phone. It further verifies the result of static analysis.