Safety Analysis Of Mobile Intelligent Terminal Software Behavior

With the rapid development of mobile communication technology and the emergence of mobile Internet, mobile intelligent terminal plays an increasingly important role in our lives, so that we enjoy more colorful and convenient services. However, with the continuous expansion of the scale of mobile intelligent terminal users and the more understanding of the technology, it faces increasingly serious security risk. The security issues on the mobile intelligent terminal has become one of the emerging frontier research hotspot with domestic and international concerned. Among the security risks, hazards caused by malware accounted for the vast majority. Currently, there are some of the more sophisticated methods of the identification and analysis of malware behavior at home and abroad, existing methods and techniques are invalid of it due to its inherent characteristics. Thus based on the in-depth research of mobile intelligent terminal and the technology of software category, software behavior analysis methods for mobile intelligent terminal is proposed for the classification and identification of malware software behavior, and to provide security for mobile intelligent terminals and mobile Internet.Taking the malware identified as mobile intelligent terminal research background, we analyze the structural characteristics and runtime software behavioral characteristics of mobile terminal software in the Android operating system environment. With the introduction of cloud processing and machine learning, static and dynamic software behavior analysis detection methods of intelligent mobile terminal is proposed:by comparing signature, static behavior analysis in the cloud analyze the software behavior by analyzing classes and functions calling information on mobile intelligent terminal; dynamic behavior detection in the cloud detect the software behavior by analyzing the calling system sensitive Native API sequence in the runtime. Taking these two methods, this paper presents a set of behavioral malware detection scheme.Weka is used to extracted attributes about subtraction with using cross-validation approach for program analysis and evaluation. The experiment result shows that comparing with traditional methods of software behavior analysis, it can effectively enhance the detection accuracy and the detection rate, and can also lower the use of the system resources accordingly for the proposed software behavior analysis program on mobile intelligent terminal.