面向Android手机平台异常入侵检测的研究

发布时间：2018-04-13 23:21

本文选题：智能手机 + Android　；参考：《山西大学》2014年硕士论文

【摘要】：自2007年以来,智能手机的发展越来越快。智能手机已也不再是仅仅只有电话和短信等基本功能,而是集多媒体、办公、娱乐、上网等于一体的多功能手机。它能通过移动网络与远程互联网的不同服务端相连,下载大量信息存入手机,而这些信息对于许多用户而言,不知道安全与否。一旦用户错误下载了恶意软件或者其它错误信息,就有可能将自身的所有正常信息全都泄露出去,给用户自身带来无可估量的后果。因此,智能手机入侵检测的研究迫在眉捷。本文根据这个现状针对智能手机异常入侵检测进行了研究。入侵检测技术作为计算机安全领域的关键技术,其发展理论已经逐步走向成熟。但由于智能手机的起步较晚,在智能手机入侵检测领域的研究可以说处于起步阶段。本文在面向Android智能手机平台下,选用异常入侵检测作为研究方向,将异常入侵检测与智能手机相结合,从而达到提高手机系统对异常入侵的防范能力。文中依据异常入侵检测的思想与原理,设计并给出了面向Android手机平台下的异常入侵检测系统的总体框架设计,并对数据特征选取、数据采集模块、数据库模型以及处理算法和响应模块进行了详细的分析。其中还给出了数据采集的详细过程及关键代码。文中在Android平台下采集CPU利用率、内存信息、进程信息、网络流量信息等智能手机系统信息以及用户操作信息,将其作为数据来源,经过筛选整理形成训练集以及测试集后,采用不同的SVM算法进行实验分析,并统计记录不同的实验结果。而本文实验采用了原始的C-SVM算法以及改进的VC-SVM算法对设计的系统采集整理的数据进行实验。实验过程中选用了不同的核函数以及不同的参数对数据进行分析计算,取得了较好的实验结果。最后对得出的结果进行了详细的分析与对比。结果表明,该算法能很好的适用于Android平台,得到较高的检测率和较低的误报率。将其与Android智能手机平台相结合,能有效改善手机的主动防御性能力,从而提高手机对异常入侵的防范能力。
[Abstract]:Since 2007, smartphones have been growing faster and faster.Smartphone is no longer just basic functions such as telephone and SMS, but a multi-functional mobile phone that integrates multimedia, office, entertainment and Internet.It can connect to different services of the remote Internet via mobile network and download a large amount of information into the mobile phone which for many users do not know whether safe or not.Once the user has downloaded malware or other wrong information wrongly, it is possible to leak all the normal information, which will bring incalculable consequences to the user himself.Therefore, the research of smart phone intrusion detection is urgent.According to this situation, this paper studies the anomaly intrusion detection of smart phone.As a key technology in the field of computer security, intrusion detection theory has gradually matured.However, due to the late start of smart phone, the research in the field of smart phone intrusion detection can be said to be in its infancy.In this paper, anomaly intrusion detection is chosen as the research direction under the platform of Android smart phone, which combines abnormal intrusion detection with smart phone so as to improve the ability of mobile phone system to prevent abnormal intrusion.According to the idea and principle of anomaly intrusion detection, this paper designs and gives the overall frame design of anomaly intrusion detection system based on Android mobile phone platform, and selects the data feature and data acquisition module.The database model, processing algorithm and response module are analyzed in detail.The detailed process and key codes of data acquisition are also given.In this paper, CPU utilization, memory information, process information, network traffic information and user operation information are collected under the Android platform, which are used as data sources. After screening, training set and test set are formed.Different SVM algorithms are used for experimental analysis and different experimental results are recorded.In this paper, the original C-SVM algorithm and the improved VC-SVM algorithm are used to test the data collected by the designed system.In the experiment, different kernel functions and different parameters are selected to analyze and calculate the data, and good experimental results are obtained.Finally, the results are analyzed and compared in detail.The results show that the algorithm can be well applied to Android platform, with high detection rate and low false alarm rate.Combining it with Android smart phone platform, it can effectively improve the mobile phone's active defensive ability, thus improve the mobile phone's ability to prevent abnormal intrusion.
【学位授予单位】：山西大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.08

【参考文献】