基于鼠标和键盘行为特征组合的用户身份认证  

USER IDENTITY AUTHENTICATION BASED ON MOUSE AND KEYBOARD BEHAVIOURAL BIOMETRICS COMBINATION

作　　者：王振辉[1] 王振铎[2] 支侃买[1] 

机构地区：[1]西安翻译学院工程技术学院,陕西西安710105 [2]西安思源学院电子信息工程学院,陕西西安710038

出　　处：《计算机应用与软件》2016年第7期308-312,共5页Computer Applications and Software

基　　金：陕西省教育厅科研计划项目(12JK1055)

摘　　要：为解决单一行为特征存在的不足和提高认证识别率,提出一种基于鼠标和键盘行为特征组合的双指标用户身份认证方法。首先分别提取鼠标和键盘两种指标的行为特征,然后利用支持向量机进行模式识别,实现特征分析和验证,以达到实时监测用户身份、检测非法用户的目的。最后通过多个用户采集鼠标和键盘行为数据进行身份识别与认证实验。结果表明,相对于单一行为特征,该方法提高了用户身份认证的识别率,降低了误识率和拒识率,而且结果优于BP和SOM方法,充分展示了双指标身份认证的高可靠性。In order to solve the drawback of single behavioural biometrics and to improve the recognition rate of authentication,we put forward a dual-indicator user identity authentication method which is based on the combination of mouse and keyboard behavioural biometrics.First,we extract the behavioural biometrics of two indicators of mouse and keyboard separately,and then use SVM for pattern recognition so as to implement biometrics analysis and verification in order to achieve the goal of timely monitoring users' identities and detecting illegitimate users. Finally,we collect through a group of users the mouse and keyboard behavioural data to carry out the experiment of identity recognition and authentication. Result indicates that compared with single behavioural biometrics,this method increases the recognition rate of user identity authentication and decreases the false accept rate and false reject rate. Besides,the result is better than the BP and SOM methods,which fully shows the high reliability of dual-indicator identity authentication.

关 键 词：行为特征 身份认证 支持向量机 击键特征 鼠标行为 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]