CEP隐私安全事件属性检测树及访问控制框架

CEP Privacy Security Event Attribute Detection Tree and Access Control Framework

为防止非法用户通过注册事件模式的方式获取复杂事件处理(Complex Event Processing,CEP)过程中所涉及到的关键信息,在不影响CEP处理效率的情况下,提供隐私安全保护的能力.文中讨论了CEP的隐私安全访问控制对象;形式化定义了完全读取、部分读取、拒绝访问及数量统计等四类事件属性操作符;给出了以事件属性检测树为运行机制的隐私安全保护引擎;提出了一种新颖可行的访问控制框架.实验结果表明:该框架隐私安全检测引擎运行效率控制在毫秒级别,事件流吞吐率达千个event·s-1以上,实现了隐私信息的高效过滤.

The study aims to prevent illegal users from accessing key information in the complex event processing in the way of event pattern registration and to provide privacy protection without affecting the processing efficiency. The paper discusses the object of CEP privacy security access control, and gives formalized definition of the four types of event attribute operators--fully reading, part reading, access rejection,and quantitative statistics. Also the privacy security engine is given, and a novel and feasible CEP privacy access control framework is proposed. Experimental results show that the framework privacy security detection engine operates only at the level of milliseconds, and the event stream throughput rate is over 1 000 events per second. It is concluded that this framework can filter privacy information with high efficiency.