中文摘要：

提出了一种可生存系统的自主可识别性机制。按此机制，可识别性通过定义面向生存性的可识别性参数实现，自主性通过自主识别单元控制进程实现。首先，定义了若干可识别性参数，依据可识别参数的累计分布函数确定了动态可变的阈值约束；然后，基于可识别参数的参考基点，给出了服务连接可识别性检测的计算方法，并通过自主识别单元来控制实现服务连接的断开和系统资源的释放；最后，在Emulab环境下使用实时的ClarkNetwww服务器的访问数据进行了仿真性能实验，结果显示上述方法可以有效地识别出非法连接并且重新分配系统资源给合法用户，提高了系统可生存性的自主认知能力和服务承载能力。

英文摘要：

An autonomous recognition mechanism for a survivable system was presented. According to the mechanism, the recognition can be achieved by definition of the recognition parameters referred to the survivability, and the au- tonomy can be realized through the process for control of the autonomous recognition unit. Firstly, a number of rec- ognition parameters were defined, and according to the cumulative distribution function, the dynamic and variable threshold constraints were determined ; Then, based on the reference mark of the recognition parameters, a calcula- tion method for recognition detection of service connections was given, and the autonomous recognition unit was used to implement service connection, disconnection and system resource release; Finally, in the Emulab environ- ment, the accessing data of the real-time ClarkNet WWW server was used to carry out the simulation experiment. The results showed that the proposed mechanism effectively recognized the illegal connections and reassigned the system resources to legitimate users, with the improvements of the survivable system＇ s service capacity and autono- mous cognitive ability.