Design And Application Of Access Control Frame Based On History Characteristic

This thesis mainly researches on security property for open network information resource sever. This kind of sever aims for the share of information .Its security requirement is not the confidentiality of information ,but the usability and controllablity of the sever. Therefor, the guidelines of security design is not to restrict user's access, but to guarantee the function of the sever while keeping the attack away. So we adopt an unusual access control policy that only restrict the users who have criminal record. The information resource sever collects the historical context of user access, and does operations with probability,and take the result as the criterion to decide whether or not to provide the service to the user.To the users who don't have criminal record,the sever can always be accessed.This paper follows the mind and theroy of access control. At first ,we analyze the early work of access contorl and the existed frameworks of access control .We try to find the frameworks that can satisfy the policy that we propose and solve the problem in practice .This thesis focuses on the probable forecast based on historical character.This method takes the historical information as main property, and implements quantitative analysis, then take the result as the criterion for decision-making.We adopt the main idea of the Flask architecture,extend the procedure of historical information collecting, storage, abstracting, and operating .We propose a model of access control based on probable forecast.With this model ,we design a framework of access control based on probable forecast,and describ the main parts of the framework detailedly.Finally we implement the framework of access control based on probable forecast in a small-scale open network ,and complete the designment of the framework ,then we show the feasibility of probable forecast.