Research On Adaptive Access Control Model Based On Risk And Data Value In Cloud Environment

With the official rollout of 5G commercial application in China,cloud computing and big data development have opened a fast lane.Cloud platforms in various industries,such as storage cloud,medical cloud,financial cloud,transportation cloud,education cloud and social security cloud,are being rapidly promoted.In order to reduce the repetitive construction and corresponding costs of resource platforms,and to make it convenient for more people to make full use of existing data resources through fewer resource platforms,improve the utilization rate of information resources,and thus improve the overall social benefits,data sharing is bound to become a future trend.With the rapid emergence of a large number of new applications and demands,cross-industry and cross-field applications will become the mainstream.In terms of cross-industry and cross-domain access control,it is difficult to meet the application scenarios of big data and cloud environment with pre-defined strict and unchanging access control policies and traditional access control technologies,and even some business authorization requirements cannot be realized at all.At present,risk-based access control technology has been favored by researchers.This paper abandons the research method of introducing risk factors based on roles and attributes,and takes the risk of access request and data value as the research entry point,supplemented by XACML to form an adaptive access control model based on risk and data value.The model defines the risk engine,data engine and risk policy module,classifies the risk access request and data value,and dynamically adjusts the risk policy module through the classification.A cyclic dynamic system is formed in which access request determines data value,data value determines risk policy and access result is determined by risk policy.On the basis of predecessors,this paper makes dynamic allocation optimization for the weight of context risk factors,improves the quantification of availability,integrity and confidentiality risk factors combined with data value,and makes in-depth research on dynamic adjustment of historical risk weight.The k-means algorithm is used to dynamically grade the risk of access request,which ensures the dynamic and high efficiency of risk grading of access request.XACML strategy is used to assist the processing of special requirements,which improves the controllability and flexibility of the system.In the end,the performance,adaptability and optimization direction of the adaptive access control model based on risk and data value are evaluated through simulated data experiments under Tencent cloud environment.