Adaptive Service Dependent Proximity Analysis Based Intrusion Detection in Cloud Environment

Intrusion attack is the most dominant threat faced by cloud environment, which targets the service behavior and QoS metrics. To handle such threats, number of approaches is defined in literature which works on data level, behavior level, and service level and so on. However, the methods face inclined performance in security as well as QoS achievement. To improve the precision of intrusion attack detection, an Adaptive Service Dependent Proximity Analysis (ASDPAM) based intrusion detection model is presented in this article. The proximity of threat against the services is the main focus in this approach. To perform this, the model starts with categorizing the services under various sensitive levels as the services would handle different data which are sensitive, highly sensitive, normal and moderate sensitive. By categorizing the services in this way, helps the model to define the threshold for the threat level and support the detection of intrusion attack. By receiving any service request, the model analyze the proximity of threat would be for the concern service request received. Using the service trace maintained, the model preprocesses the logs of service as well as user initially. Further, the method computes the User Threat Proximity Score (UTPS) for the service threat proximity Score (STPS). Considering the worth of UTPS and STPS the technique determines the intrusion attack. The suggested paradigm enhances intrusion attack detection performance and improves the QoS performance as well.