Detecting vulnerabilities in secure communication using RCEE

Security and privacy issues are important with the multi-agent systems such as pervasive computing, grid computing and P2P networks which are open, anonymous and dynamic in nature. These characteristics introduces vulnerabilities and threats to providing secured communication. To minimize these threats and to manage altering behavior of malicious agent, it evaluates the trust and reputation of the interacting agents; a dynamic trust computation model called `Secured Trust´. It uses different factors to evaluate the trust of an agent and proposes a novel load balancing algorithm and effectively manage with strategic behavioral change of malicious agents and distribute workload among the service providing agents under stable condition by finding the satisfaction level. These are implemented in client server architecture to self defense the application and to detect the threat from the server application itself and runtime code execution behavior monitoring is implemented. At runtime, a monitor compares the behavior of the variants at certain synchronization points and raises an alarm when a difference is detected.