Replication Mechanism in Cognitive Mobile Ad Hoc Networks
Dynamic trust models and replication approaches that adequately address the security needs of wireless cognitive ad hoc networks (MANET) have proven quite difficult to develop. The inherent decentralized nature of ad hoc networks and their collaborative services give rise to a series of vulnerabilities that can be exploited by malicious entities. This problem is exacerbated by the fact that risk assessment models show high levels of complexity, which suggest significant domain constraints and lack of true human-like reasoning. Recent work in the area of HTM demonstrates the ability to mimic higher level cognitive skills. This paper discusses the potential of using HTM to improve human-like reasoning in the replication of trust information in cognitive MANETs.
From a human perspective, trust is a very subjective measure. There are numerous elements or variables used to represent its value. As suggested by D’ Arienzo , the ability of a system to adapt to ever changing requirements without human intervention is a clear challenge that needs to be resolved. This lack of dynamic adaptation can also be extended to existing trust models and in particular to trust data distribution methods   , which are responsible for the accurate and timely dissemination of trust changes in order to protect the entire network. Current models acknowledge attributes changing their values, but they do not acknowledge the implication of inaccurate trust data distribution. This inability places significant restrictions on ad-hoc networks, in particular cognitive networks, to only allow for scenarios where a compromised device can successfully detect malicious behavior and isolate itself. The fact that a number of attacks utilize one or more compromised devices in the target network to further the attack clearly demonstrates that this standard approach is not enough. For example, the decision of a secondary device in an ad hoc network to establish a connection with another initiating secondary can be made by evaluating trust values . If a malicious device is trying to gain access in order to hijack another device and the attributes used to determine the trust value are compromised, the secondary device and ultimately the entire network can be compromised. Trust evaluations can be enhanced by leveraging automated risk assessment techniques. Novel approaches to accomplish have been proposed and studied in recent years. They leverage research across multiple disciplines, including, but not limited to, artificial intelligence, game theory, knowledge management, and others. However, existing risk assessment models are limited to predetermined set of mathematical models and functions, and, although some of them attempt to leverage artificial intelligence (e.g. neural networks), their capabilities are constrained to specific domains and limited by the lack of temporal awareness.