Defending Against Smart Grayhole Attack Within MANETs: A Reputation-Based Ant Colony Optimization Approach for Secure Route Discovery in DSR Protocol

Abstract

Moving randomly without any centralized authority, dynamic nodes constitute the Mobile Adhoc Networks on the basis of fully-fledged cooperation and native trustworthiness. Unfortunately, in real scenarios, the malicious nodes take advantage of this inherent trustworthiness to settle in and perform their suspicious activities. In the light of the growing concerns about security attacks in hostile environments, new challenges have emerged to thwart routing attacks including the smart grayhole attack, which adversely affects the availability and accuracy of the network by dropping data packets. This paper addresses this disturbing attack by monitoring the behavior of the participating nodes through a bio-inspired trust management model. The distributed trustworthiness assessment model is based on the beta reputation system combined with the Ant Colony Optimization (ACO) metaheuristic. The main focus of the beta reputation system is rating the nodes according to their successful tasks and their consumed energy, while the ACO metaheuristic maintains this reputation metric during the discovery process and calculates the preference value of each traversed path to select the most secure one. The proposed model improves the traditional Dynamic Source Routing (DSR) protocol by isolating the malicious nodes from participation in data packets transmission. The simulations conducted with the network simulator 2, show that despite the presence of numerous gray holes, the reputation-based ACO DSR (RACODSR) outperforms the standard DSR in terms of packet loss ratio by a decrease of 9.8%, packet delivery ratio by an increase of 0.22%, throughput by an increase of 0.4%, jitter by a decrease of 22.76%, the end to end delay by a decrease of 2.51% and the consumed energy by a decrease of 0.17%.