WRSR: wormhole-resistant secure routing for wireless mesh networks

Abstract

Wormhole attack is one of the most severe security threats in wireless mesh network that can disrupt majority of routing communications, when strategically placed. At the same time, most of the existing wormhole defence mechanisms are not secure against wormhole attacks that are launched in participation mode. In this paper, we propose WRSR, a wormhole-resistant secure routing algorithm that detects the presence of wormhole during route discovery process and quarantines it. Unlike other existing schemes that initiate wormhole detection process after observing packet loss, WRSR identifies route requests traversing a wormhole and prevents such routes from being established. WRSR uses unit disk graph model to determine the necessary and sufficient condition for identifying a wormhole-free path. The most attractive features of the WRSR include its ability to defend against all forms of wormhole (hidden and Byzantine) attacks without relying on any extra hardware like global positioning system, synchronized clocks or timing information, and computational intensive traditional cryptographic mechanisms.

Secure Multicast Routing Algorithm for Wireless Mesh Networks

“…It facilitates nodes to monitor their two-hop subpaths and selects only such routes that are verified to be wormhole-free. It relies on the observations made in [36] to distinguish out-of-band links from real links. It relies on the fact that, in a uniformly distributed network, alternate routes within a maximum of 4-hops exist between node separated by a distance of 2R, where is the transmission range of a node.…”

“…But it has been already proven that finding such an alternate path is negligible in [36]. Therefore, the only way to accept such an incorrect routing entry is by fabricating an alternate Join Request traversing a nonexisting alternate subpath.…”

“…A path is said to be free from wormhole links if and only if for each subpath of length 2R there exists an alternate subpath of maximum length 4R, where is the transmission range of a node. The proof of concept is presented in [36]. SEMRAW thrives on the fact that the probability of finding alternate paths between nodes separated by a distance ( < < 2 ) (in a network where nodes are uniformly distributed) is high.…”

“…SEMRAW allows nodes to monitor the two-hop subpath on a received Join Request and identify a JRQ that traverses an out-of-band link like a wormhole. A JRQ that traverses an out-of-band link would not satisfy the necessary wormhole-free path criterion [36], which can be detected at the neighbors of a wormhole node and is quarantined. A path is said to be free from wormhole links if and only if for each subpath of length 2R there exists an alternate subpath of maximum length 4R, where is the transmission range of a node.…”

“…The public and private keys of a node are represented by and V , respectively. It also employs a wormhole detection mechanism presented in [36] to distinguish real links from links traversing a wormhole. The proposed SEMRAW operates in following 3 phases.…”