A Cooperative Watchdog System to Detect Misbehavior Nodes in Vehicular Delay-Tolerant Networks

In vehicular delay-tolerant networks (VDTNs), an end-to-end relay path between bundle source and destination nodes may not be available. To accomplish such goal, VDTNs rely on nodes cooperation. Thus, in order to maintain network efficiency, it is very important to ensure that all network nodes follow the protocol. This is not an easy task since nodes may diverge from the protocol due to a selfish behavior or to maintain their data or resources integrity. This paper proposes a cooperative watchdog system to detect and act against misbehavior nodes in order to reduce their impact in the overall network performance. Its operation relies on a cooperative exchange of nodes reputation along the network. By detecting selfish or misbehaving nodes, it is possible to improve the overall network performance. Conducting simulation experiments considering the VDTNSim tool and the Spray-and-Wait routing protocol shows that the cooperative watchdog proposed for VDTNs is not only effective in detecting misbehaving nodes but also contributes to improve the overall network performance by increasing the bundle delivery probability and decreasing the amount of resources waste.

I. Introduction

VEHICULAR delay-tolerant networks (VDTNs) [1] were proposed as new kind of vehicular networks, whose design supports communications in environments where an end-to-end path between the source and destination may not be available. Like other ad hoc networks [2]– [4], VDTNs rely their operation on cooperation between mobile nodes (e.g., vehicles), which are exploited to store-carry-and-forward bundles. VDTNs consider three kinds of nodes: mobile, terminal, and relay nodes. Mobile nodes move along paths and may interact with the other two types of VDTN nodes. Terminal nodes are usually placed at the edge of the VDTN network and are responsible for the heavy data processing and interaction with other networks (such as the Internet), whereas relay nodes are placed at road intersections increasing the number of network contacts and storing a higher number of bundles that can be picked by any passing-by vehicle. Contrary to other vehicular networks, in VDTNs, each contact opportunity is processed in two phases: control plane and data plane phases (performing out-of-band signaling). At the beginning of a contact opportunity (using the control plane) nodes exchange signaling information (e.g., speed, buffer status, destination node) in order to setup and reserve resources for an appropriate transmission of data bundles at the data plane. In the data plane, datagrams are aggregated into bundles and forwarded to a single or multiple destination nodes. This out-of-band signaling approach offer the possibility to use different network technologies in each plane and improves the overall network performance since nodes, based on the signaling information, may decide to reject a contact opportunity in order to save resources or to prevent data from being compromised. Although all the already achieved improvements, VDTNs still dealing with the presence of misbehavior nodes that do not follow the protocol and severely affect the overall network performance. Usually, this kind of nodes exploits and consumes other nodes resources serving only their purposes. For example, a node that drops bundles without sent them at least once may be classified as a selfish node. This kind of nodes also leads to a huge waste of network resources, and may compromise the performance of well-behaved nodes. This situation makes very important to detect and take some kind of action against such nodes. However, this is a challenged task due to the high mobility of vehicles that increases the ambiguity of their detection and classification. A possible solution for this problem is to afford nodes with sophisticated mechanisms that can detect and avoid nodes with suspicious behavior.