Trust and exclusion in Vehicular Ad Hoc Networks: An economic incentive model based approach

In this body of work, we are interested in road safety applications such as advanced driver assistance systems, based on Vehicular Ad hoc Networks (VANETs). One of the particular characteristics of this kind of networks is the continuous sharing of safety information by its nodes. Since this kind of information is time sensitive, a node cannot spend much time to verify its validity with an authority. However, the presence of malicious and selfish nodes in VANETs corrupts exchanged data, and lowers the overall data reception ratio in the network. To tackle this, we propose a new incentive model with exclusion for malicious nodes called VIME. VIME is inspired from the signaling theory from economics. It is based on managing a credit count that each node receives at the initialization of the application. Straightforwardly, VIME is based on two pillars. On the one hand, a node pays an appropriate cost for each sent message, which is seen by the receivers as a guarantee from the source about the truthfulness of the information. On the other hand, nodes get rewarded for cooperating in the network. The proposed economic model allows computing the amounts to be paid and those to be awarded in order to fight selfish and malicious nodes. We validate our approach via simulations. We show that VIME is able to detect and evict gradually all malicious nodes in the network, and decreases the ratio of corrupted and false sent data until reaching zero. Moreover, it has a positive impact on the participation of selfish nodes, as our approach increases the average ratio of sent data as to equal the ideal case´s percentage, when no selfish node is present.