A critical issue for small and low-cost Internet of Things (IoT) devices facing multiple complex, advanced and interactive tasks trying to save their power resources. To reach these goals IoT devices can use the capabilities of nearby devices having suitable resources, given that they make their resources available for free or with a determined cost. In such a context, IoT devices can take significant benefits by exploiting the social attitude of software agents to mutually interact and cooperate with other agents they consider as trustworthy. However, in wide communities it is common that a lot of members are unreferenced with respect to the own trustworthiness and, therefore, the task of carrying out a reliable choice about a potential partner can be very difficult. To tackle such an issue, we propose an agent framework where each IoT device is associated with an agent that helps its device in choosing reliable partners for its tasks. To this aim, we designed a reputation model implementing some countermeasures against malicious IoT devices. To verify the efficiency and effectiveness of our proposal, we carried out some experiments in a simulated scenario, which confirmed the potential advantages deriving by its adoption.
File in questo prodotto:
Non ci sono file associati a questo prodotto.