Mitigating NDN-Based Fake Content Dissemination in Opportunistic Mobile Networks

Abstract

Listen

Translate article

In this work, we address the problem of Named Data Networking (NDN)-based fake content dissemination in Opportunistic Mobile Networks (OMNs), where nodes have intermittent connectivity and typically lack in end-to-end communication paths. It is important to mitigate the dissemination of such fake contents not only because they waste bandwidth for legitimate communication, but also because such files can be harmful for devices. However, the inherent characteristics of OMNs make such mitigation a challenging task. In this context, we consider a group of nodes, Fake Content Providers (FCPs), who, on receiving content requests, respond with fake contents, rather than the real version. In particular, we consider four different behaviors of the FCPs—referred to as threat scenarios—depending on whether or not they always respond to all requests with fake contents. We analyze these distinct threat scenarios, and characterize the relative performance degradation arising because of them. To mitigate the adverse effects of FCPs, we propose two schemes, wherein the identified FCPs are blacklisted permanently or temporarily, and communication with them is restricted. Results of simulation-based experiments using real-life connection traces show that, compared to the normal scenario with no FCP, the percentage of content requests satisfied decreases by 20–40 percent in the presence of 40 percent FCPs in the OMN. Moreover, in the same scenario, the average latency of content satisfaction relatively increases by up to 176 percent with respect to the normal scenario. However, on using the proposed mitigation schemes, the latency can be reduced by about 13–36 percent together with up to 9 percent improvement in the number of interests satisfied.