A lightweight mechanism for detection of cache pollution attacks in Named Data Networking

Content-Centric Networking (CCN) is an emerging paradigm being considered as a possible replacement for the current IP-based host-centric Internet infrastructure. In CCN, named content - rather than addressable hosts - becomes a first-class entity. Content is therefore decoupled from its location. This allows, among other things, the implementation of ubiquitous caching. Named-Data Networking (NDN) is a prominent example of CCN. In NDN, all nodes (i.e., hosts, routers) are allowed to have a local cache, used to satisfy incoming requests for content. This makes NDN a good architecture for efficient large scale content distribution. However, reliance on caching allows an adversary to perform attacks that are very effective and relatively easy to implement. Such attacks include cache poisoning (i.e., introducing malicious content into caches) and cache pollution (i.e., disrupting cache locality). This paper focuses on cache pollution attacks, where the adversary's goal is to disrupt cache locality to increase link utilization and cache misses for honest consumers. We show, via simulations, that such attacks can be implemented in NDN using limited resources, and that their effectiveness is not limited to small topologies. We then illustrate that existing proactive countermeasures are ineffective against realistic adversaries. Finally, we introduce a new technique for detecting pollution attacks. Our technique detects high and low rate attacks on different topologies with high accuracy.