A novel content eviction strategy to retain vital contents in NDN-IoT networks

Information-Centric Networking (ICN) may serve as future internet architecture in a dense, high-demand IoT environment to provide reli- able data delivery between subscribers and publishers. Named Data Networking (NDN) is a well-recognized architecture of ICN for fast data delivery and availability for both persistent and IoT data. One prominent feature of NDN is in-network caching, which stores data in the intermediate caching nodes to provide efficient data dissemination. When stored in these nodes, the cached IoT data becomes obsolete due to its ephemeral nature. Various cache content eviction policies have been proposed in the literature for replacing the existing content with new ones. However, they don’t consider essential characteristics of the IoT content while making the eviction decision, eventually leading to more cache misses. This paper proposes a Popularity, Freshness, and Recency (PFR) based cache content eviction policy to retain essential content in cache in NDN-IoT networks. Extensive simulations are performed to compare the PFR with other benchmarks NDN cache content eviction policies. Simulation results show that, on average, PFR outperforms other benchmark cache content eviction policies in terms of the cache hit ratio, server hit reduction ratio, average response delay, and average energy consumption by 27.43%, 21.56%, 15.85%, and 7.24%, respectively.