An efficient content replacement policy to retain essential content in information-centric networking based internet of things network

Abstract

In the current scenario, TCP/IP is widely used in the Internet of Things (IoT) for disseminating data, addressing, and controlling traffic. However, the exponential growth of IoT devices has incurred the TCP/IP-based network from mismanagement due to heavy traffic, address space scarcity, and low efficiency. In the last few years, Information-centric networking (ICN) has emerged as an excellent choice to address the above issues. One of the important characteristics of ICN is In-network Caching, in which contents are cached at the intermediate nodes. Most in-network caching strategies use traditional replacement policies such as First in first out, least recently used, and least frequently used, ignoring the essential content properties, eventually evicting the vital contents. This work introduces a novel content replacement scheme that proactively records the number of requests and access time of the content in the Pending interest table (PIT) and considers essential content attributes, e.g., number of requests, remaining lifetime, and upcoming request waiting time, before replacement to retain vital contents in the cache. The simulations are performed on the ndnSIM framework of NS-3 to evaluate the performance of our proposed replacement scheme with other widely used benchmark replacement schemes and tested under six different caching strategies. The experimental finding indicates that, on average, the proposed replacement scheme shows 28.94% and 27.49% increase in cache hit ratio and server hit reduction ratio while 22.30% and 30.64% reduction in response delay and energy consumption compared to NDN-FIFO, NDN-LRU, and NDN-LRFU.