An Efficient Data Communication Framework for Named Data Networking in IoT-Edge Ecosystem

Abstract

Internet of Things (IoT) has evolved as a revolutionary technology that allows smart devices and sensors to monitor the environment (like smart agriculture, fleet management, smart healthcare, intelligent transportation, etc) wherein they are deployed. Thus, a huge volume of multidimensional data is generated that must be processed promptly for the success of the underlying IoT applications. However, the traditional communication technologies (like TCP/IP) are rigid and limit the scalability of large-scale IoT deployments. Thus, a data-centric framework is required to provide flexible connectivity and data availability for IoT applications. Named Data Network (NDN) is a data-centric network paradigm that does not rely on conventional IP-based technologies and can provide answers to several challenges of IoT applications that are otherwise not resolved by conventional communication technologies. However, it requires an efficient mechanism to handle the large volume of data and store it on the cache of the NDN node. Edge Computing provides the compute (and caching) capability for handling IoT data at a location closer to its origin. Thus, the idea is to utilize the benefits of edge computing and NDN to design an efficient data communication framework. The proposed framework utilizes the edge computing nodes as a cluster head for numerous IoT devices deployed in a specific IoT application (like healthcare). It works as two subsystems, a) a sensor data processing and recommendation system, and b) an NDN-based data storage and delivery system. It was validated using ndnSIM for a hybrid topology concerning the delay, throughput, and packet rate.