A 3-Tier Architecture for Network Latency Reduction in Healthcare Internet-of-Things Using Fog Computing and Machine Learning

Abstract

Healthcare Internet-of-things comprises a huge number of wearable sensors and interconnected computers. The high volume of IoT data is transacted over servers leading to servers overloading with high traffic causing network congestion. These cloud servers are typically for analyzing, retrieving and storing the large data generated from IoT devices. There exist challenges regarding sending real-time healthcare data from cloud servers to end-users. These challenges include the high computational latency, high communication latency, and high network latency. Due to these challenges, IoTs may not be able to send data in real-time to end-users. Fog nodes can be used to play a major role in reducing the high delay and high traffic. It can be a solution to increase system performance. In this paper, we proposed a 3-tier architecture, an analytical model for healthcare IoT using a hybrid approach consisting of fuzzy logic and reinforcement learning in a fog computing environment. The aim is to minimize network latency. The proposed model and 3-tier architecture are simulated using iFogSim simulator.