Edge Computing for IoT Systems: Architecture, Applications, Challenges and Future Directions

Abstract

Internet of Things (IoT) systems are generating very large amount of data from sensors, devices and smart objects. Traditional cloud computing model is not always suitable for IoT applications because of latency, bandwidth consumption and privacy concerns. Edge computing is emerging as an effective solution by processing data near to the source rather than sending everything to distant cloud servers. This paper presents a detailed review of edge computing for IoT systems including its architecture, enabling technologies, applications, benefits and challenges. Different edge architectures such as fog computing, mist computing and cloudlets are discussed. Role of edge computing in smart cities, healthcare, agriculture, industrial IoT and transportation is also explained. The paper also includes comparison tables and figures to understand the concepts better. Finally, challenges and future research directions are presented. Edge computing is expected to play very important role in next generation IoT ecosystem.

KEYWORDS: Edge computing, IoT, fog computing, latency, smart devices, data processing, real-time systems.