Chapter 4 - An overview of low power hardware architecture for edge computing devices

Abstract

With the rise in cutting-edge artificial intelligence technology, 5G networks and the Internet of Things (IoT), enormous amounts of data are produced at the source and are required to be processed in the cloud. A large amount of power and bandwidth is required to transfer the data from the source to the cloud. Consequently, healthcare applications demanding seamless real time data processing and analysis must be able to compute the data closer to the source, that is, the edge. Edge computing devices are the hardware components that harness the application of edge computing thereby reducing the latency of data transfer and processing. Therefore this chapter proposes the hardware architecture for edge computing devices considering the power constraints and provides a survey of efficient embedded cores that can be used in edge computing. The main idea behind this chapter is to present an overview of state-of-the-art techniques and approaches involved in designing low power microcontroller units for edge computing devices which can be used in smart healthcare applications. To further build on the idea and explore its applications, examples of edge computing devices are also presented. In addition, applications of edge computing for smart healthcare systems, challenges in implementation, the impact of 5G, IoT, and advantages of edge computing in smart healthcare applications are also discussed.