A Study of the Design Architectures of Configurable Processors for the Internet of Things

Abstract

The Internet of Things involves ubiquitous diffusion of processors consisting of small Integrated Circuits having sensing and actuation capabilities. These processors combined with embedded microprocessor core(s) are used for control, communication, and information processing. An Internet of Things processor has to meet various challenges such as low power consumption, limited battery power, limited memory space, high performance, low cost, and desired security. Besides computation and communication needs, Internet of Things applications considerably differs in the scale of interaction with the environment. Specific applications enforce additional constraints for the Internet of Things processors, such as environment invisibility, ruggedness and timing properties imposed by real-time events in the environment they are embedded. The new class of processors needs a step function enhancement in power and performance efficiency composed of strong reliability and security requirements. The processors running the Internet of Things applications must evolve to sustain the varying requirements emerging out of the new use cases in the electronics market. The configurable and extensible processors are the next generation in the processor family tree. This paper discusses the processor characteristics required for the Internet of Things applications and the possible configurations to meet the growing demands of the processor market. Some possible processor configurations required for the Internet of Things applications are discussed, and architecture of two configurable processors is also explained in detail.