Edge Intelligence in the Cognitive Internet of Things: Improving Sensitivity and Interactivity

A new network paradigm, CIoT, has been proposed by applying cognitive computing technologies, which is derived from cognitive science and artificial intelligence in combination with the data generated by connected IoT devices and the actions that these devices perform. The development of cognitive computing is very important in the above process to meet key technical challenges, such as generation of big sensory data, efficient computing/storage at the CIoT edge, and integration of multiple data sources and types. On the other hand, to evolve with the new computing and communication paradigms, the CIoT ecosystem has to update by absorbing new capabilities such as deep learning, the CIoT sensing system, data analytics, and cognitiion in providing human-like intelligence.

Introduction

In the last decade, with the rapid expansion of sensors, wireless communications, mobile networks, cloud computing, edge computing, and so on, the Internet of Things (IoT) is expected to utterly change the patterns of interaction between humans and their pervasive digital environment by supporting various advanced trends, such as comprehensive interconnection, human-like intelligence, shiftable information collection, and cloud-based applications [1]. The great potential of IoT in terms of both economic and social values has attracted significant attention worldwide and constitutes one of the national development strategies of many countries.