Edge Intelligence in Wireless Networks

Overview: The development of 5G communication and internet of things (IoT) technology, more and more data are generated by a large number of IoT devices at edge network. If these massive amounts of data are transmitted to remote cloud data center, it will not only bring huge transmission pressure to the core network, but also incur long transmission delays. Edge intelligence has attracted great attention from the academic and industrial communities all over the world. It enables to run deep model locally on an IoT device to process data generated by the IoT device itself. It requires high computation capacity and power to train these data and infer model using artificial intelligence algorithms. Therefore, it needs to take advantage of a wide range of available resources across the hierarchy of end devices, edge nodes, and cloud data centers to optimize performance of training and inferencing a deep neural network model. However, it is currently in its early stage for research on edge AI and needs a special communication for the recent advances of edge AI. Therefore, it is lack of summary and discussion for the technology advancement of edge intelligence. In this special issue, we will call for the related researches about computing architectures and enabling technologies.