Scope and Topics:
With the explosive growth of smart devices and new applications, mobile traffic volume has been growing exponentially. The myriad technological advances proposed for the future 5G networks still mostly focus on capacity increase, which is fundamentally constrained by the limited radio spectrum resources and the diminishing investment efficiency for operators and, therefore, will always lag behind the growth rate of mobile traffic. Therefore, novel distributed architectures, which bring network functions (such as computing, caching and sharing) and AI capabilities to the network edge, emerge to confront the aforementioned challenges in the networks development and many emerging applications, such as AR/VR, IoT, eHealth, autonomous driving, gaming etc. Edge intelligence and distributed AI inevitably consume significant energy from wireless communications for frequent information exchanges among participating nodes, as well as computation energy from running machine learning algorithms. The energy consumption also has to be shared with non-AI tasks, demanding a holistic consideration of energy efficiency. However, wireless devices are often energy constrained, powered by battery or renewable energy. This conflict has grown into one of the most important issues to realize sustainable edge intelligence and distributed AI. In fact, even for devices as powerful as electric vehicles, in which communication and computing power may seem to be negligible as compared to mechanicals for driving, research has surprisingly shown that running AI algorithms for autonomous driving can in fact reduce the range by 10%-20%. In short, enhancing the energy efficiency and sustainability of edge intelligence and distributed AI is crucial for their practical implementation; and thus, strongly motivates inter-disciplinary research across wireless communications, networking, computing and machine learning, circuits, and hardware design. The ultimate goal of this workshop is to provide a forum where researchers from these topics can meet and find synergy, leading to discovering best practices for a future intelligent wireless network. The topics of interests related to green edge intelligence and distributed AI include (but are not limited to): Green wireless networks: green network architecture design, energy consumption modelling and analysis, etc; Network optimization: including energy harvesting, energy storage, energy transfer, etc; Green AI: light-weighted AI models for intelligent edge nodes, novel AI methods for green edge node , etc; Cloud/fog computing: cloud-edge assisted computing for sustainable devices and networks, etc; Prototyping and application areas: test-beds and field trials, vehicular networking, IoT, smart grid etc.
Submission Guidelines:
Authors are invited to submit original papers of no more than 6 pages (standard IEEE proceedings, two-column, 10 pt font, etc.), including figures, tables, and references, in PDF format. Papers should be submitted through EDAS: https://edas.info/newPaper.php?c=29710&track=112449
Workshop Organizers:
Sheng Zhou, Tsinghua University, China, sheng.zhou@tsinghua.edu.cn
Zhiyuan Jiang, Shanghai University, China, jiangzhiyuan@shu.edu.cn
Shan Zhang, Beihang University, China, zhangshan18@buaa.edu.cn
Jie Xu, University of Miami, U.S., jiexu@miami.edu
Zhisheng Niu, Tsinghua University, China, niuzhs@tsinghua.edu.cn
Important Dates:
Workshop Paper Submission: Jun. 1st, 2022 Jun. 16th, 2022
Workshop Paper Acceptance: Jul. 10th, 2022
Workshop Paper Camera-ready: Jul. 24th, 2022