Agenda

People, places, and things are said to be getting “smart.”  But what does that mean?  This session will explore a range of answers to that question.

Session Chair: Joel Myers, Chair of Smart Cities for the IEEE IoT Initiative, and CEO, DOMILA LTD (Ireland)

We live a fast urbanization process where the United Nations estimates by 2050 to have 80% of the world population living in cities. In numbers, the world is moving from 35 Megacities that are cities with more than 10 million habitants to reach 100 Megacities. And even we are facing climate change with solid consequences to the environment and all living ecosystems because of fast urbanization. Hence, cities today need to commit to reducing their carbon footprint and respect their natural resources for a better quality of life. In addition, cities are moving to a digital transformation where technology enables efficiency and better decision-making based on data to manage services and infrastructures. That is the quest of the Smart Cities, where technology is only a tool that cannot succeed if there are no local innovation ecosystems enabled with continuous cooperation from government, industry, academia, and citizens. This talk focuses on lessons learned around the world and the experience in Guadalajara on its quest to become a Smart City. We introduce local innovation and city development with different technological infrastructures and applications. Finally, we in how a strategy based on incentives with a digital coin can change citizens’ behavior can help connect better the innovation ecosystem of the cities. 

Cities are one of the most significant sources of greenhouse gas emissions and experience the harshest consequences of grid catastrophic failures due to extreme weather events. Consequently, utilities and municipalities must respond fast to restore the grid to a minimum operational state. To better identify the conditions and the nature of the damage and, therefore, plan the infrastructure’s remediation at multiple levels, it is essential to work with reliable information. Street Lighting is part of the electric grid and is considered to be a last-mile infrastructure because it is spread throughout the city, and it is also vulnerable to the damages caused by extreme weather events. The technological advancements enable the streetlights to collect data from the physical environment and to be controlled and monitored remotely. It demands a connectivity infrastructure for data transfer, which can also be leveraged by the electric grid to improve resilience. This talk will describe how the Smart Street Lighting solutions can improve power grid connectivity and data-driven decision-making in the face of extreme weather events.

The speech is to present a series of electrically related actions on transforming the ports into a sustainable electric energy hub extending the concept of smart grids in view of the decarbonization on both the transport and the energy sector. Within this frame, the components of a holistic approach called the “Proteus plan” elaborated by an inter-disciplinary scientific team in Greece are presented, while the case study of the port of Igoumenitsa in Greece where the “Proteus” plan implemented via the ALFION project funded by the European Commission is showcased.

Is it ever too soon to look at the future?  We can’t resist!  What are some of the qualities that future networks might have?

Session Chair: Hung-Yu Wei, Professor, Department of Electrical Engineering, National Taiwan University

This talk will briefly outline the expected 5G path in the next several years and the journey towards beyond 5G. It will discuss the key attributes of the global initiatives for the systems of 2030 and beyond. The vision, motivation and role, and potential use cases of 6G will be highlighted, sharing insight on the context, implications and key requirements, along with research challenges and design requirements.

With 5G availability fast expanding worldwide and a “mid-generation” evolution cycle starting in 3GPP (Rel-18), now is the right time to lay down the foundations of the next generation towards one global 6G standard. MediaTek has played a leading role in the design & standardization of 5G and its evolution, and has led the way bringing to the market mature 5G devices that can operate in new groundbreaking 5G Systems. While current industry continues 5G technology evolution to address the aforementioned challenges, 6G technology is on the horizon to not only address these issues but also to bring fundamental transformation to mobile networks. This short presentation will provide an brief overview on MediaTek’s vision on the 6G design challenges and potential trends.

As wireless networks are evolving to 6G, communications need to be considered together with sensing. Moreover, both communications and sensing can be highly enhanced via computing and AI. Thus, joint communication, sensing, and computing has emerged as a key building block of 6G networks. This talk is focused on key research problems of joint communications, sensing, and computing for 6G networks.

In 6G era, communications and computing will be closely integrated. To provide low latency services, edge/fog computing and networking technologies are needed. Management and automation for intelligent edge/fog for Next G system will be discussed.

We have very little choice as to whether we preserve earth as a quality environment in which humans can live.  Information and Communications technology can both be an energy hog and be misused for information distribution.  This session looks at how to mitigate this phenomena.

Session Chair: Charles Despins, Professor, École de Technologie Supérieure

Traffic continues to grow in communication networks at 30% to 40% per year thus doubling every two years and increasing by factors of 30x and 1000x in 10 and 20 years respectively. Power consumption in equipment tends to scale with data rate and thus this increased traffic can lead to power consumption levels that are not sustainable if there is no intervention. Power is consumed in communication networks and in data centres (clouds) that support the digital transformation. Edge processing (fog) can help reduce latency and reduce power consumption by bringing processing closer to end users. In this talk we will introduce a number of measures that can be taken to improve the energy efficiency of the communication networks and edge processing and outline how ICT can be used to improve sustainability in several sectors such as manufacturing and transport. 

Today’s energy critical infrastructure is becoming highly complex; the distributed nature of this increasingly interconnected network allows for a constantly expanding cyber-attack surface.  The objective of such malicious attempts on the critical infrastructure is usually aimed in the collection of information, taking control, shutting down, or even completely destroying the infrastructure at hand.  This talk will focus on a multi-stage data-driven ICT framework to defend against cyber-attacks and improve the power/energy network’s security and resiliency.

This talk will provide an overview on the evolution of cryptocurrency, blockchain and Distributed Ledger Technology (DLT). In 2009, Bitcoin was launched as a peer-to-peer electronic cash. In 2015, programmable blockchain/DLT projects such as Ethereum, Hyperledger Fabric, and R3 Corda were launched. Over the last decade, cryptocurrency/blockchain/DLT have evolved to encompass a collection of distributed computer network architectures implementing various decentralized consensus protocols and data structures. In this rapidly emerging field, some remarkable innovations, experimentation, and developments have been carried out by startups, enterprises, and governments. This talk will highlight various current and future applications of this technology. High (electrical) energy consumption by certain blockchains (e.g., Bitcoin) has raised serious environmental concerns. This talk will highlight several recent efforts to develop energy efficient blockchain architectures and solutions.

Is cloud computing obscuring our view of the network?  Other architectures may lead to a clearer distribution of responsibility with other parts of the network.  This session shares some ideas along these lines.

Session Chair: Rajesh Hegde, Professor and Head of the Department of Electrical Engineering, Indian Institute of Technology, Kanpur

Fog networking aims to orchestrate computing tasks across the continuum of network elements from edge/IoT devices to cloud datacenters. The growth in data-intensive, latency-sensitive applications for user and network intelligence calls for careful consideration of how these tasks will be managed throughout fog networks at scale. This talk will overview the technical challenges and opportunities for fog learning, and how the topology, heterogeneity, and proximity properties of contemporary network deployments can be leveraged to elevate the fundamental tradeoff between intelligence quality and resource efficiency.

This talk will look at history, how complexity is trending with examples, how DARPA defines it and what are its research challenges.

The very notion of the “car” is being revolutionized. Connected and Automated Vehicles (CAVs) are those with Internet access and with a variety of sensors, which enable the vehicles to sense the proximal physical environment, send and receive signals, and interact with other vehicles or entities. While the sensory data available provides tremendous opportunities for CAVs, a grand challenge is that driving decisions must take place right here right now in the vehicle, in order to meet the requirements for safety, accuracy, and latency.  In this talk, I will talk about challenges and opportunities in establishing the convergence of sensing, communication, control and computing towards building the brain of CAVs; and present our studies in theory, algorithm design and testbed in this exciting area.

Everything has a cost, especially new networks.  Every network also has some benefits.  How do we calculate the economic benefits of new network technologies and architectures so that we can determine whether the benefits outweigh the costs?

Session Chair: Jianwei Huang, Presidential Chair Professor and Associate Dean, School of Science and Engineering, the Chinese University of Hong Kong, Shenzhen

Multi-arm bandit (MAB) is a classic online learning framework that studies the sequential decision-making in an uncertain environment, but often overlooks the scenario where the decision-maker cannot take actions (e.g., pulling arms) directly. In this talk, we will present an incentivized online learning (IOL) framework for this scenario. To address the challenges of tight coupling of the unknown environment learning and asymmetric information revelation, we propose a socially-optimal mechanism for the IOL framework based on a specifically designed Lagrangian function. The proposed mechanism satisfies various desirable properties such as agent fairness, incentive compatibility, and voluntary participation, and meanwhile achieves the same asymptotic performance as the state-of-art benchmark that requires extra information.

As an emerging machine learning paradigm in edge computing, federated learning has received significant attention recently due to its promising performance in mitigating privacy risks and costs. In this talk, we will present an analytical study on the optimal incentive mechanism design for federated learning, in the presence of strategic edge devices’ multi-dimensional private information. 

Many simply characterize the benefits of the edge as reduced latency enabled by endpoint proximity, but the reality is that there are numerous other benefits to edge processing, storage, and networking.  However, there are also a number of trade-offs between centralization and dispersion that can be modeled and analyzed to determine where to optimally deploy resources.