Mobihoc 2020

ACM MobiHoc Workshop on “Cooperative data dissemination in future vehicular networks” (D2VNet)

Held online on October 11, 2020

Sunday, October 11, 2020

09:30-11:30: Technical Session 1

How to Deal with Data Hungry V2X Applications?

Alessandro Bazzi (University of Bologna, Italy); Claudia Campolo (University Mediterranea of Reggio Calabria, Italy); Barbara Masini (CNR - IEIIT, Bologna, Italy); Antonella Molinaro (University Mediterranea of Reggio Calabria, Italy)

The Role of Machine Learning for Trajectory Prediction in Cooperative Driving

Luis Sequeira (King's College London); Toktam Mahmoodi (King's College London)

Distributed Resource Allocation with Multi-Agent Deep Reinforcement Learning for 5G-V2V Communication

Alperen Gündogan (Nomor Research and Technical University of Munich, Chair of Communication Networks); H. Murat Gürsu (Technical University of Munich, Chair of Communication Networks); Volker Pauli (Nomor Research, Germany); Wolfgang Kellerer (Technical University of Munich, Chair of Communication Networks)

Graph-based Model for Beam Management in Mmwave Vehicular Networks

Zana Limani Fazliu (Faculty of Electrical and Computer Engineering, University of Prishtina); Carla Fabiana Chiasserini (Politecnico di Torino); Francesco Malandrino (CNR-IEIIT); Alessandro Nordio (CNR-IEIIT)

11:30-11:45: Break

11:45-12:45: Keynote

Title : Millimeter Wave Vehicular Link Configuration Using Machine Learning
Keynote Speaker : Prof. Robert W. Heath Jr.

Bio: Robert W. Heath Jr. is a Distinguished Professor in the Department of ECE at North Carolina State University . He has received several awards including the 2019 IEEE Communications Society Stephen O. Rice Prize and then 2019 IEEE Kiyo Tomiyasu Award. He authored "Introduction to Wireless Digital Communication” (Prentice Hall in 2017) and "Digital Wireless Communication: Physical Layer Exploration Lab Using the NI USRP” (National Technology and Science Press in 2012). He co-authored “Millimeter Wave Wireless Communications” (Prentice Hall in 2014) and "Foundations of MIMO Communications" (Cambridge 2019). He is a licensed Amateur Radio Operator, a registered Professional Engineer in Texas, a Private Pilot, a Fellow of the National Academy of Inventors, and a Fellow of the IEEE.

Abstract: Millimeter-wave (MmWave) vehicular communication enables massive sensor data sharing in vehicular systems, leading to enhances in automation, safety, transportation efficiency and infotainment. Estimating and tracking beams in mmWave vehicular communication, however, is challenging due to the use of large antenna arrays and high mobility in the vehicular context. Fortunately, wireless cellular communication systems have access to many kinds of data, which can make beam training more efficient. In this talk, I introduce beam alignment solutions that work with different types of information related to link performance. Data-driven approaches are able to leverage side information and underlying channel statistics to optimize link configuration in mmWave vehicular communication with negligible overhead.

12:45-13:45: Lunch Break

13:45-15:15: Technical Session 2

Tensor Completion-Based 5G Positioning with Partial Channel Measurements

Fuxi Wen (Chalmers University of Technology, Sweden); Tommy Svensson (Chalmers University of Technology, Sweden)

Acting Selfish for the Good of All: Contextual Bandits for Resource-Efficient Transmission of Vehicular Sensor Data

Benjamin Sliwa (TU Dortmund University, Germany); Rick Adam (TU Dortmund University, Germany); Christian Wietfeld (TU Dortmund University, Germany)

Vehicular Knowledge Networking and Application to Risk Reasoning

Seyhan Ucar (InfoTech Labs, Toyota Motor North America R&D); Takamasa Higuchi (InfoTech Labs, Toyota Motor North America R&D); Chang-Heng Wang (InfoTech Labs, Toyota Motor North America R&D); Duncan Deveaux (EURECOM - Communication Systems Department, Sophia-Antipolis, France); Jérôme Härri (EURECOM - Communication Systems Department, Sophia-Antipolis, France); Onur Altintas (InfoTech Labs, Toyota Motor North America R&D)

15:15-16:45: Industry Panel

Topic : The role of machine learning for autonomous driving

Panel Moderator : Dr. Mate Boban

Mate Boban received the Diploma degree in informatics from the University of Zagreb, Croatia, and the Ph.D. degree in electrical and computer engineering from Carnegie Mellon University, Pittsburgh, PA, USA, in 2004 and 2012, respectively. He is a Principal Research Engineer with Huawei Munich Research Center, Germany. Before Huawei, he was with NEC Laboratories Europe, Carnegie Mellon University, and Apple. He is an alumni of the Fulbright Scholar program. He has co-chaired several IEEE workshops and conferences and has been involved in European Union funded projects (5G-CAR, DRIVE-C2X, and TEAM) as a Work Package leader and editor of deliverables. He is actively involved in key industry and standardization bodies dealing with V2X: 3GPP, 5GAA, and ETSI. His current research interests include resource allocation, machine learning applied to wireless communication systems in vertical industries (V2X, IIoT), and channel modeling. He coauthored three papers that received the Best Paper Award, at IEEE VTC Spring 2014, IEEE VNC 2014, and EuCAP 2019.

Panelists :
Tim Leinmüller , DENSO

Tim Leinmüller is currently heading DENSO’s European fundamental technology R&D department. His group is responsible for R&D in the domains of cybersecurity, microcontroller, in-vehicle networks, wireless communication, and cooperative, connected and automated mobility (CCAM). He is responsible for DENSO’s involvement in the 5G Automotive Association (5GAA), where he is an elected member of the board since 2018, and chairmen of WG1 since 2019. Furthermore, he is representing DENSO in CCAM related activities in ETSI, CLEPA, ERTRAC, EATA, VDA, and the EU CCAM Single Platform. In total, Tim is working in the connected vehicles domain for more than 10 years. He is/has been serving in related organizations in multiple positions, amongst others as member of the technical committee and as chair of the architecture working group in the CAR 2 CAR Communication Consortium (C2C-CC).

Tahir Sari , Ford

Tahir Sari received his Electrical and Electronics Engineering degree from the Ege University, Izmir, Turkey in 2014. He joined Ford Otosan Electrical and Electronics Engineering Team as an Active Safety and Driver Information Engineer in 2016. He worked on Ford Transit, Courier and Cargo Trucks commercial vehicle projects to develop ADAS and Instrument Cluster systems. After that, he joined to Autonomous Engineering Department as a Connected and Autonomous Vehicle Engineer in 2018. Currently, he leads vehicle connectivity related H2020 European Projects, 5GMOBIX and BEYOND5.

Andreas Festag

Andreas Festag is a professor at the Technische Hochschule Ingolstadt and with the research and test center for vehicle safety CARISSMA. Andreas Festag received a diploma degree (1996) and Ph.D. (2003) in Electrical Engineering from the Technische Universität Berlin. As researcher, he worked with the Telecommunication Networks Group (TKN) at Technische Universität Berlin, Heinrich-Hertz-Institute (HHI) in Berlin, NEC Laboratories in Heidelberg, Vodafone chair Mobile Communication Systems at Technische Universität Dresden and Fraunhofer Institute for Transportation and Infrastructure Systems (IVI). Andreas has worked on various research projects for wireless and mobile communication networks and published more than 100 papers in journals, conference proceedings and workshops. His research is concerned with architecture, design and performance evaluation of wireless and mobile communication systems and protocols, with a focus on vehicular communication and Intelligent Transportation Systems (ITS). He actively contributes to the ETSI Technical Committee ITS. He is Senior member of IEEE.

Call For Papers

Recent developments in software, hardware and communication technologies have opened the way towards Cooperative and Intelligent Transportation Systems (C-ITSs) as a means to deliver improved road safety, traffic efficiency and infotainment services. This potential will be unleashed through cooperative dissemination of context information, including observations that intelligent vehicles acquire through sophisticated sensors (e.g., radars, cameras and LiDARs).

Sending information about all of on-board sensor observations may saturate the capacity of traditional technologies for vehicular communications, especially in a dense urban environment with a large number of vehicles, thus motivating efforts towards the design of techniques that set a bound on the amount of information that is distributed over bandwidth-constrained communication channels.

A traditional approach is to monitor the age of information (AoI), so that vehicles broadcast sensory messages that are not too old. Another approach is to rank the utility of sensor data by relying on feedback messages which describe how helpful the received information was in relation with the requirements of target applications. Machine learning has also emerged as a promising option to measure the mutual information of different combinations of the sensory readings and dynamically assign them value scores which depend on the degree of correlation. Autoencoders can also be trained in an unsupervised manner to extract features from input vectors and prioritize the transmission that have greatest value for target applications.

This workshop aims at collecting original contributions that can advance the state of the art on the design challenges for data dissemination in vehicular networks, with a focus on three main areas of interest. First, it is of particular interest the analysis and design of compression techniques for sensory acquisitions, in order to facilitate storage and dissemination in vehicular applications. Second, we look forward to submissions that study how to optimize the concept of data dissemination while avoiding network congestion, enhancing average network throughput and prioritizing specific vehicles messages. Third, we encourage submissions focusing on the trade-off between on-board and edge-assisted approaches for the processing of sensors’ observations, where signal latency, power consumption, and system overhead performance is compared. We particularly welcome papers on the application of machine learning techniques in vehicular scenarios, with a focus on multi-frame perception, prediction, feature extraction, and evaluation of data correlation for autonomous driving.

Topics of interest for the workshop include, but are not limited to:

Cooperative services in vehicular networks.
V2X technologies for data dissemination.
Value of information in vehicular networks.
Data-driven techniques in anticipatory vehicular communication systems.
Machine learning for sensor-based object detection and/or tracking.
Machine learning for resource-efficient data dissemination.
Machine learning for cross-modal feature extraction in automotive sensors.
Machine learning for compression of automotive sensors’ observations.
Machine learning for mutual information maximization among automotive sensors.
Challenges of sensors’ data dissemination in a commercial vehicular system.
On-board vs. edge-assisted sensor processing in vehicular networks.
New automotive sensor technologies.

Submission Instructions

Prospective authors should submit a paper with a maximum length of six (6) pages, including all figures and references. Papers must be formatted with the ACM style sheet. If you are using LaTeX, you can make use of a simplified ACM conference template. The papers of this workshop will go through the same single-blind peer-review process of the main conference. Submitted manuscripts must be original and not be published or under review elsewhere. Papers must not infringe any copyright or third party right. The submission website is https://d2vnet20.hotcrp.com.

Important Dates

June 14, 2020 [Extended]	Workshop paper submission
June 28, 2020	Notification of papers acceptance
July 28, 2020 [Extended]	Camera-ready submission

Contact Workshop Chairs

Committees

Workshop Organizers

Marco Giordani, University of Padova
Michele Zorzi, University of Padova

TPC Members

Onur Altintas, TOYOTA North America, Infotech Labs
Takamasa Higuchi, TOYOTA North America, Infotech Labs
Antonella Molinaro, Università Mediterranea di Reggio Calabria
Toktam Mahmoodi, King's College London
Robert Piechocki, University of Bristol
Marco Centenaro, Fondazione Bruno Kessler (FBK)
Falko Dressler, Paderborn University
Christian Wietfeld, TU Dortmund University
Robert W. Heath Jr., University of Texas at Austin
Carla Fabiani Chiasserini, Politecnico di Torino
Tommy Svensson, Chalmers University of Technology