Keynote Speakers
Keynote I
Prof. Kazuya Tsukamoto, Kyushu Institute of Technology, Japan
Title: Flexible Cyber-Physical Systems for Geolocation-centric Services
In recent years, realizing Cyber-Physical Systems (CPS) has become necessary to solve social problems using IoT and Beyond 5G/6G technologies. In IoT and B5G/6G era, the growth of data variety is driven by cross-domain data fusion. Therefore, collecting and processing data generated from the various IoT devices become crucial for developing service and control policies. In our study, we advocate that the ''local production for local consumption (LPLC) paradigm'' can be an innovative approach to cross-domain data fusion. Then, we propose a new network infrastructure that fully utilizes Mobile Edge Computing architecture and Artificial Intelligence (AI) technologies by Beyond 5G and 6G, thereby realizing flexible CPS. The first part of my talk will introduce the concept of a geolocation-centric information platform (GCIP) that can produce and deliver diverse Geolocation-centric Services. In the GCIP, (1) an infrastructure-based geographic hierarchy edge network and (2) an adhoc-based service retention system are interplayed to provide geolocation awareness and resiliency. However, since both the computational and communication, including wireless radio resources, in the B5G/6G edge network are relatively limited, we propose a new concept, Floating CPS (F-CPS), as the extended version of GCIP. The second part of my talk will introduce the details of F-CPS that can utilize local computing and communication resources in a distributed and autonomous manner. Finally, I will share some new research directions of the F-CPS, simulation results, and results of the preliminary experiments on the NICT Beyond 5G testbed deployed at our campus.
Biography
Prof. Kazuya Tsukamoto received a DE degree in computer science from the Kyushu Institute of Technology (Kyutech), Japan, in 2006. From April 2006 to March 2007, he was a Japan Society for the Promotion of Science (JSPS) research fellow at Kyutech. In 2007, he was an assistant professor in the Department of Computer Science and Electronics, Kyutech, and then was an associate professor in the same department in 2013. He has been a professor in the same department since January 2022. His research interests include performance evaluation of computer networks and wireless networks. He is a member of the ACM, IPSJ, IEICE, and IEEE.
Keynote II
Prof. Pruet Boonma, Chiang Mai University, Thailand
Title: Improving Distributed PageRank with Balancing Partitioning
PageRank is a well-known web-page ranking algorithm. It performs iterative calculations of PageRank's value of each web page based on the summation of PageRank's values of all the web pages with an incoming link to the page. The applications of PageRank are not limited to web page ranking but are also in biology, chemistry, ecology, and physics. However, to make the algorithm scalable for real-life applications, the algorithm is often performed distributedly. In particular, the global web graph is partitioned, and a sub-graph is assigned to each computation node. However, because each node maintains only a partial graph, they must frequently exchange data for the PageRank calculation. This talk discusses an observation on the graph properties that can be used to improve distributed PageRank. In particular, a new mathematical model based on minimal cut and density balance partitioning is discussed, and efficient algorithms are considered and evaluated.
Biography
Pruet Boonma received this B.Eng in Computer Engineering from Chiang Mai University, M.IT in Information Systems from Monash University, and PhD. in Computer Science from the University of Massachusetts, Boston. He has worked with the Faculty of Engineering at Chiang Mai University as a faculty since 1997. Since 2015, he has been a science advisor for the PODD project, a participatory digital surveillance platform to prevent zoonotic spillovers. His publication is on graph algorithm, bio-inspired algorithm, wireless sensor network, and data analytics.