Abstract: In the last years, advancements in data-processing power and availability of large amount of data have led rise of machine learning applications in different areas such as computer vision, natural language processing to name a few. We are now several years into explosion of machine learning (ML) in wireless networks, used to enrich decision-making by finding structures in data – knowledge discovery – as means to extract models that describe the user behavior and/or wireless network performance. The fifth (5G) generation of wireless systems offers new modes of cell-based broadband connectivity to diverse devices types with three major aspects: 1) massive infrastructure densification; 2) universal resource management and 3) network function virtualization. Moreover, the communication engineering transforms from a traditional model-based toward data-driven discipline with machine learning. While the trend of densification continues beyond 5G, with a consequence of diminishing cell sizes, there is a common understanding that future wireless will require more autonomous operations. Today, wireless operations rely on optimization frameworks with independent heuristics per highly dependent problems such as coverage optimization, load balancing, interference management, to name just a few. To partially address these issues, a joint optimization (e.g. maximum likelihood) with multiple objectives was considered. However, because of inherent complexity of algorithms and wireless system dynamics, an optimal solution even for simple multi-objective functions, is analytically and practically not tractable. We need to revisit today’s approaches and discuss novel directions to support distributed and autonomous wireless.
This talk explores challenges and future research directions related to future wireless with machine intelligence by introducing wireless intelligent agents. We start with discussion on ongoing transformation toward data-driven wireless with machine learning, its benefits and shortcomings. A paradigm shift from contemporary offline and data-driven wireless with ML toward online and autonomous wireless with AI is discussed. We highlight challenges and requirements toward network self-organization of complex system-of-systems. We briefly discuss principles of intelligent agent design with knowledge management (by AI disciplines such as perception, reasoning, searching, decision-making, learning). Finally, we end with an example of case study with wireless AI prototype using commodity Wi-Fi access points. The talk provokes new coming challenges and unveil interesting future directions across multi-disciplinary research areas.
Biography: Haris Gačanin received his Dipl.-Ing. degree in Electrical engineering from the University of Sarajevo in 2000. In 2005 and 2008, respectively, he received MSc and PhD from Tohoku University in Japan. He was with Tohoku University from 2008 until 2010 first as Japan Society for Promotion of Science postdoctoral fellow and later, as Assistant Professor. In 2010, he joined Alcatel-Lucent (now Nokia), where he is currently Department Head at Nokia Bell Labs. He is adjunct professor at University of Leuven (KU Leuven). His professional interests are related to applications of artificial intelligence with machine learning in autonomous wireless networks. He has 200+ scientific publications (journals, conferences and patent applications) and invited/tutorial talks. He is senior member of the Institute of Electrical and Electronics Engineers (IEEE) and the Institute of Electronics, Information and Communication Engineering (IEICE). He is a recipient of IEICE Communication System Study Group Best Paper Award (joint 2014, 2015, 2017), The 2013 Alcatel-Lucent Award of Excellence, the 2012 KDDI Foundation Research Award, the 2009 KDDI Foundation Research Grant Award, the 2008 Japan Society for Promotion of Science (JSPS) Postdoctoral Fellowships for Foreign Researchers, the 2005 Active Research Award in Radio Communications, 2005 Vehicular Technology Conference (VTC 2005-Fall) Student Paper Award from IEEE VTS Japan Chapter and the 2004 Institute of IEICE Society Young Researcher Award. He was awarded by Japanese Government (MEXT) Research Scholarship in 2002.