Interview with George C. Alexandropoulos
At GLOBECOM 2018 in Abu Dhabi, we have given the IEEE ComSoc YP Industry Award to George C. Alexandropoulos, who is currently an Assistant Professor for Communication Systems and Signal Processing at the National and Kapodistrian University of Athens. We were inspired by his work, so we asked him a few questions about his research and also asked for some directions for young researchers. We hope you will enjoy reading his answers.
ComSoc YP: Hi George, could you please tell us about your education background and professional working experience?
Answer: My educational background includes the Engineering Diploma degree in computer engineering and informatics, the M.A.Sc. degree in signal processing and communication systems, and the Ph.D. degree in wireless communications from the Computer Engineering and Informatics Department (CEID), School of Engineering (SE), University of Patras (UoP), Rio–Patras, Greece received in 2003, 2005, and 2010, respectively. My professional career started in 2001 as Research Assistant at CEID’s Signal Processing and Communications Laboratory, where I was affiliated till 2005. During my CEID’s postgraduate studies, I served as Teaching Assistant for various laboratory courses on signal processing systems. After my one year mandatory military service, I joined the National Center for Scientific Research–”Demokritos,” Athens, Greece, as Research Fellow and Ph.D. scholar of the Institute of Informatics and Telecommunications for the period 2006–2010. From 2007–2011, I also collaborated with the National Observatory of Athens, Institute for Astronomy, Astrophysics, Space Applications and Remote Sensing, Athens, Greece in the framework of several national and European research and development projects. During the academic summer semester of 2011, I was Adjunct Lecturer at the Department of Informatics and Telecommunications, School of Economy, Management, and Informatics, University of the Peloponnese, Tripolis, Greece. For the period 2011–2014, I was Senior Researcher at the Athens Information Technology (AIT) Center for Research and Education, Athens, where I was involved with the technical management of several European research and innovation projects, and lectured several mathematics courses for AIT’s joint B.Sc. program in Computer Engineering with Aalborg University in Denmark. Within the year 2012, I also collaborated with the Telecommunication Systems Research Institute, Technical University of Crete, Chania, Greece in the framework of an ARTEMIS joint undertaking project on underwater cooperative robotic systems. From November 2014 till the beginning of January 2019, I was Senior Research Engineer at the Mathematical and Algorithmic Sciences Lab, Paris Research Center, Huawei Technologies France SASU. From January 2019, I am Assistant Professor for Communication Systems and Signal Processing at the Department of Informatics and Telecommunications, School of Sciences, National and Kapodistrian University of Athens, Athens, Greece, as well as associated faculty with ATHENA Research and Innovation Center. Currently, I am serving as Editor for IEEE Transactions on Wireless Communications, IEEE Communications Letters, and Elsevier Computer Networks, as well as acting as Reviewer, Member of the Technical Program Committee, Symposia Chair, and Special Sessions Organizer for many top-level international conferences and journals on signal processing and wireless communications. I am Senior Member of IEEE from 2015, as well as of IEEE Communications and Signal Processing Societies, and registered Professional Engineer at the Technical Chamber of Greece.
ComSoc YP: Could you describe what you are working on?
Answer: I am currently working on hardware and algorithmic designs as well as various performance analyses for transceiver architectures exhibiting lower cost and power consumption footprints compared to conventional fully digital multi-antenna transceiver architectures. In the latter conventional architectures, each antenna element is accompanied with a dedicated radio frequency chain enabling baseband processing. Among the architectures that are lately considered as candidate practical enablers for transceivers with extreme numbers of antenna elements belong the different versions of hybrid analog and digital beamforming, electronically steerable parasitic antenna radiators, load modulated arrays, electromagnetic lens antennas, and the very recent concept of reconfigurable metasurfaces. The metasurfaces are cost-efficient planar structures of very few centimeters thickness, whose hardware design may differ with the operating frequency band, and are comprised of nearly passive unit elements that can be tuned online in order to influence the impinging electromagnetic field (e.g., reflection, absorption, and scattering). Hence, their potential when coated on large objects of the wireless propagation environment, thus transforming them to intelligent reconfigurable beamformers from otherwise passive reflectors, is significant for various communication objectives (e.g., throughput, coverage, positioning, and security). Despite the cost and energy efficiency aspects of all aforementioned transceiver architectures, all of them impose certain limitations when it comes to algorithmic design, which become more severe when the transceivers are considered for mobile and vehicular scenarios, and millimeter wave frequencies. I am lately working on signal processing algorithms (including machine learning and artificial intelligence) for channel parameters’ estimation and online tuning with hybrid analog and digital beamformers and metasurfaces, as well as channel modeling of metasurfaces and the analysis of their potential when deployed in large quantities in dense heterogeneous networks.
ComSoc YP: How did your work evolve over the last couple of years reaching to your current research?
Answer: During the third year of my undergraduate studies, I started working on various speech signal processing algorithms and the implementation and demonstration of speech coding techniques using real digital signal processors. Afterwards, I got attracted by wireless communications taking various courses on the topic together with image processing ones, and it was during my postgraduate thesis when I decided to focus on wireless sensor networks, and specifically in the design and implementation of fast and robust source tracking algorithms considering arbitrarily distributed sensors. Having got intrigued by array signal processing techniques and statistical signal processing algorithms, during my Ph.D. years, I focused on various performance analysis methodologies and algorithmic designs for multi-antenna wireless communication systems. Considering generalized fading channel models, models for transceiver impairments and channel estimation imperfections, and arbitrarily correlated fading, I worked on analyzing the performance of numerous antenna diversity schemes, Multiple Input Multiple Output (MIMO) systems, and relaying protocols. During my first postdoc years, I conducted research on multi-user MIMO systems and specifically on designing distributed transceiver algorithms for the K-user MIMO interference channel, as well as evaluating their performance both via link-level simulations and proof of concepts. I also worked on the interference alignment technique with electronically steerable parasitic antenna radiators and load modulated arrays. Then, I started investigating the performance of multi-user MIMO techniques over large dense networks working on models for uncoordinated interference and applying stochastic geometry tools. In the meanwhile, the applications of MIMO and beamforming techniques had attracted me and, in parallel, I worked on the performance analysis of cognitive radios and vehicular systems. In the framework of a development project, I also worked on simulating underwater communication channel models and designing underwater cooperative robotic systems. During my industrial years, I worked on low complexity transceiver architectures and relevant algorithms for full duplex MIMO systems, energy harvesting MIMO, and hybrid analog and digital beamforming. For the latter topic in the case of millimeter wave communications, my team’s beam tracking algorithms followed the 5G NR specifications and that of the project IEEE 802.11ay. I also worked on fast beam alignment techniques for fixed microwave systems. The last one year, I am also working on signal processing algorithms with reconfigurable metasurfaces, metasurfaces’ modeling, as well as the analysis of large heterogeneous networks incorporating them. In the framework of this research, machine learning and artificial intelligence techniques have kicked out. The applications of machine learning and artificial intelligent in effectively tuning the physical and medium access layers at large, and specifically the design of distributed algorithms, are also topics that I am recently involved with.
ComSoc YP: What would you suggest to young researchers to look at in the wireless communications area?
Answer: Wireless communications have transformed our lives and I strongly believe that they will continue doing this for the years to come. Their undoubtable and continuous accomplishments over the last 20 years have spurred numerous wireless communication system applications ranging from voice and video communication to telemetry and industrial automation, as well as some sort of autonomous driving. New hardware structures enable embedding smart functionalities to the wireless environment (to its 3D objects, to humans via wearable transceivers, or even implanted body sensors), and machine learning and artificial intelligence promise intelligent softwarization of the wireless infrastructure for diverse applications. Interestingly, many industrial verticals (e.g., automotive and manufacturing) have focused on 5G infrastructure, and already challenging requirements are being sketched for 6G smart connectivity enabling futuristic applications like remote surgery and holographic-type communication. New hardware transceiver architectures and materials enabling wireless communication at any entity in the 3D space will be a hot topic for wireless communication together with architectures for millimeter and THz communication. New mathematical tools to model their behavior and their performance in 3D are also needed to provision the performance of future wireless networks. Last but not least, network analytics tools and machine learning and artificial intelligence algorithms for wireless connectivity will continue being a hot research and development direction.
ComSoc YP: Can you share any relevant books, journals or videos that have caught your interest in recent times and can be helpful for young professionals to develop their understanding about changing technologies and challenges?
Answer: It’s difficult to create a short list of impactful material for the readers of this discussion. Indeed, there exist various books, journals, and videos with significant technical and visionary contribution for researchers working on wireless communications. There is also enough material with questionable content that has not followed any rigorous reviewing process, and can possibly mislead its readers. My suggestion to young professionals is to read certified material ideally with the coordination of their supervisors, managers, mentors, or colleagues they trust. In this type of material belong IEEE ComSoc journals and magazines, companies’ white papers, and documents from wireless standards bodies and international wireless communications associations.
Interview created by Richard Cziva. Uploaded 5 November, 2019.