13:30 Panel with participating professors
Derya Malak, EURECOM
Ana García Armada, Madrid
Ana Perez-Neira, Barcelona
Derya Malak, EURECOM
Ana García Armada, Madrid
Ana Perez-Neira, Barcelona
Distributed Computation over Networks
Derya Malak, Assistent Professor at EURECOM, France
Large-scale distributed computing systems, such as MapReduce, Spark, or distributed deep networks, are critical for parallelizing the execution of computational tasks. Nevertheless, a struggle between computation and communication complexity lies at the heart of distributed computing.
There has been recently a substantial effort to address this problem for a class of functions, such as distributed matrix multiplication, distributed gradient coding, linearly separable functions, etc. The optimal cost has been achieved under some constraints, based mainly on ideas of linear separability of the tasks and linear space intersections.
Motivated by the same challenge, we propose a novel distributed computing framework where a master seeks to compute an arbitrary function of distributed datasets in an asymptotically lossless manner.
Our approach exploits the notion of characteristic graphs, which have been widely utilized by Shannon, Körner, and Witsenhausen to derive the rate lower bounds for computation, and later by Alon-Orlitsky, Orlitsky-Roche, Doshi-Shah-Médard, and Feizi-Médard, to resolve some well-known distributed coding and communication problems, allowing for lowered communication complexity and even for a) correlated data, b) a broad class of functions, and c) well-known topologies.
The novelty of our approach lies in accurately capturing the communication-computation cost tradeoff by melding the notions of characteristic graphs and distributed placement, to provide a natural generalization of distributed linear function computation, thus elevating distributed gradient coding and distributed linear transform to the realm of distributed computing of any function.
In toy scenarios, we demonstrate gains up to %70 over fully distributed solutions and an approximation ratio of 2 within the optimal centralized rate.
Bio: Derya Malak is an Assistant Professor (Maître de Conférence) in Communication Systems at
Eurecom, France. Previously, she was a tenure track Assistant Professor in the Department of ECSE at RPI between 2019-2021, and a Postdoctoral Associate at MIT between 2017-2019. She received her Ph.D. in ECE at the University of Texas at Austin in 2017, B.S. in Electrical and Electronics Engineering (EEE) with a minor in Physics at Middle East Technical University, in 2010, and M.S. in EEE at Koc University, in 2013. Dr. Malak has held visiting positions in INRIA and LINCS, Paris, and at Northeastern University, and summer internships at Huawei, Plano, TX, and Bell Labs, Murray Hill, NJ. Her expertise is in information theory, communication theory, and networking areas. She has developed novel distributed computation solutions, and wireless caching algorithms by capturing the confluence of storage, communication, and computation aspects. Dr. Malak was awarded the Graduate School fellowship by UT Austin between 2013-2017. She was selected to participate in the Rising Stars Workshop for women in EECS, MIT, in 2018. She received the best paper awards in WiOpt 2022 and WiOpt 2023. Her research has been funded by the Huawei Chair program on Advanced Wireless Systems (lead 2022-), NSF, the Rensselaer-IBM AI Research Collaboration, and the DARPA Dispersive Computing Programs. She is the recipient of the ERC Starting Grant 2023-2028 on computing nonlinear functions over communication networks (SENSIBILITÉ).
Multicarrier waveforms for the next G
Ana García Armada, Professor at Universidad Carlos III de Madrid
Orthogonal Frequency Division Multiplexing (OFDM) is widely used today in wireless communications standards. Its success is due to its robustness to multipath, its ability to leverage multiuser diversity and the low-complexity implementation with the Fast Fourier Transform (FFT).
However, it has several disadvantages, such as sensitivity to synchronization, phase noise or rapidly varying channels, and a very high dynamic range. Even though these disadvantages have been over-compensated by the clear advantages, the question arises whether OFDM is still a good waveform for the next generation of wireless communications.
With the growing needs of ubiquitous communications, the evolved radio access needs to work in environments that may be subject to very high speed movement (vehicles, trains, drones, LEO satellites), while the quest for spectrum availability is taking the operating carrier frequency to unprecedently high values, where amplification and radio-frequency impairments are more difficult to handle.
This talk presents some of the existing solutions to these problems and offers a perspective of recently proposed multicarrier waveforms that may be more suited to these challenges while maintaining the desirable features of OFDM.
Bio: Ana García Armada is a Professor at University Carlos III of Madrid, where she has held many management positions (Head of Signal Theory and Communications Department, Vice-dean of Electrical Engineering, Deputy Vice-Chancellor of International Relations). She is serving on the editorial boards of IEEE Transactions on Communications and IEEE Open Journal of the Communications Society. She has been a member of the organizing committee of many conferences and is the General Chair of IEEE Globecom 2021. She has been the Secretary and Chair of the IEEE ComSoc Women in Communications Engineering Standing Committee (WICE). She is a Member at Large of the Board of Governors and Director of Online Content for IEEE ComSoc. She has published around 150 papers in journals and conferences and she holds four patents. She received the outstanding service award from IEEE ComSoc Signal Processing and Communications Electronics TC and from IEEE ComSoc WICE.
CTTC and the revolution of New Space towards 6G communication networks
Ana Pérez-Neira, Barcelona, Director CTTC
We imagine the future communication networks as 3-Dimensional, fully integrating the terrestrial and the satellite infrastructure, so that communication services can be everywhere and anytime available.
We are most familiar with terrestrial radio communications, but what about satellite communications? They are currently experiencing the New Space revolution.
This talk will explain what does this revolution really mean. It’s goal is a democratization of space that suggests that access to launch, manufacturing of satellites, and/or leveraging Low-Earth Orbit (LEO) connectivity to deliver data transmission capability to remote locations is widely available because these services are commoditized and affordable, enabling them to be equitably distributed to diverse user groups, rich and poor alike.
The reality, however, is that the delivery of space-based value-added services to remote locations around the world will require groundbreaking innovations.
This talk will introduce a number of key technologies powering the New Space age and the integration with the terrestrial segment: i) advances in signal processing and computing on board satellites to enable edge in space, ii) the emergence of commoditised, high bandwidth optical Inter Satellite Links (ISL) that enable joint communication and computing, and iii) advances in antenna technology.
The talk will pay special attention to this latter detailing the innovative possibilities that mega-constellations and swarm satellite offer for distributed spatial processing.
Bio: Ana Isabel Pérez-Neira (born 1967) is a Spanish telecommunications engineer, a professor in the Signal Theory and Communication Department of the Polytechnic University of Catalonia, the former vice rector for research of the university, the director of the Centre Tecnológic de Telecomunicacions de Catalunya, and the coordinator of the European Space Agency Networks of Excellence on Satellite Communications. Her research concerns multibeam satellite communications, beamforming, and the signal processing needed in that application. Pérez-Neira was born in 1967 in Zaragoza. She studied electrical engineering, earning a bachelor’s degree in 1989, and then a master’s degree in 1991 at University Ramon Llull. She completed a Ph.D. in 1995 at the Polytechnic University of Catalonia, working there with Miguel Àngel Lagunas Hernández . She was appointed as a full professor at the Polytechnic University of Catalonia in 2006, and served as vice rector for research from 2010 to 2014. She was appointed as director of the Centre Tecnológic de Telecomunicacions de Catalunya in 2021. Pérez-Neira was named an IEEE Fellow, in the 2020 class of fellows, “for contributions to signal processing for satellite communications and systems”. She became a full member of the Royal Academy of Sciences and Arts of Barcelona in 2021. In 2022 she was named a EURASIP Fellow by the European Association for Signal Processing, “for her contributions to signal processing for satellite and wireless communication systems.