lu.se

Electrical and Information Technology

Faculty of Engineering LTH | Lund University

Event archive

5G - An Antenna and Measurements Perspective

Published: 2019-01-25

An exciting event oreganised by The Antenna Measurement Techniques Association. Spend the day with the Antenna Measurement Techniques Association listening to top experts present the most recent developments in the industry.

Technical Tour

May 6, 2019, 18:00?20:00

Arrive a day early so you can plan to join us as we take a tour of the MAX IV Laboratory. Transportation and tour are included in the price of registration.
Technical Program

Programme May 7, 08:00?19:00

DTU-ESA Spherical Near-Field Antenna Test Facility ? Past, Present, and Future Activities
by Prof. Olav Breinbjerg, Technical University of Denmark, Lyngby, Denmark

Near-Field Measurement Technique for Electromagnetic Exposure of 5G Devices
by Prof. Mats Gustafsson, Lund University, Sweden

Far-Field OTA Testing of User Equipment Using Plane Wave Generators
by Mr. Lars Foged, Microwave Vision Group (MVG), Italy

5G Over-The-Air Conformance Testing
by Dr. Jonas Fridén, Ericsson Research, Gothenburg, Sweden

5G: Challenges for Human Exposure Assessment and Virtual-Drive Over-the-Air Testing
by Dr. Christian Bornkessel, Technische Universität Ilmenau, Germany

High-Resolution Dynamic Characterization of mm-Wave Channels
by Prof. Fredrik Tufvesson, Lund University, Sweden

Organizing Committee

Christer Larsson
Donnie Gray
Manuel Sierra Castañer
Michael Havrilla
Michelle Taylor
Fredrik Tufvesson
Lars Foged
Jan Zackrisson

Registrationis now open at: www.amta.org

The full program and information about fees (PDF)

 

When: 2019-05-07 08:00 to 2019-05-07 19:00
Location: Lund University Student Union (Kårhuset) LTH John Ericssons väg 3, Lund Sweden
Contact: christer.larsson@eit.lth.se


 


AIML@LU WS: AI & ML Technologies POSTPONED to August

Published: 2019-01-16

This sixth AIML@LU fika-to-fika workshop since May 2018 focuses on the development of the technologies that form the basis of Artificial Intelligence and Machine Learning. Possible topics to discuss are the research front for different types of AI, but also to look at different techniques for machine learning.

This fika-to-fika workshop is postponed  ? preliminary to 30 August 2019 Pleas save-the-date

 

Contact

If you have any questions, suggestions or would like to contribute to the program please contact one of:

More AIML@LU events at http://aiml.lu.se/events/

 

 

 

When: 2019-04-11 09:30 to 2019-04-11 09:30
Location: E:A, E-building, Ole Römers väg 3, LTH, Lund University
Contact: Jonas.Wisbrant@cs.lth.se


 


PhD dissertation by William Tärneberg: The confluence of Cloud computing, 5G, and IoT in the Fog

Published: 2019-03-11

Author: William Tärneberg, Department of EIT

Location:  E:C, E-Building, John Ericssons väg 4, Lund University, Faculty of Engineering LTH

Faculty opponent: Professor Maarten van Steen, University of Twente, Nederländerna

Thesis for download (PDF)

Abstract: 

In the wake of the arrival of cloud computing, future applications are poised to be- come more resilient and adaptive by embracing elasticity in an osmotic manner. Although cloud computing is a strong attractor for application developers, there
are still unconquered performance frontiers. Latency-sensitive and mission-critical ap- plications make up a significant portion of all software systems, and their owners are eager to reap the benefits of cloud computing. However, they are hindered by signific- ant delay, jitter in the delay, and relatively low resilience when operating on traditional, distant, cloud data centres.

Fog computing is emerging as a remedy. Fog computing is a heterogeneous hyper- distributed cloud infrastructure paradigm, ranging from small compute nodes close to the end-users to traditional distant data centres. With greater proximity to the end- users, delay and jitter in the delay can be reduced, and intermediate network reliability improved. Additionally, with increased heterogeneity of resources, applications have a richer tapestry of resources to take advantage of for their objectives. However, man- aging and taking advantage of this heterogeneity in resources and objectives is a chal- lenge for both the infrastructure providers and application owners alike. Only where to place and scale application components and how to manage system resources to meet the objectives of both parties, is non-trivial. Application placement implies elaborate optimisation objectives, hard-to-find solutions, and operational conflicts.
The objective of this thesis is to investigate the performance-related properties of fog computing, how such an infrastructure can be managed while applications can osmotic- ally take advantage of the infrastructure, and what Fog computing?s potential practical performance gains are. These are fundamental topics that need to be answered for pro- viders and application owners alike to be able to invest in fog computing. In general terms, the work in this thesis seeks the trade-offs between infrastructure, applications, and software platform in contrast to the traditional cloud offering.

The thesis provides modelling and simulation tools for evaluating the performance and feasibility of Fog computing. Based on which, the thesis goes on to propose holistic infrastructure management algorithms. The requirements of latency-sensitive and mission-critical applications and use cases are discussed for a fog computing paradigm. These requirements are then translated to Fifth Generation Wireless Spe- cifications (5G) Massive Multiple Input Multiple Output (MIMO) specifications. An original 5G-based fog computing test-bed for time-sensitive and mission-critical ap- plications is implemented. The test-bed is used to evaluate the potential application performance gains of fog computing and to what extent the applications can practic- ally take advantage of a fog infrastructure. The thesis also investigates the architecture of the applications that are proposed to benefit from fog computing and how they per- form in traditional cloud offerings.

The included works show that fog computing indeed has a performance advantage over the traditional distant cloud, not only in latency but also in robustness. The be- nefits of 5G on a time-sensitive application deployed in a fog computing infrastructure are shown to be significant. It is also shown that a fog computing infrastructure with a high degree of heterogeneity and with multiple objectives can be successfully managed scalably. Additionally, the thesis sheds some light on the challenges of implementing latency-sensitive and mission-critical applications with traditional cloud service offerings.

When: 2019-03-29 09:15 to 2019-03-29 09:15
Location: Lecture Hall E:1406, E-Building, Ole Römers väg 3, Lund University, Faculty of Engineering LTH


 


Digit@LTH breakfast seminar: Modelling Intelligent Robot behaviour with Behavioural Trees by Volker Krueger

Published: 2019-01-16

Professor Volker KruegerTitle: Modelling Intelligent Robot behaviour with Behavioural Trees

Speaker:  Professor Volker Krueger, Department of Computer Science  

Location:  M-house, Ole Römers väg 1. M:E

Abstract: Behavioral Trees (BT) have been used for a long time in computer games to give artificial characters a certain level of intelligence: With BTs the programmer can easily describe very complex behaviour patterns of  the artificial game character, i.e., what the character should do in a given situation or context. This correlates very much with the problems in robotics where modern robots are expected to handle tasks in very different conditions and contexts. 

Traditionally, finite state machines were used in robotics to describe complex behavioural patterns, the use of BTs is relatively new in the robotics community. In my talk, I want to discuss what BTs are, their definition, how they are constructed and how they are used. Then, in order to get a deeper understanding, I want to put the BTs into the context of FSM: Finite state machines (FSMs) are constructs from Theoretical computer science. They are well known from the area of formal language. We know they model the class of regular languages and we can related each FSM directly to a regular expression and vice versa. The question is: can BTs be analysed in the same way as FSMs? Can we also express a BT in form of a formal language? Are BTs even equivalent to FSMs?

Bio: Volker Krueger has studied computer science at the University of Kiel, Germany where he graduated with a Diploma degree (M.Sc.) in 1995. Dr. Krueger has completed his Dr. Ing. (PhD) in 2000 at Kiel University in the area of computer vision with a thesis of Gabor Wavelet Networks. He spent is PostDoc time in the Lab of Azriel Rosenfeld and Rama Chellappa at CFAR, University of Maryland before joining Aalborg University (DK) in 2002 as an Associate Professor. Until 2005 Dr. Krueger was researching in the area of Biometrics, Face recognition and Gait recognition. Since 2005, he has focused on cognitive robot in general and since 2012 on cognitive robotics particularly for manufacturing. He became Full Professor at Aalborg University in 2014. He was PI in the EU projects Paco-Plus(FP6), Tapas(FP7) and Scalable (H2020), and he was coordinator of the projects GISA (ECHORD/FP7) and STAMINA (FP7). Dr. Krueger has joined LTH in August 2018 as  WASP professor for Autonomous Systems.

Registration: Please register no later than 27 March at 12.00 at https://www.lth.se/digitalth/events/register0328/

 

 

When: 2019-03-28 09:00 to 2019-03-28 10:00
Location: E-huset, Ole Römers väg 3. Start in EIT Lunch-room (E:2328) 
Contact: Jonas.Wisbrant@cs.lth.se


 


PhD dissertation by Jakob Helander: Millimeter Wave Imaging and Phased Array Antennas for 5G and Aerospace Applications

Published: 2019-02-28

Author: Jakob Helander, Department of EIT

Location:  E:C, E-Building, John Ericssons väg 4, Lund University, Faculty of Engineering LTH

Faculty opponent: Professor Andrea Massa

Thesis for download (PDF)

Abstract: 

Phased array antennas are cornerstones in many proposed antenna solutions concerning the next generation of both airborne radar systems and wireless communication systems (5G). Additionally, millimeter wave (mm-wave) frequencies are expected to play an integral role in 5G, and are deemed well-suited for inspecting structural components used in the aerospace industry.

This dissertation consists of a general introduction (Part I) and six scientific papers (Part II) - of which four have been published and two are under review in peer-reviewed international journals. The introduction comprises the background, the motivation and the subject-specific technical foundation on which the research presented in the included papers is based on. Fundamental theory on antenna arrays, mm-wave imaging systems and computational electromagnetics are presented together with the specific performance metrics, experimental setups, and computational acceleration algorithms that are of interest for the contained research work. The included papers can be divided into three tracks with two distinct applicational overlaps. 

Papers I and II concern electrically large phased arrays for airborne systems, and the numerical techniques that alleviate time-efficient and accurate simulations of such antennas. Paper I investigates the performance of two different approaches to the macro basis function (MBF) method for interconnected subdomains under the harsh electromagnetic conditions that endfire operation implies. Paper II presents a synthesis technique for endfire operation of large scale arrays that utilizes convex optimization to improve the impedance matching performance. 

Papers III and IV concern phased arrays for 5G applications. In Paper III, various array configurations of two microstrip antenna designs are evaluated with respect to two radiation performance metrics introduced specifically for evaluating the beam steering capabilities of phased array systems in the UE. A novel near field measurement technique for running electromagnetic field (EMF) exposure compliance tests of mm-wave phased arrays for future 5G devices is presented in Paper IV. 

Papers V and VI deal with mm-wave imaging systems developed for non-destructive testing (NDT) of composite materials used in the aerospace industry. A transmission-based bistatic imaging system is presented in Paper V, whereas Paper VI presents a further development of this system in a reflection-based measurement scenario. Data is retrieved using a planar scan, and the image retrieval algorithms comprise a numerical technique to separate the sources that contribute to the measured data, and an L1-minimization formulation to exploit potential sparsity of the sought-after solution.

When: 2019-03-26 09:15 to 2019-03-26 09:15
Location: E:C, E-Building, John Ericssons väg 4, Lund University, Faculty of Engineering LTH


 


Workshop on Sensing, Imaging, and Machine Learning

Published: 2019-02-11

We are arranging a Workshop on Sensing, Imaging, and Machine Learning at Lund University, March 5, 2019.
This workshop is organized in cooperation between Lund University and Saab.

The aim is to provide the participants with an opportunity to discuss and learn about theory and applications in Sensing,
Imaging, and Machine Learning.

Presentations will be held by researchers from academia and industry. 

Program overview

09.00 Welcome

09.15 IEEE Distinguished Lecturer Professor Carey Rappaport from Northeastern University will hold the keynote presentation.

10.05 Presenations

11.45 Lunch

13.15 Preentations

16.30 End of workshop

Please contact the organisers as soon as possible if you would like to contribute with a presentation.

Organisation:

christer [dot] Larsson [at] eit [dot] lth [dot] se (subject: WS_5_March)
mats [dot] gustafsson [at] eit [dot] lth [dot] se (subject: WS_5_March)

 

 

When: 2019-03-05 09:00 to 2019-03-05 16:30
Location: E:2311, The department of Electrical and Information Technology, E-building, Ole Römers väg 3, LTH, Lund University


 


PhD dissertation by Muris Sarajlic: Hardware-Conscious Wireless Communication System Design

Published: 2019-02-08

Author: Muris Sarajlic, Department of EIT

Location:  E:1406, E-building, Ole Römers väg 3, LTH, Lund University

Faculty opponent: Professor Christoph Studer

Thesis for download (PDF)

Abstract: 

The work at hand is a selection of topics in efficient wireless communication system design, with topics logically divided into two groups.

One group can be described as hardware designs conscious of their possibilities and limitations. In other words, it is about hardware that chooses its configuration and properties depending on the performance that needs to be delivered and the influence of external factors, with the goal of keeping the energy consumption as low as possible. Design parameters that trade off power with complexity are identified for analog, mixed signal and digital circuits, and implications of these tradeoffs are analyzed in detail. An analog front end and an LDPC channel decoder that adapt their parameters to the environment (e.g. fluctuating power level due to fading) are proposed, and it is analyzed how much power/energy these environment-adaptive structures save compared to non-adaptive designs made for the worst-case scenario. Additionally, the impact of ADC bit resolution on the energy efficiency of a massive MIMO system is examined in detail, with the goal of finding bit resolutions that maximize the energy efficiency under various system setups.

In another group of themes, one can recognize systems where the system architect was conscious of fundamental limitations stemming from hardware.
Put in another way, in these designs there is no attempt of tweaking or tuning the hardware. On the contrary, system design is performed so as to work around an existing and unchangeable hardware limitation. As a workaround for the problematic centralized topology, a massive MIMO base station based on the daisy chain topology is proposed and a method for signal processing tailored to the daisy chain setup is designed. In another example, a large group of cooperating relays is split into several smaller groups, each cooperatively performing relaying independently of the others. As cooperation consumes resources (such as bandwidth), splitting the system into smaller, independent cooperative parts helps save resources and is again an example of a workaround for an inherent limitation.

From the analyses performed in this thesis, promising observations about hardware consciousness can be made. Adapting the structure of a hardware block to the environment can bring massive savings in energy, and simple workarounds prove to perform almost as good as the inherently limited designs, but with the limitation being successfully bypassed. As a general observation, it can be concluded that hardware consciousness pays off.

When: 2019-02-25 09:15 to 2019-02-25 09:15
Location: E:1406, E-building, Ole Römers väg 3, LTH, Lund University


 


Seminar Accommodating Plane Wave Reflections in the kDB System by Professor Michael Havrilla

Published: 2019-02-13

Time and Location: 14:15-15:00, Monday, February 18, E:2311 (orangeriet)

Title: Accommodating Plane Wave Reflections in the kDB System

Speaker: Professor Michael Havrilla, Air Force Institute of Technology, WPAFB, OH, USA

Abstract: Recent developments in fabrication capabilities, such as 3D printing, has made it possible to rapidly prototype anisotropic and bianisotropic materials. There is much interest in these materials since they provide more control (i.e., magnitude, phase and polarization control) over scattered electromagnetic fields. Measurement-based techniques for determining material tensor values are often used due to the popularity and availability of network analyzers that are capable of acquiring reflection and transmission coefficients from material samples. Material tensor values are determined via comparison of the experimental and theoretical reflection and transmission coefficients. The goal of this talk is to discuss how the kEH and kDB systems1 can be used to compute theoretical reflection and transmission coefficients for planar bianisotropic samples. The advantages and drawbacks of each system is discussed. The main contribution of this work is provided by demonstrating how reflections from planar interfaces can be accommodated in the kDB system. Scattering examples and future work are also provided. 

1. J. A. Kong, Electromagnetic Wave Theory, Second Edition, New York, John Wiley and Sons, Inc., 1990.

Bio: Michael J. Havrilla received B.S. degrees in Physics and Mathematics in 1987, the M.S.E.E degree in 1989 and the Ph.D. degree in electrical engineering in 2001 from Michigan State University, East Lansing, MI. From 1990-1995, he was with General Electric Aircraft Engines, Evendale, OH, and Lockheed Skunk Works, Palmdale, CA, where he worked as an electrical engineer. He is currently a Professor in the Department of Electrical and Computer Engineering at the Air Force Institute of Technology, Wright-Patterson AFB, OH. He is a member of URSI Commission B, a senior member of the IEEE and AMTA, and a member of the Eta Kappa Nu and Sigma Xi honor societies. His current research interests include electromagnetic and guided-wave theory, electromagnetic propagation and radiation in complex media and structures, electromagnetic characterization of complex media and quantum field theory.

When: 2019-02-18 14:15 to 2019-02-18 15:00
Location: E:2311, The department of Electrical and Information Technology, E-building, Ole Römers väg 3, LTH, Lund University
Contact: christer.larsson@eit.lth.se


 


Seminar: Graph Kernels and Applications by Michalis Varziagiannis from Ecole Polytechnique

Published: 2019-02-13

Speaker:  Michalis Varziagiannis, Ecole Polytechnique

Title: Graph Kernels and Applications

Abstract: Graphs are becoming a dominant structure in current information management with many domains involved, including social networks, chemistry, biology, NLP etc. Then machine learning tasks involving graphs need valid similarity metrics that in the case of graphs pose significant challenges going beyond the simple vector based similarities.  In our group we have devoted significant efforts on the topic of graph similarity as cornerstone element of machine learning for graphs mostly in supervised tasks. We will present the different kernels we designed capitalising on degeneracy or successive embeddings and also how graph kernels can be exploited in diverse tasks including text mining and NLP. Also we will present briefly our ongoing efforts for deep learning based methods for graph and node classification. Finally we will present the Grakel   library, developed in our group,  that unifies several graph kernels into a common framework, written in Python and  build on top of scikit-learn. It can be naturally combined with scikit-learn's modules to build a complete machine learning pipeline for tasks such as graph classification and clustering.

Bio: Dr. Vazirgiannis is a Distinguished Professor at LIX, Ecole Polytechnique in France and leads the Data Science and Mining (SaSciM group. He holds a degree in Physics and a PhD in Informatics from Athens University(Greece) and a Master degree in AI from HerioWatt Univ Edinburgh. He has conducted research in GMD-IPSI, Max Planck MPI (Germany), in INRIA/FUTURS (Paris). He has been a teaching in AUEB (Greece), Ecole Polytechnique, Telecom-Paristech, ENS (France), Tsinghua, Jiaotong Shanghai (China) and in Deusto University (Spain).  His current research interests are on machine learning and combinatorial methods for Graph analysis (including community detection, graph clustering and embeddings, influence maximization), Text mining including Graph of Words, word embeddings with applications to web advertising and marketing, event detection and summarization. He has active cooperation with industrial partners in the area of data analytics and machine learning for large scale data repositories in different application domains. He has supervised fifteen completed PhD theses. He has published three books and more than a 160 papers in international refereed journals and conferences, earning best paper awards (CIKM2013 and IJCAI2018). He has organized large scale conferences in the area of Data Mining and Machine Learning (such as ECML/PKDD) while he participates in the senior PC of AI and ML conferences ? such as AAAI and IJCAI, He has received the ERCIM and the Marie Curie EU fellowships, the Tencent ?Rhino-Bird International Academic Expert Award? in 2017 and since 2015 he leads the AXA Data Science chair.

More information at: http://www.lix.polytechnique.fr/dascim
 

When: 2019-02-13 13:15 to 2019-02-13 14:00
Location: E:1406, E-building, Ole Römers väg 3, LTH, Lund University
Contact: volker.krueger@cs.lth.se


 


How to keep ESS running on time - Digit@LTH seminar by Anders Johansson, EIT

Published: 2019-01-08

Karta: Ole Römers Väg 3 in Lund, E:2358Title: How to keep ESS running on time...Nytt ärende CHG0137655 har öppnats för dig

Speaker:  Anders Johansson, Associate Professor at department of Electrical and Information Technology, LTH, Lund University

Location:  E-huset, Ole Römers väg 3. Start in EIT Lunch-room (E:2328)

Abstract
LTH, Lund University, have been deeply involved in the development of the accelerator for ESS since its establishment. Among the activities we have designed a couple of its subsystems, such as the low-level RF system and the master oscillator. One important challenge of the design is to keep everything synchonized, where parts of it has to be accurate to within pico-seconds. The talk will present how this is done, and how it is connected to the systems that have been designed at LTH.

Bio:
Anders J Johansson was born in Malmö, Sweden, in July 1968. He received the Masters, Lic. Eng. and Ph.D. degrees in electrical engineering from Lund University, Lund, Sweden, in 1993, 2000 and 2004 respectively.
From 1994 to 1997 he was with Ericsson Mobile Communications AB developing transceivers and antennas for mobile phones. Since 2005 he is an Associate Professor at the department of Electroscience at Lund University.

His research interests include antennas and wave propagation for medical implants as well as antenna systems and propagation modelling for MIMO systems. One of his main research areas is now also the design and implementation of high precision control systems for linear accelerators, especially the LLRF system for the European Spallation Source.

Please register at: https://www.lth.se/digitalth/events/register/ no later than 30 Januari at 12.00

 

 

When: 2019-01-31 09:00 to 2019-01-31 10:00
Location: E-huset, Ole Römers väg 3. Start in EIT Lunch-room (E:2328) 
Contact: Jonas.Wisbrant@cs.lth.se


 


WASP seminar: Collaborative General Problem-Solving AI Systems by Michael Thielscher

Published: 2019-01-23

Title: Collaborative General Problem-Solving AI Systems

Speaker: Michael Thielscher, professor of computer science at UNSW Sydney

Where: E:B, E-huset, Ole Römers väg 3, LTH, Lund University

Abstract: General problem-solving AI systems can understand descriptions of new tasks and successfully tackle them without human intervention. As an example of a general problem-solving technique, I will present and discuss approaches to collaborative acting and planning, which requires AI systems to reason about, and plan with, the knowledge and capabilities of their human users and other systems they need to cooperate with. General problem-solving robots moreover require architectures for cognitive robotics that integrate symbolic and sub-symbolic representations. I will present a formal framework for the design of control hierarchies along with an instantiation for a collaborative Baxter robot that combines high-level reasoning and planning with a physics simulator and low-level control nodes for motors and sensor processing. 

Bio: Michael Thielscher is a professor of computer science at UNSW Sydney, where he is also affiliated with the iCinema Research Centre. Michael Thielscher received his postgraduate diploma, Ph.D. and Higher Doctorate (Habilitation) in computer science from Darmstadt University in Germany. He then joined Dresden University, where he was an associate professor before he moved to his present position. His Habilitation thesis was honoured with the Award for Research Excellence by the alumni of Darmstadt University, and in 2009 he won a Future Fellowship Award from the Australian Research Council. His current research is mainly in Knowledge Representation, Cognitive Robotics, and General Problem-Solving AI. He is author of over 160 refereed papers and five books, including a new textbook on General Game Playing, and he has co-authored the award-winning system FLUXPLAYER, which in 2006 was crowned the World Champion at the AAAI General Game Playing Competition.

When: 2019-01-28 11:15 to 2019-01-28 12:00
Location: E:B, E-huset, Ole Römers väg 3, LTH, Lund University
Contact: Jacek.Malec@cs.lth.se


 


"Hardware/Algorithm Codesign for Energy Efficiency and Robustness: From Error-tolerant Computing to Approximate and Brain-inspi

Published: 2019-01-16

Title: Hardware/Algorithm Codesign for Energy Efficiency and Robustness: From Error-tolerant Computing to Approximate and Brain-inspired Computing

Speaker: Dr Abbas Rahimi, ETHZ

E:2311, E-huset; ole Römers väg 3Place: E:2311, The department of Electrical and Information Technology, E-building, Ole Römers väg 3, LTH, Lund University

Date and time: Tuesday, January 22, 2019, at 15:15

Abstract: Scaling model of semiconductors has been immensely successful in providing exponentially increasing computational performance at an ever-reducing cost and energy footprint. Underlying this evolution is a set of well-defined abstraction layers, starting from robust switching devices to a scalable and stored program architecture, which is Turing complete. Unfortunately, this abstraction chain is being challenged as scaling continues to nanometer dimensions. Maintaining the current deterministic computational model ultimately puts a lower bound on the energy scaling, set in place by uncertainty (arising from process variations, temporal changes, and data statistics). On the other hand, the nature of computation itself is changing with data and learning-based paradigm taking primacy. Both these trends force us to rethink functionality to cope with uncertainty by adopting energy-efficient computational approaches that are inherently robust to uncertainty and ?approximate? in nature.

We entail the formulation, analysis, and development of a unified hardware/software environment that addresses the challenge of uncertainty in deeply scaled CMOS processes. Specifically, we devise codesigned methods to predict and prevent, detect and correct, and opportunistically accept impact of uncertainty and the resulting errors at many layers in the system abstraction. This discussion naturally leads to use of these methods into area of approximate computing where errors and approximations are becoming acceptable as long as the outcomes have a well-defined statistical behavior. Going one step further, we take inspiration from the very size of the brain?s circuits, to compute with points of a hyperdimensional (HD) space that thrives on randomness and mediocre components. HD computing provides a novel look at data representations (holographic and pseudorandom HD vectors), associated operations, and materials and substrates that enable them. This novel computing paradigm is closely intertwined with properties of emerging monolithically 3-D integrated and nonvolatile nanotechnologies. This synergy enables codesigned solutions to overcome large variability in both data and computing platform leading to fast learning and robust decision making with extreme energy efficiency. This offers a unique opportunity for the next-generation nanoscalable fabrics especially for cognitive and perceptive applications.

About the speaker: Abbas Rahimi received his B.S. in computer engineering from the University of Tehran, Tehran, Iran (2010) and his M.S. and Ph.D. in computer science and engineering from the University of California San Diego, CA, USA (2015), followed by two years postdoctoral research in the Department of Electrical Engineering and Computer Sciences at the University of California Berkeley, Berkeley, CA, USA. Dr. Rahimi has been awarded an ETH Zurich Postdoctoral Fellowship, and subsequently joined the Department of Information Technology and Electrical Engineering at ETHZ in June 2017. He is also affiliated with the Berkeley Wireless Research Center. His research interests include embedded systems and software, brain-inspired computing, approximate computing, and massively parallel integrated architectures with an emphasis on improving energy efficiency and robustness. His doctoral dissertation has received the 2015 Outstanding Dissertation Award in the area of ?New Directions in Embedded System Design and Embedded Software? from the European Design and Automation Association (EDAA). He has also received the Best Paper at BioCAS (2018), BICT (2017), and the Best Paper Candidate at DAC (2013).

When: 2019-01-22 15:15 to 2019-01-22 15:15
Location: E:2311, The department of Electrical and Information Technology, E-building, Ole Römers väg 3, LTH, Lund University
Contact: ove.edfors@eit.lth.se


 


Page Manager: |