2016 ieee workshop on statistical signal...

2016 IEEE Workshop on Statistical Signal Processing

26 - 29 June 2016

Palma de Mallorca, Spain

Office of Naval Research Global2 Providence CourtMayfairLondon W1K 6PR


Copyright and Reprint Permission: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limit of U.S. copyright law for

private use of patrons those articles in this volume that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through

Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For other copying, reprint or republication permission, write to IEEE Copyrights Manager, IEEE

Operations Center, 445 Hoes Lane, Piscataway, NJ 08854.

All rights reserved.

Copyright © 2016 by IEEE.

IEEE catalog number: CFP16SAP-USB

ISBN: 978-1-4673-7802-4

3

SSP 2016 Palma de Mallorca

SSP 2016 SPONSORS

WELCOME MESSAGE

COMMITTEES SSP 2016 Organizing Committee

SSP 2016 Technical Program Committee

PRACTICAL INFORMATION Mallorca

Social Program Venue

TECHNICAL PROGRAM

Keynote SpeakersProgram at a Glance

Poster Presentation GuidelinesDetailed Technical Program

AUTHOR INDEX

NOTES

4

5

67

99

1011

1314202236

73

79

4


SSP 2016 SPONSORS


5


WELCOME MESSAGEGeneral Chairs of SSP 2016

On behalf of the organizing committee, it is our greatest pleasure to welcome you to the IEEE Workshop on Statistical Signal Processing 2016 (SSP’16), which will be held in Palma de Mallorca (Spain) from the 26th to the 29th of June, 2016. It will feature 6 plenary sessions that will serve as cornerstones of an exciting program including 8 special and 11 regular poster sessions. The plenary talks will address cutting-edge topics and will be delivered by our renowned colleagues Petar M. Djuric’, Zoubin Ghahramani, Pablo Laguna, José Moura, Wolfgang Utschick and Antonio Torralba.

We have been exceedingly fortunate to work with a highly-committed team, who has made SSP’16 possible. Sergios Theodoridis and Konstantinos Slavakis handled the review process and put together the technical program. Mónica F. Bugallo organized the special sessions. Matilde Sánchez wrestled the workshop finances. Pau Closas handled the workshop publicity, the travel grants and compiled the proceedings. Guillem Femenías and Felip Riera-Palou were in charge of the local arrangements. Harold Molina set up the workshop website and Ana Hernando managed its contents, besides finding time to liaise with our sponsors. Furthermore, we express our sincere gratitude to all the authors, session organizers, area chairs, and reviewers for their valuable contributions.

The venue of the workshop is the Museum of Modern and Contemporary Art of Palma, Es Baluard, a unique space in a unique city. Palma de Mallorca, recently singled-out by The Times as ‘the best place to live in the world’, is the capital of the Balearic Islands, a Mediterranean archipelago off the eastern coast of Spain. It displays an outstanding mixture of historic and cultural wealth, magnificent landscapes and a benign Mediterranean climate. We encourage all SSP participants to spare some time and enjoy a taste of life in Palma.

Finally, we take this opportunity to thank our sponsors: the IEEE Signal Processing Society, the Office of Naval Research Global, Universidad Carlos III de Madrid and Universitat de les Illes Balears. Without their support, this 19th IEEE SSP workshop would not have been possible.

We look forward to meeting you in Palma and we wish you enjoy the workshop.

Antonio Artés and Joaquín MíguezGeneral Chairs

6


SSP 2016Organizing Committee

General chairsAntonio Artés-Rodríguez (Universidad Carlos III de Madrid, Spain)Joaquín Míguez Arenas (Universidad Carlos III de Madrid, Spain; Queen Mary University of London, UK)

Technical chairsSergios Theodoridis (National and Kapodistrian University of Athens, Greece)Konstantinos Slavakis (University of Buffalo, USA)

Finance chairMatilde Sánchez-Fernández (Universidad Carlos III de Madrid, Spain)

Special Sessions chairMónica F. Bugallo (Stony Brook University, USA)

Local Arrangements chairGuillem Femenias (Universitat de les Illes Balears, Spain)Felip Riera-Palou (Universitat de les Illes Balears, Spain)

Publications chairPau Closas (Centre Tecnològic de Telecomunicacions de Catalunya, Spain)

7


SSP 2016Technical Program Committee

Yuri Abramovich; W R Systems, Ltd, USA.Patrice Abry; Ecole Normale Superieure, Lyon, France.Jerónimo Arenas-García; Universidad Carlos III de Madrid, Spain.Kostas Berberidis; University of Patras, Greece.Alberto Carini; University of Urbino, Italy.Luis Castedo; University of A Coruña; Spain.Jonathon Chambers; Newcastle University, UK.Caroline Chaux; Aix-Marseille Université; France.Yuejie Chi; The Ohio State University, USA.Daniel Clark; Heriot Watt, UK.Mark Coates; McGill University, Canada.Romain Couillet; Centrale Supélec, France.Rodrigo de Lamare; Pontifical Catholic University of Rio de Janeiro, Brazil.Subhrakanti Dey; Uppsala University, Sweden.Paulo Diniz; Universidade Federal do Rio de Janeiro; Brazil.Petar Djuric’; Stony Brook University, USA.Kutluyıl Dogançay; University of South Australia, Australia.Aleksandar Dogandžic’; Iowa State University; USA.Marco Duarte; University of Massachusetts Amherst, USA.Bogdan Dumitrescu; Tampere University of Technology, Finland.Alper Erdogan; Koc University, Turkey.Mounir Ghogho; University of Leeds, UK.Fulvio Gini; University of Pisa, Italy.Simon Godsill; University of Cambridge, UK.Neil Gordon; DSTO, Australia.Maria Greco; University of Pisa, Italy.Rémi Gribonval; Centre de Recherche INRIA, France.Yuantao Gu; Tsinghua University, China.Jarvis Haupt; University of Minnesota, USA.Braham Himed; Air Force Research Laboratory, AFRL, USA.Reza Hoseinnezhad; RMIT University, Australia.Jérôme Idier; Institut de Recherche en Communications et Cybernétique de Nantes, IRCCyN,France.Hamid Krim; North Carolina State University, USA.Hongbin Li; Stevens Institute of Technology, USA.Athanasios Liavas; Technical University of Crete, Greece.Yue Lu; Harvard University, USA.Antonio Marques; Universidad Rey Juan Carlos, Spain.Pina Marziliano; Nanyang Technological University, Singapore.

8


Gonzalo Mateos; University of Rochester, USA.Gerald Matz; Vienna University of Technology, Austria.Steve McLaughlin; Heriot Watt University, UK.Magnus Mossberg; Karlstad University, Sweden.Jose Carlos Moreira Bermudez; Federal University of Santa Catarina; Brazil.Juan José Murillo-Fuentes; Universidad de Sevilla, Spain.Vitor Nascimento; Universidade de São Paulo, USP, Brazil.Arye Nehorai; Washington University in St. Louis, USA.Fernando Pérez-González; Universidad de Vigo, Spain.Jean-Christophe Pesquet; Université Paris-Est, France.Anh Huy Phan; RIKEN, Japan.Tomasz Piotrowski; Nicolaus Copernicus University, Poland.Eric Plourde; Université de Sherbrooke, Canada.Mark Plumbley; University of Surrey, UK.Nelly Pustelnik; ENS Lyon, France.Raviv Raich; Oregon State University, USA. Branko Ristic; RMIT University, Australia.Athanasios Rontogiannis; National Observatory of Athens, Greece.Ignacio Santamaría; Universidad de Cantabria, Spain.Karin Schnass; University of Innsbruck, Austria.Abd-krim Seghouane; University of Melbourne, Australia.Erchin Serpedin; Texas A&M University, USA.Dirk Slock; EURECOM, France.Hing Cheung So; City University of Hong Kong, China.Nordholm Sven; Curtin University of Technology, Australia.Toshihisa Tanaka; Tokyo University of Agriculture and Technology, Japan.Jean-Yves Tourneret; University of Toulouse, France.Alle-Jan van der Veen; TUDelft, The Netherlands.Kush Varshney; IBM Thomas J. Watson Research Center, USA.Javier Vía; University of Cantabria, Spain.Ba-Tuong Vo; Curtin University, Australia.Wenwu Wang; University of Surrey, UK.Stefan Werner; Aalto University, Finland.Ami Wiesel; Hebrew University in Jerusalem, Israel.Isao Yamada; Tokyo Institute of Technology, Japan.Abdelhak Zoubir; Darmstadt University of Technology, Germany.

9


PRACTICAL INFORMATION Mallorca

Imagine an island with turquoise-blue seas, secret sparkling coves, soft golden sands, lush green foliage, soaring rugged mountains, sleepy honey-stone villages, and a vibrant capital city. A place steeped in history, with a rich artisan & cultural scene, and featuring gastrono-mic excellence. Imagine this island also has 300 days of sunshine a year, with hot summers and mild winters. Welcome to Mallorca.

It is one of the Balearic Islands in the western Mediterra-nean, off the east coast of Spain. Its location means it is no more than a three hour flight from northern Europe, making it easily accessible for those in search of a little piece of paradise.

Mallorca really does have it all - glorious beaches and majestic mountains, a wealth of historical buildings and cultural experiences, a buzzing and cosmopolitan city, theme parks for all the family and all the shopping you could care to do whilst on holiday. Whether you want to visit one of the beautiful country houses, enter one of the remarkable caves, or chill out in a beach club, Mallorca has it.

Each region in Mallorca has its own particular appeal - the northeast for history, the east coast for beaches and caves, the north and west for spectacular mountains and picture-postcard villages. If you want to know about the real Mallorca you should drive across Es Pla, the fertile plain at the center of the island, with its almond gro-ves, windmills and old market towns. Choosing where to stay depends on what you want to see and do, from the bustle of Palma, to the coast, to deep inland with many luxury retreats being in places you can get away from it all.

Of course, the island has also been well-known for its purpose built resorts and package holidays. Very popular they were too, but Mallorca has changed since their hey-day in the later part of the 20th century. Local town halls on Mallorca have invested time and energy into smarte-ning up their resorts and making them a lot more family-friendly. Mallorca is now a destination where everyone can find their own spot, whether that be a quiet coastal village, a party-central beach resort, a mountain hi-deaway or a rural idyll. MALLORCA HAS IT ALL.

10


Sunday, 26 June 2016

Visit to Palma de MallorcaDeparture time: 18:00 h (6:00 P.M.)Departure place: Es Baluard Museu d’Art Modern i Contemporani de PalmaEstimated Arrival time: 20:30h. (8:30 P.M.)Arrival Place: The Bartolomé March Foundation (Welcome Cocktail)

Welcome CocktailPlace: The Bartolomé March Foundation (Palau Reial, 18)Time: 20:30 – 22:30 h (8:30 – 10:30 P.M.).Dress Code: Casual DressTuesday, 28th June 2016

Tuesday, 28 June 2016

Visit to ValldemossaDeparture time: 18:00 h (6:00 P.M.)Departure place: Es Baluard Museu d’Art Modern i Contemporani de PalmaEstimated Arrival time: 21:00h (9:00 P.M.)Arrival place: Balneario Illetas Beach Club (Banquet Dinner)

Banquet DinnerPlace: Balneario Illetas Beach Club (Passeig Illetes, 52, 07181 Calvià)Time: 21:00h – 23:30h (9:00 to 11:30 P.M.)Dress Code: Casual Dress

SSP 2016Social Program

11


The 2016 IEEE Workshop on Statistical Signal Processing (SSP 16) will be held at Palma de Mallorca, Spain (Es Baluard Museu d’Art Modern i Contemporani) from 26 June to 29 June 2016.

Es Baluard, Museu d’Art Modern i Contemporani de PalmaPlaça Porta de Santa Catalina Nº 10, Palma 07012Tel. (34) 971 908 200Fax (34) 971 908 [email protected]

Es Baluard lies in a privileged corner of the city of Palma. Situated on Plaça Porta de Santa Catalina at the end of Paseo Mallorca, one of the city’s main arteries, its wat-chtower commands one of the finest views of the bay of Palma, with the Gothic cathedral and Bellver Castle as reference points. Its location in the centre of town connects it to the city’s other major infrastructures and charming historic district and its proximity to Avenida Jaime III links it to the city’s main commercial street.

WIFI ACCESSA username and a password will be provided at the re-gistration desk.

REGISTRATION DESKThe registration will be located at Es Baluard, Museu d’Art Modern i Contemporani de Palma on Sunday from 17:00 to 21:00, Monday from 08:00 to 18:00, Tuesday from 8:00 to 18:00 and Wednesday from 8:00 to 12:00.

ACCESSBuses: 1, 2, 3, 4, 5, 6, 7, 20, 46 and 50 (tourist bus)Car Parks: Paseo Mallorca, 2 minutes from the Mu-seum, Parc de la Mar, 10 minutes from the Museum, Via Roma, 5 minutes from the Museum

LUNCHEONSDate: Monday 27 and Tuesday 28Time: 13:00 – 15:00Place: Es Baluard Museu d’Art Modern i Contemporani de Palma

PRACTICAL INFORMATION Venue

Foto

C.C

. Ola

f Tau

sch

12


Aljub RoomPlenary Talks and poster presentations will take place in the Aljub Room (Floor -1, with indepen-dent entrances from the central courtyard -direct lift- and the Seafront Promenade.)

1. Conference Entrance (Aljub Room)

2. Luncheons: Baluard Restaurant & Lounge

1

2

13


TECHNICAL PROGRAM

14


KEYNOTE SPEAKERSJose M. F. Moura

Monday, June 27, 09:00 – 10:00

“Data Science: Analytics for Unstructured and Distributed Data”

José Moura is the Philip L. and Marsha Dowd University Professor at CMU, with interests in signal processing and data science. He coinvented (with ALEK Kavcic) a patented detector found in at least 60% of the disk drives of all computers sold worldwide in the last 12 years (over 3 billion and counting). He is (2016) IEEE VP for Technical Activities, IEEE Board Director, and was President of the IEEE Signal Processing Society (SPS), and Editor in Chief for the Transactions on SP. Moura received the IEEE SPS Technical Achievement Award and Society Award. He is Fellow of the IEEE and of AAAS, corresponding member of the Academy of Sciences of Portugal, Fellow of the US National Academy of Innovators, and member of the US National Academy of Engineering.

AbstractSignal Processing has traditionally dealt with time series, images, video where data is indexed by time ticks and pixels. The structure of the indexing set is taken for granted. In the last few years, new opportunities for signal and data processing have arisen, except data is now indexed by social agents, genes, customers of service providers, or by some other arbitrary enumeration suggested by the application. We develop Signal Processing on Graphs by revisiting the fundamentals of Signal Processing, developing for data (signals) arising from these various domains the essential concepts and methods of traditional Signal Processing. We illustrate the approach with data drawn from a number of different applications including social networks and customer data. Ours is an attempt to identify structure in unstructured data and theory and modeling in the “data deluge[1].”

[1] The End of Theory: The Data Deluge Makes The Scientific Method Obsolete, Chris Anderson, Wired Magazine, June 23, 2008.

15


KEYNOTE SPEAKERSPetar M. Djuric’

Monday, June 27, 12:00 – 13:00

“The past, present, and future of particle filtering”

Petar M. Djuric’ received the B.S. and M.S. degrees in electrical engineering from the University of Belgrade, Belgrade, Yugoslavia, and the Ph.D. degree in electrical engineering from the University of Rhode Island, Kingston, RI, USA. He is currently a Professor with the Department of Electrical and Computer Engineering, Stony Brook University, Stony Brook, NY, USA. His research has been in the area of signal and information processing with primary interest in signal analysis, modeling, detec-tion, and estimation; Monte Carlo-based methods; Bayesian theory; machine learning; network science; applications to biomedicine; Radio Frequency Identification. He has been invited to lecture at many universities in the United States and overseas. Prof. Djuric’ was a recipient of the IEEE Signal Processing Magazine Best Paper Award in 2007 and the EURASIP Technical Achievement Award in 2012. He is the Editor-in-Chief of the IEEE Transactions on Signal and Information Processing over Networks. Prof. Djuric’ is a Fellow of the IEEE and EURASIP.

AbstractParticle filtering has emerged as one of the most exciting developments in sequential signal proces-sing in the last two decades. It has become a methodology of choice for resolving most complex problems of stochastic filtering. In this presentation, the basics of particle filtering and its properties will be briefly reviewed. Then an overview of various advances of the methodology in recent years will be provided. Current challenges and the future of particle filtering will also be discussed.

16


KEYNOTE SPEAKERSAntonio Torralba

Monday, June 27, 15:00 – 16:00

“Visual Scene Understanding”

Antonio Torralba received the degree in telecommunications engineering from Telecom BCN, Spain, in 1994 and the Ph.D. degree in signal, image, and speech processing from the Institut National Polytechnique de Grenoble, France, in 2000. From 2000 to 2005, he spent postdoctoral training at the Brain and Cognitive Science Department and the Computer Science and Artificial Intelligence Laboratory, MIT. He is now a Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT). Prof. Torralba is an Associate Editor of the International Journal in Computer Vision, and program chair for the Computer Vision and Pattern Recognition conference in 2015. He received the 2008 National Science Foundation (NSF) Career award, the best student paper award at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) in 2009, and the 2010 J. K. Aggarwal Prize from the International Association for Pattern Recognition (IAPR).

AbstractIt is an exciting time for computer vision. With the success of new computational architectures for visual processing, such as deep neural networks (e.g., CNNs) and access to image databases with millions of labeled examples (e.g., ImageNet, Places), the state of the art in computer vision is advancing rapidly. Computer vision is now present among many commercial products, such as digital cameras, web applications, security applications, etc.

In this talk I will describe some of the challenges faced by computer vision and some of our recent work on visual scene understanding that try to build integrated models for scene and object recognition, emphasizing the power of large database of annotated images in computer vision.

17


KEYNOTE SPEAKERSWolfgang Utschick

Tuesday, June 28, 09:00 – 10:00

“Model-Aware Compressive Sensing with Applications to Channel Estimation”

Wolfgang Utschick completed several accredited industrial training programs before he received the diploma (’93) and doctoral degrees (’98) in electrical engineering, both with honors, from Technische Universität München (TUM). Since 2002 Dr. Utschick has been appointed Professor at TUM where he is director of the Signal Processing Chair (Professur für Methoden der Signalverarbeitung). Dr. Utschick teaches courses on Signal Processing, Stochastic Processes, and Optimization Theory in the field of Wireless Communications, Signal Processing Applications and Power Transmission Systems. Since 2011 he is serving as a regular guest professor at Singapore’s new autonomous university, Singapore Institute of Technology (SIT). He holds several patents in the field of multi-antenna signal processing and has authored and co-authored a great many of technical articles in international journals and conference proceedings. He edited several books and is founder and editor of the Springer book series Foundations in Signal Processing, Communications and Networking. Dr. Utschick has been principal investigator in multiple research projects funded by the German Research Fund (DFG). He is currently the coordinator and spokesman of the German DFG focus program Communications over Interference limited Networks (COIN) which is devoted to topics as cooperative communications, crosslayer design, ad-hoc wireless networks, etc. He is a member of the VDE and senior member of the IEEE, where he currently serves as an elected member for the IEEE SPS Technical Committee on Signal Processing for Communications and Networking and as the Chair of the German Signal Processing Section. In 2016 Dr. Utschick has been appointed Vice Dean for the TUM Department for Electrical and Computer Engineering.

AbstractCompressive sensing (CS) methods are currently embraced to solve nonlinear parameter estimation problems. By discretizing the parameter space, it is possible to write the original estimation problem as a high-dimensional, linear problem with sparsity constraints. Many algorithms have been developed that can solve such sparsity-constrained linear systems even if they are severely underdetermined. A crucial difficulty measure associated with a given problem is the restricted isometry constant (RIC) of the measurement matrix, which relates the observations with the unknowns. The algorithm is successful if the constant is below a threshold, which depends on the algorithm. The RIC quantifies the maximal amount of distortion that a sparse vector undergoes in the measurement process, i.e., when multiplied with the measurement matrix.In this presentation we focus on the exploitation of additional structure in the vector of the unknowns. In particular, the RIC of a matrix is determined by those sparse vectors that are maximally distorted. If these harmful vectors are removed from the class of allowed vectors, the RIC shrinks. Within the context of block CS and model-based CS, it has been shown that such a procedure is indeed possible if the additional structural information can be expressed in terms of the support of the unknown vector. For example, it is possible to enforce a minimum separation of adjacent nonzero entries of the unknown vector. We propose a framework, model-aware compressive sensing (MA-CS), with which certain manifold structural constraints can be incorporated into CS algorithms. Such constraints arise in multi-dimensional parameter estimation problems. Our main application example is channel estimation in wireless communications and radar. In this estimation problem, the goal is to estimate several delay and angle parameters. If both parameters are discretized, the RIC of the corresponding measurement matrix grows very quickly. Within the MA-CS framework, it is possible to discretize only the delay parameters and use conventional estimators for the angle parameters and still perform a joint estimation.

18


KEYNOTE SPEAKERSZoubin Ghahramani

Tuesday, June 28, 15:00 – 16:00

“Automating Machine Learning”

Zoubin Ghahramani is Professor of Information Engineering at the University of Cambridge, where he leads a group of about 30 researchers. He studied computer science and cognitive science at the University of Pennsylvania, obtained his PhD from MIT in 1995, and was a postdoctoral fellow at the University of Toronto. His academic career includes concurrent appointments as one of the founding members of the Gatsby Computational Neuroscience Unit in London, and as a faculty member of CMU’s Machine Learning Department for over 10 years. His current research focuses on nonparametric Bayesian modelling and statistical machine learning. He has also worked on applications to bioinformatics, econometrics, and a variety of large-scale data modelling problems. He has published over 200 papers, receiving 25,000 citations (an h-index of 68). His work has been funded by grants and donations from EPSRC, DARPA, Microsoft, Google, Infosys, Facebook, Amazon, FX Concepts and a number of other industrial partners. In 2013, he received a $750,000 Google Award for research on building the Automatic Statistician. He serves on the advisory boards of Opera Solutions and Microsoft Research Cambridge, on the Steering Committee of the Cambridge Big Data Initiative, and in a number of leadership roles as programme and general chair of the leading international conferences in machine learning: AISTATS (2005), ICML (2007, 2011), and NIPS (2013, 2014). More information can be found at http://mlg.eng.cam.ac.uk

AbstractI will describe the “Automatic Statistician“, a project which aims to automate the exploratory analysis and modelling of data. Our approach starts by defining a large space of related probabilistic models via a grammar over models, and then uses Bayesian marginal likelihood computations to search over this space for one or a few good models of the data. The aim is to find models which have both good predictive performance, and are somewhat interpretable. The Automatic Statistician generates a natural language summary of the analysis, producing a 10-15 page report with plots and tables describing the analysis. I will also link this to recent work we have been doing in the area of Probabilistic Programming (including an new system in Julia) to automate inference, and on the rational allocation of computational resources (and our entry in the AutoML conference). The theme is: automate, automate, automate!

19


KEYNOTE SPEAKERSPablo Laguna

Wednesday, June 29, 09:00 – 10:00

“Physiologically driven, model-based, multi-modal biomedical signal processing approaches”

I was born in Hoz de Jaca, Spain, in 1962. Received the M.S. degree in physics and the Ph.D. from the Science Faculty at the University of Zaragoza, Zaragoza, Spain, in 1985 and 1990, respectively. Currently, I am Full Professor of Signal Processing and Communications in the Department of Electrical Engineering at the Engineering School, and a Researcher at the Aragón Institute for Engineering Research (I3A), both at the University of Zaragoza, also a member of the Spanish Center for Biomedical Engineering, Biomaterial and Nano-medicine Research CIBER-BBN. From 1992 to 2005, I was an Associate Professor at the same University and from 1987 to 1992, I was Assistant Professor of Automatic Control in the Department of Control Engineering at the Polytechnic University of Catalonia (U.P.C.), Barcelona, Spain, and a Researcher in the Biomedical Engineering Division of the Institute of Cybernetics at U.P.C.–C.S.I.C where I develop the PhD thesis. My professional research interests are in signal processing, in particular with biomedical applications.

AbstractBiomedical signals convey information about biological systems, and can emanate from sources of as varied origins as electrical, mechanical or chemical. In particular, biomedical signals can provide relevant information on the function of the human body. This information, however, may not be apparent in the signal due to measurement noise, presence of signals coming from other interacting subsystems, or simply because it is not visible to the human eye. Signal processing is usually required to extract the relevant information from biomedical signals and convert it into meaningful data that physicians can interpret. In this respect, knowledge of the physiology behind the biomedical measurements under analysis is fundamental. Not considering the underlying physiology may lead, in the best case, to processing methods that do not fully exploit the biomedical signals being analyzed and thus extract only partially their meaningful information and, in the worst case, to processing methods that distort or even remove the information of interest in those signals.

Biomedical signal processing (BSP) tools are typically applied on just one particular signal recorded at a unique level of the functional system under investigation and with limited knowledge of the interrelationships with other components of that system. In most instances though, BSP can benefit from an analysis in which more than one signal is evaluated at a time (multi-modal processing), different levels of function are considered (multi-scale processing) and scientific input from different disciplines is incorporated (multi-disciplinary processing). For each problem at hand, the BSP researcher should decide up to which extent information from a number of signals, functional levels or disciplines needs to be incorporated to solve the problem.

As an example, a multi-scale model may be necessary in cases where, for instance, a deeper knowledge of the cell and tissue mechanisms underpinning alterations in a body surface signal is required, whereas a simplified single-scale model may be sufficient in other cases, as when investigating the relationship between two signals measured on the whole human body. At present, there are many biomedical signals that can be acquired and processed using relatively low-cost systems, which makes their use in the clinics very extensive. In particular, non-invasive signals readily accessible to physicians are increasingly being used to improve the diagnosis, treatment and monitoring of a variety of diseases. The presentation aims to illustrate the role played by BSP in the analysis of cardiovascular signals. A set of applications will be presented where BSP contributes to improve our knowledge on atrial and ventricular arrhythmias, the modulation of cardiac activity by the autonomic nervous system (ANS) and the interactions between cardiac and respiratory signals.

20


Program at a Glance

START TIME SUNDAY JUNE 26 MONDAY JUNE 27 TUESDAY JUNE 28 WEDNESDAY JUNE 29

08:00 8:00 - 18:00

Registration8:00 - 18:00

Registration8:00 - 12:00

Registration

08:45 8:45 - 9:00

Opening Ceremony

09:00 9:00 - 10:00

Plenary TalkJose Moura

9:00 - 10:00

Plenary TalkWolfgang Utschick

9:00 - 10:00

Plenary TalkPablo Laguna

10:00 10:00 - 11:30

Poster Session 1 & Special Session(s)

10:00 - 11:15


10:00 - 10:30 Coffee Break

10:30 10:30 - 12:00


11:00

11:15 11:15 - 11:45 Coffee Break

11:30 11:30 - 12:00 Coffee Break

11:45 11:45 - 13:00


12:00 12:00 - 13:00Plenary TalkPetar Djuric’

12:00 - 12:15 Closing

12:15

12:30

12:45

13:00 13:00 - 15:00 Lunch 13:00 - 15:00 Lunch

15:00 15:00 - 16:00Plenary Talk

Antonio Torralba

15:00 - 16:00

Plenary Talk Zoubin Ghahramani

16:00 16:00 - 16:30 Coffee Break 16:00 - 16:30 Coffee Break

16:30 16:30 - 18:00


16:30 - 18:00


17:00 17:00 - 21:00 RegistrationEs Baluard

18:00 18:00 - 20:30

Tour to Palma de Mallorca 18:00 - 21:00

Visit to Valldemossa

20:30 20:30 - 23:00

Welcome Reception21:00 21:00 - 23:00

Banquet Dinner23:00

21


SUNDAY JUNE 26, 2016

17:00 – 21:00 Registration Desk. Es Baluard, Museu d’Art Modern i Contemporani de Palma

18:00 - 20:30 Tour to Palma de Mallorca

20:30 - 22:30 Welcome Reception

MONDAY JUNE 27, 2016

08:00 Registration Desk opening

08:45 - 09:00 Opening Ceremony

09:00-10:00 Plenary Talk : José Moura

10:00-11:30 Mon-Ia: Detection and estimation theory IMon-Ib: Signal processing over graphs and networks IMon-Ic SS: Random matrices in signal processing and machine learning

11:30-12:00 Coffee break

12:00-13:00 Plenary talk: Petar Djuric’

13:00-15:00 Lunch

15:00-16:00 Plenary talk: Antonio Torralba


16:30-18:00 Mon-IIa: Machine learning and pattern recognition I,Mon-IIb SS: Advanced robust techniques for signal processing applications,Mon-IIc SS: Recent advances in Monte Carlo methods for multi-dimensional signal processing and machine learning

TUESDAY JUNE 28, 2016


09:00-10:00 Plenary talk: Wolfgang Utschick

10:00-11:15 Tue-Ia: Array processing, radar and sonar,Tue-Ib: Detection and estimation theory II,Tue-Ic SS: Multivariate statistical signal modeling and analysis


11:45-13:00 Tue-IIa: Compressed sensing,Tue-IIb: Signal processing over graphs and networks II,Tue-IIc SS: Bayesian detection and estimation techniques for radar applications,Tue-IId SS: Making sense out of multi-channel physiological data for pervasive health applications

13:00-14:00 Lunch

15:00-16:00 Plenary talk: Zoubin Ghahramani


16:30-18:00 Tue-IIIa: Machine learning and pattern recognition II,Tue-IIIb: Signal processing for communications,Tue-IIIc SS: Statistical signal processing and learning in smart grid

18:00-21:00 Visit to Valldemossa

21:00-23:30 Banquet Dinner

WEDNESDAY JUNE 29, 2016


09:00-10:00 Plenary talk: Pablo Laguna


10:30-12:00 Wed-Ia : Applications (biomedical, energy, security),Wed-Ib: Detection and estimation theory III,Wed-Ic SS: Optimization and simulation for image processing

12:00-12:15 Closing

Program at a Glance

22


Poster Presentation Guidelines

All the accepted contributions will be presented in poster format. Poster presentations will take place in the Aljub Room.

The Aljub, a structure dating from the 17th century, is now one of the Museum’s most emblematic spaces. It has been restored and fitted out with the necessary infrastructure for holding almost any kind of event.

It is located on the floor -1, with independent entrances from the central courtyard (direct lift) and the Seafront Promenade.

Poster boards (1.00m wide, 2.45m tall) will be available. Please prepare your poster in a format respecting these dimensions. Recommended poster size: A0 portrait

The poster sessions will be in general 1 hour and a half long. Authors should arrive to ensure their poster is in place before the session starts. Authors are requested to be present during the poster session only. All poster materials should be removed by the end of the scheduled session.

ENTRANCE / EXIT

23


* The panels for placing the posters are located at the back of the Room Aljube.

* The final location of the posters may vary.

24


Mon-I: Detection and estimation theory I,Mon-I: Signal processing over graphs and networks I,Mon-I SS: Random matrices in signal processing and machine learning

Poster Presentation GuidelinesMonday, June 27 10:00-11:30h

Nº SESION TITLE

1 Mon-1a.1 Efficient Distributed Estimation of Inverse Covariance Matrices

2 Mon-1a.2 Sampling schemes and parameter estimation for nonlinear Bernoulli-Gaussian sparse models

3 Mon-1a.3 Measure transformed quasi likelihood ratio test for Bayesian binary hypothesis testing

4 Mon-1a.4 Detecting the dimension of the subspace correlated across multiple data sets in the sample poor regime

5 Mon-1a.5 Nonparametric estimation of a shot-noise process

6 Mon-1a.6 Finite sample performance of least squares estimation in sub-Gaussian noise

7 Mon-1a.7 Weighting a resampled particle in Sequential Monte Carlo

8 Mon-1a.8 Structure-Induced Complex Kalman Filter for Decentralized Sequential Bayesian Estimation

9 Mon-1a.9 An EM Algorithm for Maximum Likelihood Estimation of Barndorff-Nielsen's Generalized Hyperbolic Distribution

10 Mon-1a.10 Markov-tree Bayesian Group-sparse Modeling with Wavelets

11 Mon-1b.1 Inferring Network Properties From Fixed-Choice Design with Strong and Weak Ties

12 Mon-1b.2 Estimating Signals over Graphs via Multi-kernel Learning

13 Mon-1b.3 Network Topology Identification from Spectral Templates

14 Mon-1b.4 Bayesian Inference of Diffusion Networks with Unknown Infection Times

15 Mon-1b.5 Clustering time-varying connectivity networks by Riemannian geometry: The brain-network case

16 Mon-1b.6 Temporal Network Tracking based on Tensor Factor Analysis of Graph Signal Spectrum

17 Mon-1b.7 Multitask Diffusion LMS with Optimized Inter-Cluster Cooperation

18 Mon-1c SS.1 Robust Shrinkage M-estimators of Large Covariance Matrices

19 Mon-1c SS.2 Training Performance of Echo State Neural Networks

20 Mon-1c SS.3 Optimal adaptive Normalized Matched Filter for Large Antenna Arrays

21 Mon-1c SS.4 Linear receivers for Massive MIMO FBMC/OQAM under strong channel frequency selectivity

22 Mon-1c SS.5 On the statistical performance of MUSIC for distributed sources

23 Mon-1c SS.6 Optimization of the loading factor of regularized estimated spatial-temporal Wiener filters in large system case

24 Mon-1c SS.7 On the Eigenvalue Distribution of Column Sub-sampled Semi-unitary Matrices

25


Nº SESION TITLE

1 Mon-1a.1 Efficient Distributed Estimation of Inverse Covariance Matrices

2 Mon-1a.2 Sampling schemes and parameter estimation for nonlinear Bernoulli-Gaussian sparse models

3 Mon-1a.3 Measure transformed quasi likelihood ratio test for Bayesian binary hypothesis testing

4 Mon-1a.4 Detecting the dimension of the subspace correlated across multiple data sets in the sample poor regime

5 Mon-1a.5 Nonparametric estimation of a shot-noise process

6 Mon-1a.6 Finite sample performance of least squares estimation in sub-Gaussian noise

7 Mon-1a.7 Weighting a resampled particle in Sequential Monte Carlo

8 Mon-1a.8 Structure-Induced Complex Kalman Filter for Decentralized Sequential Bayesian Estimation

9 Mon-1a.9 An EM Algorithm for Maximum Likelihood Estimation of Barndorff-Nielsen's Generalized Hyperbolic Distribution

10 Mon-1a.10 Markov-tree Bayesian Group-sparse Modeling with Wavelets

11 Mon-1b.1 Inferring Network Properties From Fixed-Choice Design with Strong and Weak Ties

12 Mon-1b.2 Estimating Signals over Graphs via Multi-kernel Learning

13 Mon-1b.3 Network Topology Identification from Spectral Templates

14 Mon-1b.4 Bayesian Inference of Diffusion Networks with Unknown Infection Times

15 Mon-1b.5 Clustering time-varying connectivity networks by Riemannian geometry: The brain-network case

16 Mon-1b.6 Temporal Network Tracking based on Tensor Factor Analysis of Graph Signal Spectrum

17 Mon-1b.7 Multitask Diffusion LMS with Optimized Inter-Cluster Cooperation

18 Mon-1c SS.1 Robust Shrinkage M-estimators of Large Covariance Matrices

19 Mon-1c SS.2 Training Performance of Echo State Neural Networks

20 Mon-1c SS.3 Optimal adaptive Normalized Matched Filter for Large Antenna Arrays

21 Mon-1c SS.4 Linear receivers for Massive MIMO FBMC/OQAM under strong channel frequency selectivity

22 Mon-1c SS.5 On the statistical performance of MUSIC for distributed sources

23 Mon-1c SS.6 Optimization of the loading factor of regularized estimated spatial-temporal Wiener filters in large system case

24 Mon-1c SS.7 On the Eigenvalue Distribution of Column Sub-sampled Semi-unitary Matrices

Mon-1b.3

Mon-1b.6

Mon-1c SS.1

Mon-1c SS.3

Mon-1c SS.5

Mon-1c SS.7

Mon-1b.4

Mon-1b.5

Mon-1b.7

Mon-1c SS.2

Mon-1c SS.4

Mon-1c SS.6

1

3

5

8

10

2

4

6

7

9

11

12

14

16

20

22

13

15

17

19

21

23

18

24

Mon-1a.1

Mon-1a.4

Mon-1a.6

Mon-1a.8

Mon-1a.10

Mon-1b.2

Mon-1a.2

Mon-1a.3

Mon-1a.5

Mon-1a.7

Mon-1a.9

Mon-1b.1

26


Mon-II: Machine learning and pattern recognition I,Mon-II SS: Advanced robust techniques for signal processing applications,Mon-II SS: Recent advances in Monte Carlo methods for multi-dimensional signal processing and machine learning

Poster Presentation GuidelinesMonday, June 27 16:30-18:00h

Nº SESION TITLE

1 Mon-2a.1 Multi-Scale Sparse Coding with Anomaly Detection and Classification

2 Mon-2a.2 Learning Rank Reduced Mappings using Canonical Correlation Analysis

3 Mon-2a.3 Indian Buffet Process Dictionary Learning for image inpainting

4 Mon-2a.4 Online low-rank subspace learning from incomplete data using rank revealing l1 / l2 regularization

5 Mon-2a.5 Video denoising via online sparse and low-rank matrix decomposition

6 Mon-2a.6 Sparse Multivariate Factor Regression

7 Mon-2a.7 Binary stable embedding via paired comparisons

8 Mon-2a.8 Group-Sparse Subspace Clustering with Missing Data

9 Mon-2a.9 Joint segmentation of multiple images with shared classes: a Bayesian nonparametrics approach

10 Mon-2a.10 Fast Convergent Algorithms for Multi-Kernel Regression

12 Mon-2b SS.2 A robust signal subspace estimator

13 Mon-2b SS.3 The Impact of Unknown Extra Parameters on Scatter Matrix Estimation and Detection Performance in Complex t-Distributed Data

14 Mon-2b SS.4 Mean Square Error performance of sample mean and sample median estimators

15 Mon-2b SS.5 A robust estimation approach for fitting a PARMA model to real data

16 Mon-2b SS.6 Recursive Bayesian Tracking in big-data: Analysis of Estimation Accuracy with respect to Sensor Reliability

17 Mon-2b SS.7 Automatic diagonal loading for Tyler's robust covariance estimator

18 Mon-2c SS.1 NLOS Mitigation in TOA-based Indoor Localization by nonlinear filtering under Skew t-distributed measurement noise

19 Mon-2c SS.2 A Partially Collapsed Gibbs Sampler with Accelerated Convergence for EEG Source Localization

20 Mon-2c SS.3 Multiple Importance Sampling with Overlapping Sets of Proposals

21 Mon-2c SS.4 An improved SIR-based sequential Monte Carlo algorithm

22 Mon-2c SS.5 Sticky proposal densities for adaptive MCMC methods

23 Mon-2c SS.6 Sequential Monte Carlo methods under model uncertainty

24 Mon-2c SS.7 Use of Particle Filtering and MCMC for Inference in Probabilistic Acoustic Tube Model

27


Mon-2b SS.3

Mon-2b SS.6

Mon-2c SS.1

Mon-2c SS.3

Mon-2c SS.5

Mon-2c SS.7

Mon-2b SS.4

Mon-2b SS.5

Mon-2b SS.7

Mon-2c SS.2

Mon-2c SS.4

Mon-2c SS.6

Mon-2a.1

Mon-2a.4

Mon-2a.6

Mon-2a.8

Mon-2a.10

Mon-2b SS.2

Mon-2a.2

Mon-2a.3

Mon-2a.5

Mon-2a.7

Mon-2a.9

1

3

5

7

9

2

4

6

8

10

12

13

15

17

21

23

14

16

18

20

22

24

19

28


Tue-I: Array processing, radar and sonar,Tue-I: Detection and estimation theory II,Tue-I SS: Multivariate statistical signal modeling and analysis

Poster Presentation GuidelinesTuesday, June 28 10:00-11:15h

Nº SESION TITLE

1 Tue-1a.1 Online Angle of Arrival Estimation in the Presence of Mutual Coupling

2 Tue-1a.2 Robust Adaptive Subspace Detection in Impulsive Noise

3 Tue-1a.3 A sparse approach for DOA estimation with a multiple spatial invariances sensor array

4 Tue-1a.4 Improved RARE Methods for DOA Estimation in Uniform Circular Arrays with Unknown Mutual Coupling

5 Tue-1a.5 Elevation and azimuth estimation in arbitrary planar mono-static MIMO radar via tensor decomposition

6 Tue-1a.6 Comparison of Passive Radar Detectors with Noisy Reference Signal

7 Tue-1a.7 Fast Convolution Formulations for Radar Detection using LASSO

8 Tue-1a.8 SINR Analysis in Persymmetric Adaptive Processing

9 Tue-1a.9 Stochastic Resolution Analysis of Co-prime Arrays in Radar

10 Tue-1a.10 Multiple target tracking in the fully adaptive radar framework

11 Tue-1a.11 Direct Data Domain Based Adaptive Beamforming for FDA-MIMO Radar

12 Tue-1b.1 Wald-Kernel: A Method for Learning Sequential Detectors

13 Tue-1b.2 Signal Reconstruction for Multi-source Variable-rate Samples with Autocorrelated Errors in Variables

14 Tue-1b.3 Inferring High-Dimensional Poisson Autoregressive Models

15 Tue-1b.4 Generalized Legendre Transform Multifractal Formalism for Nonconcave Spectrum Estimation

16 Tue-1b.5 An Auxiliary Variable Method for Langevin based MCMC algorithms

17 Tue-1b.6 Alternative Effective Sample Size measures for Importance Sampling

18 Tue-1b.7 Robust Markov Random Field Outlier Detection and Removal in Subsampled Images

19 Tue-1b.8 Translation Invariant DWT based Denoising using Goodness of Fit Test

20 Tue-1b.9 A wavelet based likelihood ratio test for the homogeneity of Poisson processes

21 Tue-1c SS.1 Distributed Multivariate Regression with Unknown Noise Covariance in the presence of Outliers: An MDL Approach

22 Tue-1c SS.2 Optimal transport vs. Fisher-Rao distance between copulas for clustering multivariate time series

23 Tue-1c SS.3 Combining EEG source connectivity and network similarity: Application to object categorization in the human brain

24 Tue-1c SS.4 A Generalized Multivariate Logistic Model and EM Algorithm based on the Normal Variance Mean Mixture Representation

25 Tue-1c SS.5 Approximating Bayesian confidence regions in convex inverse problems

26 Tue-1c SS.6 Balanced Least Squares: Estimation in Linear Systems with Noisy Inputs and Multiple Outputs

27 Tue-1c SS.7 Multimodal Metric Learning with Local CCA

29


Tue-1b.2

Tue-1b.5

Tue-1b.7

Tue-1b.9

Tue-1c SS.2

Tue-1c SS.4

Tue-1b.3

Tue-1b.4

Tue-1b.6

Tue-1b.8

Tue-1c SS.1

Tue-1c SS.3

Tue-1c SS.5

Tue-1c SS.6

Tue-1c SS.7

Tue-1a.1

Tue-1a.4

Tue-1a.6

Tue-1a.8

Tue-1a.10

Tue-1b.1

Tue-1a.2

Tue-1a.3

Tue-1a.5

Tue-1a.7

Tue-1a.9

Tue-1a.11

1

3

5

7

9

2

4

6

8

10

12

15

17

21

23

16

18

20

22

27

19

11

1413

24

25

26

30


Tue-II: Compressed sensing,Tue-II: Signal processing over graphs and networks II,Tue-II SS: Bayesian detection and estimation techniques for radar applications,Tue-II SS: Making sense out of multi-channel physiological data for pervasive health applications

Poster Presentation GuidelineTuesday, June 28 11:45-13:00h

Nº SESION TITLE

1 Tue-2a.1 Evolutionary Spectral Graph Clustering Through Subspace Distance Measure

2 Tue-2a.2 Diffusion estimation of mixture models with local and global parameters

3 Tue-2a.3 Translation on Graphs: An Isometric Shift Operator

4 Tue-2a.4 Identifying Rumor Sources with Different Start Times

5 Tue-2a.5 Estimation of Time-Varying Mixture Models: An Application to Traffic Estimation

6 Tue-2a.6 Synchronization for classical blind source separation algorithms in wireless acoustic sensor networks

7 Tue-2a.7 A Hawkes' eye view of network information flow

8 Tue-2b.1 Oracle Performance Estimation of Bernoulli-distributed Sparse Vectors

9 Tue-2b.2 Bound on the estimation grid size for sparse reconstruction in direction of arrival estimation

10 Tue-2b.3 On Sparse Recovery Using Finite Gaussian Matrices: RIP-Based Analysis

11 Tue-2b.4 Bilevel Feature Selection in Nearly-Linear Time

12 Tue-2b.5 Sparse Error Correction with Multiple Measurement Vectors

13 Tue-2b.6 A Refined Analysis for the Sample Complexity of Adaptive Compressive Outlier Sensing

14 Tue-2b.7 Low Rank Matrix Recovery From Column-Wise Phaseless Measurements

15 Tue-2c SS.1 Weiss-Weinstein bound for an abrupt unknown frequency change

16 Tue-2c SS.2 Adaptive Waveform Design for Target Detection with Sequential Composite Hypothesis Testing

17 Tue-2c SS.3 Estimation and compensation of I/Q imbalance for FMCW radar receivers

18 Tue-2c SS.4 Improving Separating Function Estimation Tests Using Bayesian Approaches

19 Tue-2c SS.5 Velocity ambiguity mitigation of off-grid range migrating targets via Bayesian sparse recovery

20 Tue-2c SS.6 Bayesian Framework and Radar: On Misspecified Bounds and Radar-Communication Cooperation

21 Tue-2c SS.7 Detection in Multiple Channels Having Unequal Noise Power

22 Tue-2d SS.1 Accelerometer-based gait assessment: pragmatic deployment on an international scale

23 Tue-2d SS.2 Cortical Distribution of N400 Potential in Response to Semantic Priming with Visual Non-Linguistic Stimuli

24 Tue-2d SS.3 Reconstruction of Brain Activity in EEG / MEG Using Reduced-Rank Nulling Spatial Filter

25 Tue-2d SS.4 Efficient Low-Rank Spectrotemporal Decomposition using ADMM

26 Tue-2d SS.5 Dimensionality Reduction of Sample Covariance Matrices by Graph Fourier Transform for Motor Imagery Brain-Machine Interface

27 Tue-2d SS.6 Estimation of High-Dimensional Connectivity in fMRI Data via Subspace Autoregressive Models

28 Tue-2d SS.7 Hierarchical Online SSVEP Spelling Achieved With Spatiotemporal Beamforming

31


Tue-2b.6

Tue-2c SS.2

Tue-2c SS.4

Tue-2c SS.6

Tue-2d SS.1

Tue-2d SS.3

Tue-2b.7

Tue-2c SS.1

Tue-2c SS.3

Tue-2c SS.5

Tue-2c SS.7

Tue-2d SS.2

Tue-2d SS.4

Tue-2d SS.5

Tue-2d SS.6

Tue-2a.1

Tue-2a.4

Tue-2a.6

Tue-2b.1

Tue-2b.3

Tue-2b.5

Tue-2a.2

Tue-2a.3

Tue-2a.5

Tue-2a.7

Tue-2b.2

Tue-2b.4

Tue-2d SS.7

1

3

5

7

9

2

4

6

8

10

12

15

17

21

23

16

18

20

22

27

19

11

1413

24

2526

28

32


Tue-III: Machine learning and pattern recognition II,Tue-III: Signal processing for communications,Tue-III SS: Statistical signal processing and learning in smart grid

Poster Presentation GuidelinesTuesday, June 28 16:30-18:00h

Nº SESION TITLE

1 Tue-3a.1 A Finite Moving Average Test for Transient Change Detection in GNSS Signal Strength Monitoring

2 Tue-3a.2 Statistical Analysis and Optimization of FFR/SFR-aided OFDMA-based Multi-cellular Networks

3 Tue-3a.3 Analog joint source channel coding over MIMO fading channels with imperfect CSI

4 Tue-3a.4 A new approach for solving anti-jamming games in stochastic scenarios as pursuit-evasion games

5 Tue-3a.5 Weighted Sum Rate Maximization of MISO Interference Broadcast Channels via Difference of Convex Functions Programming: A Large System Analysis

6 Tue-3a.6 Recursive End-To-End Distortion Estimation for Error-Resilient Adaptive Predictive Compression Systems

7 Tue-3a.7 Study of Statistical Robust Closed Set Speaker Identification with Feature and Score-Based Fusion

8 Tue-3a.8 Analog Distributed Coding of Correlated Sources for Fading Multiple Access Channels

9 Tue-3a.9 Generalized Integration techniques for high-sensitivity GNSS receivers affected by oscillator phase noise

10 Tue-3a.10 Measurement Matrix Design For Compressive Sensing With Side Information at the Encoder

11 Tue-3b.1 Efficient KLMS and KRLS Algorithms: A Random Fourier Feature Perspective

12 Tue-3b.2 Designing Classifier Architectures using Information Theory

13 Tue-3b.3 Democratic prior for anti-sparse coding

14 Tue-3b.4 Group invariant subspace learning for outlier detection

15 Tue-3b.5 Human Authentication From Ankle Motion Data Using Convolutional Neural Networks

16 Tue-3b.6 Unsupervised segmentation of piecewise constant images from incomplete, distorted and noisy data

17 Tue-3b.7 Multilinear Subspace Clustering

18 Tue-3b.8 Order-based Generalized Multivariate Regression

19 Tue-3b.9 Hierarchical Bayesian variable selection in the probit model with mixture of nominal and ordinal responses

20 Tue-3b.10 Jeffreys Prior Regularization for Logistic Regression

21 Tue-3c SS.1 An Approximation Algorithm for future wind scenarios

22 Tue-3c SS.2 Online Learning and Pricing for Demand Response in Smart Distribution Networks

23 Tue-3c SS.3 Decentralized MMSE Attacks in Electricity Grids

24 Tue-3c SS.4 A Graphical Approach to Quickest Outage Localization in Power Grids

25 Tue-3c SS.5 Estimating Treatment Effects in Demand Response

26 Tue-3c SS.6 Dynamic Decentralized Voltage Control for Power Distribution Networks

27 Tue-3c SS.7 Learning to Infer: a New Variational Inference Approach for Power Grid Topology Identification

33


Tue-3b.2

Tue-3b.5

Tue-3b.7

Tue-3b.9

Tue-3c SS.2

Tue-3c SS.4

Tue-3b.3

Tue-3b.4

Tue-3b.6

Tue-3b.8

Tue-3c SS.1

Tue-3c SS.3

Tue-3c SS.5

Tue-3c SS.6

Tue-3c SS.7

Tue-3a.1

Tue-3a.4

Tue-3a.6

Tue-3a.8

Tue-3a.10

Tue-3b.1

Tue-3a.2

Tue-3a.3

Tue-3a.5

Tue-3a.7

Tue-3a.9

Tue-3a.11

1

3

5

7

9

2

4

6

8

10

12

15

17

21

23

16

18

20

22

27

19

11

1413

24

25

26

34


Wed: Applications (biomedical, energy, security),Wed: Detection and estimation theory III,Wed SS: Optimization and simulation for image processing

Poster Presentation GuidelinesWednesday, June 29 10:30-12:00h

Nº SESION TITLE

1 Wed-1a.1 Sparse Genomic Structural Variant Detection: Exploiting Parent-Child Relatedness for Signal Recovery

2 Wed-1a.2 Multiscale Time Irreversibility to predict Orthostatic Intolerance in Older People

3 Wed-1a.3 Pose estimation of cyclic movement using inertial sensor data

4 Wed-1a.4 Genomic Transcription Regulatory Element Location Analysis via Poisson weighted LASSO

5 Wed-1a.5 Design of Data-Injection Adversarial Attacks against Spatial Field Detectors

6 Wed-1a.6 Security of (n,n)-threshold audio secret sharing schemes encrypting audio secrets

7 Wed-1a.7 Secure Estimation Against Complex-valued Attacks

8 Wed-1a.8 A skewed exponential power distribution to measure value at risk in electricity markets

9 Wed-1a.9 Accelerometer calibration using sensor fusion with a gyroscope

10 Wed-1b.1 Joint range estimation and spectral classification for 3D scene reconstruction using multispectral Lidar waveforms

11 Wed-1b.2 Regularised Estimation of 2D-Locally Stationary Wavelet Processes

12 Wed-1b.3 Fast filtering with new sparse transition Markov chains

13 Wed-1b.4 On the estimation of many closely spaced complex sinusoids

14 Wed-1b.5 A Multiscale Approach for Tensor Denoising

15 Wed-1b.6 Two-Stage Estimation after Parameter Selection

16 Wed-1b.7 An order fitting rule for optimal subspace averaging

17 Wed-1b.8 Block-Wise MAP Inference for Determinantal Point Processes with Application to Change-Point Detection

18 Wed-1c SS.1 Spatial regularization for nonlinear unmixing of hyperspectral data with vector-valued functions

19 Wed-1c SS.2 A Regularized Sparse Approximation Method for Hyperspectral Image Classification

20 Wed-1c SS.3 Unbiased Injection of Signal-Dependent Noise in Variance-Stabilized Range

21 Wed-1c SS.4 Bayesian Multifractal Analysis of Multi-temporal Images Using Smooth Priors

22 Wed-1c SS.5 Robust hyperspectral unmixing accounting for residual components

23 Wed-1c SS.6 Analysis Dictionary Learning for Scene Classification

24 Wed-1c SS.7 Weakly-supervised Analysis Dictionary Learning with Cardinality Constraints

35


Wed-1b.4

Wed-1b.7

Wed-1c SS.1

Wed-1c SS.3

Wed-1c SS.5

Wed-1c SS.7

Wed-1b.5

Wed-1b.6

Wed-1b.8

Wed-1c SS.2

Wed-1c SS.4

Wed-1c SS.6

Wed-1a.1

Wed-1a.4

Wed-1a.6

Wed-1a.8

Wed-1b.1

Wed-1b.3

Wed-1a.2

Wed-1a.3

Wed-1a.5

Wed-1a.7

Wed-1a.9

Wed-1b.2

1

3

5

7

9

2

4

6

8

10

12

13

15

17

21

23

14

16

18

20

22

24

19

11

36


SSP 2016Detailed Technical Program

POSTER SESSION 1& SPECIAL SESSIONSMonday, June 27, 10:00 – 11:30

MON-I: DETECTION AND ESTIMATION THEORY I

Efficient Distributed Estimation of Inverse Covariance MatricesJesus Daniel Arroyo Relion and Elizabeth Hou (University of Michigan, USA)

In distributed systems, communication is a major concern due to issues such as its vulnerability or efficiency. In this paper, we are interested in estimating sparse inverse covariance matrices when samples are distributed into different machines. We ad-dress communication efficiency by proposing a method where, in a single round of communication, each machine transfers a small subset of the entries of the inverse covariance matrix. We show that, with this efficient distributed method, the error rates are comparable with estimation in a non-distributed setting, and correct model selection is still possible. Practical performance is shown through simulations.

Sampling schemes and parameter estimation for nonlinear Bernoulli-Gaussian sparse modelsMegane Boudineau (Université Toulouse III Paul Sabatier, Fran-ce); Hervé Carfantan (Université de Toulouse, UPS-OMP / CNRS, IRAP, France); Sébastien Bourguignon (Ecole Centrale de Nan-tes, IRCCyN, France); Michaël Bazot (New York University Abu Dhabi, UAE)

We address the sparse approximation problem in the case where the data are approximated by the linear combination of a small number of elementary signals, each of these signals depending non-linearly on additional parameters. Sparsity is explicitly expressed through a Bernoulli-Gaussian hierarchical model in a Bayesian framework. Posterior mean estimates are computed using Markov Chain Monte-Carlo algorithms. We generalize the partially marginalized Gibbs sampler proposed in the linear case in [1], and build an hybrid Hastings-within-Gibbs algorithm in order to account for the nonlinear parame-ters. All model parameters are then estimated in an unsupervi-sed procedure. The resulting method is evaluated on a sparse spectral analysis problem. It is shown to converge more effi-ciently than the classical joint estimation procedure, with only a slight increase of the computational cost per iteration, con-sequently reducing the global cost of the estimation procedure.

Measure transformed quasi likelihood ratio test for Ba-yesian binary hypothesis testingNir Halay and Koby Todros (Ben Gurion University of the Negev, Israel); Alfred Hero III (University of Michigan, USA)

In this paper, a generalization of the Gaussian quasi likelihood ratio test (GQLRT) for Bayesian binary hypothesis testing is de-veloped. The proposed generalization, called measure-transfor-med GQLRT (MT-GQLRT), selects a Gaussian probability model that best empirically fits a transformed conditional probability measure of the data. By judicious choice of the transform we show that, unlike the GQLRT, the proposed test is resilient to outliers and involves higher-order statistical moments leading to significant mitigation of the model mismatch effect on the decision performance. Under some mild regularity conditions

1

37


we show that the test statistic of the proposed MT-GQLRT is asymptotically normal. A data driven procedure for optimal selection of the measure transformation parameters is deve-loped that minimizes an empirical estimate of the asymptotic Bayes risk. The MT-GQLRT is applied to signal classification in a simulation example that establishes significantly improved pro-bability of error performance relative to the standard GQLRT.

Detecting the dimension of the subspace correlated across multiple data sets in the sample poor regimeTanuj Hasija (Universität Paderborn, Germany); Yang Song (Hong Kong Polytechnic University, Hong Kong); Peter J. Schreier (Universitaet Paderborn, Germany); David Ramírez (University Carlos III of Madrid, Spain)

This paper addresses the problem of detecting the number of signals correlated across multiple data sets with small sample support. While there have been studies involving two data sets, the problem with more than two data sets has been less explo-red. In this work, a rank-reduced hypothesis test for more than two data sets is presented for scenarios where the number of samples is small compared to the dimensions of the data sets.

Nonparametric estimation of a shot-noise processPaul Ilhe (Télécom ParisTech, France); Francois Roueff (Telecom ParisTech, Finland); Eric Moulines (Télécom Paris Tech, Fran-ce); Antoine Souloumiac (CEA, LIST, Laboratoire Outils pour l’Analyse de Données, France)

We propose an efficient method to estimate in a nonparametric fashion the marks’ density of a shot-noise process in presence of pileup from a sample of low-frequency observations. Based on a functional equation linking the marks’ density to the characteris-tic function of the observations and its derivative, we propose a new time-efficient method using B-splines to estimate the densi-ty of the underlying gamma-ray spectrum which is able to han-dle large datasets used in nuclear physics. A discussion on the numerical computation of the algorithm and its performances on simulated data are provided to support our findings.

Finite sample performance of least squares estimation in sub-Gaussian noiseMichael Krikheli (Bar Ilan University, Israel); Amir Leshem (Bar-Ilan University, Israel)

In this paper we analyze the finite sample performance of the least squares estimator. In contrast to standard performance analysis which uses bounds on the mean square error together with asymptotic normality, our bounds are based on large de-viation and concentration of measure results.This allows for accurate bounds on the tail of the estimator. We show the fast exponential convergence of the number of samples required to ensure accuracy with high probability. We analyze a sub-Gaussian setting with fixed or random mixing matrix of the least squares problem. We provide probability tail bounds on the L infinity norm of the error of the finite sample approximation of the true parameter. Our method is simple and uses simple analysis for L infinity type bounds of the es-timation error. The tightness of the bound is studied through simulations.

Weighting a resampled particle in Sequential Monte CarloLuca Martino (University of Helsinki, Finland); Víctor Elvira (Uni-versity Carlos III of Madrid, Spain); Francisco Louzada (Univer-sidade de São Paulo (USP), Brazil)

The Sequential Importance Resampling (SIR) method is the core of the Sequential Monte Carlo (SMC) algorithms (a.k.a., particle filters). In this work, we point out a suitable choice for weighting properly a resampled particle. This observation en-tails several theoretical and practical consequences, allowing also the design of novel sampling schemes. Specifically, we describe one theoretical result about the sequential estimation of the marginal likelihood. Moreover, we suggest a novel resam-pling procedure for SMC algorithms called partial resampling, involving only a subset of the current cloud of particles. Clearly, this scheme attenuates the additional variance in the Monte Carlo estimators generated by the use of the resampling.

Structure-Induced Complex Kalman Filter for Decentra-lized Sequential Bayesian EstimationArash Mohammadi (Concordia University, Canada); Konstanti-nos N Plataniotis (University of Toronto, Canada)

The letter considers a multi-sensor state estimation problem configured in a decentralized architecture where local complex statistics are communicated to the central processing unit for fusion instead of the raw observations. Naive adaptation of

38


the augmented complex statistics to develop a decentralized state estimation algorithm results in increased local compu-tations, and introduces extensive communication overhead, making it practically unattractive. The letter proposes a struc-ture-induced complex Kalman filter framework with reduced communication overhead. In order to further reduce the local computations, the letter proposes a non-circularity criterion which allows each node to examine the non-circularity of its local observations. A local sensor node disregards its extra second-order statistical information when the non-circularity coefficient is small. In cases where the local observations are highly non-circular, an intuitively pleasing circularization ap-proach is proposed to avoid computation and communication of the pseudo-covariance matrices. Simulation results indicate that the proposed structured-induced complex Kalman filter (SCKF) provides significant performance improvements over its traditional counterparts.

An EM Algorithm for Maximum Likelihood Estimation of Barndorff-Nielsen’s Generalized Hyperbolic Distribu-tionJason Palmer (University of California San Diego, USA); Ken Kreutz-Delgado (University of California, San Diego, USA); Scott Makeig (University of California San Diego, USA)

We present an EM algorithm for Maximum Likelihood (ML) estimation of the location, structure matrix, skew or drift, and shape parameters of Barndorff-Nielsen’s Generalized Hyperbo-lic distribution, which is the Gaussian Location Scale mixture (GLSM) (or Normal Variance Mean Mixture) with Generalized Inverse Gaussian (GIG) scale mixing distribution. We use the GLSM representation along with the closed form posterior ex-pectations possible with the GIG distribution to derive an EM algorithm for computing ML parameter estimates.

Markov-tree Bayesian Group-sparse Modeling with WaveletsGanchi Zhang and Nick Geoffrey Kingsbury (University of Cam-bridge, United Kingdom)

In this paper, we propose a new Markov-tree Bayesian mode-ling of wavelet coefficients. Based on a group-sparse GSM mo-del with 2-layer cascaded Gamma distributions for the varian-ces, the proposed method effectively exploits both intrascale

and interscale relationships across wavelet subbands. To deter-mine the posterior distribution, we apply Variational Bayesian inference with a subband adaptive majorization-minimization method to make the method tractable for large problems.

MON-I: SIGNAL PROCESSING OVER GRAPHS AND NETWORKS I

Inferring Network Properties From Fixed-Choice Design with Strong and Weak TiesNaghmeh Momeni and Michael Rabbat (McGill University, Canada)

Typically studies of networked systems begin with obtaining information about the network structure. In many settings it is impractical or impossible to directly observe the network, and sampling is used. Sampling the structure of offline social net-works is especially costly and time-consuming. Respondents are asked to name close friends and acquaintances (strong and weak ties). However, because a person may have a large number of acquaintances, surveys use a fixed-choice design, where respondents are asked to name a small, fixed number of their weak ties. Surprisingly, studies based on fixed-choice de-signs then directly use the network derived from the responses without correcting for the bias introduced by fixed-choice sam-pling. In this paper we demonstrate how to account for fixed-choice sampling when inferring network characteristics. Our approach is based on application of the generalized method of moments. We verify the accuracy of our results via simulation and discuss immediate applications and consequences of our work to existing results.

Estimating Signals over Graphs via Multi-kernel Lear-ningDaniel Romero, Meng Ma and Georgios B. Giannakis (Univer-sity of Minnesota, USA)

Estimating functions on graphs finds well-documented applica-tions in machine learning and, more recently, in signal processing. Given signal values on a subset of vertices, the goal is to estimate the signal on the remaining ones. This task amounts to estima-ting a function (or signal) over a graph. Most existing techniques either rely on parametric signal models or require costly cross-validation. Leveraging the framework of multi-kernel learning,

39


a data-driven non-parametric is developed here. Instead of a sin-gle kernel, the algorithm relies on a dictionary of candidate ker-nels and efficiently selects the most suitable ones by minimizing a convex criterion using a group Lasso module. Numerical tests demonstrate the superior estimation performance of the novel approach over competing alternatives.

Network Topology Identification from Spectral TemplatesSantiago Segarra (University of Pennsylvania, USA); Antonio G. Marques (Universidad Rey Juan Carlos, Spain); Gonzalo Mateos (University of Rochester, USA); Alejandro Ribeiro (University of Pennsylvania, USA)

Network topology inference is a cornerstone problem in statisti-cal analyses of complex systems. In this context, the fresh look advocated here permeates benefits from convex optimization and graph signal processing, to identify the so-termed graph shift operator (encoding the network topology) given only the eigenvectors of the shift. These spectral templates can be ob-tained, for example, from principal component analysis of a set of graph signals defined on the particular network. The novel idea is to find a graph shift that while being consistent with the provided spectral information; it endows the network structure with certain desired properties such as sparsity. The focus is on developing efficient recovery algorithms along with identifiabili-ty conditions for two particular shifts, the adjacency matrix and the normalized graph Laplacian. Application domains include network topology identification from steady-state signals gene-rated by a diffusion process, and design of a graph filter that facilitates the distributed implementation of a prescribed linear network operator. Numerical tests showcase the effectiveness of the proposed algorithms in recovering synthetic and struc-tural brain networks.

Bayesian Inference of Diffusion Networks with Unk-nown Infection TimesShohreh Shaghaghian and Mark Coates (McGill University, Ca-nada)

The analysis of diffusion processes in different propagation sce-narios often involve estimating variables that are not directly ob-served in real-world scenarios. These hidden variables include pa-rental relationships, strength of connections between nodes, and the moment of time that the infection happens for each node.

In this paper, we propose a framework in which all the three sets of parameters are assumed to be hidden and develop a Bayesian approach to infer them. After justifying the model assumptions, we evaluate the performance efficiency of our proposed approach through numerical simulations on datasets from synthetic and real-world diffusion processes.

Clustering time-varying connectivity networks by Rie-mannian geometry: The brain-network caseKonstantinos Slavakis, Shiva Salsabilian, David Wack and Sarah Muldoon (University at Buffalo (SUNY), USA)

In response to the demand on data-analytic tools that monitor time-varying connectivity patterns within brain networks, the present paper introduces a framework for clustering (unsuper-vised learning) of dynamically evolving connectivity states of networks. This work advocates learning of network dynamics on Riemannian manifolds, capitalizing on the well-known fact that popular features in statistics enjoy that structure: (Partial) correlations or covariances can be mapped to the manifold of positive (semi-)definite symmetric matrices, while low-rank linear subspaces can be considered as points of the Grassman-nian. Sequences of such features, collected over time and across a network, are mapped to sequences of points on a Riemannian manifold, and a sequence that corresponds to a specific state of the network forms a cluster or submanifold. Geometry is exploited in a novel way to demonstrate the rich potential of the proposed learning method for monitoring ti-me-varying network patterns by outperforming state-of-the-art techniques on synthetic brain-network data.

Temporal Network Tracking based on Tensor Factor Analysis of Graph Signal SpectrumMarisel Villafañe-Delgado (Michigan State University, USA); Se-lin Aviyente (Electrical and Computer Engineering, Michigan State University, MI, USA)

Wide varieties of networks, ranging from biological to social, evolve, adapt and change over time. Recent methods emplo-yed in the assessment of temporal networks include tracking topological graph metrics, evolutionary clustering, tensor based anomaly methods and, more recently, graph to signal transformations. In this paper, we propose to assess the tem-poral evolution of networks by first transforming networks into

40


signals through Classical Multidimensional Scaling based on the resistance distance and then constructing a tensor ba-sed on the spectra of each signal across time. The proposed method is first evaluated on simulated temporal networks with varying structural properties. Next, the method is applied to temporal functional connectivity networks constructed from multichannel electroencephalogram (EEG) data collected du-ring a study of cognitive control. This analysis shows that the proposed method is more sensitive to changes in the network structure and more robust to variations in edge weights.

Multitask Diffusion LMS with Optimized Inter-Cluster CooperationYuan Wang and Wee Peng Tay (Nanyang Technological Univer-sity, Singapore); Hu Wuhua (Institute for Infocomm Research, Singapore)

We consider a multitask network where nodes are divided into several connected clusters, with each cluster performing a least mean squares estimation of a different random parameter vec-tor. Inspired by the adapt-then-combine strategy, we propose a multitask diffusion strategy whose mean and mean-square sta-bility can be achieved independent of the inter-cluster coopera-tion weights. We develop a distributed optimization algorithm that allows each node in the network to locally optimize its inter-cluster cooperation weights. Simulation results demons-trate that our approach leads to a lower average steady-state network MSD, compared with the multitask diffusion strategy using an averaging rule for the inter-cluster cooperation.

MON-I SS: RANDOM MATRICES IN SIGNAL PRO-CESSING AND MACHINE LEARNING

Robust Shrinkage M-estimators of Large Covariance MatricesNicolas Auguin, David Morales and Matthew R McKay (Hong Kong University of Science and Technology, Hong Kong); Ro-main Couillet (CentraleSupélec, France)

Robust high dimensional covariance estimators are conside-red, comprising regularized (linear shrinkage) modifications of Maronna’s classical M-estimators. Such estimators aim to pro-vide robustness to outliers, while simultaneously giving well-

defined solutions under high dimensional scenarios where the number of samples does not exceed the number of variables. By applying tools from random matrix theory, we characterize the asymptotic performance of such estimators when the num-ber of samples and variables grow large together. In particular, our results show that, when outliers are absent, many estima-tors of the shrinkage-Maronna type share the same asymptotic performance, and for such estimators we present a data-driven method for choosing the asymptotically optimal shrinkage pa-rameter. Although our results assume an outlier-free scenario, simulations suggest that certain estimators perform substan-tially better than others when subjected to outlier samples.

Training Performance of Echo State Neural NetworksRomain Couillet (CentraleSupélec, France); Gilles Wainrib (ENS Ulm, Paris, France); Harry Sevi (ENS Lyon, Lyon, France); Hafiz Tiomoko Ali (CentraleSupélec, Gif-sur-Yvette, France)

This article proposes a first theoretical performance analysis of the training phase of large dimensional linear echo-state net-works. This analysis is based on advanced methods of random matrix theory. The results provide some new insights on the core features of such networks, thereby helping the practitio-ner when using them.

Optimal adaptive Normalized Matched Filter for Large Antenna ArraysAbla Kammoun (Kaust, Saudi Arabia); Romain Couillet and Fre-deric Pascal (CentraleSupélec, France); Mohamed-Slim Alouini (King Abdullah University of Science and Technology (KAUST), Saudi Arabia)

This paper focuses on the problem of detecting a target in the presence of a compound Gaussian clutter with unknown statis-tics. To this end, we focus on the design of the adaptive norma-lized matched filter (ANMF) detector which uses the regularized Tyler estimator (RTE) built from N -dimensional observations x 1 , • • • , x n in order to estimate the clutter covariance matrix. The choice for the RTE is motivated by its possessing two major attri-butes: first its resilience to the presence of outliers, and second its regularization parameter that makes it more suitable to hand-le the scarcity in observations. In order to facilitate the design of the ANMF detector, we consider the regime in which n and N are both large. This allows us to derive closed-form expressions for

41


the asymptotic false alarm and detection probabilities. Based on these expressions, we propose an asymptotically optimal setting for the regularization parameter of the RTE that maximizes the asymptotic detection probability while keeping the asymptotic false alarm probability below a certain threshold. Numerical re-sults are provided in order to illustrate the gain of the proposed detector over a recently proposed setting of the regularization parameter.

Linear receivers for Massive MIMO FBMC/OQAM under strong channel frequency selectivityXavier Mestre (CTTC, Spain); François Rottenberg (Université Catholique de Louvain, Belgium); Monica Navarro (Centre Tec-nològic de Telecomunicacions de Catalunya (CTTC), Spain)

Filterbank Multicarrier (FBMC) modulations based on OQAM (FBMC/OQAM) have become a promising alternative to con-ventional OFDM because of their higher spectral efficiency and their improved selectivity in the frequency domain. Unfortuna-tely, the orthogonality of these modulations is lost when the channel presents strong frequency selectivity, meaning that it cannot be approximated as frequency flat within each subca-rrier bandwidth. In this paper, this effect is analyzed in the massive MIMO setting, whereby the number of transmit and receive antennas is asymptotically large (but not as large as the number of subcarriers). It is formally shown that, under these asymptotic conditions, the output mean squared error (MSE) at each subcarrier converges to a constant independent of the subcarrier index. This was previously referred to as “self-equa-lization” principle in the FBMC/OQAM literature. It is demons-trated here that this phenomenon is a direct consequence of channel hardening effect in large scale MIMO configurations.

On the statistical performance of MUSIC for distributed sourcesOuiame Najim (University of Bordeaux, France); Pascal Vallet (Bordeaux INP & IMS, France); Guillaume Ferré (University of Bordeaux, France); Xavier Mestre (CTTC, Spain)

This paper addresses the statistical behaviour of the MUSIC method for DoA estimation, in a scenario where each source signal direct path is disturbed by a clutter spreading in an an-gular neighborhood around the source DoA. In this scenario, it is well-known that subspace methods performance suffers

from an additional clutter subspace, which breaks the orthogo-nality between the source steering vectors and noise subspace. To perform a statistical analysis of the MUSIC DoA estimates, we consider an asymptotic regime in which both the number of sensors and the sample size tend to infinity at the same rate, and rely on classical random matrix theory results. We establish the consistency of the MUSIC estimates and provide numerical results illustrating their performance in this nons-tandard scenario.

Optimization of the loading factor of regularized esti-mated spatial-temporal Wiener filters in large system caseGia-Thuy Pham (Université Paris Est Marne-La-Vallée, France); Philippe Loubaton (Université de Marne La Vallée, France)

In this paper, it is established that the signal to interferen-ce plus noise ratio (SINR) produced by a trained regularized Wiener spatio-temporal filter can be estimated consistently in the asymptotic regime where the number of receivers and the number of snapshots converge to infinity at the same rate. The optimal regularization parameter is estimated as the argument of the maximum of the estimated SINR. Numerical simulations show that the proposed optimum regularized Wiener filter outperforms the existing regularized spatio-temporal Wiener filters.

On the Eigenvalue Distribution of Column Sub-sampled Semi-unitary MatricesRaviv Raich and Jinsub Kim (Oregon State University, USA)

Random matrix theory is applied in areas of signal processing, communications, and machine learning. One aspect of random matrix theory involves the study of the eigenvalues of random matrices. For example, in communications, the eigenvalues associated with a channel matrix are used in the analysis of channel capacity and in compressive sensing the eigenvalues of sub-matrices of the sampling matrix can be used to study its restricted isometry property. This paper focuses on a theo-retical analysis of the limiting empirical spectral distribution of column sub-sampled semi-unitary matrices. A key contribution of this paper is a closed-form expression for the limiting empi-rical spectral distribution and the identification of the sufficient conditions for this result.

42


POSTER SESSION 2& SPECIAL SESSIONSMonday, June 27, 16:30 - 18:00

MON-II: MACHINE LEARNING AND PATTERN RE-COGNITION I

Multi-Scale Sparse Coding with Anomaly Detection and ClassificationHojjat Akhondi-Asl and James D B Nelson (University College London, United Kingdom)

We here place a recent joint anomaly detection and classifica-tion approach based on sparse error coding methodology into multi-scale wavelet basis framework. The model is extended to incorporate an overcomplete wavelet basis into the dictionary matrix whereupon anomalies at specified multiple levels of sca-le are afforded equal importance. This enables, for example, subtle transient anomalies at finer scales to be detected which would otherwise be drowned out by coarser details and missed by the standard sparse coding techniques. Anomaly detection in power networks provides a motivating application and tests on a real-world data set corroborate the efficacy of the pro-posed model.

Learning Rank Reduced Mappings using Canonical Co-rrelation AnalysisChristian Conrad (Goethe University, Germany); Rudolf Mester (Goethe University, Frankfurt & VSI Lab, Germany)

Correspondence relations between different views of the same scene can be learnt in an unsupervised manner. We address autonomous learning of arbitrary fixed spatial (point-to-point) mappings. Since any such transformation can be represented by a permutation matrix, the signal model is a linear one, whereas the proposed analysis method, mainly based on Canonical Correlation Analysis (CCA) is based on a generalized eigensystem problem, i.e. a nonlinear operation. The learnt transformation is represented implicitly in terms of pairs of learned basis vectors and does neither use nor requi-re an analytic / parametric expression for the latent mapping. We show how the rank of the signal that is shared among views may be determined from canonical correlations and

how the overlapping (=shared) dimensions among the views may be inferred.

Indian Buffet Process Dictionary Learning for image inpaintingHong Phuong Dang (Centrale Lille, University of Lille, CRIStAL CNRS, France); Pierre Chainais (Ecole Centrale Lille & CRIStAL CNRS, France)

Ill-posed inverse problems call for adapted models to define relevant solutions. Dictionary learning for sparse represen-tation is often an efficient approach. In many methods, the size of the dictionary is fixed in advance and the noise level as well as regularization parameters need some tuning. Indian Buffet process dictionary learning (IBP-DL) is a Bayesian non parametric approach which permits to learn a dictionary with an adapted number of atoms. The noise and sparsity levels are also inferred so that the proposed approach is really non parametric: no parameters tuning is needed. This work adapts IBP-DL to the problem of image inpainting by proposing an accelerated collapsed Gibbs sampler. Experimental results illus-trate the relevance of this approach.

Online low-rank subspace learning from incomplete data using rank revealing l2 / l1 regularizationParis Giampouras, Athanasios A. Rontogiannis and Konstan-tinos Koutroumbas (National Observatory of Athens, Greece)

Massive amounts of data (also called big data) generated by a wealth of sources such as social networks, satellite sensors etc., necessitate the deployment of efficient processing tools. In this context, online subspace learning algorithms that aim at retrieving low-rank representations of data constitute a mains-tay in many applications. Working with incomplete (partially observed) datums has recently become commonplace. Mo-reover, the knowledge of the real rank of the sought subspace is rarely at our disposal {\it a priori}. Herein, a novel low-rank subspace learning algorithm from incomplete data is presen-ted. Its main premise is the online processing of incomplete da-tums along with the imposition of low-rankness on the sought subspace via a sophisticated utilization of the group sparsity inducing l2 / l1 norm. As is experimentally shown, the resulting scheme is efficient in accurately learning the subspace as well as in unveiling its real rank.

2

43


Video denoising via online sparse and low-rank matrix decompositionHan Guo and Namrata Vaswani (Iowa State University, USA)

Video denoising refers to the problem of removing “noise” from a video sequence. Here the term “noise” is used in a broad sense to refer to any corruption or outlier or interference that is not the quantity of interest. In this work, we develop a novel approach to video denoising that is based on the idea that most noisy or corrupted videos can be split into two parts – the approximate “low-rank” layer and the “sparse layer”. We first splitting the given video into these two layers, and then apply an existing state-of-the-art denoising algorithm on each layer. We show, using extensive experiments, that our denoising ap-proach outperforms the state-of-the art denoising algorithms.

Sparse Multivariate Factor RegressionMilad Kharratzadeh and Mark Coates (McGill University, Ca-nada)

We introduce a sparse multivariate regression algorithm which simultaneously performs dimensionality reduction and para-meter estimation. We decompose the coefficient matrix into two sparse matrices: a long matrix mapping the predictors to a set of factors and a wide matrix estimating the responses from the factors. We impose an elastic net penalty on the former and an `1 penalty on the latter. Our algorithm simultaneously performs dimension reduction and coefficient estimation and automatically estimates the number of latent factors from the data. Our formulation results in a non-convex optimization problem, which despite its flexibility to impose effective low-dimensional structure, is difficult, or even impossible, to solve exactly in a reasonable time. We specify a greedy optimization algorithm based on alternating minimization to solve this non-convex problem and provide theoretical results on its conver-gence and optimality. Finally, we demonstrate the effectiveness of our algorithm via experiments on simulated and real data.

Binary stable embedding via paired comparisonsAndrew Massimino and Mark Davenport (Georgia Institute of Technology, USA)

Suppose that we wish to estimate a vector x from a set of binary paired comparisons of the form “x is closer to p than

to q” for various choices of vectors p and q . The problem of estimating x from this type of observation arises in a variety of contexts, including nonmetric multidimensional scaling, “un-folding,” and ranking problems, often because it provides a powerful and flexible model of preference. The main contribu-tion of this paper is to show that under a randomized model for p and q, a suitable number of binary paired comparisons yield a stable embedding of the space of target vectors.

Group-Sparse Subspace Clustering with Missing DataDaniel L Pimentel-Alarcon (University of Wisconsin-Madison, USA); Laura Balzano (University of Michigan, USA); Roummel Marcia (University of California, Merced, USA); Rob Nowak (University of Wisconsin, Madison, USA); Rebecca Willett (Uni-versity of Wisconsin-Madison, USA)

This paper explores algorithms for subspace clustering with missing data. In many high-dimensional data analysis set-tings, data points lie in or near a union of subspaces. Subs-pace clustering is the process of estimating these subspaces and assigning each data point to one of them. However, in many modern applications the data are severely corrupted by missing values. This paper describes two novel methods for subspace clustering with missing data: (a) group-sparse subs-pace clustering (GSSC), which is based on group-sparsity and alternating minimization, and (b) mixture subspace clustering (MSC), which models each data point as a convex combination of its projections onto all subspaces in the union. Both of these algorithms are shown to converge to a local minimum, and experimental results show that they outperform the previous state-of-the-art, with GSSC yielding the highest overall clus-tering accuracy.

Joint segmentation of multiple images with shared classes: a Bayesian nonparametrics approachJessica Sodjo (University of Bordeaux, France); Audrey Giremus (Université de Bordeaux, France); François Caron (University of Oxford, United Kingdom); Jean-François Giovannelli (IMS, UMR CNRS 52 18, Université Bordeaux 1, France); Nicolas Dobigeon (University of Toulouse, France)

A combination of the hierarchical Dirichlet process (HDP) and the Potts model is proposed for the joint segmentation/clas-sification of a set of images with shared classes. Images are

44


first divided into homogeneous regions that are assumed to belong to the same class when sharing common characteris-tics. Simultaneously, the Potts model favors configurations de-fined by neighboring pixels belonging to the same class. This HDP-Potts model is elected as a prior for the images, which allows the best number of classes to be selected automatically. A Gibbs sampler is then designed to approximate the Bayesian estimators, under a maximum a posteriori (MAP) paradigm. Preliminary experimental results are finally reported using a set of synthetic images.

Fast Convergent Algorithms for Multi-Kernel RegressionLiang Zhang, Daniel Romero and Georgios B. Giannakis (Uni-versity of Minnesota, USA)

Kernel ridge regression plays a central role in various signal processing and machine learning applications. Suitable kernels are often chosen as linear combinations of “basis kernels” by optimizing criteria under regularization constraints. Although such approaches offer reliable generalization performance, sol-ving the associated min-max optimization problems face major challenges, especially with big data inputs. After analyzing the key properties of a convex reformulation, the present paper introduces an efficient algorithm based on a generalization of Nesterov’s acceleration method, which achieves order-optimal convergence rate among first-order methods. Closed-form up-dates are derived for common regularizers. Experiments on real datasets corroborate considerable speedup advantages over competing algorithms.

MON-II SS: ADVANCED ROBUST TECHNIQUES FOR SIGNAL PROCESSING APPLICATIONS

A robust signal subspace estimatorArnaud Breloy (LEME – Université Paris X, France); Ying Sun and Prabhu Babu (HKUST, Hong Kong); Guillaume Ginolhac (Univer-site de Savoie & LISTIC, France); Daniel P Palomar (Hong Kong University of Science and Technology, Hong Kong); Frederic Pascal (CentraleSupélec, France)

An original estimator of the orthogonal projector onto the signal subspace is proposed. This estimator is derived as the maximum likelihood estimator for a model of sources plus

orthogonal outliers, both with varying power (modeled by Compound Gaussians process), embedded in a white Gaussian noise. Validity and interest – in terms of performance and ro-bustness – of this estimator is illustrated through simulation results on a low rank STAP filtering application.

The impact of unknown extra parameters on scatter matrix estimation and detection performance in com-plex t-distributed data Stefano Fortunati, Maria S. Greco and Fulvio Gini (University of Pisa, Italy)

Scatter matrix estimation and hypothesis testing in Complex Elliptically Symmetric (CES) distributions often relies on the knowledge of additional parameters characterizing the distri-bution at hand. In this paper, we investigate the performance of optimal estimation and detection algorithms exploiting low-complexity but suboptimal estimates of the extra parameters under the assumption of t-distributed data. Their performance is also compared with that of robust algorithms, which do not rely on such estimates.

Mean Square Error performance of sample mean and sample median estimatorsAdrià Gusi-Amigó and Pau Closas (Centre Tecnològic de Tele-comunicacions de Catalunya (CTTC), Spain); Luc Vandendorpe (Université catholique de Louvain, Belgium)

Based on the Ziv-Zakai methodology to bound estimators, we derived an estimation bound able to predict the mean square error degradation due to model mismatches. In this article, we build upon this result to provide a performance comparison bet-ween mean and median estimators in the presence of outliers. The latter is well known to be statistically more robust than the mean in the presence of outliers. Here we show this superiority by comparing their theoretical error bounds. Analytical results are obtained, which are validated by computer simulations.

A robust estimation approach for fitting a PARMA mo-del to real dataAlessandro Jose Queiroz Sarnaglia (Federal University of Espiri-to Santo, Brazil); Valderio Anselmo Reisen (Federal University of Espírito Santo, Brazil); Pascal Bondon (LSS CNRS, France); Céline Lévy-leduc (AgroParisTech, France)

45


This paper proposes an estimation approach of the Whittle estimator to fit periodic autoregressive moving average (PAR-MA) models when the process is contaminated with additive outliers and/or has heavy-tailed noise. It is derived by repla-cing the ordinary Fourier transform with the non-linear M-regression estimator in the harmonic regression equation that leads to the classical periodogram. A Monte Carlo experiment is conducted to study the finite sample size of the proposed estimator under the scenarios of contaminated and non-con-taminated series. The proposed estimation method is applied to fit a PARMA model to the sulfur dioxide (SO2) daily average pollutant concentrations in the city of Vitória (ES), Brazil.

Recursive Bayesian Tracking in big-data: Analysis of Estimation Accuracy with respect to Sensor ReliabilityThanuka Wickramarathne (University of Notre Dame & University of Miami, USA); Haiqiao Zhang (University of Notre Dame, USA)

Recursive bayesian filtering methods provide a well-understood class of techniques for `tracking’ the behavior of a dynamic system that is observed via a sequence of noisy measurements. While tracking performance is defined by the estimation ac-curacy, the latter in turn depends heavily on what’s being measured by the sensors and how accurately these measure-ments are modeled. In particular, with statistical signal proces-sing community taking an interest in big-data and potential application of recursive bayesian tracking methods therein, one must clearly understand the ramifications of using `non-ideal’ sensors on estimation accuracy. The existing approach to characterizing the system states and observations solely via a measurement model may turn out to be inadequate in the-se applications, mainly due to the difficulties associated with capturing highly uncertain, imperfect and subject nature of these environments. By deriving from first principles to include explicit sensor reliability terms into estimation equations, we explore the impact of non-ideal sensors on estimation accura-cy. Multiple non-ideal sensor case is also explore. A numerical example is utilized for illustration of results.

Automatic diagonal loading for Tyler’s robust covarian-ce estimatorTeng Zhang (University of Central Florida, USA); Ami Wiesel (Hebrew University in Jerusalem, Israel)

An approach of regularizing Tyler’s robust M-estimator of the covariance matrix is proposed. We also provide an automatic choice of the regularization parameter in the high-dimensional regime. Simulations show its advantage over the sample cova-riance estimator and Tyler’s M-estimator when data is heavy-tailed and the number of samples is small. Compared with the previous approaches of regularizing Tyler’s M-estimator, our approach has a similar performance and a much simpler way of choosing the regularization parameter automatically.

MON-II SS: RECENT ADVANCES IN MONTE CAR-LO METHODS FOR MULTI-DIMENSIONAL SIGNAL PROCESSING AND MACHINE LEARNING

NLOS Mitigation in TOA-based Indoor Localization by nonlinear filtering under Skew t-distributed measure-ment noisePau Closas (Centre Tecnològic de Telecomunicacions de Cata-lunya (CTTC), Spain); Jordi Vilà-Valls (Universitat Politècnica de Catalunya (UPC), Spain)

Wireless localization by time-of-arrival (TOA) measurements is typically corrupted by non-line-of-sight (NLOS) conditions, causing biased range measurements that can degrade the ove-rall positioning performance of the system. In this article, we propose a localization algorithm that is able to mitigate the impact of NLOS observations by employing a heavy-tailed noi-se statistical model. Modeling the observation noise by a skew t-distribution allows us to, on the one hand, employ a compu-tationally light sigma-point Kalman filtering method while, on the other hand, be able to effectively characterize the positive skewed non-Gaussian nature of TOA observations under LOS/NLOS conditions. Numerical results show the enhanced perfor-mance of such approach.

A Partially Collapsed Gibbs Sampler with Accelerated Convergence for EEG Source LocalizationFacundo Costa (University of Toulouse & ENSEEIHT, France); Hadj Batatia and Thomas Oberlin (University of Toulouse, France); Jean-Yves Tourneret (University of Toulouse & ENSEEIHT, France)

This paper addresses the problem of designing efficient sampling moves in order to accelerate the convergence of MCMC methods.

46


The Partially collapsed Gibbs sampler (PCGS) takes advantage of variable reordering, marginalization and trimming to accelera-te the convergence of the traditional Gibbs sampler. This work studies two specific moves which allow the convergence of the PCGS to be further improved. It considers a Bayesian model whe-re structured sparsity is enforced using a multivariate Bernoulli Laplacian prior. The posterior distribution associated with this model depends on mixed discrete and continuous random vec-tors. Due to the discrete part of the posterior, the conventional PCGS gets easily stuck around local maxima. Two Metropolis-Hastings moves based on multiple dipole random shifts and inter-chain proposals are proposed to overcome this problem. The resulting PCGS is applied to EEG source localization. Experi-ments conducted with synthetic data illustrate the effectiveness of this PCGS with accelerated convergence.

Multiple Importance Sampling with Overlapping Sets of ProposalsVíctor Elvira (University Carlos III of Madrid, Spain); Luca Mar-tino (University of Helsinki, Finland); David Luengo (Universidad Politecnica de Madrid (UPM), Spain); Monica F. Bugallo (Stony Brook University, USA)

In this paper, we introduce multiple importance sampling (MIS) approaches with overlapping (i.e., non-disjoint) sets of proposals. We derive a novel weighting scheme, based on the deterministic mixture methodology, that leads to unbiased es-timators. The proposed framework can be seen as a generali-zation of other well-known MIS algorithms available in the li-terature. Furthermore, it allows to achieve any desired trade-off between the variance of the estimators and the computational complexity through the definition of the sets of proposals. Pre-liminary numerical results on a bimodal target density show the good performance of the proposed approach.

An improved SIR-based sequential Monte Carlo algo-rithmRoland Lamberti (Télécom SudParis, France); Yohan Petetin (Telecom SudParis); François Septier (Telecom Lille, Univ Lille, CNRS – CRIStAL, France); François Desbouvries (Telecom Su-dParis, France)

Sequential Monte Carlo (SMC) algorithms are based on impor-tance sampling (IS) techniques. Resampling has been intro-

duced as a tool for fighting the weight degeneracy problem. However, for a fixed sample size N, the resampled particles are dependent, are not drawn exactly from the target distribution, nor are weighted properly. In this paper, we revisit the resam-pling mechanism and propose a scheme where the resampled particles are (conditionally) independent and weighted pro-perly. We validate our results via simulations.

Sticky proposal densities for adaptive MCMC methodsLuca Martino (University of Helsinki, Finland); Roberto Casarin (University Ca’ Foscari of Venice, Italy); David Luengo (Universi-dad Politecnica de Madrid (UPM), Spain)

Monte Carlo (MC) methods are commonly used in Bayesian signal processing to address complex inference problems. The performance of any MC scheme depends on the similarity bet-ween the proposal (chosen by the user) and the target (which depends on the problem). In order to address this issue, many adaptive MC approaches have been developed to construct the proposal density iteratively. In this paper, we focus on adapti-ve Markov chain MC (MCMC) algorithms, introducing a novel class of adaptive proposal functions that progressively “stick” to the target. This proposed class of sticky MCMC methods converge very fast to the target, thus being able to generate virtually independent samples after a few iterations. Numerical simulations illustrate the excellent performance of the sticky proposals when compared to other adaptive and non-adaptive schemes.

Sequential Monte Carlo methods under model uncer-taintyIñigo Urteaga, Monica F. Bugallo and Petar M. Djuric’ (Stony Brook University, USA)

We propose a Sequential Monte Carlo (SMC) method for filtering and prediction of time-varying signals under mo-del uncertainty. Instead of resorting to model selection, we fuse the information from the considered models within the proposed SMC method. We achieve our goal by dynamically adjusting the resampling step according to the posterior pre-dictive power of each model, which is updated sequentially as we observe more data. The method allows the models with better predictive powers to explore the state space with more resources than models lacking predictive power.

47


3

This is done autonomously and dynamically within the SMC method. We show the validity of the presented method by eva-luating it on an illustrative application.

Use of Particle Filtering and MCMC for Inference in Pro-babilistic Acoustic Tube ModelRuobai Wang (Tsinghua University, P.R. China); Yang Zhang (University of Illinois, Urbana-Champaign, USA); Zhijian Ou (Tsinghua University, P.R. China); Mark Hasegawa-Johnson (Uni-versity of Illinois, USA)

Speech modeling has a wide range of applications in speech processing. Probabilistic Acoustic Tube (PAT) is a probabilistic generative model for speech which has a potential advantage in many speech processing tasks. In this paper, we model AM-FM effect in voiced source via autoregressive process. Based on Auxiliary Particle Filtering (APF) and Taylor expansion assisted Markov Chain Monte Carlo (MCMC), we successfully develop an effective inference algorithm for such improved and compli-cated model, which produces satisfactory performance in our experiments.

POSTER SESSION 3& SPECIAL SESSIONSTuesday June 28, 10:00 - 11:15

TUE-I: ARRAY PROCESSING, RADAR AND SONAR

Online Angle of Arrival Estimation in the Presence of Mutual CouplingAhmad Bazzi (EURECOM & RivieraWaves-A CEVA COMPANY, France); Dirk Slock (EURECOM, France); Lisa Meilhac (CEVA-RivieraWaves, France)

A novel algorithm for estimating the Angles of Arrival (AoA) of multiple sources in the presence of mutual coupling is derived. We first formulate an “Equality Constrained Quadratic Optimi-sation” problem, then derive a suitable MUSIC-like algorithm to solve the aforementioned problem, and thus obtain good estimates of the AoA parameters. Identifiability conditions of the proposed algorithm are also derived. Finally, simulation results demonstrate the Root-Mean-Square Error (RMSE) per-

formance of the algorithm as a function of Signal-to-Noise Ratio (SNR) and number of snapshots, with comparison to an existing method.

Robust Adaptive Subspace Detection in Impulsive NoiseIsmail Ben Atitallah (KAUST, Saudi Arabia); Abla Kammoun (Kaust, Saudi Arabia); Mohamed-Slim Alouini (King Abdullah University of Science and Technology (KAUST), Saudi Arabia); Tareq Y. Al-Naffouri (King Abdullah University of Science and Technology, USA)

This paper addresses the design of the Adaptive Subspace Matched Filter (ASMF) detector in the presence of compound Gaussian clutters and a mismatch in the array steering vector. In particular, we consider the case wherein the ASMF uses the regularized Tyler estimator (RTE) to estimate the clutter cova-riance matrix. Under this setting, a major question that needs to be addressed concerns the setting of the threshold and the regularization parameter. To answer this question, we consider the regime in which the number observations used to estima-te the RTE and their dimensions grow large together. Recent results from random matrix theory are then used in order to approximate the false alarm and detection probabilities by deterministic quantities. The latter are optimized in order to maximize an upper bound on the asymptotic detection pro-bability while keeping the asymptotic false alarm probability at a fixed rate.

A sparse approach for DOA estimation with a multiple spatial invariances sensor arrayMarc-Abel Bisch and Sebastian Miron (CRAN, Université de Lorraine, CNRS, France); David Brie (CRAN, Nancy Université, CNRS, France)

In this paper, we introduce a sparse direction-of-arrival (DOA) estimation algorithm for sensor arrays presenting multiple sca-les of spatial invariance. We exploit the Khatri-Rao structure of the over-complete steering vector dictionary, corresponding to this array geometry, in order to devise a computationally effi-cient sparse estimation approach. This approach is based on a iterative refinement and pruning strategy of the dictionary. We show, in numerical simulations, that our approach outperforms the state-of-the art approach based on a Candecomp/Parafac (CP) decomposition, proposed by Miron et al. in 2015.

48


Improved RARE Methods for DOA Estimation in Uni-form Circular Arrays with Unknown Mutual CouplingShu Cai (Nanjing University of Posts and Telecommunications, P.R. China)

Mutual coupling between array elements is known to seriously degrade the direction of arrival estimation performance of most super-resolution techniques. The rank-reduction (RARE) methods, which are variations of the MUSIC algorithm, can re-duce these performance loss. But their spatial spectra may still be disturbed by the spurious peaks caused by unknown mutual coupling. In this work, we try to mitigate the influence of these peaks to the self-calibration performance of RARE methods. According to our analysis, the spurious peaks can be divided into predictable and unpredictable sets, and the former can be identified and located by utilizing the special structure of mutual coupling matrix. Based on this, three different methods are proposed to improve existing RARE methods. The validity and effectiveness of our analysis is demonstrated via numerical experiments.

Elevation and azimuth estimation in arbitrary planar mono-static MIMO radar via tensor decompositionMing-Yang Cao (Harbin Institute of Technology, P.R. China); Sergiy A. Vorobyov (Aalto University, Finland); Xingpeng Mao (Harbin Institute of Technology, P.R. China)

Elevation and azimuth estimation in arbitrary planar monos-tatic multiple-input-multiple-output (MIMO) radar via tensor decomposition is proposed. Transmit beamspace design is used to map a planar array into a Khatri-Rao product of two perpendicular desired uniform linear arrays within a range bin of interest, and to suppress sidelobes outside of the range bin of interest. Each desired array has rotational invariance pro-perty with respect to elevation and azimuth separately. Then the received MIMO radar data are folded into a fourth-order tensor along each spatial and temporal dimension. A compu-tationally efficient tensor-based target localization method is proposed. Our simulation results demonstrate the effectiveness of the proposed method and its superiority over the matrix-based counterpart.

Comparison of Passive Radar Detectors with Noisy Re-ference SignalSandeep Gogineni (Wright State University, USA); Pawan Set-lur (Wright State University & Wright State Research Institute, USA); Muralidhar Rangaswamy (AFRL, USA); Raj Rao Nadakudi-ti (University of Michigan, USA)

Traditional passive radar systems with a noisy reference sig-nal use the cross-correlation statistic for detection. However, owing to the composite nature of this hypothesis testing problem, no claims can be made about the optimality of this detector. Therefore, exploiting the low-rank structure of most passive radar illuminators, we recently proposed singular value decomposition based detectors that outperform the CC detec-tor. In this paper, we derive the generalized likelihood ratio tests for this signal model and compare with our proposed SVD based detectors. We demonstrate the near CFAR behavior (highly desirable) of our SVD detectors. We show that on the other hand, the GLRT detectors have a varying probability of false alarm with changing reference channel characteristics making it impractical to use them in a passive radar system.

Fast Convolution Formulations for Radar Detection using LASSOZeyi Lee, Yanjun Zhan and Han Lun Yap (DSO National Labo-ratories, Singapore); Radmila Pribic’ (Thales Nederland BV Delft, The Netherlands)

Sparse reconstruction has recently been shown to perform bet-ter than conventional detection methods in multi-target sce-narios. However, algorithms performing sparse reconstruction have computational complexity that far exceeds that of con-ventional detection techniques. To bridge this computational gap, we present three methods of exploiting fast operations in the convolution transform model for the detection problem using LASSO. We empirically show that these methods do not compromise on the statistical power provided by LASSO.

SINR Analysis in Persymmetric Adaptive ProcessingJun Liu (National Laboratory of Radar Signal Processing & Xidian University, P.R. China); Liu Hongwei and Bo Chen (Xidian Uni-versity, P.R. China); Xiang-Gen Xia (University of Delaware, USA)

49


We study the normalized output signal-to-interference-plus-noise ratio (SINR) of a sample matrix inversion (SMI) bea-mformer with exploiting a priori information on persymmetric structures in the received signal. An exact expression for the expectation of the normalized output SINR (i.e., average SINR loss) of the persymmetric SMI beamformer is obtained. Simu-lation results reveal that the exploitation of the persymmetric structure is equivalent to doubling the amount of training data.

Stochastic Resolution Analysis of Co-prime Arrays in RadarRadmila Pribic’ (Thales Nederland BV Delft, The Netherlands); Mario Coutino and Geert Leus (Delft University of Technology, The Netherlands)

Resolution from co-prime arrays and from a full ULA of the size equal to the virtual size of co-prime arrays is investigated. We take into account not only the resulting beam width but also the fact that fewer measurements are acquired by co-prime arrays. This fact is relevant in compressive acquisition typical for compressive sensing. Our stochastic approach to resolution uses information distances computed from the geometrical structure of data models that is characterized by the Fisher information. The probability of resolution is assessed from a likelihood ratio test by using information distances. Based on this information-geometry approach, we compare stochastic resolution from active co-prime arrays and from the full-size ULA. This novel stochastic resolution analysis is applied in a one-dimensional angle processing. Results demonstrate the suitability in radar-resolution analysis.

Multiple target tracking in the fully adaptive radar fra-meworkLuis Úbeda-Medina and Jesús Grajal (Universidad Politécnica de Madrid, Spain)

The fully adaptive radar framework aims to use some known information, or cognition, about the environment in which the system is deployed to obtain some improvement in its perfor-mance, typically either reducing the uncertainty of the obtai-ned results or optimizing the use of the available resources. In this paper, the extension of the fully adaptive radar framework for the case of multiple target tracking is introduced. In or-der to exemplify the proposed framework use, a simulation is

carried out in an scenario comprising multiple targets and a sensor network with resource constraints. Results show a re-markable performance improvement when the proposed fully adaptive radar approach is used.

Direct Data Domain Based Adaptive Beamforming for FDA-MIMO RadarJingwei Xu, Yanhong Xu and Guisheng Liao (Xidian University, P.R. China)

As frequency diverse array (FDA) introduces range-angle-dependent beamforming, it is capable to handle the range-dependent interference. However, the underlying independent and identically distributed condition of the interference is vio-lated, which induces performance degradation of interference suppression. In this paper, we propose a robust adaptive bea-mforming approach based on direct data domain technique for the multiple-input multiple-output (MIMO) radar with FDA as transmit array, referred to as FDA-MIMO. In this approach, the data is smoothed once in transmit and receive domains to mitigate the influence of target, which results in three homoge-neous samples. In the sequel, the data collected from different pulses is utilized as secondary data. Basically, the interference can be isolated from the target signal in the joint transmit and receive domains of the FDA-MIMO radar. Simulation results de-monstrate the effectiveness of the proposed approach.

TUE-I: DETECTION AND ESTIMATION THEORY II

Wald-Kernel: A Method for Learning Sequential DetectorsEmre Ertin and Diyan Teng (The Ohio State University, USA)

We consider the problem of training a binary sequential clas-sifier under an error rate constraint. It is well known that for known densities, accumulating the likelihood ratio statistics is time optimal under a fixed error rate constraint. For the case of unknown densities, we formulate the learning for sequential detection problem as a constrained density ratio estimation pro-blem. Specifically, we show that the problem can be posed as a convex optimization problem using a Reproducing Kernel Hilbert Space (RKHS) representation for the log-density ratio function. The proposed binary sequential classifier is tested on a synthetic data set and four real world data sets, together with previous

50


approaches for density ratio estimation. Our empirical results show that the classifier trained through the proposed technique achieves smaller average sampling cost than previous classifiers proposed in the literature for the same error rate.

Signal Reconstruction for Multi-source Variable-rate Samples with Autocorrelated Errors in VariablesAndrew Fox (Carnegie Mellon University, USA); Vinod Sharma (Indian Institute of Science, India); B. V. K. Vijaya Kumar (Car-negie Mellon University, USA)

Aggregating data from multiple sensors has become a critical requirement in cyberphysical systems (CPS) to increase the effec-tive sampling rate for signal reconstruction. Depending on the application, these sensors can be geo-distributed, mobile, or only intermittently functional. These factors cause the aggrega-ted sample set to be nonuniformly spaced with varying amounts of data collected per sensor. Due to the nature of how the timing or location measurements are made from the different sensors (e.g., indexed by GPS location), the samples may have significant errors in variables (EIV), where the location error from the di-fferent sensors follows an exponential autocorrelation function. In this work we demonstrate how to reconstruct signals using such noisy multi-source, variable-rate (MSVR) data samples, and show that the proposed approach improves the error over existing EIV signal reconstruction algorithms.

Inferring High-Dimensional Poisson Autoregressive ModelsEric C Hall (University of Wisconsin-Madison, USA); Garvesh Raskutti (University of Melbourne, Australia); Rebecca Willett (University of Wisconsin-Madison, USA)

Consider observing a series of events associated with a group of interacting nodes in a network, where the interactions among those nodes govern the likelihood of future events. Such data are common in spike trains recorded from biological neural networks, interactions within a social network, and pri-cing changes within financial networks. Vector autoregressive point processes accurately model these settings and are widely used in practice. This paper addresses the inference of the network structure and autoregressive parameters from such data. A sparsity-regularized maximum likelihood estimator is proposed for a Poisson autoregressive process. While sparsi-

ty- regularization is well-studied in the statistics and machine learning communities, common assumptions from that lite-rature are difficult to verify here because of correlations and heteroscedasticity inherent in the problem. Novel performance guarantees characterize how much data must be collected to ensure reliable inference depending on the size and sparsity of the autoregressive parameters, and these bounds are suppor-ted by several simulation studies.

Generalized Legendre Transform Multifractal Forma-lism for Nonconcave Spectrum EstimationRoberto Leonarduzzi (Université de Lyon, France); Hugo Tou-chette (Stellenbosch University, South Africa); Herwig Wendt (University of Toulouse & IRIT – ENSEEIHT, CNRS, France); Pa-trice Abry (Ecole Normale Superieure, Lyon, France); Stephane Jaffard (University of Paris-Est Creteil, France)

Despite widespread adoption of multifractal analysis as a sig-nal processing tool, most practical multifractal formalisms su-ffer from a major drawback: since they are based on Legendre transforms, they can only yield concave estimates for multi-fractal spectrum that are, in most cases, only upper bounds on the (possibly nonconcave) true spectrum. Inspired by ideas bo-rrowed from statistical physics, a procedure is devised for the estimation of not a priori concave spectra that retains the sim-ple and efficient Legendre transform formalism structure. The potential and interest of the proposed procedure are illustrated and assessed on realizations of a synthetic multifractal process, with theoretically known nonconcave multifractal spectrum.

An Auxiliary Variable Method for Langevin based MCMC algorithmsYosra Marnissi and Emilie Chouzenoux (Université Paris-Est Marne-la-Vallée, France); Jean-Christophe Pesquet (Université Paris-Est, France); Amel Benazza (Carthage University, Tunisia)

Markov Chain Monte Carlo sampling algorithms are efficient Bayesian tools to explore complicated posterior distributions. However, sampling in large scale problems remains a cha-llenging task since the Markov chain is very sensitive to the dependencies between the signal samples. In this paper, we are mainly interested in Langevin based MCMC sampling algo-rithms that allow us to speed up the convergence by contro-lling the direction of sampling and/or exploiting the correlation

51


structure of the target signal. However, these techniques may sometimes fail to explore efficiently the target space becau-se of poor mixing properties of the chain or the high cost of each iteration. By adding some auxiliary variables, we show that the resulting conditional distribution of the target signal is much simpler to explore by using these algorithms. Expe-riments performed in the context of multicomponent image restoration illustrate that the proposed approach can achieve substantial performance improvement compared with stan-dard algorithms.

Alternative Effective Sample Size measures for Impor-tance SamplingLuca Martino (University of Helsinki, Finland); Víctor Elvira (Uni-versity Carlos III of Madrid, Spain); Francisco Louzada (Univer-sidade de São Paulo (USP), Brazil)

The Effective Sample Size (ESS) is an important measure of efficiency in the Importance Sampling (IS) technique. A well-known approximation of the theoretical ESS definition, invol-ving the inverse of the sum of the squares of the normalized importance weights, is widely applied in literature. This expression has become an essential piece within Sequen-tial Monte Carlo (SMC) methods, using adaptive resampling procedures. In this work, first we show that this ESS approxi-mation is related to the Euclidean distance between the probability mass function (pmf) described by the normalized weights and the uniform pmf. Then, we derive other possible ESS functions based on different discrepancy measures. In our study, we also include another ESS measure called perplexity, already proposed in literature, that is based on the discrete entropy of the normalized weights. We compare all of them by means of numerical simulations.

Robust Markov Random Field Outlier Detection and Removal in Subsampled ImagesPaul McCool (NHS, United Kingdom); Yoann Altmann and An-tonios Perperidis (Heriot-Watt University, United Kingdom); Ste-ve McLaughlin (Heriot Watt University, United Kingdom)

Certain imaging technologies, such as fibred optical microsco-py, operate with irregularly-spaced sparse sub-samples from their field of view. In this work, we address the problem of data restoration for applications where the observed sub-samples

are corrupted by additive observation noise and sparse outliers (such as broken and damaged fibre cores). This problem is for-mulated as joint outlier detection and de-noising of irregularly sampled data. A fully Bayesian approach is used. A Markov Random Field is considered to capture the intrinsic spatial correlation of the underlying intensity field and binary labels are used to locate the spatial position of the outliers. Markov Chain Monte Carlo is then used to perform Bayesian inference, which uses the posterior distribution associated with the resul-ting Bayesian model. Simulations conducted on simulated data shows the potential benefits of the proposed method in terms of image reconstruction and outlier identification.

Translation Invariant DWT based Denoising using Goodness of Fit TestNaveed ur Rehman (COMSATS Institute of Information Tech-nology, Pakistan); Ubaid ur Rehman (University of Kassel, Ger-many); Syed Zain Abbas, Anum Asif and Anum Javed (COM-SATS Institute of Information Technology, Pakistan)

A novel signal denoising method based on discrete wavelet trans-form (DWT) and goodness of fit (GOF) statistical tests employing empirical distribution function (EDF) statistics is proposed. We formulate the denoising problem into a hypothesis testing pro-blem with a null hypothesis H0 corresponding to the presence of noise, and alternate hypothesis H1 representing the presence of only desired signal in the samples being tested. The decision process involves GOF tests being applied directly on multiple scales obtained from DWT. Cycle spinning approach is next em-ployed on the denoised data to render translation invariance property to the proposed method. We evaluate the performance of the resulting method against standard and modern wavelet shrinkage denoising methods through extensive repeated simu-lations performed on standard test signals.

A wavelet based likelihood ratio test for the homoge-neity of Poisson processesYoussef Taleb and Edward Cohen (Imperial College London, United Kingdom)

Estimating the rate (first-order intensity) of a point process is a task of great interest in the understanding of its nature. In this work we first address the estimation of the rate of an orderly point process on the real line using a multiresolution wavelet expansion approach.

52


Implementing Haar wavelets, we find that in the case of a Poisson process the piecewise constant wavelet estimator of the rate has a scaled Poisson distribution. We apply this result in the design of a likelihood ratio test for a multiresolution formulation of the homogeneity of a Poisson process. We demonstrate this method with simulations and provide Type 1 error and empirical power plots under specific models.

TUE-I SS: MULTIVARIATE STATISTICAL SIGNAL MO-DELING AND ANALYSIS

Distributed Multivariate Regression with Unknown Noise Covariance in the presence of Outliers: An MDL ApproachRoberto López-Valcarce (Universidad de Vigo, Spain); Daniel Romero (University of Minnesota, USA); Josep Sala (Universitat Politècnica de Catalunya, Spain); Alba Pagès-Zamora (Technical University of Catalonia & UPC, Spain)

We consider the problem of estimating the coefficients in a multivariate linear regression model by means of a wireless sensor network which may be affected by anomalous measu-rements. The noise covariance matrices at the different sensors are assumed unknown. Treating outlying samples, and their support, as additional nuisance parameters, the Maximum Likelihood estimate is investigated, with the number of outliers being estimated according to the Minimum Description Length principle. A distributed implementation based on iterative con-sensus techniques is then proposed, and it is shown effective for managing outliers in the data.

Optimal transport vs. Fisher-Rao distance between co-pulas for clustering multivariate time seriesGautier Marti (Ecole Polytechnique & Hellebore Capital, Fran-ce); Sébastien Andler (Ecole Normale Supérieure de Lyon, France); Frank Nielsen (Ecole Polytechnique, France); Philippe Donnat (Hellebore Capital, United Kingdom)

We present a methodology for clustering N objects which are described by multivariate time series, i.e. several sequences of real-valued random variables. This clustering methodology leverages copulas which are distributions encoding the depen-dence structure between several random variables. To take fully

into account the dependence information while clustering, we need a distance between copulas. In this work, we compa-re renowned distances between distributions: the Fisher-Rao geodesic distance, related divergences and optimal transport, and discuss their advantages and disadvantages. Applications of such methodology can be found in the clustering of financial assets. A tutorial, experiments and implementation for repro-ducible research can be found at www.datagrapple.com/Tech.

Combining EEG source connectivity and network similari-ty: Application to object categorization in the human brainAhmad Mheich and Mahmoud Hassan (University of Rennes1, France); Olivier Dufor (Télécom Bretagne, Institut Mines-Té-lécom & UMR CNRS 6285 Lab-STICC, France); Mohamad Khalil (Lebanese University & Doctoral School of Sciences and Tech-nology, Lebanon); Fabrice Wendling (INSERM, France)

A major challenge in cognitive neuroscience is to evaluate the ability of the human brain to categorize or group visual stimuli based on common features. This categorization process is very fast and occurs in few hundreds of millisecond time scale. Howe-ver, an accurate tracking of the spatiotemporal dynamics of large-scale brain networks is still an unsolved issue. Here, we show the combination of recently developed method called ‘dense-EEG source connectivity’ to identify functional brain networks with ex-cellent temporal and spatial resolutions and an algorithm, called SimNet, to compute brain networks similarity. Two categories of visual stimuli were analysed in this study: immobile and mobile. Networks similarity was assessed within each category (intra-con-dition) and between categories (inter-condition). Results showed high similarity within each category and low similarity between the two categories. A significant difference between similarities computed in the intra and inter-conditions was observed at the period of 120-190ms supposed to be related to visual recognition and memory access. We speculate that these observations will be very helpful toward understanding the object categorization in the human brain from a network perspective.

A Generalized Multivariate Logistic Model and EM Al-gorithm based on the Normal Variance Mean Mixture RepresentationJason Palmer (University of California San Diego, USA); Ken Kreutz-Delgado (University of California, San Diego, USA); Scott Makeig (University of California San Diego, USA)

53


4

We present an EM algorithm for Maximum Likelihood estima-tion of the location, scale, and skew, and shape parameters of the z distribution, also known as the generalized logistic function (type IV). We use the Barndorff-Nielsen, Kent, and Sorensen representation of the z distribution as a Gaussian location-scale mixture to derive an EM algorithm for estima-ting the location, scale, skew, and shape parameters. We use a variational bound on the likelihood function to determine a monotonically converging closed form update for the skew (or drift) parameter. The algorithm also extends naturally to multivariate GLSM estimation using the Kolmogorov-Smirnov mixing density in odd dimensions.

Approximating Bayesian confidence regions in convex inverse problemsMarcelo Pereyra (University of Bristol, United Kingdom)

Solutions to inverse problems that are ill-conditioned or ill-posed have significant intrinsic uncertainty. Unfortunately, analysing and quantifying this uncertainty is very challenging, particularly in high-dimensional settings. As a result, while most modern statistical signal processing methods achieve impressive point estimation results, they are generally unable to quantify the uncertainty in the solutions delivered. This work presents a new general methodology for approximating Baye-sian high-posterior-density (confidence or credibility) regions in inverse problems that are convex and potentially very high-dimensional. A remarkable property of the approximations is that they can be computed very efficiently, even in large-scale problems, by using standard convex optimisation techniques. The proposed methodology is demonstrated on a high-dimen-sional image restoration problem, where the approximation error is assessed by using proximal Markov chain Monte Carlo as benchmark.

Balanced Least Squares: Estimation in Linear Systems with Noisy Inputs and Multiple OutputsJavier Vía and Ignacio Santamaria (University of Cantabria, Spain)

This paper revisits the linear model with noisy inputs, in which the performance of the total least squares (TLS) method is far from acceptable. Under the assumption of Gaussian noises, the maximum likelihood (ML) estimation of the system response is

reformulated as a general balanced least squares (BLS) pro-blem. Unlike TLS, which minimizes the trace of the product between the empirical and inverse theoretical covariance matrices, BLS promotes solutions with similar values of both the empirical and theoretical error covariance matrices. The general BLS problem is reformulated as a semidefinite program with a rank constraint, which can be relaxed in order to obtain polynomial time algorithms. Moreover, we provide new theo-retical results regarding the scenarios in which the relaxation is tight, as well as additional insights on the performance and interpretation of BLS. Finally, some simulation results illustrate the satisfactory performance of the proposed method.

Multimodal Metric Learning with Local CCAOr Yair and Ronen Talmon (Technion – Israel Institute of Te-chnology, Israel)

In this paper, we address the problem of multimodal signal processing from a kernel-based manifold learning standpoint. We propose a data-driven method for extracting the common hidden variables from two multimodal sets of nonlinear high-dimensional observations. To this end, we present a metric based on local canonical correlation analysis (CCA). Our ap-proach can be viewed both as an extension of CCA to a non-linear setting as well as an extension of manifold learning to multiple data sets. We test our method in simulations, where we show that it indeed discovers the common variables hidden in high-dimensional nonlinear observations without assuming prior rigid model assumptions.


TUE-II: COMPRESSED SENSING

Oracle Performance Estimation of Bernoulli-distributed Sparse VectorsRémy Boyer (Université Paris-Sud (UPS), CNRS, CentraleSupelec, France); Pascal Larzabal (ENS-Cachan, PARIS, France); Bernard Henri Fleury (Aalborg University, Denmark)

54


Compressed Sensing (CS) is now a well-established research area and a plethora of applications has emerged in the last decade. In this context, assuming N available noisy measu-rements, lower bounds on the Bayesian Mean Square Error (BMSE) for the estimated entries of a sparse amplitude vector are derived in the proposed work for (i) a Gaussian overcom-plete measurement matrix and (ii) for a random support, assu-ming that each entry is modeled as the product of a continuous random variable and a Bernoulli random variable indicating that the current entry is non-zero with probability P. A closed-form expression of the Expected CRB (ECRB) is proposed. In the second part, the BMSE of the Linear Minimum MSE LMMSE estimator is derived and it is proved that the LMMSE estima-tor tends to be statistically efficient in asymptotic conditions, i.e., if product (1-P)2 \ SNR is maximized. This means that in the context of the Gaussian CS problem, the LMMSE estimator gathers together optimality in the low noise variance regime and a simple derivation (as opposed to the derivation of the MMSE estimator). This result is original because the LMMSE estimator is generally sub-optimal for CS when the measurement matrix is a single realization of a given random process.

Bound on the estimation grid size for sparse recons-truction in direction of arrival estimationMario Coutino (Delft University of Technology, The Nether-lands); Radmila Pribic’ (Thales Nederland BV Delft, The Nether-lands); Geert Leus (Delft University of Technology, The Nether-lands)

A bound for sparse reconstruction involving both the signal-to-noise ratio (SNR) and the estimation grid size is presented. The bound is illustrated for the case of a uniform linear array (ULA). By reducing the number of possible sparse vectors present in the feasible set of a constrained l1 norm minimi-zation problem, ambiguities in the reconstruction of a single source under noise can be reduced. This reduction is achieved by means of a proper selection of the estimation grid, which is naturally linked with the mutual coherence of the sensing matrix. Numerical simulations show the performance of spar-se reconstruction with an estimation grid meeting the provi-ded bound demonstrating the effectiveness of the proposed bound.

On Sparse Recovery Using Finite Gaussian Matrices: RIP-Based AnalysisAhmed Elzanaty, Andrea Giorgetti and Marco Chiani (Univer-sity of Bologna, Italy)

We provide a probabilistic framework for the analysis of the restricted isometry constants (RICs) of finite dimensional Gaus-sian measurement matrices. The proposed method relies on the exact distribution of the extreme eigenvalues of Wishart matrices, or on its approximation based on the gamma dis-tribution. In particular, we derive tight lower bounds on the cumulative distribution functions (CDFs) of the RICs. The pre-sented framework provides the tightest lower bound on the maximum sparsity order, based on sufficient recovery condi-tions on the RICs, which allows signal reconstruction with a given target probability via different recovery algorithms.

Bilevel Feature Selection in Nearly-Linear TimeChinmay Hegde (Iowa State University, USA)

Selection of a small subset of informative features from data is a basic technique in signal processing, machine learning, and statistics. Joint selection of entire groups of features is desirable if the data features exhibit shared grouping structures. Bilevel feature selection constitutes a refinement of these ideas, pro-ducing a small subset of data features that themselves belong to a small number of feature groups. However, algorithms for bilevel feature selection suffer a computational cost that can be cubic in the size of the data, hence impeding their utility.In this paper, we propose an approach for bilevel feature se-lection that resolves this computational challenge. The core component of our approach is a novel fast algorithm for bilevel hard thresholding for a specific non-convex, discrete optimi-zation problem. Our algorithm produces an approximate so-lution to this problem, but only incurs a nearly-linear running time. We extend this algorithm into a two-stage thresholding method that performs statistically as well as the best available methods for bilevel feature selection, but that also scales extre-mely well to massive dataset sizes.

Sparse Error Correction with Multiple Measurement VectorsSharmin Kibria, Jinsub Kim and Raviv Raich (Oregon State Uni-versity, USA)

55


Exploiting a sparse nature of gross errors in sensor system mea-surements, feasibility of gross error identification and robust estimation of system state is studied.Under the practical assumption that potential locations of gross errors stay the same during multiple measurement pe-riods, gross error correction based on multiple measurement vectors is proposed.Our feasibility analysis shows that the maximum number of identifiable gross errors can double compared to the identi-fication based on a single measurement vector, if gross error values are diverse across different measurement periods.A convex optimization framework is proposed to identify gross error locations and calculate an accurate state estimate.The proposed state estimator is applied for power system DC state estimation of IEEE 14-bus network and shown to outperform ben-chmark techniques that are based on a single measurement vector.

A Refined Analysis for the Sample Complexity of Adap-tive Compressive Outlier SensingXingguo Li and Jarvis D. Haupt (University of Minnesota, USA)

The Adaptive Compressive Outlier Sensing (ACOS) method, proposed recently in (Li & Haupt, 2015), is a randomized se-quential sampling and inference method designed to locate column outliers in large, otherwise low rank, matrices. While the original ACOS work established conditions on the sample complexity sufficient to enable accurate outlier localization (with high probability), the guarantees required a minimum sample complexity that grew linearly (albeit slowly) in the number of matrix columns. This work presents a refined analy-sis of the sampling complexity of ACOS that overcomes this limitation; we show that the sample complexity of ACOS is su-blinear in both of the matrix dimensions — on the order of the squared rank of the low-rank component plus the number of outliers, times constant and logarithmic factors.

Low Rank Matrix Recovery From Column-Wise Phase-less MeasurementsSeyedehsara Nayer and Namrata Vaswani (Iowa State Uni-versity, USA); Yonina C. Eldar (Technion-Israel Institute of Te-chnology, Israel)

We study the problem of recovering a low-rank matrix, X, from magnitude-only observations of random linear projections of

its columns (phaseless measurements). We develop a new approach, called Phase Retrieval Low Rank (PReLow), that borrows ideas from a very recent non-convex phase retrieval approach, called truncated Wirtinger flow (TWF). We show via extensive numerical experiments that, when the rank of X is small compared to its dimensions, PReLow significantly outper-forms TWF which operates on one column of X at a time (does not use its low-rank structure).

TUE-II: SIGNAL PROCESSING OVER GRAPHS AND NETWORKS II

Evolutionary Spectral Graph Clustering Through Subs-pace Distance MeasureEsraa Al-sharoa (Michigan State University, USA); Selin Aviyente (Elec-trical and Computer Engineering, Michigan State University, MI, USA)In the era of Big Data, massive amounts of high-dimensional data are increasingly gathered. Much of this is streaming big data that is either not stored or stored only for short perio-ds of time. Examples include cell phone conversations, texts, tweets, network traffic, changing Facebook connections, mo-bile video chats or video surveillance data. It is important to be able to reduce the dimensionality of this data in a streaming fashion. One common way of reducing the dimensionality of data is through clustering. Evolutionary clustering provides a framework to cluster the data at each time point such that the cluster assignments change smoothly across time. In this pa-per, an evolutionary spectral clustering approach is proposed for community detection in dynamic networks. The proposed method tries to obtain smooth cluster assignments by minimi-zing the subspace distance between consecutive time points, where the subspaces are defined through spectral embedding. The algorithm is evaluated on several synthetic and real data sets, and the results show the improvement in performance over traditional spectral clustering and state of the art evolu-tionary clustering algorithms.

Diffusion estimation of mixture models with local and global parametersKamil Dedecius (Institute of Information Theory and Automa-tion, Czech Academy of Sciences, Czech Republic); Vladimira Seckarova (Institute of Information Theory and Automation, Czech Academy of Sciences)

56


The state-of-art methods for distributed estimation of mixtures assume the existence of a common mixture model. In many practical situations, this assumption may be too restrictive, as a subset of parameters may be purely local, e.g., if the num-bers of observable components differ across the network. To reflect this issue, we propose a new online Bayesian method for simultaneous estimation of local parameters, and diffu-sion estimation of global parameters. The algorithm consists of two steps. First, the nodes perform local estimation from own observations by means of factorized prior/posterior distri-butions. Second, a diffusion optimization step is used to merge the nodes’ global parameters estimates. A simulation example demonstrates improved performance in estimation of both pa-rameters sets.

Translation on Graphs: An Isometric Shift OperatorBenjamin Girault (ENS Lyon, France); Paulo Gonçalves (INRIA, France); Eric Fleury (ENS Lyon / INRIA, France)

In this letter, we propose a new shift operator for graph signals, enforcing that our operator is isometric. Doing so, we ensure that as many properties of the time shift as possible get carried over. Finally, we show that our operator behaves reasonably for graph signals.

Identifying Rumor Sources with Different Start TimesFeng Ji and Wee Peng Tay (Nanyang Technological University, Singapore)

We study the problem of identifying multiple rumor or infection sources in a network under the susceptible-infected (SI) model. We do not assume that the sources start infection spreading at the same time. We introduce the notion of a quasi-regular tree as the basic model, and an abstract estimator, which includes seve-ral of the single source estimators developed in the literature. We develop a general two source joint estimation algorithm based on any abstract estimator, and show that it converges to a local optimum of the estimation function if the underlying network is a quasi-regular tree. We further extend our algorithm to more than two sources, and heuristically to general graphs.

Estimation of Time-Varying Mixture Models: An Appli-cation to Traffic EstimationSean Lawlor and Michael Rabbat (McGill University, Canada)

Time varying mixture models can be a useful tool for modelling complex data collections. However the additional complexity of letting the number of mixture components vary over time adds even more difficulty in inference of the distribution para-meters. We propose the automatic k-means algorithm to infer the parameters of these complex, time-varying mixture models. We demonstrate its performance using simulated and real data in a traffic estimation scenario.

Synchronization for classical blind source separation algorithms in wireless acoustic sensor networksCosme Llerena, Roberto Gil-Pita, David Ayllón and Héctor Adrián Sánchez-Hevia (University of Alcalá, Spain); Inmacula-da Mohino-Herranz (University of Alcala, Spain); Manuel Rosa (University of Alcalá, Spain)

The use of wireless acoustic sensor networks is becoming very popular since they entail many advantages. However, this type of distributed sensor networks has an important drawback for many signal processing algorithms, the synchronization pro-blem. Broadly speaking, in those networks, signals received at the different nodes are not synchronized due to two main factors, the clock problem and the important differences in propagation delays between sources and microphones. In this work we introduce a synchronization solution for mixtures of two and three speech sources in the framework of blind source separation. This proposal of synchronization has a mixture alig-nment stage prior to apply the separation method. Obtained results demonstrate that this synchronization method aligns speech mixtures correctly since it improves the performance of the classical separation algorithm in terms of both speech quality and speech intelligibility.

A Hawkes’ eye view of network information flowMichael G Moore and Mark Davenport (Georgia Institute of Technology, USA)

An important problem that arises in the analysis of many com-plex networks is to identify the common pathways that enable the flow of information (or other quantities) through the network. This is a particularly challenging problem when the only infor-mation observed consists of the timing of events in the network. We develop a framework based on multidimensional Hawkes processes that can be used to determine how events are related.

57


This extends the capability of Hawkes process-based models to infer how network events relate. We then show how a simple dy-namic program can exploit this data to recognize chains of events and provide a much deeper insight into the behavior of nodes within the network. Simulations are provided to demonstrate the capabilities and limitations of this framework.

TUE-II SS: BAYESIAN DETECTION AND ESTIMA-TION TECHNIQUES FOR RADAR APPLICATIONS

Weiss-Weinstein bound for an abrupt unknown fre-quency changeLucien Bacharach (Université Paris 11, France); Mohammed Nabil El Korso (Paris 10 University & LEME-EA 4416, France); Alexandre Renaux (Universite Paris 11, France)

In this paper, we derive analytical expressions of the Weiss-Weinstein bound (WWB) in the context of observations whose frequency abruptly changes at an unknown time instant. Since both frequencies before and after the change are assumed to be unknown as well, it is appropriate to consider the multi-parameter version of the WWB. Furthermore, numerical simu-lations are provided in order to illustrate the tightness of the proposed bound expressions regarding to the estimates errors.

Adaptive Waveform Design for Target Detection with Sequential Composite Hypothesis TestingShahar Bar and Joseph Tabrikian (Ben-Gurion University of the Negev, Israel)

This paper addresses the problem of adaptive waveform design for target detection with composite sequential hypothesis tes-ting. We begin with an asymptotic analysis of the generalized sequential probability ratio test (GSPRT). The analysis is based on Bayesian considerations, similar to the ones used for the Bayesian information criterion (BIC) for model order selection. Following the analysis, a novel test, named penalized GSPRT (PGSPRT), is proposed on the basis of restraining the exponential growth of the GSPRT with respect to (w.r.t.) the sequential probability ratio test (SPRT). The performance measures of the PGSPRT in terms of average sample number (ASN) and error probabilities are also investigated. In the proposed waveform design scheme, the transmit spatial waveform is adaptively determined at each

step based on observations in the previous steps. The waveform is determined to minimize the ASN of the PGSPRT. Simulations demonstrate the performance measures of the new algorithm for target detection in a multiple input, single output (MISO) channel.

Estimation and compensation of I/Q imbalance for FMCW radar receiversAngelo Coluccia, Vincenzo Dodde and Antonio Masciullo (Uni-versity of Salento, Italy); Giuseppe Ricci (University of Salento, Lecce, Italy)

This paper deals with adaptive compensation of gain and pha-se errors possibly present in FMCW radars. In particular, we show that previously-proposed compensation procedures can be related to the LMMSE approach. In addition, we assess by Monte Carlo simulation gain and phase estimators and, even-tually, the effectiveness of the compensation procedure.

Improving Separating Function Estimation Tests Using Bayesian ApproachesAli Ghobadzadeh and Raviraj Adve (University of Toronto, Canada)

Separating function estimation tests (SFETs) replace detection problems with an estimation problem. In this paper, we study the relationship between improving the estimation of unknown parameters using Bayesian approaches and the performance of the corresponding SFET. Although the estimation method in the SFET is deterministic, we show that applying Bayesian methods to estimate the rest of unknown parameters that are not involved in the SF provide improved SFET performance. We illustrate this idea using two important problems. In the first example, we consider a sinusoid signal with unknown parameters in white noise. We show that a softmax function using the Fourier transform of the signal is a proper probability density function (pdf) for the frequency to improve the performance of the SFET. In the second example, a more accurate estimation of the unknown parameters of the signal is achieved, using the Minimum Mean Square Error (MMSE) esti-mation of the random signal corrupted by white noise.

Velocity ambiguity mitigation of off-grid range migra-ting targets via Bayesian sparse recoveryMarie Lasserre and Stéphanie Bidon (University of Toulouse / ISAE, France); François Le Chevalier (Thales Air Systems & TU Delft, France)

58


Within the scope of sparse signal representation, we consider the problem of velocity ambiguity mitigation for wideband ra-dar signal. We present a Bayesian robust algorithm based on a new sparsifying dictionary suited for range-migrating targets possibly straddling range-velocity bins. Numerical simulations on experimental data demonstrate the ability of the proposed algorithm in mitigating velocity ambiguity.

Bayesian Framework and Radar: On Misspecified Bounds and Radar-Communication CooperationChrist D. Richmond and Prabahan Basu (MIT Lincoln Labora-tory, USA)

The Bayesian framework is versatile as it allows for incorpo-ration of prior knowledge or experience in making inference. The case of no prior knowledge at all is likewise seamlessly supported. The Bayesian framework is naturally suited to many fields of science and engineering including the discipline of radar design, analysis, and function. This paper will highlight recent findings on (i) Bayesian bounds under model misspeci-fication that include the impact of using incorrect prior infor-mation, and (ii) parameter estimation for a cooperative radar-communication system.

Detection in Multiple Channels Having Unequal Noise PowerSongsri Sirianunpiboon and Stephen D Howard (Defence Scien-ce and Technology Group, Australia); Douglas Cochran (Arizo-na State University, USA)

A bayesian detector is formulated for the problem of detecting a signal of known rank using data collected at multiple sen-sors. The noise on each sensor channel is white and gaussian, but its variance is unknown and may be different from channel to channel. A low-SNR assumption that enables approximation of one of the marginalization integrals in the likelihood ratio, yielding a tractable approximate bayesian detector for this re-gime. Performance of this detector is evaluated and compared to other recently introduced detectors.

TUE-II SS: MAKING SENSE OUT OF MULTI-CHANNEL PHYSIOLOGICAL DATA FOR PERVASIVE HEALTH APPLICATIONS

Accelerometer-based gait assessment: pragmatic de-ployment on an international scaleSilvia Del Din, Aodhan Hickey, Simon Woodman, Hugo Hiden and Rosie Morris (Newcastle University, United Kingdom); Paul Watson (University of Newcastle, United Kingdom); Kianoush Nazarpour, Michael Catt, Lynn Rochester and Alan Godfrey (Newcastle University, United Kingdom)

Gait is emerging as a powerful tool to detect early disease and monitor progression across a number of pathologies. Typically quantitative gait assessment has been limited to specialised laboratory facilities. However, measuring gait in home and community settings may provide a more accurate reflection of gait performance because: (1) it will not be confounded by attention which may be heightened during formal testing; and (2) it allows performance to be captured over time. This work addresses the feasibility and challenges of measuring gait cha-racteristics with a single accelerometer based wearable device during free-living activity. Moreover, it describes the current methodological and statistical processes required to quantify those sensitive surrogate markers for ageing and pathology. A unified framework for large scale analysis is proposed. We present data and workflows from healthy older adults and those with Parkinson’s disease (PD) while presenting current algorithms and scope within modern pervasive healthcare. Our findings suggested that free-living conditions heighten between group differences showing greater sensitivity to PD, and provided encouraging results to support the use of the suggested framework for large clinical application.

Cortical Distribution of N400 Potential in Response to Semantic Priming with Visual Non-Linguistic StimuliElvira Khachatryan (KULeuven & YSMU, Belgium); Nikolay Chumerin (Katholieke Universiteit Leuven, Belgium); Evelien Ca-rrette (University Hospital Ghent, Belgium); Flavio Camarrone (KU Leuven, Belgium); Leen De Taeye (UZ Ghent, Belgium); Al-fred Meurs, Paul Boon and Dirk Van Roost (University Hospital Ghent, Belgium); Marc Van Hulle (KU Leuven, Belgium)

We conducted a study on visual semantic priming using rela-ted and unrelated image pairs while simultaneously recording electroencephalography (EEG) from 27 scalp electrodes and electrocorticography (ECoG) from a mixture of deep brain and subdural grid/strip electrodes in the left and right hippocam-

59


pus, the right temporobasal and temporo-lateral cortices, and the left temporal cortex. The EEG data showed a clear centro-parietal, bi-hemispheric N400 effect in response to unrelated image-pairs compared to related ones. Although with ECoG the N400 effect was more widely spread across both hemis-pheres, compared to linguistic stimuli, it was relatively loca-lized within each ECoG grid as it was present only in some electrodes and, in some cases, even had its polarity reversed. We advocate this could be due to some grids gauging dipoles at different positions when covering sulci and gyri.

Reconstruction of Brain Activity in EEG / MEG Using Reduced-Rank Nulling Spatial FilterTomasz Piotrowski (Nicolaus Copernicus University, Poland); Jan Nikadon (Nicolaus Copernicus Univerity, Poland); Dania Gutie-rrez (Center for Research and Advanced Studies & Monterrey’s Unit, Mexico)

We consider the problem of reconstruction of brain activity from electroencephalography (EEG) and magnetoencephalography (MEG) using spatial filtering (beamforming). We assume the pre-sence of interfering sources, whose activity may be highly corre-lated with activity of sources to be reconstructed. Such situation causes the celebrated linearly constrained minimum-variance (LCMV) filter to produce erroneous estimates due to source can-celation. This problem is especially acute if signal-to-noise ratio (SNR) is low. We propose a robust reduced-rank nulling spatial filter designed to overcome these drawbacks of the LCMV fil-ter. The proposed filter is a natural reduced-rank extension of the recently introduced nulling spatial filter, allowing efficient implementation of nulling constraints in low SNR regime. It is equipped with rank-selection criterion minimizing mean-square-error (MSE) of the resulting estimate. We verify its improved performance in the challenging conditions described above in comparison with established spatial filtering techniques.

Efficient Low-Rank Spectrotemporal Decomposition using ADMMGabriel Schamberg (University of California, San Diego, USA); Demba E Ba (Harvard University); Mark Wagner (University of California, San Diego, USA); Todd Coleman (UCSD, USA)

Classical time-evolving spectral analysis techniques utilize a sliding window approach that fails to exploit overarching

spectrotemporal structures that are known to occur in many real-world signals. In particular, many biological signals have the distinct quality of having few defining spectral characte-ristics. We propose an algorithm for efficiently estimating a low-rank spectrotemporal decomposition. While existing approaches yield such representations by penalizing a group norm of successive differences, our insight is that a penalty that promotes low-rank yields more flexible representations and an efficient distributed implementation. We demonstrate on simulated time series data and human electroencephalo-gram (EEG) recordings that this low-rank spectrotemporal de-composition can provide a spectral representation of time series that highlights salient features and reduces the effects of noise.

Dimensionality Reduction of Sample Covariance Ma-trices by Graph Fourier Transform for Motor Imagery Brain-Machine InterfaceToshihisa Tanaka, Takashi Uehara and Yuichi Tanaka (Tokyo University of Agriculture and Technology, Japan)

An efficient method for dimensionality reduction in classifica-tion of multi-class electroencephalogram (EEG) during motor imagery (MI) aiming at brain-machine interfacing is proposed. In this method, the reduction of dimensions is achieved by spectral decomposition of a given graph, which is defined by a geometrical distribution of electrodes on the head surface. The resulting subspace reduces the dimension of EEG signals, and therefore, the size of the sample covariance matrix (SCM) of EEG can also be reduced. The reduction method is combined with a differential geometry-based approach called tangent space mapping (TSM) that can map a SCM in a Riemannian manifold onto an element in an Euclidean space called a tan-gent space. Thus, any machine learning algorithm that works in the Euclidean space can be applied. Results of two-class and four-class classification of EEG during MI show that the proposed method of dimensionality reduction increases the recognition accuracy even in the case of a training dataset ha-ving a small number of elements.

Estimation of High-Dimensional Connectivity in fMRI Data via Subspace Autoregressive ModelsChee-Ming Ting (Universiti Teknologi Malaysia, Malaysia); Abd-krim Seghouane (The University of Melbourne, Australia); Sh-Hussain Salleh (Universiti Teknologi Malaysia, Malaysia)

60


We consider the challenge in estimating effective connectivity of brain networks with a large number of nodes from fMRI data. The classical vector autoregressive (VAR) modeling tends to pro-duce unreliable estimates for large dimensions due to the huge number of parameters. We propose a subspace estimator for large-dimensional VAR model based on a latent variable model. We derive a subspace VAR model with the observational and noise process driven by a few latent variables, which allows for a lower-dimensional subspace of the dependence structure. We introduce a fitting procedure by first estimating the latent space by principal component analysis (PCA) of the residuals and then reconstructing the subspace estimators from the PCs. Simula-tion results show superiority of the subspace VAR estimator over the conventional least squares (LS) under high-dimensional set-tings, with improved accuracy and consistency. Application to estimating large-scale effective connectivity from resting-state fMRI shows the ability of our method in identifying interesting modular structure of human brain networks during rest.

Hierarchical Online SSVEP Spelling Achieved With Spa-tiotemporal BeamformingBenjamin Wittevrongel and Marc Van Hulle (KU Leuven, Belgium)

Steady-State Visual Evoked Potentials (SSVEP) are widely adopted in brain-computer interface (BCI) applications. To increase the number of selectable targets, joint frequency- and phase-coding is sometimes used but it has only been tested in offline settings.In this study, we report on an online, hierarchical SSVEP spelling application that relies on joint frequency/ phase coded targets, and, in addition, propose a new decoding scheme based on spatiotemporal beamforming combined with time-domain EEG analysis. Experiments on 17 healthy subjects confirm that with our new decoding scheme, accurate spelling can be performed in an online setting, even when using short stimulation lengths (1 sec) and closely separated stimulation frequencies (1 Hz).

TUE-III: MACHINE LEARNING AND PATTERN RE-COGNITION II

Efficient KLMS and KRLS Algorithms: A Random Fourier Feature PerspectivePantelis Bouboulis, Spyridon Pougkakiotis and Sergios Theodo-ridis (University of Athens, Greece)

We present a new framework for online Least Squares algorithms for nonlinear modeling in RKH spaces (RKHS). Instead of impli-citly mapping the data to a RKHS (e.g., kernel trick), we map the data to a finite dimensional Euclidean space, using random features of the kernel’s Fourier transform. The advantage is that, the inner product of the mapped data approximates the kernel function. The resulting “linear” algorithm does not require any form of sparsification, since, in contrast to all existing algorithms, the solution’s size remains fixed and does not increase with the iteration steps. As a result, the obtained algorithms are computa-tionally significantly more efficient compared to previously derived variants, while, at the same time, they converge at similar speeds and to similar error floors.

Designing Classifier Architectures using Information TheoryJohn Cortese (Massachusetts Institute of Technology, USA)

An architecture for hypothesis testing systems is analyzed using tools from information theory. The classifier architectu-re consists of a Markov Chain of statistical signal processing blocks. As part of the analysis, a mathematical framework for tradeoff studies among system components is introduced.

Democratic prior for anti-sparse codingClément Elvira and Pierre Chainais (Ecole Centrale Lille & CRIStAL CNRS, France); Nicolas Dobigeon (University of Toulouse, France)

Anti-sparse coding aims at spreading the information uniformly over representation coefficients and can be naturally expressed through an l∞ norm regularization. This paper derives a proba-bilistic formulation of such a problem. To that, a new probabi-lity distribution is introduced. This so-called \emph{democratic} distribution is then used as a prior to promote anti-sparsity in a linear Gaussian inverse problem. A Gibbs sampler is de-signed to generate samples asymptotically distributed accor-ding to the joint posterior distribution of interest. To scale to higher dimension, a proximal Markov chain Monte Carlo algorithm is proposed as an alternative to Gibbs sampling.


61


Simulations on synthetic data illustrate the performance of the proposed method for anti-sparse coding on a complete dictionary. Results are compared with the recent deterministic variational FITRA algorithm.

Group invariant subspace learning for outlier detectionBo Fan and Shuchin Aeron (Tufts University, USA)

In this paper, we present a novel method for detecting outliers when the images are misaligned by action of a finite group. Our approach rests on robust learning of group-invariant subspaces in presence of outliers. By group-invariant subspaces, we mean subspaces of a vector space that are invariant to action of a finite (Abelian) group. Such scenarios naturally arise in computer vision problems when one is interested in shift (translation) or rotation in-variant image processing. While the proposed methods are gene-ral, we will focus on misalignment by the group of circular shifts on 2-D images and show that our methods are effective in detecting outliers in real data sets (YaleB and MNIST database) and outper-form methods that do not take the group-invariance into account.

Human Authentication From Ankle Motion Data Using Convolutional Neural NetworksMatteo Gadaleta, Luca Merelli and Michele Rossi (University of Padova, Italy)

We present a data acquisition and signal processing framework for the authentication of users from their gait signatures (acce-lerometer and gyroscope data). An ankle-worn inertial measu-rement unit (IMU) is utilized to acquire the raw motion data, which is pre-processed and used to train a number of signal processing tools, including a convolutional neural network (CNN) for the extraction of features as well as one-class sin-gle- and multi-stage classifiers. The CNN is trained (offline and only once) using a representative set of subjects and is then exploited as a universal feature extractor, i.e., to extract relevant features of walking patterns from previously unseen subjects. The one-class classifier is trained on the subject that we intend to authenticate and employed to gauge new motion data. Sco-res from the one-class classifier are finally fed into a multi-stage decision maker, which performs a sequential decision testing for improved accuracy. The system operates in an online fashion, delivering excellent results, while requiring in the worst case fewer than five walking cycles to reliably authenticate the user.

Unsupervised segmentation of piecewise constant images from incomplete, distorted and noisy dataJean-François Giovannelli (IMS, UMR CNRS 52 18, Université Bordeaux 1, France); Andrei Barbos (IMS, France)

The paper tackles the problem of piecewise constant image segmentation. A triple degradation model is assumed for the observation system: missing data, non-linear gain and addi-tive noise. The proposed solution follows a Bayesian strategy that yields optimal decisions and estimations. A numerical ap-proach is used to explore the intricate posterior distribution: a Gibbs sampler including a Metropolis-Hastings step. The pos-terior samples are subsequently used in computing the esti-mates and the decisions. A first numerical evaluation provided encouraging results despite the triple degradation.

Multilinear Subspace ClusteringEric Kernfeld (University of Washington, USA); Nathan Majum-der, Shuchin Aeron and Misha Kilmer (Tufts University, USA)

In this paper we present a new model and an algorithm for unsupervised clustering of 2-D data such as images. We assu-me that the data comes from a union of multilinear subspaces (UOMS) model, which is a specific structured case of the much studied union of subspaces (UOS) model. For segmentation under this model, we develop Multilinear Subspace Clustering (MSC) algorithm and evaluate its performance on the YaleB and Olivietti image data sets. We show that MSC is highly com-petitive with existing algorithms employing the UOS model in terms of clustering performance while enjoying improvement in the computational complexity.

Order-based Generalized Multivariate RegressionMilad Kharratzadeh and Mark Coates (McGill University, Ca-nada)

In this paper, we consider a generalized multivariate regres-sion problem where the responses are monotonic functions of linear transformations of predictors. We propose a semi-parametric algorithm based on the ordering of the responses which is invariant to the functional form of the transformation function. We prove that our algorithm, which maximizes the rank correlation of responses and linear transformations of pre-dictors, is a consistent estimator of the true coefficient matrix.

62


We also identify the rate of convergence and show that the squared estimation error decays with a rate of 0 (1/ √ n). We then propose a greedy algorithm to maximize the highly non-smooth objective function of our model and examine its performance through simulations. Finally, we compare our algorithm with traditional multivariate regression algorithms over synthetic and real data.

Hierarchical Bayesian variable selection in the probit model with mixture of nominal and ordinal responsesEleftheria Kotti, Ioanna Manolopoulou and Tom Fearn (Univer-sity College London (UCL), United Kingdom)

Multi-class classification problems have been studied for pure nominal and pure ordinal responses. However, there are some cases where the multi-class responses are a mixture of nomi-nal and ordinal. To address this problem we build a hierarchi-cal multinomial probit model with a mixture of both types of responses using latent variables. The nominal responses are each associated to distinct latent variables whereas the or-dinal responses have a single latent variable. Our approach first treats the ordinal responses as a single nominal category and then separates the ordinal responses within this category. We introduce sparsity into the model using Bayesian variable selection within the regression in order to improve variable se-lection classification accuracy. Two indicator vectors (indicating presence of the covariate) are used, one for nominal and one for ordinal responses. We develop efficient posterior sampling. Using simulated data, we compare the classification accuracy of our method to existing ones.

Jeffreys Prior Regularization for Logistic RegressionTam Nguyen, Raviv Raich and Phung Lai (Oregon State Uni-versity, USA)

Logistic regression is a statistical model commonly used for sol-ving classification problems. Maximum likelihood is used train the model parameters. When data from two classes is linearly separable, maximum likelihood is ill-posed for logistic regres-sion. To address this problem as well as to handle over-fitting issues, regularization is considered. A regularization coefficient is used to balance the trade-off between model complexity and data fit. Cross-validation is commonly used to determine this parameter. In this paper, we develop a regularization fra-

mework for logistic regression using Jeffreys prior, which is free of any tuning parameters. Our experiments show that our pro-posed method achieved promising results in comparison with recent well-known regularization techniques.

TUE-III: SIGNAL PROCESSING FOR COMMUNICA-TIONS

A Finite Moving Average Test for Transient Change De-tection in GNSS Signal Strength MonitoringDaniel Egea-Roca (Universitat Autònoma de Barcelona, Spain); Gonzalo Seco-Granados (Universitat Autonoma de Barcelona, Spain); José A. López-Salcedo (Universitat Autònoma de Barce-lona, Spain); H. Vincent Poor (Princeton University, USA)

Due to the increasing interest in Global Navigation Satellite Systems (GNSS) for safety-critical applications, one of the ma-jor challenges to be solved is the provision of integrity to urban environments. In the past years, it has been noted that to do so, the integrity of the received signal must be analyzed with the aim of detecting any local effect disturbing the GNSS sig-nal. Moreover, the detection of such disturbing effects must be done with a bounded delay. This is desirable because the presence of any local effect may cause large position errors. This work addresses the signal integrity problem as a transient change detection problem by proposing a stopping time based on a Finite Moving Average. The statistical performance of this stopping time is investigated and compared, in the context of multipath detection relying on the C/N0 monitoring, to diffe-rent methods available in the literature. Numerical results are presented in order to assess their performance.

Statistical Analysis and Optimization of FFR/SFR-aided OFDMA-based Multi-cellular NetworksJan Garcia-Morales, Guillem Femenias and Felip Riera-Palou (University of the Balearic Islands, Spain)

Interference coordination techniques are incorporated in OFDMA-based multi-cellular networks allowing near univer-sal frequency reuse while preserving reasonably high spectral efficiencies over the whole coverage area. Two very represen-tative strategies are fractional frequency reuse (FFR) and soft frequency reuse (SFR), which are deemed to play a key role in

63


current and next generation networks. This paper presents an statistical characterization of FFR/SFR-aided networks that is subsequently used to optimize various operational parameters. The proposed design is capable of trading off throughput per-formance and fairness by suitably dimensioning the inner and outer cellular areas, the frequency allocation to each of these regions and the corresponding transmit power.

Analog joint source channel coding over MIMO fading channels with imperfect CSIJosé P González-Coma, Pedro Suárez-Casal, Óscar Fresnedo and Luis Castedo (University of A Coruña, Spain)

In this work, analog Joint Source Channel Coding (JSCC) tech-niques are considered for the transmission of independent data over fading Multiple Input Multiple Output (MIMO) channels assuming imperfect Channel State Information (CSI). In general, analog JSCC schemes require channel knowledge for both the design of the receive filter and the optimization of the encoder parameters. The inaccuracy of this information leads to a severe performance degradation. Robust strategies are required to mi-tigate the impact of the imperfect CSI. In this work, a statistical characterization of the channel estimation error is employed to design a robust linear MMSE receiver and to select adequate encoder parameters. Simulation results are presented to eva-luate the performance loss due to imperfect CSI and to illustrate the gain of the proposed robust strategy. In addition, the opti-mal distortion-cost tradeoff with imperfect CSI is determined.

A new approach for solving anti-jamming games in stochastic scenarios as pursuit-evasion gamesJuan Parras and Jorge del Val (Universidad Politécnica de Ma-drid, Spain); Santiago Zazo (Universidad Politécnica Madrid, Spain); Javier Zazo (Universidad Politécnica de Madrid, Spain); Sergio Valcarcel Macua (Universidad Politecnica de Madrid (UPM), Spain)

We solve a communication problem between a UAV and a set of relays, in the presence of a jamming UAV, using differential game theory tools. The standard solution involves a set of cou-pled Bellman equations which are hard to solve. We propose a new approach in which this kind of games can be approxima-ted as pursuit-evasion games. The problem is posed in terms of optimizing capacity and it is approximated as a zero-sum,

pursuit-evasion game. This game is solved using a set of diffe-rential equations known as Isaacs equations and simulations are run in order to validate the results.

Weighted Sum Rate Maximization of MISO Interference Broadcast Channels via Difference of Convex Functions Programming: A Large System AnalysisWassim Tabikh and Dirk Slock (EURECOM, France); Yi Yuan-Wu (Orange Labs, France)

The weighted sum rate (WSR) maximizing linear precoder algo-rithm is studied in large correlated multiple-input single-output (MISO) interference broadcast channels (IBC). We consider an iterative WSR design via successive convex approximation as in [1], [2] and [3], focusing on the version in [3]. We propose an asymptotic approximation of the signal-to-interference plus noise ratio (SINR) at every iteration. Simulations show that the asymptotic approximations are accurate.

Recursive End-To-End Distortion Estimation for Error-Resilient Adaptive Predictive Compression SystemsSina Zamani (University of California Santa Barbara, USA); Te-jaswi Nanjundaswamy and Kenneth Rose (University of Califor-nia, Santa Barbara, USA)

Linear prediction is widely used in speech, audio and video coding systems. Predictive coders often operate over unreliable channels or networks prone to packet loss, wherein errors pro-pagate through the prediction loop and may catastrophically degrade the reconstructed signal at the decoder. To mitigate this problem, end-to-end distortion (EED) estimation, accoun-ting for error propagation and concealment at the decoder, has been developed for video coding, and enables optimal rate-distortion (RD) decisions at the encoder. However, this approach was limited to the video coder’s simple setting of a single tap constant coefficient temporal predictor. This paper considerably generalizes the framework to account for: i) high order prediction filters, and ii) filter adaptation to local signal statistics. Specifically, we propose to simultaneously track the decoder statistics of the reconstructed signal and the predic-tion parameters, which enable effective estimation of the ove-rall EED. We first demonstrate the accuracy of the EED estimate in comparison to extensive simulation of transmission through a lossy network. Finally, experimental results demonstrate how

64


this EED estimate can be leveraged, by an encoder with short and long term linear prediction, to improve RD decisions and achieve major performance gains.

Study of Statistical Robust Closed Set Speaker Identifi-cation with Feature and Score-Based FusionMusab Al-Kaltakchi and Wai Lok Woo (Newcastle Universi-ty, United Kingdom); Satnam Dlay (University of Newcastle, United Kingdom); Jonathon Chambers (Newcastle University, United Kingdom)

In this paper, the statistical combination of Power Normalization Cepstral Coefficient (PNCC) and Mel Frequency Cepstral Coeffi-cient (MFCC) features in robust closed set speaker identification is studied. Feature normalization and warping together with late score-based fusion are also exploited to improve perfor-mance in the presence of channel and noise effects. In addition, combinations of score and feature-based approaches are consi-dered with early and/or late fusion; these systems use different feature dimensions (16, 32). A 4th order G.712 type IIR filter is employed to represent handset degradation in the channel. Simulation studies based on the TIMIT database confirm the improvement in Speaker Identification Accuracy (SIA) through the combination of PNCC and MFCC features in the presence of handset and Additive White Gaussian Noise (AWGN) effects.

Analog Distributed Coding of Correlated Sources for Fading Multiple Access ChannelsÓscar Fresnedo, Pedro Suárez-Casal and Luis Castedo (Uni-versity of A Coruña, Spain); Javier Garcia-Frias (University of Delaware, USA)

In this work, we address the analog transmission of correlated information over fading Multiple Access Channels (MACs) using analog Joint Source Channel Coding (JSCC). We consider module-like mappings to encode the source data and the utilization of different orthogonal access schemes. The user information is indi-vidually mapped at each transmitter and the receiver exploits the source correlation and the properties of the module mappings to decode the received symbols. We also propose a Maximum-A-Posteriori (MAP) method which achieves similar performance to that of the optimal decoding with significantly lower complexity. The obtained results confirm the potential of analog JSCC techni-ques to transmit correlated data over fading MACs.

Generalized Integration techniques for high-sensitivity GNSS receivers affected by oscillator phase noiseDavid Gómez-Casco and José A. López-Salcedo (Universitat Autònoma de Barcelona, Spain); Gonzalo Seco-Granados (Uni-versitat Autonoma de Barcelona, Spain)

This paper addresses the use of generalized correlations in the context of High-Sensitivity Global Navigation Satellite System (HS-GNSS) receivers. Generalized correlations are also referred to as post-detection integration (PDI) techniques or simply as non-coherent integration methods. The contributions of this work are twofold. On the one hand, a novel PDI method is presented, which improves the performance of methods found in the literature for small errors of the frequency offset. On the other hand, an exhaustive comparative performance analysis is provided between the proposed technique and the existing ones in the presence of phase noise coming from the local osci-llator. To this end, results have been obtained for two different clocks, namely a temperature compensated crystal oscillator (TCXO) and an oven-controlled crystal oscillator (OCXO). In both cases, the proposed technique outperforms the existing ones.

Measurement Matrix Design For Compressive Sensing With Side Information at the EncoderPingfan Song (University College London); Joao Mota (Univer-sity College London, United Kingdom); Nikos Deligiannis (Vrije Universiteit Brussel, Belgium); Miguel Raul Dias Rodrigues (Uni-versity College London, United Kingdom)

We study the problem of measurement matrix design for Com-pressive Sensing (CS) when the encoder has access to side information, a signal analogous to the signal of interest. We propose a novel design scheme to incorporate the side in-formation into the acquisition process in order to reduce the number of encoding measurements, while still allowing perfect signal reconstruction at the decoder. We analyse the recons-truction performance of the resulting CS system assuming the decoder reconstructs the signal via Basis Pursuit. Finally, we leverage Gaussian width related tools to establish a tight theoretical bound for the number of required measure-ments. Extensive numerical experiments not only validate our approach, but also demonstrate that it requires fewer measure-ments than alternative designs, such as an i.i.d. Gaussian matrix.

65


TUE-III SS: STATISTICAL SIGNAL PROCESSING AND LEARNING IN SMART GRID

An Approximation Algorithm for future wind scenariosBita Analui (Arizona State University & Arizona State University, USA); Anna Scaglione (Arizona State University, USA)

The success of stochastic optimization methods in helping power systems operations cope with increasing penetration of wind and solar power rests on the effective construction of scenario trees as approximations of the true probability space of the renewable power stochastic process. In this work, while analyzing the statistical properties of wind power samples of a wind farm in Washington state (WA) for years 2012-2014, we first identify gaps that exist in traditional modeling approaches and propose a possible solutions. The key idea we propose is to view scenario tree generation as a form of compression of the wind power trajectories, sidestepping completely the mo-del selection approach. We argue that to retain key features of the high order statistics, one can directly quantize realizations over the optimization horizon or to perform the quantization in a finite subspace of Morlet-wavelets. In fact, the wind time series tend to be characterized by the same set of localized features and has a sparse representation over the Morlet basis. We analyze our scenario reduction performance compared to time domain methods.

Online Learning and Pricing for Demand Response in Smart Distribution NetworksSevi Baltaoglu, Lang Tong and Qing Zhao (Cornell University, USA)

The problem of online learning of consumer response to retail pricing of electricity in a distribution network is considered. In a two-settlement market, the retailer who sets the retail price is exposed to risks from the stochastic response of its consumers and the real-time price fluctuation in the wholesale market. The optimal price maximizing the expected profit is a function of consumer’s response to prices, and any pricing scheme un-der unknown demand model accumulates regret measured by the difference between the total expected profit of the retailer under known and unknown demand model.This paper presents an online learning approach to dynamic pri-cing aimed at minimizing the regret of the retailer for consumers

with unknown Markov jumped affine demand. It is shown that the regret of the proposed policy has the lowest order of regret growth characterized by the square-root of the learning horizon.

Decentralized MMSE Attacks in Electricity GridsIñaki Esnaola (University of Sheffield, United Kingdom); Samir M. Perlaza (INRIA, France); H. Vincent Poor (Princeton Universi-ty, USA); Oliver Kosut (Arizona State University, USA)

Decentralized data injection attack constructions with mi-nimum-mean-square-error state estimation is studied in a game-theoretic setting. Within this framework, the interaction between the network operator and the set of attackers, as well the interactions among the attackers, are modeled by a game in normal form. A novel utility function that captures the trade-off between the maximum distortion that an attack can introduce and the probability of the attack being detected by the network operator is proposed. Under the assumption that the state variables can be modelled as a multivariate Gaus-sian random process, it is shown that the resulting game is a potential game. The cardinality of the corresponding set of Nash Equilibria (NEs) of the game is analyzed. It is shown that attackers can agree on a data injection vector construction that achieves the best trade-off between distortion and detection probability by sharing only a limited number of bits offline. In-terestingly, this vector construction is also shown to be an NE of the resulting game.

A Graphical Approach to Quickest Outage Localization in Power GridsJavad Heydari and Ali Tajer (Rensselaer Polytechnic Institute, USA)

Line outage detection and localization play pivotal roles in contingency analysis, power flow optimization, and situational awareness delivery in power grids. Hence, agile detection and localization of line outages enhances the efficiency of opera-tions and their resilience against cascading failures. This paper proposes a stochastic graphical framework for localizing line outages. This framework capitalizes on the correlation among the measurements generated across the grid, where the co-rrelation is induced by the connectivity topology of the grid. By formalizing a proper correlation model, this paper designs data-adaptive coupled data-acquisition and decision-making processes for the quickest localization of the line outages.

66


6

This leads to efficient outage localization by using only partial measurements and is shown to outperform the existing dimen-sionality reduction methods.

Estimating Treatment Effects in Demand ResponsePan Li and Baosen Zhang (University of Washington, USA)

Demand response is designed to motivate electricity customers to modify their loads at critical time periods. Accurately estima-ting customers response to demand response signals is crucial to the success of these programs. In this paper, we consider signals in demand response programs as a treatment to the customers and estimate the average treatment effect. Specifi-cally, we adopt the linear regression model and derive several consistent linear regression estimators. From both synthetic and real data, we show that including more information about the customers does not always improve estimation accuracy: the interaction between the side information and the demand response signal must be carefully modeled. We then apply the so-called modified covariate method to capture these interac-tions and show it can strike a balance between having more data and model correctness. Our results are validated using data collected by Pecan Street.

Dynamic Decentralized Voltage Control for Power Dis-tribution NetworksHao Jan Liu (University of Illinois at Urbana Champaign, USA); Wei Shi and Hao Zhu (University of Illinois at Urbana-Cham-paign, USA)

Voltage regulation in power distribution networks has been in-creasingly challenged by the integration of volatile and intermit-tent distributed energy resources (DERs). These resources can also provide limited reactive power resources that can be used to optimize the network-wide voltage. A decentralized voltage control scheme based on the gradient-projection (GP) method is adopted to minimize a voltage mismatch error objective under limited reactive power. Coupled by the power network flow, the local voltage directly provides the instantaneous gradient in-formation. This paper aims to quantify the performance of this decentralized GP-based voltage control under dynamic system operating conditions modeled by an autoregressive process. Our analysis offers the tracking error bound on the instanta-neous solution to the transient optimizer. Under stochastic

processes that have bounded iterative changes, the results can be extended to general constrained dynamic optimization pro-blems with smooth strongly convex objective functions. Nume-rical tests have been perform to validate our analytical results using a 21-bus network.

Learning to Infer: a New Variational Inference Ap-proach for Power Grid Topology IdentificationYue Zhao (Stony Brook University, USA); Jianshu Chen (Micro-soft Research, Redmond, WA & Microsoft, USA); H. Vincent Poor (Princeton University, USA)

Identifying arbitrary topologies of power networks is a compu-tationally hard problem due to the number of hypotheses that grows exponentially with the network size. A new variational inference approach is developed for efficient marginal inference of every line status in the network. Optimizing the variational model is transformed to and solved as a discriminative learning problem. A major advantage of the developed learning based approach is that the labeled data used for learning can be ge-nerated in an arbitrarily large amount at very little cost. As a re-sult, the power of offline training is fully exploited to offer effec-tive real-time topology identification. The proposed methods are evaluated in the IEEE 30-bus system. With relatively simple variational models and only an undercomplete measurement set, the proposed methods already achieve reasonably well performance in identifying arbitrary power network topologies.

POSTER SESSION 6& SPECIAL SESSIONSWednesday June 29, 10:30 - 12:00

WED: APPLICATIONS (BIOMEDICAL, ENERGY, SECURITY)

Sparse Genomic Structural Variant Detection: Exploi-ting Parent-Child Relatedness for Signal RecoveryMario Banuelos, Rubi Almanza, Lasith Adhikari, Roummel Mar-cia and Suzanne Sindi (University of California, Merced, USA)

Structural variants (SVs) – rearrangements of an individuals’ genome – are an important source of heterogeneity in human

67


and other mammalian species. Typically, SVs are identified by comparing fragments of DNA from a test genome to a known reference genome, but errors in both the sequencing and the noisy mapping process contribute to high false positive rates. When multiple related individuals are studied, their relatedness offers a constraint to improve the signal of true SVs. We deve-lop a computational method to predict SVs given genomic DNA from a child and both parents. We demonstrate that enforcing relatedness between individuals and constraining our solution with a sparsity-promoting l1 penalty (since SV instances should be rare) results in improved performance. We present results on both simulated genomes as well as two-sequenced parent-child trios from the 1000 Genomes Project.

Multiscale Time Irreversibility to predict Orthostatic In-tolerance in Older PeopleRaquel Cervigon (Universidad de Castilla-La Mancha, Spain)

Orthostatic intolerance (OI) is a clinical syndrome, which is characterized by symptoms and loss of consciousness before impeding syncope and that it has been reported that is caused by orthostatic hypotension (OH). Healthy subjects and people with diseases can often be distinguished by the complexity of their physiological activity. The phenomenon of irreversibility is specific for non-equilibrium system and its presence in hemo-dynamic variables results from the complexity of cardiovascular control system typical for the healthy human. This study is fo-cused to quantify the effect of the OI on the time irreversibility of the heart rate (HR), cardiac output (CO) and systolic blood pressure (SBP) time series, during six-minutes walking distance test in symptomatic older people, who were compared by the following multiscale time irreversibility indexes: Porta’s (Pm%), Guzik’s (Gm%) and Euclidean distance (Dm). We analyzed 65 older subjects, of whom 42 were women. Results show higher indexes in non-OI groups, especially during the descent phase and in the subsequent passive phase. This study shows the irreversibility indexes as useful measure to extract non-linearity properties in hemodynamic parameters in order to find out di-fferences during orthostastism.

Pose estimation of cyclic movement using inertial sen-sor dataKjartan Halvorsen and Fredrik Olsson (Uppsala University, Sweden)

We propose a method for estimating the rotation and displa-cement of a rigid body from inertial sensor data based on the assumption that the movement is cyclic in nature, meaning that the body returns to the same position and orientation at regular time intervals. The method builds on a parameteriza-tion of the movement by sums of sinusoids, and the amplitude and phase of the sinusoids are estimated from the data using measurement models with Gaussian noise. The maximum likelihood estimate is then equivalent to a weighted nonlinear least squares estimate. The performance of the method is de-monstrated on simulated data and on experimental data.

Genomic Transcription Regulatory Element Location Analysis via Poisson weighted LASSOXin Jiang (Duke University); Patricia Reynaud-Bouret (University of Nice, France); Vincent Rivoirard (Paris Dauphine University, France); Laure Sansonnet (INRA – AgroParisTech, France); Re-becca Willett (University of Wisconsin-Madison, USA)

The distances between DNA Transcription Regulatory Elements (TRE) provide important clues to their dependencies and function within the gene regulation process. However, the locations of tho-se TREs as well as their cross distances between occurrences are stochastic, in part due to the inherent limitations of Next Gene-ration Sequencing methods used to localize them, in part due to biology itself. This paper describes a novel approach to analyzing these locations and their cross distances even at long range via a Poisson random convolution. The resulting deconvolution problem is ill-posed, and sparsity regularization is used to offset this cha-llenge. Unlike previous work on sparse Poisson inverse problems, this paper adopts a weighted LASSO estimator with data-depen-dent weights calculated using concentration inequalities that ac-count for the Poisson noise. This method exhibits better squared error performance than the classical (unweighted) LASSO both in theoretical performance bounds and in simulation studies, and can easily be computed using off-the-shelf LASSO solvers.

Design of Data-Injection Adversarial Attacks against Spatial Field DetectorsRoberto López-Valcarce (Universidad de Vigo, Spain); Daniel Romero (University of Minnesota, USA)

Data-injection attacks on spatial field detection corrupt a sub-set of measurements to cause erroneous decisions.

68


We consider a centralized decision scheme exploiting spatial field smoothness to overcome lack of knowledge on system parameters such as noise variance. We obtain closed-form expressions for system performance and investigate strategies for an intruder injecting false data in a fraction of the sensors in order to reduce the probability of detection. The problem of determining the most vulnerable subset of sensors is also analyzed.

Security of (n,n)-threshold audio secret sharing sche-mes encrypting audio secretsYuto Miura and Yodai Watanabe (University of Aizu, Japan)

Secret sharing is a method of encrypting a secret into multi-ple pieces called shares so that only qualified sets of shares can be employed to reconstruct the secret. Audio secret sha-ring (ASS) is an example of secret sharing whose decryption can be performed by human ears. This paper examines the security of (n,n)-threshold ASS schemes encrypting audio se-crets by estimating the mutual information between secret and shares.

Secure Estimation Against Complex-valued AttacksArash Mohammadi (Concordia University, Canada); Konstanti-nos Plataniotis (University of Toronto, Canada)

Motivated by recent evolution of cutting-edge sensor techno-logies with complex-valued measurements, the paper exposes complex-valued (non-circular) false data injection attacks. We propose an attack model where an adversary applies widely-linear transformations on the sensor measurements to introduce correlations between the real and imaginary parts of the reported observations. Existing state estimators and attack detectors assume the measurements to have statistical properties similar to real-valued signals making them highly vulnerable to such complex-valued attacks. As a countermea-sure, we propose to transform the attack detection problem into the problem of comparing the statistical distance between the Gaussian representation of the innovation sequence under attack and its counterpart with the optimal profile. Our Monte Carlo simulations illustrate the destructive nature of complex-valued attacks and validate the effectiveness of the proposed detection concept.

A skewed exponential power distribution to measure value at risk in electricity marketsAymeric Thibault (Universite Paris-Sud, France); Pascal Bondon (LSS CNRS, France)

Interest in risk measurement for spot price has increased sin-ce the worldwide deregulation and liberalization of electricity started in the early 90’s. This paper focused on quantifying risk for the Nordic Power Exchange (Nord Pool) system price. Our analysis is based on a GARCH approach with skewed exponen-tial power innovations to model the stochastic component of the system price. Value-at-risk backtesting procedures are conduc-ted and our model performance is compared to commonly used distributions in risk measurement. We show that the skewed exponential power distribution outperforms the competitors for the upside risk (95%, 97.5% and 99% Value-at-risk), which is of high interest as electricity spot prices are positively skewed.

Accelerometer calibration using sensor fusion with a gyroscopeFredrik Olsson (Uppsala University, Sweden); Manon Kok (Linkö-ping University, Sweden); Kjartan Halvorsen and Thomas B. Schön (Uppsala University, Sweden)

In this paper, a calibration method for a triaxial accelerometer using a triaxial gyroscope is presented. The method uses a sen-sor fusion approach, combining the information from the acce-lerometers and gyroscopes to find an optimal calibration using Maximum likelihood. The method has been tested by using real sensors in smartphones to perform orientation estimation and verified through Monte Carlo simulations. In both cases, the method is shown to provide a proper calibration reducing the effect of sensor errors and improving orientation estimates.

WED: DETECTION AND ESTIMATION THEORY III

Joint range estimation and spectral classification for 3D scene reconstruction using multispectral Lidar waveformsYoann Altmann and Aurora Maccarone (Heriot-Watt University, United Kingdom); Aongus McCarthy (Heriot Watt University, Uni-ted Kingdom); Gerald Buller (Heriot-Watt University, United King-dom); Steve McLaughlin (Heriot Watt University, United Kingdom)

69


This paper presents a new Bayesian classification method to analyse remote scenes sensed via multispectral Lidar measu-rements. To a first approximation, each Lidar waveform mainly consists of the temporal signature of the observed target, which depends on the wavelength of the laser source con-sidered and which is corrupted by Poisson noise. By sensing the scene at several wavelengths, we expect a more accurate target range estimation and a more efficient spectral analysis of the scene. Thanks to its spectral classification capability, the proposed hierarchical Bayesian model, coupled with an efficient Markov chain Monte Carlo algorithm, allows the esti-mation of depth images together with reflectivity-based scene segmentation images. The proposed methodology is illustrated via experiments conducted with real multispectral Lidar data.

Regularised Estimation of 2D-Locally Stationary Wave-let ProcessesAlexander Gibberd and James D B Nelson (University College London, United Kingdom)

Locally Stationary Wavelet processes provide a flexible way of describing the time/space evolution of autocovariance structu-re over an ordered field such as an image/time-series. Classica-lly, estimation of such models assume continuous smoothness of the underlying spectra and are estimated via local kernel smoothers. We propose a new model which permits spectral jumps, and suggest a regularised estimator and algorithm which can recover such structure from images. We demons-trate the effectiveness of our method in a synthetic experiment where it shows desirable estimation properties. We conclude with an application to real images which illustrate the quali-tative difference between the proposed and previous methods.

Fast filtering with new sparse transition Markov chainsIvan Gorynin and Emmanuel Monfrini (Institut Telecom, Telecom SudParis, France); Wojciech Pieczynski (Télécom SudParis, France)

We put forward a novel Markov chain approximation method with regard to the filtering problem. The novelty consists in making use of the sparse grid theory which deals with the curse of dimensionality. Our method imitates the marginal distribution of the latent continuous process with a discrete probability distribution on a sparse grid. The grid points may be seen as the states of a Markov chain which we construct to

imitate the whole process. The transition probabilities are then chosen to preserve the joint moments of the underlying conti-nuous process. We provide a simulation study on a multivariate stochastic volatility filtering problem to compare the proposed methodology with a similar technique and the particle filtering.

On the estimation of many closely spaced complex sinusoidsRoland Jonsson (SAAB Electronic Defence Systems, Sweden)

The estimation of sinusoidal signals is a very well researched area, and it is well known that two signals can be resolved well for frequency separation below the Fourier resolution at high enough signal to noise ratio. However, in the case of many closely spaced sinusoids estimation is impaired for separations well above the Fourier resolution, and the dependence on sig-nal to noise ratio is involved. The problem is analyzed by con-sidering the Hessian of the log-likelihood function. When there is some direction in the parameter space, where the curvature of its deterministic part (i.e. the Fisher information matrix) is less than the curvature of its stochastic part, this is an indica-tion of problems for correct estimation. An expression for the probability of this to occur is presented.

A Multiscale Approach for Tensor DenoisingAlp Ozdemir and Mark Iwen (Michigan State University, USA); Selin Aviyente (Electrical and Computer Engineering, Michigan State University, MI, USA)

As higher-order datasets become more common, researchers are primarily focused on how to analyze and compress them. Howe-ver, the most common challenge encountered in any type of data, including tensor data, is noise. Furthermore, the methods developed for denoising vector or matrix type datasets cannot be applied directly to higher-order datasets. This motivates the development of denoising methods for tensors. In this paper, we propose the use of a multiscale approach for denoising general higher-order datasets. The proposed approach works by decomposing the higher-order data into subtensors, and then denoises the subtensors by recursively exploiting filtered residuals. The method is validated on both hyperspectral image and brain functional connectivity network data.

Two-Stage Estimation after Parameter SelectionTirza Routtenberg (Ben Gurion University of the Negev, Israel)

70


In many practical multiparameter estimation problems, no a-priori information exists regarding which parameters are more relevant within a group of candidate unknown parameters. This paper considers the estimation of a selected “parameter of inter-est”, where the selection is conducted according to a data-based selection rule, Psi. The selection process introduces a selection bias and creates coupling between decoupled parameters. We propose a two-stage data-acquisition approach that can remo-ve the selection bias and improve estimation performance. We derive a two-stage Cramer-Rao-type bound on the post-selection mean squared error (PSMSE) of any Psi-unbiased estimator, whe-re the Psi-unbiasedness is in the Lehmann sense. In addition, we present the two-stage post-selection maximum-likelihood (PSML) estimator. The proposed Psi-Cramer-Rao bound (CRB), PSML es-timator and other existing estimators are examined for a linear Gaussian model, which is widely used in clinical research.

An order fitting rule for optimal subspace averagingIgnacio Santamaria (University of Cantabria, Spain); Louis Scharf (Colorado State, USA); Chris Peterson and Michael Kirby (Colorado State University, USA); Joseph M. Francos (Ben Gu-rion University, Israel)

The problem of estimating a low-dimensional subspace from a collection of experimentally measured subspaces arises in many applications of statistical signal processing. In this paper we address this problem, and give a solution for the avera-ge subspace that minimizes an extrinsic mean-squared error, defined by the squared Frobenius norm between projection matrices. The solution automatically returns the dimension of the optimal average subspace, which is the novel result of the paper. The proposed order fitting rule is based on thresholding the eigenvalues of the average projection matrix, and thus it is free of penalty terms or other tuning parameters commonly used by other rank estimation techniques. Several numerical examples demonstrate the usefulness and applicability of the proposed criterion, showing how the dimension of the average subspace captures the variability of the measured subspaces.

Block-Wise MAP Inference for Determinantal Point Pro-cesses with Application to Change-Point DetectionMartin Jinye Zhang (Stanford University, P.R. China); Zhijian Ou (Tsinghua University, P.R. China)

Studies of change-point detection (CPD) often focus on develo-ping similarity metrics that quantify how likely a time point is to be a change point. After that, the process of selecting true change points among those high-score candidates is less well-studied. This paper proposes a new CPD method that uses determinantal point processes to model the process of change-point selection. Particularly, this paper explores the special kernel structure arose in such modelling, i.e. almost block diagonal, and show that the maximum a posteriori task, requiring at least O(N2.4) in general, can be achieved using O(N)under such structure. The resulting al-gorithm, BwDPP-MAP and BwDppCpd, are empirically validated through simulation and five real-world data experiments.

WED SS: OPTIMIZATION AND SIMULATION FOR IMAGE PROCESSING

Spatial regularization for nonlinear unmixing of hy-perspectral data with vector-valued functionsRita Ammanouil (University of the Cote d’Azur, Lebanon); André Ferrari and Cédric Richard (Université de Nice Sophia-Antipolis, Fran-ce); Jean-Yves Tourneret (University of Toulouse & ENSEEIHT, France)

This communication introduces a new framework for incorpo-rating spatial regularization into a nonlinear unmixing proce-dure dedicated to hyperspectral data. The proposed model pro-motes smooth spatial variations of the nonlinear component in the mixing model. The spatial regularizer and the nonlinear contributions are jointly modeled by a vector-valued function that lies in a reproducing kernel Hilbert space (RKHS). The un-mixing problem is strictly convex and reduces to a quadratic programming problem. Simulations on synthetic data illustrate the effectiveness of the proposed approach.

A Regularized Sparse Approximation Method for Hy-perspectral Image ClassificationLeila Belmerhnia (CRAN, Université de Lorraine, CNRS, France); El-Hadi Djermoune (Université de Lorraine & CRAN UMR 7039 CNRS, France); David Brie (CRAN, Nancy Université, CNRS, Fran-ce); Cédric Carteret (LCPME, France)

This paper presents a new technique for hyperspectral images classification based on simultaneous sparse approximation.

71


The proposed approach consists in formulating the problem as a convex multi-objective optimization problem which incorporates a term favoring the simultaneous sparsity of the estimated coeffi-cients and a term enforcing a regularity constraint along the rows of the coefficient matrix. We show that the optimization problem can be solved efficiently using FISTA (Fast Iterative Shrinkage-Thresholding Algorithm). This approach is applied to a wood wastes classification problem using NIR hyperspectral images.

Unbiased Injection of Signal-Dependent Noise in Va-riance-Stabilized RangeLucas Borges (UFSC, Brazil); Marcelo A. C. Vieira (University of São Paulo, Brazil); Alessandro Foi (Tampere University of Tech-nology, Finland)

The design, optimization, and validation of many image pro-cessing or image-based analysis systems often requires testing of the system performance over a dataset of images corrupted by noise at different signal-to-noise ratio regimes. A noise-free ground-truth image may not be available, and different SNRs are simulated by injecting extra noise into an already noisy image. However, noise in real-world systems is typically signal-dependent, with variance determined by the noise-free image. Thus, also the noise to be injected shall depend on the unk-nown ground-truth image. To circumvent this issue, we con-sider the additive injection of noise in variance-stabilized ran-ge, where no previous knowledge of the ground-truth signal is necessary. Specifically, we design a special noise-injection operator that prevents the errors on expectation and variance that would otherwise arise when standard variance-stabilizing transformations are used for this task. Thus, the proposed ope-rator is suitable for accurately injecting signal-dependent noise even to images acquired at very low counts.

Bayesian Multifractal Analysis of Multi-temporal Ima-ges Using Smooth PriorsSébastien Combrexelle (University of Toulouse & IRIT-ENSEEIHT, Fran-ce); Herwig Wendt (University of Toulouse & IRIT – ENSEEIHT, CNRS, France); Jean-Yves Tourneret (University of Toulouse & ENSEEIHT, France); Patrice Abry (Ecole Normale Superieure, Lyon, France); Ste-ve McLaughlin (Heriot Watt University, United Kingdom)

Texture analysis can be conducted within the mathematical framework of multifractal analysis (MFA) via the study of the

regularity fluctuations of image amplitudes. Successfully used in various applications, however MFA remains limited to the independent analysis of single images while, in an increasing number of applications, data are multi-temporal. The present contribution addresses this limitation and introduces a Baye-sian framework that enables the joint estimation of multifractal parameters for multi-temporal images. It builds on a recently proposed Gaussian model for wavelet leaders parameterized by the multifractal attributes of interest. A joint Bayesian mo-del is formulated by assigning a Gaussian prior to the second derivatives of time evolution of the multifractal attributes asso-ciated with multi-temporal images. This Gaussian prior ensures that the multifractal parameters have a smooth temporal evo-lution. The associated Bayesian estimators are then approxi-mated using a Hamiltonian Monte-Carlo algorithm. The bene-fits of the proposed procedure are illustrated on synthetic data.

Robust hyperspectral unmixing accounting for residual componentsAbderrahim Halimi (Heriot-Watt University, United Kingdom); Paul Honeine (Université de Rouen, France); José Bioucas-Dias (Instituto Superior Técnico, Portugal)

This paper presents a new hyperspectral mixture model jointly with a Bayesian algorithm for supervised hyperspectral un-mixing. Based on the residual component analysis model, the proposed formulation assumes the linear model to be corrup-ted by an additive term that accounts for mismodelling effects (ME). The ME formulation takes into account the effect of outliers, the propagated errors in the signal processing chain and copes with some types of endmember variability (EV) or nonlinearity (NL). The known constraints on the model para-meters are modeled via suitable priors. The resulting posterior distribution is optimized using a coordinate descent algorithm which allows us to compute the maximum a posteriori estima-tor of the unknown model parameters. The proposed model and estimation algorithm are validated on both synthetic and real images showing competitive results regarding the quality of the inferences and the computational complexity when com-pared to the state-of-the-art algorithms.

Analysis Dictionary Learning for Scene ClassificationWen Tang, Ives Rey Otero and Hamid Krim (North Carolina State University, USA)

72


This paper presents a new framework for scene classification based on an analysis dictionary learning approach. Despite their tremendous success in various image processing tasks and so called sparse coding, synthesis-based and analysis-based sparse models fall short in classification tasks. This is partly due to the linear dependence of the dictionary atoms. In this work, we aim at improving classification performances by compensating for such dependence. The proposed methodology consists in grou-ping the atoms of the dictionary using clustering methods. This allows to sparsely model images from various scene classes and use such a model for classification. Experimental evidence shows the benefit of such an approach. Finally, we propose supervised way to train the baseline representation for each class-specific dictionary, and achieve multiple classification by finding the minimum distance between the learned baseline representation and the data’s sub-dictionary representation. The achieved re-sults in scene classification are better than the state-of-the-art.

Weakly-supervised Analysis Dictionary Learning with Cardinality ConstraintsZeyu You, Raviv Raich, Xiaoli Fern and Jinsub Kim (Oregon State University, USA)

In synthesis dictionary learning, data is compactly represented as sparse combination over a dictionary. In analysis dictionary learning, a sparsifying analysis dictionary is learned from data. In this paper, we consider the problem of analysis dictionary learning under the weak supervision setting. We introduce a discriminative probabilistic model and present a novel ap-proach to enforce sparsity using probabilistic cardinality cons-traints. A detailed derivation of the expectation maximization procedure for maximum likelihood estimation with a compu-tationally efficient E-step implementation is introduced. We illustrate the performance of the model on synthetic data.

73


A

Abry, Patrice Tue-1b.4 50

Wed-1c SS.4 71

Adhikari, Lasith Wed-1a.1 66

Adve, Raviraj Tue-2c SS.4 57

Aeron, Shuchin Tue-3b.4 61

Tue-3b.7 61

Akhondi-Asl, Hojjat Mon-2a.1 42

Al-Kaltakchi, Musab Tue-3a.7 64

Al-Naffouri, Tareq Y. Tue-1a.2 47

Al-sharoa, Esraa Tue-2a.1 55

Almanza, Rubi Wed-1a.1 66

Alouini, Mohamed-Slim Mon-1c SS.3 40

Tue-1a.2 47

Altmann, Yoann Tue-1b.7 51

Wed-1b.1 68

Ammanouil, Rita Wed-1c SS.1 70

Analui, Bita Tue-3c SS.1 65

Andler, Sébastien Tue-1c SS.2 52

Arroyo, Jesús Mon-1a.1 36

Asif, Anum Tue-1b.8 51

Auguin, Nicolas Mon-1c SS.1 40

Aviyente, Selin Mon-1b.6 39

Tue-2a.1 55

Wed-1b.5 69

Ayllón, David Tue-2a.6 56

B

Ba, Demba Tue-2d SS.4 59

Babu, Prabhu Mon-2b SS.2 44

Bacharach, Lucien Tue-2c SS.1 57

Baltaoglu, Sevi Tue-3c SS.2 65

Balzano, Laura Mon-2a.8 43

Banuelos, Mario Wed-1a.1 66

Bar, Shahar Tue-2c SS.2 57

Barbos, Andrei Tue-3b.6 61

Basu, Prabahan Tue-2c SS.6 58

Batatia, Hadj Mon-2c SS.2 45

Bazot, Michaël Mon-1a.2 36

Bazzi, Ahmad Tue-1a.1 47

Belmerhnia, Leila Wed-1c SS.2 70

Ben Atitallah, Ismail Tue-1a.2 47

Benazza, Amel Tue-1b.5 50

Bidon, Stéphanie Tue-2c SS.5 57

Bioucas-Dias, José Wed-1c SS.5 71

Bisch, Marc-Abel Tue-1a.3 47

Bondon, Pascal Mon-2b SS.5 44

Wed-1a.8 68

Boon, Paul Tue-2d SS.2 58

Borges, Lucas Wed-1c SS.3 71

Bouboulis, Pantelis Tue-3b.1 60

Boudineau, Megane Mon-1a.2 36

Bourguignon, Sébastien Mon-1a.2 36

Boyer, Rémy Tue-2b.1 53

Breloy, Arnaud Mon-2b SS.2 44

Brie, David Tue-1a.3 47

Wed-1c SS.2 70

Bugallo, Monica Mon-2c SS.3 46

Mon-2c SS.6 46

Buller, Gerald Wed-1b.1 68

C

Cai, Shu Tue-1a.4 48

Camarrone, Flavio Tue-2d SS.2 58

Cao, Ming-Yang Tue-1a.5 48

Carfantan, Hervé Mon-1a.2 36

Caron, François Mon-2a.9 43

Carrette, Evelien Tue-2d SS.2 58

Carteret, Cédric Wed-1c SS.2 70

Casarin, Roberto Mon-2c SS.5 46

Castedo, Luis Tue-3a.3 63

Tue-3a.8 64

Catt, Michael Tue-2d SS.1 58

Cervigon, Raquel Wed-1a.2 67

Chainais, Pierre Mon-2a.3 42

Tue-3b.3 60

Chambers, Jonathon Tue-3a.7 64

Chen, Bo Tue-1a.8 48

SSP 2016Author index

AUTHOR SESSION PAGE

74


Chen, Jianshu Tue-3c SS.7 66

Chiani, Marco Tue-2b.3 54

Chouzenoux, Emilie Tue-1b.5 50

Chumerin, Nikolay Tue-2d SS.2 58

Closas, Pau Mon-2b SS.4 44

Mon-2c SS.1 45

Coates, Mark Mon-1b.4 39

Mon-2a.6 43

Tue-3b.8 61

Cochran, Douglas Tue-2c SS.7 58

Cohen, Edward Tue-1b.9 51

Coleman, Todd Tue-2d SS.4 59

Coluccia, Angelo Tue-2c SS.3 57

Combrexelle, Sébastien Wed-1c SS.4 71

Conrad, Christian Mon-2a.2 42

Cortese, John Tue-3b.2 60

Costa, Facundo Mon-2c SS.2 45

Couillet, Romain Mon-1c SS.1 40

Mon-1c SS.2 40

Mon-1c SS.3 40

Coutino, Mario Tue-1a.9 49

Tue-2b.2 54

D

Dang, Hong Phuong Mon-2a.3 42

Davenport, Mark Mon-2a.7 43

Tue-2a.7 56

De Taeye, Leen Tue-2d SS.2 58

Dedecius, Kamil Tue-2a.2 55

Del Din, Silvia Tue-2d SS.1 58

del Val, Jorge Tue-3a.4 63

Deligiannis, Nikos Tue-3a.10 64

Desbouvries, François Mon-2c SS.4 46

Djermoune, El-Hadi Wed-1c SS.2 70

Djuric’, Petar Mon-2c SS.6 46

Dlay, Satnam Tue-3a.7 64

Dobigeon, Nicolas Mon-2a.9 43

Tue-3b.3 60

Dodde, Vincenzo Tue-2c SS.3 57

Donnat, Philippe Tue-1c SS.2 52

Dufor, Olivier Tue-1c SS.3 52

E

Egea-Roca, Daniel Tue-3a.1 62

El Korso, Mohammed Nabil Tue-2c SS.1 57

Eldar, Yonina Tue-2b.7 55

Elvira, Clément Tue-3b.3 60

Elvira, Víctor Mon-1a.7 37

Mon-2c SS.3 46

Tue-1b.6 51

Elzanaty, Ahmed Tue-2b.3 54

Ertin, Emre Tue-1b.1 49

Esnaola, Iñaki Tue-3c SS.3 65

F

Fan, Bo Tue-3b.4 61

Fearn, Tom Tue-3b.9 62

Femenias, Guillem Tue-3a.2 62

Fern, Xiaoli Wed-1c SS.7 72

Ferrari, André Wed-1c SS.1 70

Ferré, Guillaume Mon-1c SS.5 41

Fleury, Bernard Tue-2b.1 53

Fleury, Eric Tue-2a.3 56

Foi, Alessandro Wed-1c SS.3 71

Fortunati, Stefano Mon-2b SS.3 44

Fox, Andrew Tue-1b.2 50

Francos, Joseph Wed-1b.7 70

Fresnedo, Óscar Tue-3a.3 63

Tue-3a.8 64

G

Gadaleta, Matteo Tue-3b.5 61

Garcia-Frias, Javier Tue-3a.8 64

Garcia-Morales, Jan Tue-3a.2 62

Ghobadzadeh, Ali Tue-2c SS.4 57

Giampouras, Paris Mon-2a.4 42

Giannakis, Georgios B. Mon-1b.2 38

Mon-2a.10 44

Gibberd, Alexander Wed-1b.2 69

Gil-Pita, Roberto Tue-2a.6 56

Gini, Fulvio Mon-2b SS.3 44

Ginolhac, Guillaume Mon-2b SS.2 44

Giorgetti, Andrea Tue-2b.3 54

Giovannelli, Jean-François Mon-2a.9 43

Tue-3b.6 61

Girault, Benjamin Tue-2a.3 56

Giremus, Audrey Mon-2a.9 43

Godfrey, Alan Tue-2d SS.1 58

Gogineni, Sandeep Tue-1a.6 48

Gómez-Casco, David Tue-3a.9 64

Gonçalves, Paulo Tue-2a.3 56

González-Coma, José Tue-3a.3 63

75


Gorynin, Ivan Wed-1b.3 69

Grajal, Jesús Tue-1a.10 49

Greco, Maria Mon-2b SS.3 44

Guo, Han Mon-2a.5 43

Gusi-Amigó, Adrià Mon-2b SS.4 44

Gutierrez, David Tue-2d SS.3 59

H

Halay, Nir Mon-1a.3 36

Halimi, Abderrahim Wed-1c SS.5 71

Hall, Eric Tue-1b.3 50

Halvorsen, Kjartan Wed-1a.3 67

Wed-1a.9 68

Hasegawa-Johnson, Mark Mon-2c SS.7 47

Hasija, Tanuj Mon-1a.4 37

Hassan, Mahmoud Tue-1c SS.3 52

Haupt, Jarvis Tue-2b.6 55

Hegde, Chinmay Tue-2b.4 54

Hero III, Alfred Mon-1a.3 36

Heydari, Javad Tue-3c SS.4 65

Hickey, Aodhan Tue-2d SS.1 58

Hiden, Hugo Tue-2d SS.1 58

Honeine, Paul Wed-1c SS.5 71

Hongwei, Liu Tue-1a.8 48

Hortelano, Marcos Wed-1a.2 67

Hou, Elizabeth Mon-1a.1 36

Howard, Stephen Tue-2c SS.7 58

I

Ilhe, Paul Mon-1a.5 37

Iwen, Mark Wed-1b.5 69

J

Jaffard, Stephane Tue-1b.4 50

Javed, Anum Tue-1b.8 51

Ji, Feng Tue-2a.4 56

Jiang, Xin Wed-1a.4 67

Jonsson, Roland Wed-1b.4 69

K

Kammoun, Abla Mon-1c SS.3 40

Tue-1a.2 47

Kernfeld, Eric Tue-3b.7 61

Khachatryan, Elvira Tue-2d SS.2 58

Khalil, Mohamad Tue-1c SS.3 52

Kharratzadeh, Milad Mon-2a.6 43

Tue-3b.8 61

Kibria, Sharmin Tue-2b.5 54

Kilmer, Misha Tue-3b.7 61

Kim, Jinsub Mon-1c SS.7 41

Tue-2b.5 54

Wed-1c SS.7 72

Kingsbury, Nick Mon-1a.10 38

Kirby, Michael Wed-1b.7 70

Kok, Manon Wed-1a.9 68

Kosut, Oliver Tue-3c SS.3 65

Kotti, Eleftheria Tue-3b.9 62

Koutroumbas, Konstantinos Mon-2a.4 42

Kreutz-Delgado, Ken Mon-1a.9 38

Tue-1c SS.4 52

Krikheli, Michael Mon-1a.6 37

Krim, Hamid Wed-1c SS.6 71

Kumar, B. V. K. Vijaya Tue-1b.2 50

L

Lai, Phung Tue-3b.10 62

Lamberti, Roland Mon-2c SS.4 46

Larzabal, Pascal Tue-2b.1 53

Lasserre, Marie Tue-2c SS.5 57

Lawlor, Sean Tue-2a.5 56

Le Chevalier, François Tue-2c SS.5 57

Lee, Zeyi Tue-1a.7 48

Leonarduzzi, Roberto Tue-1b.4 50

Leshem, Amir Mon-1a.6 37

Leus, Geert Tue-1a.9 49

Tue-2b.2 54

Lévy-Leduc, Céline Mon-2b SS.5 44

Li, Pan Tue-3c SS.5 66

Li, Xingguo Tue-2b.6 55

Liao, Guisheng Tue-1a.11 49

Liu, Hao Jan Tue-3c SS.6 66

Liu, Jun Tue-1a.8 48

Llerena, Cosme Tue-2a.6 56

López-Salcedo, José A. Tue-3a.1 62

Tue-3a.9 64

López-Valcarce, Roberto Tue-1c SS.1 52

Wed-1a.5 67

Loubaton, Philippe Mon-1c SS.6 41

Louzada, Francisco Mon-1a.7 37

Tue-1b.6 51

Luengo, David Mon-2c SS.3 46

Mon-2c SS.5 46

M

Ma, Meng Mon-1b.2 38

76


Maccarone, Aurora Wed-1b.1 68

Majumder, Nathan Tue-3b.7 61

Makeig, Scott Mon-1a.9 38

Tue-1c SS.4 52

Manolopoulou, Ioanna Tue-3b.9 62

Mao, Xingpeng Tue-1a.5 48

Marcia, Roummel Mon-2a.8 43

Wed-1a.1 66

Marnissi, Yosra Tue-1b.5 50

Marques, Antonio Mon-1b.3 39

Marti, Gautier Tue-1c SS.2 52

Martino, Luca Mon-1a.7 37

Mon-2c SS.3 46

Mon-2c SS.5 46

Tue-1b.6 51

Masciullo, Antonio Tue-2c SS.3 57

Massimino, Andrew Mon-2a.7 43

Mateos, Gonzalo Mon-1b.3 39

McCarthy, Aongus Wed-1b.1 68

McCool, Paul Tue-1b.7 51

McKay, Matthew Mon-1c SS.1 40

McLaughlin, Steve Tue-1b.7 51

Wed-1b.1 68

Wed-1c SS.4 71

Meilhac, Lisa Tue-1a.1 47

Merelli, Luca Tue-3b.5 61

Mester, Rudolf Mon-2a.2 42

Mestre, Xavier Mon-1c SS.4 41

Mon-1c SS.5 41

Meurs, Alfred Tue-2d SS.2 58

Mheich, Ahmad Tue-1c SS.3 52

Miron, Sebastian Tue-1a.3 47

Miura, Yuto Wed-1a.6 68

Mohammadi, Arash Mon-1a.8 37

Wed-1a.7 68

Mohino-Herranz, Inmaculada Tue-2a.6 56

Momeni, Naghmeh Mon-1b.1 38

Monfrini, Emmanuel Wed-1b.3 69

Moore, Michael Tue-2a.7 56

Morales, David Mon-1c SS.1 40

Morris, Rosie Tue-2d SS.1 58

Mota, Joao Tue-3a.10 62

Moulines, Eric Mon-1a.5 37

Muldoon, Sarah Mon-1b.5 39

N

Nadakuditi, Raj Rao Tue-1a.6 48

Najim, Ouiame Mon-1c SS.5 41

Nanjundaswamy, Tejaswi Tue-3a.6 63

Navarro, Monica Mon-1c SS.4 41

Nayer, Seyedehsara Tue-2b.7 55

Nazarpour, Kianoush Tue-2d SS.1 58

Nelson, James Mon-2a.1 42

Wed-1b.2 69

Nguyen, Tam Tue-3b.10 62

Nielsen, Frank Tue-1c SS.2 52

Nikadon, Jan Tue-2d SS.3 59

Nowak, Rob Mon-2a.8 43

O

Oberlin, Thomas Mon-2c SS.2 45

Olsson, Fredrik Wed-1a.3 67

Wed-1a.9 68

Ou, Zhijian Mon-2c SS.7 47

Wed-1b.8 70

Ozdemir, Alp Wed-1b.5 69

P

Pagès-Zamora, Alba Tue-1c SS.1 52

Palmer, Jason Mon-1a.9 38

Tue-1c SS.4 52

Palomar, Daniel Mon-2b SS.2 44

Parras, Juan Tue-3a.4 63

Pascal, Frederic Mon-1c SS.3 40

Mon-2b SS.2 44

Pereyra, Marcelo Tue-1c SS.5 53

Perlaza, Samir Tue-3c SS.3 65

Perperidis, Antonios Tue-1b.7 51

Pesquet, Jean-Christophe Tue-1b.5 50

Peterson, Chris Wed-1b.7 70

Petetin, Yohan Mon-2c SS.4 46

Pham, Gia-Thuy Mon-1c SS.6 41

Pieczynski, Wojciech Wed-1b.3 69

Pimentel-Alarcon, Daniel Mon-2a.8 43

Piotrowski, Tomasz Tue-2d SS.3 59

Plataniotis, Konstantinos Mon-1a.8 37

Wed-1a.7 68

Poor, H. Vincent Tue-3a.1 62

Tue-3c SS.3 65

Tue-3c SS.7 66

Pougkakiotis, Spyridon Tue-3b.1 60

77


Pribic’, Radmila Tue-1a.7 48

Tue-1a.9 49

Tue-2b.2 54

R

Rabbat, Michael Mon-1b.1 38

Tue-2a.5 56

Raich, Raviv Mon-1c SS.7 41

Tue-2b.5 54

Tue-3b.10 62

Wed-1c SS.7 72

Ramírez, David Mon-1a.4 37

Rangaswamy, Muralidhar Tue-1a.6 48

Raskutti, Garvesh Tue-1b.3 50

Raul Dias Rodrigues, Miguel Tue-3a.10 64

Rehman, Naveed ur Tue-1b.8 51

Rehman, Ubaid ur Tue-1b.8 51

Reilly, Richard Wed-1a.2 67

Reisen, Valderio Mon-2b SS.5 44

Renaux, Alexandre Tue-2c SS.1 57

Rey Otero, Ives Wed-1c SS.6 71

Reynaud-Bouret, Patricia Wed-1a.4 67

Ribeiro, Alejandro Mon-1b.3 39

Ricci, Giuseppe Tue-2c SS.3 57

Richard, Cédric Wed-1c SS.1 70

Richmond, Christ Tue-2c SS.6 58

Riera-Palou, Felip Tue-3a.2 62

Rivoirard, Vincent Wed-1a.4 67

Rochester, Lynn Tue-2d SS.1 58

Romero, Daniel Mon-1b.2 38

Mon-2a.10 44

Tue-1c SS.1 52

Wed-1a.5 67

Rontogiannis, Athanasios Mon-2a.4 42

Rosa, Manuel Tue-2a.6 56

Rose, Kenneth Tue-3a.6 63

Rossi, Michele Tue-3b.5 62

Rottenberg, François Mon-1c SS.4 41

Roueff, Francois Mon-1a.5 37

Routtenberg, Tirza Wed-1b.6 69

S

Sala, Josep Tue-1c SS.1 52

Salleh, Sh-Hussain Tue-2d SS.6 59

Salsabilian, Shiva Mon-1b.5 39

Sánchez-Hevia, Héctor Adrián Tue-2a.6 56

Sansonnet, Laure Wed-1a.4 67

Santamaria, Ignacio Tue-1c SS.6 53

Wed-1b.7 70

Sarnaglia, Alessandro Mon-2b SS.5 44

Scaglione, Anna Tue-3c SS.1 65

Schamberg, Gabriel Tue-2d SS.4 59

Scharf, Louis Wed-1b.7 70

Schön, Thomas Wed-1a.9 68

Schreier, Peter Mon-1a.4 37

Seckarova, Vladimira Tue-2a.2 55

Seco-Granados, Gonzalo Tue-3a.1 62

Tue-3a.9 64

Segarra, Santiago Mon-1b.3 39

Seghouane, Abd-krim Tue-2d SS.6 59

Septier, François Mon-2c SS.4 46

Setlur, Pawan Tue-1a.6 48

Sevi, Harry Mon-1c SS.2 40

Shaghaghian, Shohreh Mon-1b.4 39

Sharma, Vinod Tue-1b.2 50

Shi, Wei Tue-3c SS.6 66

Sindi, Suzanne Wed-1a.1 66

Sirianunpiboon, Songsri Tue-2c SS.7 58

Slavakis, Konstantinos Mon-1b.5 39

Slock, Dirk Tue-1a.1 47

Tue-3a.5 63

Sodjo, Jessica Mon-2a.9 43

Song, Pingfan Tue-3a.10 64

Song, Yang Mon-1a.4 37

Souloumiac, Antoine Mon-1a.5 37

Suárez-Casal, Pedro Tue-3a.3 63

Tue-3a.8 64

Sun, Ying Mon-2b SS.2 44

T

Tabikh, Wassim Tue-3a.5 63

Tabrikian, Joseph Tue-2c SS.2 57

Tajer, Ali Tue-3c SS.4 65

Taleb, Youssef Tue-1b.9 51

Talmon, Ronen Tue-1c SS.7 53

Tanaka, Toshihisa Tue-2d SS.5 59

Tanaka, Yuichi Tue-2d SS.5 59

Tang, Wen Wed-1c SS.6 71

Tay, Wee Peng Mon-1b.7 40

Tue-2a.4 56

Teng, Diyan Tue-1b.1 49

Theodoridis, Sergios Tue-3b.1 60

78


Thibault, Aymeric Wed-1a.8 68

Ting, Chee-Ming Tue-2d SS.6 59

Tiomoko Ali, Hafiz Mon-1c SS.2 40

Todros, Koby Mon-1a.3 36

Tong, Lang Tue-3c SS.2 65

Touchette, Hugo Tue-1b.4 50

Tourneret, Jean-Yves Mon-2c SS.2 45

Wed-1c SS.1 70

Wed-1c SS.4 71

U

Úbeda-Medina, Luis Tue-1a.10 49

Uehara, Takashi Tue-2d SS.5 59

Urteaga, Iñigo Mon-2c SS.6 46

V

Valcarcel Macua, Sergio Tue-3a.4 63

Vallet, Pascal Mon-1c SS.5 41

Van Hulle, Marc Tue-2d SS.2 58

Tue-2d SS.7 60

Van Roost, Dirk Tue-2d SS.2 58

Vandendorpe, Luc Mon-2b SS.4 44

Vaswani, Namrata Mon-2a.5 43

Tue-2b.7 55

Vía, Javier Tue-1c SS.6 53

Vieira, Marcelo Wed-1c SS.3 71

Vilà-Valls, Jordi Mon-2c SS.1 45

Villafañe-Delgado, Marisel Mon-1b.6 39

Vorobyov, Sergiy Tue-1a.5 48

W

Wack, David Mon-1b.5 39

Wagner, Mark Tue-2d SS.4 59

Wainrib, Gilles Mon-1c SS.2 40

Wang, Ruobai Mon-2c SS.7 47

Wang, Yuan Mon-1b.7 40

Watanabe, Yodai Wed-1a.6 68

Watson, Paul Tue-2d SS.1 58

Wendling, Fabrice Tue-1c SS.3 52

Wendt, Herwig Tue-1b.4 50

Wed-1c SS.4 71

Wickramarathne, Thanuka Mon-2b SS.6 45

Wiesel, Ami Mon-2b SS.7 45

Willett, Rebecca Mon-2a.8 43

Tue-1b.3 50

Wed-1a.4 67

Wittevrongel, Benjamin Tue-2d SS.7 60

Woo, Wai Lok Tue-3a.7 64

Woodman, Simon Tue-2d SS.1 58

Wuhua, Hu Mon-1b.7 40

X

Xia, Xiang-Gen Tue-1a.8 48

Xu, Jingwei Tue-1a.11 49

Xu, Yanhong Tue-1a.11 49

Y

Yair, Or Tue-1c SS.7 53

Yap, Han Lun Tue-1a.7 48

You, Zeyu Wed-1c SS.7 72

Yuan-Wu, Yi Tue-3a.5 63

Z

Zain Abbas, Syed Tue-1b.8 51

Zamani, Sina Tue-3a.6 63

Zazo, Javier Tue-3a.4 63

Zazo, Santiago Tue-3a.4 63

Zhan, Yanjun Tue-1a.7 48

Zhang, Baosen Tue-3c SS.5 66

Zhang, Ganchi Mon-1a.10 38

Zhang, Haiqiao Mon-2b SS.6 45

Zhang, Liang Mon-2a.10 44

Zhang, Martin Wed-1b.8 70

Zhang, Teng Mon-2b SS.7 45

Zhang, Yang Mon-2c SS.7 47

Zhao, Qing Tue-3c SS.2 65

Zhao, Yue Tue-3c SS.7 66

Zhu, Hao Tue-3c SS.6 66

79


SSP 2016Notes

2016 ieee workshop on statistical signal...

Documents