Cristina Masoller’s research projects
|
Marie Curie ITN CAFE: Climate advanced forecasting of sub-seasonal
extremes (March 2019-February 2023,
H2020-813844) Web page: http://www.cafes2se-itn.eu/ Partners: Centre de Recerca Matemàtica
(Spain, Coordinator), Univ. Politècnica de Catalunya (Spain), Technische Universitaet Bergakademie
Freiberg (Germany), Potsdam Institut für Klimafolgenforschung(Germany),
Aria Technologies (France), Meteo-France, CSIC
(Spain), Max-Planck-Institute For Complex Systems
(Germany), Universidad de la República (Uruguay),
European Centre for Medium-Range Weather Forecasts (UK). Principal Investigator in UPC: Prof. Cristina Masoller Summary: Forecasting
climatic extreme events on the sub-seasonal time scale (from 10 days to about
3 months) is very challenging because of the poor understanding of phenomena
that may increase predictability in this time scale. The CAFE network will
provide top-level training to 12 young researchers that will enable them to
advance the state of the art of subseasonal
predictability. The students will receive training on a wide range of
disciplines, including climate science, meteorology, statistics and
non-linear physics.
Kick-off meeting of the CAFE project (March 2019)
|
|
Complex dynamical
systems and advanced data analysis tools (ComDynSys) 2019-2021 Ministerio de Ciencia, Innovación y Universidades, PGC2018-099443-B-I00 Description: The aim
of this project is to improve the understanding of nonlinear phenomena in
complex systems by the use of various data analysis tools, with the long-term
goal of exploiting nonlinear phenomena for innovative applications. The
phenomena to be studied are interdisciplinary and include the spiking output
of laser diodes, neuronal signal encoding and processing, extreme events in
atmospheric dynamics, brain signals and retina images, among many others. The
research involves experiments, numerical simulations and data analysis. |
|
ICREA
ACADEMIA (2016-2020) Funded by Institució Catalana de
Recerca i Estudis Avançats
(ICREA), Generalitat de Catalunya.
Principal Investigator: Cristina Masoller The research funded
by ICREA ACADEMIA is aimed at i) exploiting
nonlinear dynamics for novel applications and ii) developing advanced
analysis tools for studying the output signals of complex systems. A first
research objective is aimed at exploiting the optical spikes generated by a
semiconductor laser with feedback, for implementing photonic neurons that
mimic biological ones. A second goal is to exploit the analysis tools
developed by the LINC project to advance a crucial problem of climate
dynamics: predictability in the sub-seasonal time-scale, which lies between
the well-established application of weather forecast and that of the seasonal
forecast.
|
Previous projects:
|
BE-OPTICAL: Advanced biomedical imaging
and data analysis EU-H2020 Marie Skłodowska-Curie Innovative Training Network October 2015-September 2019 H2020-675512 Coordinator: Prof. Cristina Masoller Project Manager: Dr. Jordi Tiana PhD
students: D. Halpaap, P. Amil Summary: The main goals of the BE-OPTICAL project were 1) To provide
top-level multi-skill training to 14 early stage researchers (ESRs) in a wide
range of optical imaging technologies and signal processing tools, including
fluorescence spectroscopy and microscopy, optical coherence tomography (OCT),
optogenetics, engineered nanomaterials and signal
processing tools. 2) To develop
cutting-edge research in the fields of optical imaging, nanotechnology,
computer science, complex systems and data analysis (the specific research
goals of the ESR projects are listed below). 3) To bring together
interdisciplinary team of physicists, engineers and medical doctors, with
complementary expertise, and engage them in successful long-term
collaborations. The research projects
of the two PhD students supervised by Prof. C. Masoller
were -
Donatus Halpaap: to
develop new approaches for reducing speckle in double pass images. PhD thesis: Experimental
study of speckle generated by semiconductor light sources: application in
double pass imaging (2019, co-supervised with M. Vilaseca, DF-UPC
News) Publications: D. Halpaap, C. E. Garcia-Guerra, M. Vilaseca, C. Masoller, “Speckle reduction
in double-pass retinal images”, Sci. Rep. 9, 4469 (2019). D. Halpaap, J. Tiana-Alsina, M. Vilaseca, C. Masoller, “Experimental
characterization of the speckle pattern at the output of a multimode optical
fiber”, Opt. Express 27, 27738 (2019). -
Pablo Amil: to develop novel algorithms for the analysis and
classification of complex biomedical images. PhD thesis: Machine
learning methods for the characterization and classification of complex data
(February 2020) Publications: P.
Amil, L. Gonzalez, E. Arrondo, et al. Unsupervised feature extraction of anterior
chamber OCT images for ordering and classification Scientific
Reports 9, 1157 (2019). This work was featured in local and national
newspapers (read more here). P. Amil,
F. Reyes-Manzano, L. Guzmán-Vargas, I. Sendiña-Nadal
and C. Masoller Network-based methods for retinal fundus
image analysis and classification PLoS ONE 14, e0220132 (2019). P.
Amil, N. Almeira, C. Masoller Frontiers in
Physics 7, 194 (2019). Outlier mining methods based on
network structure analysis
Members of BE-OPTICAL team at UPC (Oct. 2016)
Picture taken during the first BE-OPTICAL school (Nov. 2016) BE-OPTICAL was featured in national
and local newspapers (Nov. 2016): La Vanguardia, El Periodico, Diario de Terrassa BE-OPTICAL was featured in the Success
Stories web page of the European Comission (Sep. 2017).
|
|
|
|
Complex physical and biophysical systems: towards a
comprehensive view of their dynamics and fluctuations (ComPhysBio) Ministerio de Economía y Competitividad, Spain (2016 – 2018, FIS2015-66503-C3-2-P) Principal Investigator at UPC: Cristina Masoller. Coordinator at UPF: J. Garcia Ojalvo This coordinated
project (which involves three research teams at UPF, UB and UPC) studies
nonlinear and stochastic phenomena in a broad class of systems including information
processing by optical networks, extreme events in complex systems, neuronal
excitability and brain dynamics, among many others.
Workshop on Recent Advances on Stochastic and Nonlinear Dynamics of
Complex Systems, organized at the end of the project, in honor of Prof. Carme
Torrent (February 4, 2019).
|
|
LINC: Marie
Curie Initial Training Network (ITN) Learning about Interacting Networks in
Climate Marie Curie Initial
Training Network (2011-2015, FP7-289447) Principal investigator and Coordinator: Cristina Masoller The results
of the LINC project were published in the European Commission
web page (Oct. 2016). Summary: EU-funded LINC project was aimed at transferring modelling and analysis
techniques used in other disciplines to climate science in a bid to improve
predictions of climate events like El Niño. The research results contributed
to better understand complex weather patterns and their impact on the
environment, economic activities and society. Description: The Earth’s climate is a highly complex system
that scientists are trying to unpick. Research by the EU-funded project LINC
was aimed at advancing state-of-the-art climate know-how with a new approach
that draws from fields such as transport and neural networks. LINC joined the dots
between techniques used in different fields and applied them to climate
science. It aimed to improve the forecasting of major climate events,
including El Niño. The researchers developed a new approach for climate
modelling and data analysis. They also developed new modelling software for
use by the scientific community and for assessing the predictability of
extreme weather. Modelling El Niño A main innovation of
the project was to apply complex
networks and nonlinear data
analysis tools already used to model systems including transport
networks, social networks, brain networks, and the internet and ecosystems,
and to study climate phenomena. Using the complex network approach combined
with nonlinear time-series analysis, the project analysed
how El Niño — Southern Oscillation (ENSO) the most important dynamic
phenomena in our climate — affects climate. While El Niño
originates in the Pacific Ocean, its effects are global and include altering the
frequency of hurricanes in the Atlantic and the monsoon in India. The new
methodology helped model this, potentially improving predictions of ENSO’s
effects. LINC researchers came
up with new methods to identify geographical regions that share similar
climate dynamics. These regions — known as climatic communities — are not
necessarily geographically close. Extreme weather forecasting LINC also developed
new software — PyUnicorn and Par@Graph
— which will be used extensively by the complex system scientific community.
These tools could also be used to assess the sub-seasonal predictability of
extreme events such as flooding, heatwaves and cold surges. Under the EU’s Marie Skłodowska-Curie fellowship programme,
LINC trained 12 early stage researchers and 3 young experienced researchers
in the complete set of skills required to undertake a career in physics and
geosciences with expertise in climatology, networks and complex systems.
Final LINC conference, Viena, Austria, 2015
|
NETT: Neural Engineering
Transformative Technologies
EU
FP7 Marie Curie Initial Training Network (2012-2016) –Coordinator: Prof. Coombes (Nottingham); PI Prof. García
Ojalvo
NETT
will provide training to young researchers in Neural Engineering, a new
discipline that coalesces engineering, physics and neuroscience for the design
and development of brain–computer interface systems, cognitive computers and
neural prosthetics. In our lab, the work of the PhD student Carlos Quintero is
funded by NETT.
SILICOS:
Stochasticity in Nonlinear Complex Systems (2013-2015)
This
project involved the study of a wide range of stochastic and nonlinear
phenomena in physical and biological systems, including transport processes,
intracellular coupling, bacterial stress, signaling in immune system pathways,
pattern formation in biological tissues, neuronal oscillations, visco-elasticity, mixed-mode fracture, symbolic dynamics,
polarization dynamics and extreme events in lasers. Funded by the Ministerio de Economía y Competitividad (Spain, FIS2012-37655-C02-01) –PI: J. García Ojalvo
Semiconductor laser complex dynamics: from optical neurons to optical
rogue waves (2014-2015)
This
project, funded by the European Office of Aerospace Research & Development
(EOARD, grant number FA9550-14-1-0359) was aimed at improving our understanding
of the complex dynamics in semiconductor lasers induced by external
perturbations. A specific goal was to exploit the spiking behavior of
semiconductor lasers induced by optical feedback or optical injection for
implementing optical neurons which could be building blocks of optical
information processing systems. This project was the continuation of previous
grants (FA9550-07-1-0238, FA-8655-10-1-3075, FA8655-12-1-2140) which are
described below.
ICREA
ACADEMIA (2009-2014)
This
project was aimed at studying the interplay of noise, delays and nonlinearities
in semiconductor lasers, with the ultimate goal of exploiting noise, delays and
nonlinearities for novel applications, using them for enhancing the laser
performance. In recent years there has been a growing interest in the potential
exploitation of noise and delays for controlling the operation of nonlinear
devices; from nanoscale systems to opto-electronic
devices. In the specific case of vertical-cavity surface-emitting lasers
(VCSELs), which are key devices in photonics, nanotechnology, optical signal
processing and optical networks, important nonlinear dynamic processes of
interaction of light with matter take place, and the unavoidable presence of
noise, due to spontaneous emission, and delays, due to finite light propagation
time, can significantly degrade the dynamic response of the laser, but on the
other hand, these stochastic and nonlinear effects can also be exploited for
novel applications.
Funded by
Institució Catalana de
Recerca i Estudis Avançats
(ICREA), Generalitat de Catalunya.
Spiking excitable semiconductor laser as optical
neurons: dynamics, clustering and global emerging behaviors (2012-2013)
This
project was aimed at exploiting the nonlinear dynamics of semiconductor lasers
for novel applications, such as all-optical switching and brain-inspired
photonics information processing.
Detailed
experimental and numerical studies have been performed, focusing on the
interplay of noise and nonlinear dynamics. We used a method of time-series
analysis, referred to as symbolic ordinal analysis, to demonstrate that serial
correlations present in the output intensity of a semiconductor laser with
optical feedback operating in the low-frequency fluctuations regime share
common features with serial correlations present in the inter-spike-intervals (ISIs)
of biological neuronal systems. The symbolic dynamics underlying the sequence
of inter-dropout-intervals in the laser intensity has the same statistical
features, in terms of distribution of symbolic patterns, as in ISI sequences of
biological neurons. Therefore, semiconductor laser-based optical neurons could
provide a novel, inexpensive and controllable experimental set up that could
allow for improving our understanding of neuronal activity.
Other
topics of research included the analysis of extreme events in the form of
ultra-high pulses in the output intensity of semiconductor lasers with
continuous-wave (cw) external optical injection or
optical feedback. Rogue waves, earthquakes of high magnitude and financial
crises are all rare and extreme events corresponding to abrupt changes of
environmental or socio-economic conditions with potentially catastrophic
consequences. Semiconductor lasers under external perturbations can display a
dynamical regime where extreme pulses occasionally occur and thus, they are
optimal devices for investigating novel methods for predicting extreme events
(revealing early warning signals) and novel methods for controlling these
events.
Regarding
novel applications, we proposed the implementation of an all-optical stochastic
logic gate based in the interplay of polarization bistability
and noise in optically injected vertical-cavity surface emitting lasers
(VCSELs). We also demonstrated a novel method of sub-wavelength position
sensing that exploits the regime of quasiperiodic dynamics of semiconductor
lasers with optical feedback from two external cavities.
Funded
EOARD, grant number FA8655-12-1-2140.
Stochastic and nonlinear effects in semiconductor
lasers (2010-2011)
This
project was aimed at contributing to a better understanding of the interplay of
noise and nonlinearities in nano-cavity semiconductor
lasers, which might result in intrinsic (internal) noise as well as external
noise being exploited for novel applications of these devices, and using the
potentially beneficial role of noise for optimization of the performance of the
next generation of nano-cavity lasers.
In
recent years there has been a growing interest in the potential exploitation of
noise for controlling the operation of nonlinear devices; from nanoscale
systems to opto-electronic devices, noise is gaining
a lot of attention for its potentially beneficial role.
In
the specific case of vertical cavity semiconductor lasers (VCSELs), which are
key devices in photonics, nanotechnology, optical signal processing and optical
networks, important nonlinear dynamic processes of interaction of light with
matter take place, and the unavoidable presence of noise, due to spontaneous
emission, can significantly degrade the dynamic response of the laser, but on
the other hand, noise can also be exploited for novel applications.
A
first objective of the research project was to study the possibility of
controlling the polarization of the optical pulses emitted by a VCSEL by using
a properly chosen asymmetric shape for the modulation current and/or by tuning
the amount of externally added noise (current or optical noise).
The
interplay of nonlinearity and noise can yield logic behavior, and specifically,
a noise-controlled logic gate. A second project objective was demonstrating
that an optical noise-controlled logic gate can be implemented using a VCSEL,
which is naturally a two-state system because the light emitted by a VCSEL can
be linearly polarized along two orthogonal directions, usually associated with
crystalline orientation or stress.
A
third objective of the research project was to investigate the generation of
square-wave optical pulses by coupling two VCSELs, such that the mutually
injected light has a polarization that is orthogonal to polarization of the
natural lasing mode in each VCSEL. The study of the dynamics of such coupled
lasers is relevant not only for applications in optical communications but also
for analyzing coupling phenomena between nonlinear oscillators.
Funded
by EOARD, grant number FA-8655-10-1-3075.
Nonlinear dynamics of novel types of semiconductor
lasers (2007-2009)
This
project focused on modeling VCSEL nonlinear dynamics, with the objective of
improving our understanding of some of nonlinear features. Specific objectives
of the project were the study of 1) polarization instabilities due to cavity
anisotropies (such as dichroism and birefringence), 2) the interplay of
polarization and transverse effects when VCSELs are subjected to external
perturbations that induce instabilities (such as current modulation, optical
injection, optical feedback) and 3) synchronization phenomena in unidirectional
and mutually coupled VCSELs with different types of coupling (through one
polarization, both polarizations, and polarization-rotated orthogonal coupling).
Funded
by AFOSR, grant number FA9550-07-1-0238.