iCal exportI have read and understood the above.
I have read and understood the above.
Download current event:Event calendar file
In order to enable an iCal export link, your account needs to have a key created. This key enables other applications to access data from within Indico even when you are neither using nor logged into the Indico system yourself with the link provided. Once created, you can manage your key at any time by going to 'My Profile' and looking under the tab entitled 'HTTP API'. Further information about HTTP API keys can be found in the Indico documentation.
In conjunction with a having a key associated with your account, to have the possibility of exporting private event information necessitates the creation of a persistent key. This new key is also associated with your account and whilst it is active the data which can be obtained through using this key can be obtained by anyone in possession of the link provided. Due to this reason, it is extremely important that you keep links generated with this key private and for your use only. If you think someone else may have acquired access to a link using this key in the future, you must immediately remove it from 'My Profile' under the 'HTTP API' tab and generate a new key before regenerating iCalendar links.
Permanent link for public information only:
Permanent link for all public and protected information:
Please use CTRL + C to copy this URLDetailed timetable
Theory and applications of signal processing methods (in GW detection, medical science and engineering)chaired by Elena Cuoco
from to (Europe/Rome)
at EGO, Cascina ( Auditorium )
at EGO, Cascina ( Auditorium )
The European Gravitational Observatory (http://www.ego-gw.it), site of the Virgo interferometer, is located in the countryside of the Comune of Cascina, a few kilometres from town of Pisa.
Different scientific environments (GW communities, Medical Science, Engineering, Telecommunications..etc.) use the same techniques of signal processing. In this workshop we want to put in contact scientists with different backgrounds to exchange ideas and learn new techniques. The worskhop is addressed at young researchers, PhD students and anyone wishing to update their knowledge of signal processing.
Go to day
Speaker: Dr. Elena Cuoco (EGO) Material: Slides Video
Speaker: Prof. Federico Ferrini (EGO) Material: Slides Video
Understanding (and Taming) Glitches: Threads in a Tapestry
A cartoon introduction to Gravitational waves. Non ideal features of noise in interferometric GW detectoers Glitches Data quality and Vetoing GW interferometers as MIMOs Lessons from PEM-injections; bilinear pseudochannels Wiener-Volterra description Channel-Channel Transfer functions GW interferometers as MISOs "Input" disturbances vs environmental ones. Glitch structure. Modal decomposition.
Speaker: Prof. Innocenzo Pinto Material: Slides Video
- 11:10 - 11:30 Coffe break
Signals and noises in neuroimaging
In the framework of neurodegenerative diseases, and in particular the Alzheimer's disease (AD), we are witnessing an increasing presence of neuroimaging data, magnetic resonant images (MRI) and positron emission tomography (PET) above all. Despite the fact that clinical scanners have been around for some decades, MRI and PET images have only recently become a dependable support in diagnosing early and prodromal AD. The purpose of image analysis is to find supporting evidence to help in early clinical assessment. In the case of neurodegenration leading to AD, we would ideally look for a measure (marker) able to discriminate between normalcy versus pathology at a pre-clinical stage, easy to implement in clinical practice and possibly based on quick, low- cost, and widely available procedures. In addition, a “good” marker should have predictive value as to whether a subject with current unknown or unclear clinical assessment will or will not develop the pathological condition in a given timeframe. From a data analysis point of view, the problem can be restated as a measure of a “signal” (the pathology marker) on a “background” of normalcy. Whether a marker can have predictive value at all implies that it must be sensitive to some key aspects of the pathology process well before their effects have a clear clinical counterpart. We shall give an overview of some relevant branches in brain image analysis and the requirements to define a suitable signal over the confounding variability (noises) peculiar to life-science studies.
Speaker: Dr. Andrea Chincarini (INFN) Material: Slides Video
- 13:00 - 13:59 Lunch
Bayesian Inference in Digital Communication.Part I: Theory
Given a hidden state process with memory and a measurement process that is memoryless given the state, one has to make inference on the hidden state sequence given the measurement sequence. This activity can be performed by the so-called Bayesian tracking, where the probability of the state at time k is recursively computed from the probability of the state at time k-1 and from the measurement at time k. The talk will recall the basics of Bayesian tracking: two-step recursion based on predict-update, forward-backward recursion for making inference on the state at time k given the entire measurement sequence. The Kalman filter is then presented as an instance of Bayesian tracking for a non-stationary linear state transition model and Gaussian processes. Wiener filter is then derived for the stationary case. The general case of non-linear and non-Gaussian model can be difficult to deal with. To get good results, the approach to be adopted should fit the specific state transition and measurement model at hand. As examples, the talk will present the non-Gaussian parametric approach, the state-space discretization approach, and the sequential importance sampling approach based on particle filters.
Speaker: Maurizio Magarini Material: Slides Video
- 16:00 - 16:20 coffee break
Bayesian Inference in Digital Communication.Part II: Applications
A classical problem in synchronization, an activity that plays a central role in signal processing for digital communication, is that of extracting the phase of a sinusoid affected by phase noise and embedded in additive noise. In the talk, the problem is presented as an instance of Bayesian inference. Four variants of the basic problem are treated. 1) Sinusoid with constant amplitude, where the phase is extracted by filtering the sinusoid through the Wiener filter. 2) Sinusoid affected by known amplitude modulation. Here the system model is no more stationary, and the phase should be extracted by filtering the sinusoid through the Kalman filter. 3) Sinusoid affected by unknown amplitude modulation. Here non-parametric methods, such as quantization of the phase space, should be adopted. 4) A message is transmitted through a channel that includes up/down conversion, that is multiplication of the message by a sinusoid. In this context, the phase noise affecting the sinusoid is a source of disturbance that compromises channel capacity. Computation of channel capacity can be approached by Bayesian methods that track the phase noise that affect the sinusoid.
Speaker: Luca Barletta Material: Slides Video
- 09:05 - 09:15 Welcome 10'
Searches for GW transients with the ground-based detectors.
In this lecture I give an overview of the gravitational-wave experiment and methods used in the analysis of data from the ground-base detectors. The purpose of the lecture is to describe how a worldwide network of GW detectors works, with focus on signal processing techniques and reconstruction methods, rather than to describe details of specific search algorithms used so far in the GW experiment. The lecture is organized in three parts: 1) Short introduction into the gravitational-wave experiment. 2) Signal processing methods and their applications to the analysis of single detector data, including: - time-frequency transformations - data conditioning with linear prediction error (LPE) filters - regression of GW data, Wiener-Kolmogorov filters - time-frequency analysis, clustering in TF domain - multi-resolution analysis 3) Analysis of data from multiple GW detectors (coherent network analysis), including - characterization of detector networks - Inverse problem for bursts, likelihood methods - matched filters, weakly modeled and un-modeled searches - detection statistics - reconstruction of polarization states - reconstruction of GW waveforms - sky localization - factors affecting GW network performance.
Speaker: Sergey Klimenko Material: Slides Video
- 11:15 - 11:45 coffee break
Radar Clutter Modeling and Analysis
The first part of the talk focuses on the statistical analysis and modeling of radar clutter echoes. This is a central issue for the design and performance evaluation of radar systems. Main goal of this talk is to describe the state-of-the-art approaches to the modeling and understanding of land and sea clutter echoes and their implications on performance prediction and signal processors design. The talk will first introduce radar sea and ground clutter phenomena, measurements and measurement limitations, at high and low resolution, high and low grazing angles with some references to classical model for Radar Cross Section prediction. Then the main part of the talk will be dedicated to modern statistical and spectral models for high resolution sea and ground clutter and, particularly, to the methods of experimental validation using recorded data sets. A plethora of experimental results will be shown and commented.
Speaker: Dr. Maria Sabrina Greco Material: Slides
- 13:00 - 14:30 Lunch
Radar Clutter Modeling and Analysis
Speaker: Sabrina Maria GRECO Material: Video
- 15:30 - 15:50 Coffee break
Coherent Target Detection in Heavy-Tailed Compound-Gaussian Clutter
Coherent radar target detection is the subject of the second part of the talk. In high resolution radar systems the disturbance cannot be modelled as Gaussian distributed and the classical detectors suffer from high losses. Then, according to the adopted disturbance model, optimum and sub-optimum detectors are derived and their performance analyzed against a non-Gaussian background. First, the problem of detecting completely known and fluctuating random signals, possibly with unknown parameters, against correlated non-Gaussian clutter modelled as a compound-Gaussian process is then described exploiting different degrees of knowledge on target and clutter statistical characteristics. A generalized likelihood ratio test (GLRT) detector and a fully adaptive constant false alarm rate (CFAR) detector are derived and different novel interpretations of the detection algorithms provided in order to highlight the relationships and the differences among them and the links with the Gaussian clutter case. Each interpretation suggested different ways to implement the optimum detector and allowed to derive sub-optimum detectors easier to implement than the optimum and with very low losses. Modern radar systems generally operate in non-homogeneous and non-stationary clutter environment. In this condition the amplitude statistics and the power spectral density of the disturbance are unknown. Therefore, the adaptive versions of the algorithms previously described, which estimate the clutter covariance matrix, are then introduced. An estimation algorithm is described which guarantees the CFAR property and its performance are investigated and compared to that of the maximum likelihood (ML) estimator. The important cases of target signal partially unknown or modelled as a subspace random process are also analyzed. The proposed detectors are tested against both simulated data and measured high resolution sea clutter data to investigate the dependence of their performance on the various clutter and signal parameters.
Speaker: Prof. Fulvio Gini Material: Slides Video
- 09:15 - 11:15 Searches for GW transients with the ground-based detectors. 2h0'
Understanding (and Taming) Glitches: Threads in a Tapestry
Retrieving glitch constituents. The Hilbert-Huang transform; Prony (robustified) decomposition Glitch taxonomy and entomology: correlation graphs, clustering and classification TF representations: Q and Wigner-Ville transform Getting rid of intermodulation artifacts Retrieving TF skeletons. Sparse representations in time-domain Compressed coding and adaptive dictionary learning. Non ideal features of the residual noise floor: Gaussianity tests; SIRP model. Glitch noise statistics, via Middleton model Locally Optimum Detector (LOD) and its robust implementation Numerical experiments on real data.
Speaker: Innocenzo Pinto Material: Slides Video
- 11:00 - 11:15 coffee break
Many-core computing in experimental gravitational wave physics.
With the advance of cheaply available many-core architectures (such as GPU-s) on the computing market scientific computing entered into a new era. There are plenty of mathematical algorithms and data analysis methods which can be easily, massively parallelized and implemented on these architectures resulting a dramatic increase of speed (several order of magnitude) and - at the same time - decrease of computational costs! The presenataion will give a brief overview of currently available hardwares and programing techniques with their application focused on experimental gravitational wave physics.In the last third of the lecture an introductory hands-on tutorial will be given to the audiance. Bring your laptop with you, if possible.
Speaker: Gergeley Debreczeni Material: Slides Video
- 13:00 - 14:15 Lunch
14:15 - 16:00
- 09:00 - 11:00 Understanding (and Taming) Glitches: Threads in a Tapestry 2h0'