Έμβλημα Πολυτεχνείου Κρήτης
Το Πολυτεχνείο Κρήτης στο Facebook  Το Πολυτεχνείο Κρήτης στο Instagram  Το Πολυτεχνείο Κρήτης στο Twitter  Το Πολυτεχνείο Κρήτης στο YouTube   Το Πολυτεχνείο Κρήτης στο Linkedin

Νέα / Ανακοινώσεις / Συζητήσεις

Ομιλία κας Αγγελικής Ξενάκη "Array Signal Processing and Machine Learning Methods for Applications in Underwater Acoustics"

  • Συντάχθηκε 30-04-2023 13:28 Πληροφορίες σύνταξης

    Ενημερώθηκε: 30-04-2023 13:30

    Τόπος:
    Σύνδεσμος τηλεδιάσκεψης
    Έναρξη: 04/05/2023 11:00
    Λήξη: 04/05/2023 12:00

    Who

    Angeliki Xenaki, Research Scientist, PhD
    Centre for Maritime Research and Experimentation, STO-NATO

    Abstract

    Array signal processing aims to infer information about a signal from spatio-temporal measurements of the associated wavefield on an array of sensors. Combining the outputs of the sensors on an array enhances the signal over noise and attributes directionality to the system, allowing to characterize not only the spectral content of the recorded wavefield but also the number and locations of the sources that produce it. Therefore, array signal processing is an active research area in diverse fields, e.g., in radar, seismic, and acoustic imaging. In acoustics, array signal processing is used for visualization of sound fields, identification and localization of sound sources, or acoustic imaging of objects that scatter sound. Machine learning methods allow parameter inference of increasingly complex systems by capitalizing on large amount of data. Hence, data-driven inference with machine learning methods has gained popularity over traditional model-based methods as it requires no particular assumptions or decisions about the underlying physics. This presentation will cover some model-based and data-driven methods for sonar signal processing. Synthetic aperture sonar processing, sparse reconstruction and representation learning for high-resolution acoustic imaging will be discussed in more detail, highlighting the experimental results and the scientific contributions.

    About the speaker

    Angeliki Xenaki received the Diploma degree in electrical engineering and computer science from the National Technical University of Athens, Greece, in 2007, and the M.Sc. and Ph.D. degrees in acoustics from the Technical University of Denmark (DTU), in 2010 and 2015, respectively.  From 2012 to 2014 she was a visiting researcher at the Scripps Institution of Oceanography, University of California San Diego and from 2015 to 2016 she was a postdoctoral researcher at DTU. From 2016 to 2018 she was a research scientist at GN Hearing A/S, Denmark, specializing in array signal processing for hearing devices. In 2018, she joined the Centre for Maritime Research and Experimentation, Italy, as a Scientist working in the field of synthetic aperture sonar. Her research interests include signal processing, statistical modeling and machine learning.

     

    Σύνδεσμος εκδήλωσης

     

     


© Πολυτεχνείο Κρήτης 2012