Scientific Production

SEARCH

Search

PhD Thesis

Estimation Parcimonieuse et Apprentissage de Dictionnaires pour la Détection d'Anomalies Multivariées dans des Données Mixtes de Télémesure Satellite

Author: Pilastre Barbara

Defended on November 6, 2020.

DOWNLOAD DOCUMENT

La surveillance automatique de systèmes et la prévention des pannes sont des enjeux majeurs dans de nombreux secteurs et l’industrie spatiale ne fait pas exception. Par exemple, le succès des missions des satellites suppose un suivi constant de leur état de santé réalisé à travers la surveillance de la télémesure. Les signaux de télémesure sont des données issues de capteurs embarqués qui sont reçues sous forme de séries temporelles décrivant l’évolution dans le temps de différents paramètres. Chaque paramètre est associé à une grandeur physique telle qu’une température, une tension ou une pression, ou à un équipement dont il reporte le fonctionnement à chaque instant. Alors que les approches classiques de surveillance atteignent leurs limites, les méthodes d’apprentissage automatique (machine learning en anglais) s’imposent afin d’améliorer la surveillance de la télémesure via un apprentissage semi-supervisé : les signaux de télémesure associés à un fonctionnement normal du système sont appris pour construire un modèle de référence auquel sont comparés les signaux de télémesure récemment acquis. Les méthodes récentes proposées dans la littérature ont permis d’améliorer de manière significative le suivi de l’état de santé des satellites mais elles s’intéressent presque exclusivement à la détection d’anomalies univariées pour des paramètres physiques traités indépendamment. L’objectif de cette thèse est de proposer des algorithmes pour la détection d’anomalies multivariées capables de traiter conjointement plusieurs paramètres de télémesure associés à des données de différentes natures (continues/discrètes), et de prendre en compte les corrélations et les relations qui peuvent exister entre eux. L’idée motrice de cette thèse est de supposer que la télémesure fraîchement reçue peut être estimée à partir de peu de données décrivant un fonctionnement normal du satellite. Cette hypothèse justifie l’utilisation de méthodes d’estimation parcimonieuse et d’apprentissage de dictionnaires qui seront étudiées tout au long de cette thèse. Une deuxième forme de parcimonie propre aux anomalies satellites a également motivé ce choix, à savoir la rareté des anomalies satellites qui affectent peu de paramètres en même temps. Dans un premier temps, un algorithme de détection d’anomalies multivariées basé sur un modèle d’estimation parcimonieuse est proposé. Une extension pondérée du modèle permettant d’intégrer de l’information externe est également présentée ainsi qu’une méthode d’estimation d’hyperparamètres qui a été développée pour faciliter la mise en œuvre de l’algorithme. Dans un deuxième temps, un modèle d’estimation parcimonieuse avec un dictionnaire convolutif est proposé. L’objectif de cette deuxième méthode est de contourner le problème de non-invariance par translation dont souffre le premier algorithme. Les différentes méthodes proposées sont évaluées sur plusieurs cas d’usage industriels associés à de réelles données satellites et sont comparées aux approches de l’état de l’art.

Signal and image processing / Other

READ MORE

PhD Defense Slides

Estimation Parcimonieuse et Apprentissage de Dictionnaires pour la Détection d'Anomalies Multivariées dans des Données Mixtes de Télémesure Satellite

Author: Pilastre Barbara

Defended on November 6, 2020.

DOWNLOAD DOCUMENT

La surveillance automatique de systèmes et la prévention des pannes sont des enjeux majeurs dans de nombreux secteurs et l’industrie spatiale ne fait pas exception. Par exemple, le succès des missions des satellites suppose un suivi constant de leur état de santé réalisé à travers la surveillance de la télémesure. Les signaux de télémesure sont des données issues de capteurs embarqués qui sont reçues sous forme de séries temporelles décrivant l’évolution dans le temps de différents paramètres. Chaque paramètre est associé à une grandeur physique telle qu’une température, une tension ou une pression, ou à un équipement dont il reporte le fonctionnement à chaque instant. Alors que les approches classiques de surveillance atteignent leurs limites, les méthodes d’apprentissage automatique (machine learning en anglais) s’imposent afin d’améliorer la surveillance de la télémesure via un apprentissage semi-supervisé : les signaux de télémesure associés à un fonctionnement normal du système sont appris pour construire un modèle de référence auquel sont comparés les signaux de télémesure récemment acquis. Les méthodes récentes proposées dans la littérature ont permis d’améliorer de manière significative le suivi de l’état de santé des satellites mais elles s’intéressent presque exclusivement à la détection d’anomalies univariées pour des paramètres physiques traités indépendamment. L’objectif de cette thèse est de proposer des algorithmes pour la détection d’anomalies multivariées capables de traiter conjointement plusieurs paramètres de télémesure associés à des données de différentes natures (continues/discrètes), et de prendre en compte les corrélations et les relations qui peuvent exister entre eux. L’idée motrice de cette thèse est de supposer que la télémesure fraîchement reçue peut être estimée à partir de peu de données décrivant un fonctionnement normal du satellite. Cette hypothèse justifie l’utilisation de méthodes d’estimation parcimonieuse et d’apprentissage de dictionnaires qui seront étudiées tout au long de cette thèse. Une deuxième forme de parcimonie propre aux anomalies satellites a également motivé ce choix, à savoir la rareté des anomalies satellites qui affectent peu de paramètres en même temps. Dans un premier temps, un algorithme de détection d’anomalies multivariées basé sur un modèle d’estimation parcimonieuse est proposé. Une extension pondérée du modèle permettant d’intégrer de l’information externe est également présentée ainsi qu’une méthode d’estimation d’hyperparamètres qui a été développée pour faciliter la mise en œuvre de l’algorithme. Dans un deuxième temps, un modèle d’estimation parcimonieuse avec un dictionnaire convolutif est proposé. L’objectif de cette deuxième méthode est de contourner le problème de non-invariance par translation dont souffre le premier algorithme. Les différentes méthodes proposées sont évaluées sur plusieurs cas d’usage industriels associés à de réelles données satellites et sont comparées aux approches de l’état de l’art.

Signal and image processing / Other

READ MORE

Conference Paper

Constrained Bundle Adjustment Applied to Wing 3D Reconstruction with Mechanical Limitations

Authors: Demoulin Quentin, Lefebvre-Albaret François, Basarab Adrian, Kouamé Denis and Tourneret Jean-Yves

In Proc. IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates, October 25-28, 2020.

DOWNLOAD DOCUMENT

Aircraft certification procedures require the estimation of wing deformation, which is a very challenging problem in photogrammetry applications. Indeed, in real flight conditions with varying environment, 3D reconstruction is strongly degraded. To cope with this issue, we propose to introduce prior knowledge about the wing mechanical limits in the photogrammetry reconstruction method. These mechanical limits are expressed as appropriate regularizations that are included into the classical bundle adjustment step. The proposed approach is evaluated using data acquired on a real aircraft yielding promising results.

Signal and image processing / Aeronautical communication systems

READ MORE

QUIC: Opportunities and threats in SATCOM

Authors: Kuhn Nicolas, Michel François, Thomas Ludovic, Dubois Emmanuel and Lochin Emmanuel

In Proc. Advanced Satellite Multimedia Systems (ASMS), Graz, Austria, October 20-21, 2020.

DOWNLOAD DOCUMENT

This article proposes a discussion on the strengths, weaknesses, opportunities and threats related to the deployment of QUIC end-to-end from a satellite-operator point-of-view. The deployment of QUIC is an opportunity for improving the quality of experience when exploiting satellite broadband accesses. Indeed, the fast establishment of secured connections reduces the short files transmission time. Moreover, removing transport layer performance enhancing proxies reduces the cost of network infrastructures and improves the integration of satellite systems. However, the congestion and flow controls at end points are not always suitable for satellite communications due to the intrinsic high bandwidth-delay product. Further acceptance of QUIC in satellite systems would be guaranteed if its performance in specific use-cases is increased. We propose a running code for an IETF document, and based on an emulated platform and on open-source software, this paper proposes values of performance metrics just as one piece of the puzzle. The final performance objective requires consensus among the different actors. The objective should be challenging enough for satellite operators to allow QUIC traffic but reasonable enough to keep QUIC deployable on the Internet.

Networking / Space communication systems

READ MORE

Improving the estimation of the sea level anomaly slppe

Authors: Mailhes Corinne, Besson Olivier, Guillot Amandine and Le Gac Sophie

in Proc. IEEE Int. Geosci. Remote Sens. Symp. (IGARSS), Hawaï, USA, 26 September - 2 October 2020.

DOWNLOAD DOCUMENT

Satellite altimeters provide sea level measurements along satellite track. A mean profile based on the measurements averaged over a time period is then subtracted to estimate the sea level anomaly (SLA). In the spectral domain, SLA is characterized by a power spectral density of the form one over a power of the frequency where the power (the slope) is a parameter of great interest for ocean monitoring. However, this information lies in a narrow frequency band, located at very low frequencies, which calls for some specific spectral analysis methods. This paper studies a new parametric method based on an autoregressive model combined with a warping of the frequency scale (denoted as ARWARP). A statistical validation is proposed on simulated SLA signals, showing the performance of slope estimation using this ARWARP spectral estimator, compared to classical Fourier-based methods. Application to Sentinel-3 real data highlights the main advantage of the ARWARP model, making possible SLA slope estimation on a short signal segment, i.e., with a high spatial resolution.

Signal and image processing / Earth observation

READ MORE

Journal Paper

New multiplexing method to add a new signal in the Galileo E1 band

Authors: Ortega Espluga Lorenzo, Poulliat Charly, Boucheret Marie-Laure, Aubault-Roudier Marion and Al Bitar Hanaa

IET Radar, Sonar & Navigation, E-First, Print pp.1751-8784, Online pp. 1751-8792, September, 2020.

DOWNLOAD DOCUMENT

This work addresses the problem of integrating a new signal in the Galileo E1 band. Thus, the arising question is how the existing multiplexing methods can be efficiently used or modified to integrate a new binary signal in the Galileo E1 band with the existing Galileo E1 signals. To this end, in this study, the authors first select three efficient multiplexing methods from the state of the art (i.e. interplexing, POCET and CEMIC methods) to multiplex a new Galileo signal along with the Galileo E1 legacy signals in a constant envelope modulation. Moreover, they evaluate their performance and main advantages and drawbacks. Secondly, in order to improve both performance and flexibility/adaptability of the multiplexing method, a modified CEMIC method, called ACEMIC, is proposed. This method allows to design modulations which maximise the power efficiency with respect to a given peak-to-average-power-ratio constraint. Finally, they compare the previous multiplexing methods in terms of power signal distribution, constant envelope fluctuation and power efficiency.

Signal and image processing / Localization and navigation and Space communication systems

READ MORE

Conference Paper

An Assessment Methodology of Smartphones Positioning Performance for Collaborative Scenarios in Urban Environment

Authors: Verheyde Thomas, Blais Antoine, Macabiau Christophe and Marmet François-Xavier

In Proc. ION GNSS+, St Louis, Missouri, USA, September 21-25, 2020.

DOWNLOAD DOCUMENT

The release of Android Global Navigation Satellite Systems (GNSS) raw measurements in late 2016 unlocked the access of smartphones' embedded positioning chipset capabilities for developers and the scientific community. This groundbreaking announcement was followed by technical innovations, made by smartphone brands and chipset manufacturers, in order to obtain the world's most precise smartphone on the market. In recent years, several studies investigated the development of advanced positioning techniques (e.g. Precise Point Positioning (PPP), Real-Time Kinematic (RTK)) using Android raw data measurements. However, most studies drawn their conclusions based on one smartphone brand and model in optimal open-sky conditions despite the fact that most smartphone-based positioning activities are achieved in urban and sub-urban areas. In order to overcome urban smartphone-based positioning issues, we ambition to develop a collaborative user’s network taking advantage of the tremendous numbers of connected Android devices in today's busy city centers. A throughout study has been conducted in the city center of Toulouse in France for characterizing smartphone positioning performance in both nominal and urban conditions. Various limiting factors were exposed during our data collection campaign. Nevertheless, the investigation conducted on Android GNSS raw measurement uncovered smartphone positioning potential for navigation applications in constraint environment. A methodology assessment has been implemented in order to identify, characterize and compare smartphones’ positioning performances. A classification of key parameters has been determined focusing on the implementation of collaborative algorithms, revealing the attributes and components for smartphone-based collaborative methods. Thereafter, a comprehensive state of the art review on existing cooperative positioning techniques, has been achieved. An evaluation of the feasibility and the applicability of those methods into the smartphone domain has been made. We present a method based on simple assumptions, without third-party equipment and data, only relying on smartphones’ own data combination. Our cooperative network can be described as a low-cost embedded structure aiming at providing positioning assistance to its users.

Digital communications / Localization and navigation

READ MORE

Hybrid Navigation Filters Performances Between GPS, Galileo and 5G TOA Measurements in Multipath Environment

Authors: Tobie Anne-Marie, Garcia Pena Axel, Thevenon Paul, Vezinet Jérémy and Aubault-Roudier Marion

In Proc. ION GNSS+, St Louis, Missouri, USA, September 21-25, 2020.

DOWNLOAD DOCUMENT

In this paper, the performance of different hybrid navigation filters exploiting GPS, Galileo and 5G Time Of Arrival (TOA) measurements in multipath environment are studied. For the realism of the study, realistic propagation channels must be considered and their impacts on the received signals processing must be accurately modelled. GNSS signal mathematical models in multipath environment have been analyzed for a long time. However, 5G mathematical models in a realistic multipath environment are still in its early stages of analysis. This article is divided in three main parts. The first part is dedicated to the identification of compliant GNSS and 5G signal propagation channel models; SCHUN is selected for GNSS and QuaDRiGa is selected for 5G. Based on this, the correlator output mathematical models for 5G signals and GNSS signals are derived. The second part tackles the accurate characterization of the pseudo range errors due to propagation channels shadowing and multipath effect as well as thermal noise. This step is required for the correct derivation of the navigation filters. Indeed, the study will focus on Extended Kalman Filters (EKF) and Unscented Kalman Filters (UKF); both assume a Gaussian distribution of the errors. Therefore, by optimally characterizing the errors, the performances of the filters are expected to be improved. The last part consists in validating through simulations the theory and mathematical models developed in the first and second parts.

Digital communications / Localization and navigation

READ MORE

Hybrid Navigation Filters Performances Between GPS, Galileo and 5G TOA Measurements in Multipath Environment

Authors: Tobie Anne-Marie, Garcia Pena Axel, Thevenon Paul, Vezinet Jérémy and Aubault-Roudier Marion

In Proc. ION GNSS+, St Louis, Missouri, USA, September 21-25, 2020.

DOWNLOAD DOCUMENT

In this paper, the performance of different hybrid navigation filters exploiting GPS, Galileo and 5G Time Of Arrival (TOA) measurements in multipath environment are studied. For the realism of the study, realistic propagation channels must be considered and their impacts on the received signals processing must be accurately modelled. GNSS signal mathematical models in multipath environment have been analyzed for a long time. However, 5G mathematical models in a realistic multipath environment are still in its early stages of analysis. This article is divided in three main parts. The first part is dedicated to the identification of compliant GNSS and 5G signal propagation channel models; SCHUN is selected for GNSS and QuaDRiGa is selected for 5G. Based on this, the correlator output mathematical models for 5G signals and GNSS signals are derived. The second part tackles the accurate characterization of the pseudo range errors due to propagation channels shadowing and multipath effect as well as thermal noise. This step is required for the correct derivation of the navigation filters. Indeed, the study will focus on Extended Kalman Filters (EKF) and Unscented Kalman Filters (UKF); both assume a Gaussian distribution of the errors. Therefore, by optimally characterizing the errors, the performances of the filters are expected to be improved. The last part consists in validating through simulations the theory and mathematical models developed in the first and second parts.

Digital communications / Localization and navigation

READ MORE

On the Time-Delay Estimation Accuracy Limit of GNSS Meta-Signals

Authors: Ortega Espluga Lorenzo, Vilà-Valls Jordi, Chaumette Eric and Vincent François

In Proc. Intelligent Transportation Systems Conference (IEEE/ITSC), Rhodes, Greece, September 20-23, 2020.

DOWNLOAD DOCUMENT

In standard two-step Global Navigation Satellite Systems (GNSS) receiver architectures the precision on the position, velocity and time estimates is driven by the precision on the intermediate parameters, i.e., delays and Dopplers. The estimation of the time-delay is in turn driven by the baseband signal resolution, that is, by the type of broadcasted signals. Among the different GNSS signals available the socalled AltBOC modulated signal, appearing in the Galileo E5 band and the new GNSS meta-signal concept, is the one which may provide the better time-delay precision. In order to meet the constraints of safety-critical applications such as Intelligent Transportation Systems or automated aircraft landing, it is fundamental to known the ultimate code-based precision achievable by standalone GNSS receivers. The main goal of this contribution is to assess the time-delay precision of AltBOC type signals. The analysis is performed by resorting to a new compact closed-form Cramér-Rao bound expression for time-delay estimation which only depends on the signal samples. In addition, the corresponding time-delay maximum likelihood estimate is also provided to assess the minimum signal-to-noise ratio that allows to be in optimal receiver operation.

Signal and image processing / Localization and navigation and Space communication systems

READ MORE

Activity Report

READ MORE

Apply for a PhD in Safran

Codirection by Toulouse INP & ISAE-Supaero

READ MORE

PhD positions available at TeSA

PhD subjects available on the CNES site.
On line application before the 2nd of April.

READ MORE

EUSIPCO 2020

Conference talk of Julien Lesouple at EUSIPCO 2020

READ MORE