One of the main objectives of hyperspectral image processing is to detect a given target among an unknown background. The standard data to conduct such a detection is a reflectance map, where the spectral signatures of each pixel’s components, known as endmembers, are associated with their abundances in the pixel. Due to the low spatial resolution of most hyperspectral sensors, such a target occupies a fraction of the pixel. A widely used model in case of subpixel targets is the replacement model. Among the vast number of possible detectors, algorithms matched to the replacement model are quite rare. One of the few examples is the Finite Target Matched Filter, which is an adjustment of the well-known Matched Filter. In this paper, we derive the exact Generalized Likelihood Ratio Test for this model. This new detector can be used both with a local covariance estimation window or a global one. It is shown to outperform the standard target detectors on real data, especially for small covariance estimation windows.

Detecting a target of known spectral signature from an unknown background is one of main goal of hyperspectral imaging. As the majority of hyperspectral imaging systems have a poor spatial resolution, subpixel targets are usual. In this case, the so-called replacement model is commonly advocated. This model, valid for reflectance images, specifies that if a target is present, the amount of background should reduce in the same proportion. Nevertheless, the majority of the standard detectors, such as the Match Filter or the Kelly detector, have been developed for different contexts, and do not exploit this constraint. One of the rare example that is suitable for the replacement model is the Finite Target Match Filter, which is known to improve the target selectivity detection. In this paper, we develop the exact Generalized Likelihood Ratio Test for the model at hand. We show that this new detector outperforms the standard ones, on a real data experiment.

Spacecraft health monitoring from housekeeping telemetry data represents one of the main issues in space operations. Motivated by the success of machine learning or data driven-based methods in many signal and image processing applications, some of these methods have been applied to anomaly detection in housekeeping telemetry via a semi-supervised learning. This paper studies a new multivariate anomaly detection algorithm based on a sparse decomposition on a dictionary of nominal patterns. One originality of the proposed method is a multivariate framework allowing us to take into account possible relationships between different telemetry parameters, in particular through a joint processing of time-series described by mixed continuous and discrete parameters. The proposed method is tested with real satellite telemetry and evaluated on a representative anomaly dataset composed of actual anomalies that occurred on several operated satellites. The first results confirm the interest of the proposed method and demonstrate its competitiveness with respect to the state-of-the-art.

Single-photon lidar devices can acquire 3D data at very long range with high precision. Moreover, recent advances in lidar arrays have enabled acquisitions at very high frame rates. However, these devices place a severe bottleneck on the reconstruction algorithms, which have to handle very large volumes of noisy data. Recently, real-time 3D reconstruction of distributed surfaces has been demonstrated obtaining information at one wavelength. Here, we propose a new algorithm that achieves color 3D reconstruction without increasing the execution time nor the acquisition process of the realtime single-wavelength reconstruction system. The algorithm uses a coded aperture that compresses the data by considering a subset of the wavelengths per pixel. The reconstruction algorithm is based on a plug-and-play denoising framework, which benefits from off-the-shelf point cloud and image de-noisers. Experiments using real lidar data show the competitivity of the proposed method.

Named Data Networking (NDN) proposes to change the core of the Internet. Based on mechanisms successfully used in P2P or CDN, it focuses on content and thus the Quality of Experience of users. Such an ambitious plan raises great challenges : caching, multipath, multi-producers, multi-consumers and security. This paper focuses on one of them: the control of congestion. Several studies have proposed congestion control solutions that fall into three kinds: the end-to-end solution, the hop-by-hop type and the hybrid one. However, the community lacks proper evaluations of such specific algorithms. In this work, we have implemented representative solutions on ndnSIM. In a first step, we have tested them on a small scale topology to ease their performance analysis and highlight their strengths and weaknesses. We complete this study with simulations on larger networks in order to confirm our conclusions. Furthermore, all results are reproducible. Eventually, the paper drives a discussion on how application needs could be considered in the design of a NDN congestion control.

In space operations, spacecraft health monitoring and failure prevention are major issues. This important task can be handled by monitoring housekeeping telemetry time series using anomaly detection (AD) techniques. The success of machine learning methods makes them attractive for AD in telemetry via a semi-supervised learning. Semi-supervised learning consists of learning a reference model from past telemetry acquired without anomalies in the so-called learning step. In a second step referred to as test step, most recent telemetry time-series are compared to this reference model in order to detect potential anomalies. This paper presents an extension of an existing AD method based on a sparse decomposition of test signals on a dictionary of normal patterns. The proposed method has the advantage of accounting for possible relationships between different telemetry parameters and can integrate external information via appropriate weights that allow detection performance to be improved. After recalling the main steps of an existing AD method based on a sparse decomposition [1] for multivariate telemetry data, we investigate a weighted version of this method referred to as W-ADDICT that allows external information to be included in the detection step. Some representative results obtained using an anomaly dataset composed of actual anomalies that occurred on several satellites show the interest of the proposed weighting strategy using external information obtained from the correlation coecient between the tested data and its decomposition on the dictionary.

This paper aims at describing some industrial aerospace applications requiring anomaly detection. After introducing the industrial context and needs, the paper investigates two examples of oscillation/vibration detection associated with aero-elastic phenomena: (i) the first example related to aircraft structural design optimization is characterized by very stringent constraints on detection time; (ii) the second example linked to equipment condition monitoring has more relaxed constraints in terms of detection requirements.

In the context of automatic and preventive condition monitoring of rotating machines, this paper presents a case study of a naturally-worn parallel straight gear by monitoring the evolution of the modulation functions. The Hilbert demodulation is automatically performed considering only the frequency content of the signals detected by the AStrion software. The gear has been worn over 3000 hours with a constant axial load. A particular focus is set on the amplitude modulation function in order to assess its efficiency to characterize both the severity of the wear and the most worn part of the gear. The results are confronted with on-site observation of the teeth. For this purpose, the evolution of both amplitude and phase modulations over several meshing harmonics are compared, as well as demodulation on both original and residual signals. Indicators to automatically classify the wear are discussed.

The effective coverage of satellites and the technology behind have motivated many actors to develop efficient communications for Internet access, television and telephony. For a long time, reservation resources of Demand Assignment Multiple Access (DAMA) techniques have been largely deployed in the return link of satellite communications, occupying most of the frequency bandwidth. However, these resources cannot follow the technological growth with big users communities in applications like the Internet of Things and Machine to Machine communications. Especially because the Round Trip Time is significant in addition to a potential underuse of the resources. Thus, access protocols based on ALOHA took over a big part of the Random Access (RA) research area and have considerably evolved lately. CRDSA have particularly put its fingerprint in this domain, which inspired many different techniques. In this context, a complementary method, called MARSALA comes to unlock CRDSA when packets can no longer be retrieved. This actually involves a correlation complexity related to packet localization which is necessary for replicas combinations that results in a potentially higher signal power. Accordingly, the main goal of this PhD research is to seek for effective and less complex alternatives. More precisely, the core challenge focuses on the way to manage multi-user transmissions and solve interference at reception, with the smallest complexity. In addition, the loop phenomenon which occur when multiple users transmit their packets at the same positions is tackled as it creates an error floor at the packet loss ratio performance. Synchronous and asynchronous solutions are proposed in this thesis, mainly based on providing the transmitter and the receiver with a shared prior information that could help reduce the complexity, mitigate the loop phenomenon and enhance the system performance. An in-depth description and analysis of the proposed techniques are presented in this dissertation.