Koehl V.,CNRS Communication and Information Sciences Laboratories |
Paquier M.,CNRS Communication and Information Sciences Laboratories
Applied Acoustics | Year: 2013
Paradoxically, one of the hardest to measure characteristics of a sound reproduction device such as a loudspeaker is its sound quality. The perception of this subjective character is linked to numerous parameters (stimulus type, listening environment, etc.) that must be drastically controlled to lead to reliable and repeatable judgments. Industrial and academic researchers are still focusing on the design of standard assessment procedures. The conditions under which a sound reproduction system is assessed in laboratory tests is often very far from those under which it is designed to be used. As a result, the assessment task might appear unnatural to test subjects, which could possibly bias the test results. The aim of this study is to compare, on the basis of sound quality ratings, three different test procedures based on paired comparison and exhibiting procedural differences. One of the procedures consisted in comparing loudspeakers by listening to short music excerpts (5 s) at a preset level, which was assumed to be a very controllable method. In the two other procedures, the listener could compare the systems by listening to long music excerpts (30 s), which was assumed to be more natural for loudspeaker assessment. The level was either preset by expert listeners or set by the subject himself in the two latter procedures. This paper shows that the test results were very stable over the different assessment procedures, but that some of them enabled, under certain conditions, to separate between systems obtaining very close quality ratings. © 2013 Elsevier Ltd. All rights reserved.
Saucan A.-A.,CNRS Communication and Information Sciences Laboratories
IEEE Aerospace and Electronic Systems Magazine | Year: 2015
In this article, we highlight recent advances in adaptive sonar-array processing for three-dimensional (3-D) depth map reconstruction, i.e., bathymetry. Bathymetry reconstruction of underwater environments is of great importance for applications ranging from infrastructure inspection to intrusion detection. Side-scan images, that is, two-dimensional (2-D) images of the backscattered signal amplitude are affected by the layover phenomenon, i.e., superposition of echoes. This phenomenon appears in the presence of tall objects, whenever the pulse emitted by the sonar ensonifies the top of the object before the bottom. In such, the area around the object is not entirely observable and a full 3-D bathymetry reconstruction is necessary. For sidelooking sonar arrays, bathymetry is obtained by triangulating the scatter-ers corresponding to each range bin, or equivalently, estimating the direction of arrival (DOA) of backscattered echoes. © 1986-2012 IEEE.
Diguet J.-P.,CNRS Communication and Information Sciences Laboratories
Proceedings, SBCCI 2014 - 27th Symposium on Integrated Circuits and Systems Design: Chip in Aracaju | Year: 2014
NoC and highly variable trafics will be quite common in future SoCs. Self-Adaptive NoCs provide a solution to optimise energy efficiency and performances at runtime in this uncertain environment. This paper first provides a brief survey of of self-adaptive NoCs. Secondly it presents a solution based on Self-Adaptive Network Interface include configurable FIFOs and TDMA tables. Finally it draws perpectives. Copyright 2014 ACM.
Paquier M.,CNRS Communication and Information Sciences Laboratories |
Koehl V.,CNRS Communication and Information Sciences Laboratories
Applied Acoustics | Year: 2015
Several studies have shown that differences in the placement of a headphone over a listener's ears could result in large differences in the measurements of the related transfer function (HPTF). Nevertheless, because of - at least - the non-uniform frequency resolution of human hearing system, large HPTF variations at some frequencies do not necessary imply audible consequences, which were not evaluated by past studies. The present study aims at evaluating the audibility of spectral modifications introduced by slight but realistic changes in the headphone placement over a listener's ears. Recordings were performed by placing/replacing a headphone on a dummy head. Various headphone models were realistically placed eight times each on the artificial head. Music excerpts and pink noise then were played back over the headphones and recorded with microphones located at the entrance of the blocked ear canal. These recordings were then presented to expert and naïve listeners over a single test headphone. The subjects had to discriminate between stimuli standing for different headphone placements using a 3I3AFC procedure. With the exception of the naïve listeners about one given music excerpt only, subjects were always able to discriminate between the stimuli with respect to their corresponding headphone placement. This indicates that consecutive realistic headphone placements may result in audible differences for the listener. Such a result could raise several issues about the use of headphones for psychoacoustic experiments, especially for multi-session tests. © 2015 Elsevier Ltd.
Murovec T.,CNRS Communication and Information Sciences Laboratories |
Brosseau C.,CNRS Communication and Information Sciences Laboratories
Applied Physics Letters | Year: 2013
By applying an ac electric field to a pair of layered concentric shells, a prototypical model of biological cells, we demonstrate that the sign and magnitude of the time averaged electrostatic force (EF) are strongly dependent on the field frequency. Crossover frequencies (CFs) occur when EFs vanish. Unique physical features of CF bands are studied as function of field frequency and gap distance between cells. The distance between cells corresponding to the CF shows both anisotropic and universal features. The ability to control the long-range EFs and CFs opens up exciting prospective applications including the deposition of biological cells under field excitation in an architecture that could create functional tissue. © 2013 AIP Publishing LLC.
Chi J.W.D.,CNRS Communication and Information Sciences Laboratories
2016 10th International Symposium on Communication Systems, Networks and Digital Signal Processing, CSNDSP 2016 | Year: 2016
The cleanness of fast optical pulses, especially those transmitted through optical communication networks, is overwhelmingly referenced to and judged by the so-called 'transform-limited pulse', i.e., pulses with full-width at half-maximum (FWHM)-defined time-bandwidth product around 0.5. In this paper, we will demonstrate the fundamental flaw of this notion using our simulation results. Moreover, the pulse width is usually obtained by optical intensity autocorrelation plus curve-fitting, which may be misleading as well. Consequently, these standard methods of pulse characterization should be thoroughly re-considered. © 2016 IEEE.
Masse D.,CNRS Communication and Information Sciences Laboratories
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2014
Termination analyzers generally synthesize ranking functions or relations, which represent checkable proofs of their results. In , we proposed an approach for conditional termination analysis based on abstract fixpoint computation by policy iteration. This method is not based on ranking functions and does not directly provide a ranking relation, which makes the comparison with existing approaches difficult. In this paper we study the relationships between our approach and ranking functions and relations, focusing on extensions of linear ranking functions. We show that it can work on programs admitting a specific kind of segmented ranking functions, and that the results can be checked by the construction of a disjunctive ranking relation. Experimental results show the interest of this approach. © 2014 Springer-Verlag Berlin Heidelberg.
Bouchikhi A.,IRENav |
Bouchikhi A.,CNRS Communication and Information Sciences Laboratories |
Signal Processing | Year: 2012
In this paper a signal analysis framework for estimating time-varying amplitude and frequency functions of multicomponent amplitude and frequency modulated (AM-FM) signals is introduced. This framework is based on local and non-linear approaches, namely Energy Separation Algorithm (ESA) and Empirical Mode Decomposition (EMD). Conjunction of Discrete ESA (DESA) and EMD is called EMD-DESA. A new modified version of EMD where smoothing instead of an interpolation to construct the upper and lower envelopes of the signal is introduced. Since extracted IMFs are represented in terms of B-spline (BS) expansions, a closed formula of ESA robust against noise is used. Instantaneous Frequency (IF) and Instantaneous Amplitude (IA) estimates of a multicomponent AM-FM signal, corrupted with additive white Gaussian noise of varying SNRs, are analyzed and results compared to ESA, DESA and Hilbert transform-based algorithms. SNR and MSE are used as figures of merit. Regularized BS version of EMD-ESA performs reasonably better in separating IA and IF components compared to the other methods from low to high SNR. Overall, obtained results illustrate the effectiveness of the proposed approach in terms of accuracy and robustness against noise to track IF and IA features of a multicomponent AM-FM signal. © 2012 Elsevier B.V. All rights reserved.
Liu Z.-G.,Northwestern Polytechnical University |
Liu Z.-G.,CNRS Communication and Information Sciences Laboratories |
Pan Q.,Northwestern Polytechnical University |
Dezert J.,ONERA |
Mercier G.,CNRS Communication and Information Sciences Laboratories
Pattern Recognition | Year: 2014
In this paper we present a new credal classification rule (CCR) based on belief functions to deal with the uncertain data. CCR allows the objects to belong (with different masses of belief) not only to the specific classes, but also to the sets of classes called meta-classes which correspond to the disjunction of several specific classes. Each specific class is characterized by a class center (i.e. prototype), and consists of all the objects that are sufficiently close to the center. The belief of the assignment of a given object to classify with a specific class is determined from the Mahalanobis distance between the object and the center of the corresponding class. The meta-classes are used to capture the imprecision in the classification of the objects when they are difficult to correctly classify because of the poor quality of available attributes. The selection of meta-classes depends on the application and the context, and a measure of the degree of indistinguishability between classes is introduced. In this new CCR approach, the objects assigned to a meta-class should be close to the center of this meta-class having similar distances to all the involved specific classes centers, and the objects too far from the others will be considered as outliers (noise). CCR provides robust credal classification results with a relatively low computational burden. Several experiments using both artificial and real data sets are presented at the end of this paper to evaluate and compare the performances of this CCR method with respect to other classification methods. © 2014 Elsevier Ltd. All rights reserved.
Masse D.,CNRS Communication and Information Sciences Laboratories
Electronic Notes in Theoretical Computer Science | Year: 2012
In this paper, we explore the adaptation of policy iteration techniques to compute greatest fixpoint approximations, in order to find sufficient conditions for program termination. Restricting ourselves to affine programs and the abstract domain of template constraint matrices, we show that the abstract greatest fixpoint can be computed exactly using linear programming, and that strategies are related to the template constraint matrix used. We also present a first result on the relationships between this approach and methods which use ranking functions. © 2012 Elsevier B.V.