Η διαμόρφωση με πολύπλεξη συχνότητας ορθογωνίων φερουσών (Orthogonal Frequency Division Multiplexing - OFDM) έχει εδραιωθεί ως μία από τις επικρατέστερες μεθόδους διαμόρφωσης για την υψηλού ρυθμού μετάδοση πληροφορίας μέσω ασύρματων μέσων. Σε ένα σύστημα OFDM, ένα από τα βασικότερα και υπολογιστικά πολυπλοκότερα τμήματα είναι ο υπολογισμούς του Ταχύ Μετασχηματισμού Fourier. Αντικείμενο της εργασίας αυτής είναι η μελέτη της χρήσης εναλλακτικής αριθμητικής για την υλοποίηση κυκλωμάτων FFT. Τυπικά, τέτοιου είδους κυκλώματα υλοποιούνται χρησιμοποιώντας κάποια γραμμική αναπαράσταση σταθερής υποδιαστολής. Στη βιβλιογραφία έχουν προταθεί υλοποιήσεις του FFT με χρήση του Λογαριθμικού Συστήματος Αρίθμησης (Logarithmic Numbering System – LNS) και έχουν αναφερθεί κέρδη για συγκεκριμένους παράγοντες όπως το σφάλμα κβαντισμού, η επιφάνεια ολοκλήρωσης και η κατανάλωση ισχύος. Η αποδοτικότητα αυτών των λύσεων ερευνάται για τη συγκεκριμένη περίπτωση της εφαρμογής του FFT σε OFDM modems. Εστιάζοντας στην περίπτωση του FFT 64 σημείων για OFDM modem για ασύρματα δίκτυα 802.11a, μία από τις πλέον επιτυχημένες αρχιτεκτονικές που έχουν προταθεί για την υλοποίηση του, στηρίζεται στη λογική του FFT γραμμής – στήλης και παρουσιάζει έναν τρόπο πραγματοποίησης του υπολογισμού χωρίς κανένα ψηφιακό πολλαπλασιαστή. Με το βασικό πλεονέκτημα της λογαριθμικής αναπαράστασης να είναι η απλοποίηση των κυκλωμάτων πολλαπλασιασμού (με ταυτόχρονη όμως αύξηση του κόστους για την πραγματοποίηση προσθέσεων), δείχνεται ότι τελικά η υλοποίηση ενός FFT αμιγώς σε LNS δεν είναι προτιμητέα. Αν και η αρχιτεκτονική του FFT γραμμής – στήλης μπορεί να προσφέρει υψηλή απόδοση με χαμηλό κόστος υλοποίησης, παρουσιάζει μια σειρά από αδυναμίες, που σχετίζονται κυρίως με τη χρήση ειδικών κυκλωμάτων για την εκτέλεση των πολλαπλασιασμών με τις σταθερές που εμφανίζονται στον FFT (twiddle factors). Για την αντιμετώπιση αυτών των περιορισμών προτείνεται η εισαγωγή του LNS σε κάποια τμήματα του κυκλώματος του FFT, οδηγώντας έτσι στη δημιουργία ενός συστήματος μικτής αναπαράστασης. Σε τέτοιου είδους υβριδικά συστήματα τίθενται δύο βασικά ζητήματα. Το πρώτο αφορά τον ορισμό της ισοδυναμίας μεταξύ των διαφορετικών αναπαραστάσεων και το δεύτερο τον αποδοτικό τρόπο υλοποίησης των κυκλωμάτων μετατροπής από το ένα αριθμητικό σύστημα στο άλλο. Τυπικά, τα κριτήρια ισοδυναμίας που επιλέγονται είναι αυστηρά μαθηματικά ορισμένα, όπως για παράδειγμα ο Λόγος Σήματος προς Θόρυβο (Signal-to-Noise Ratio - SNR) ή το Μέσο Σχετικό Σφάλμα Αναπαράστασης (Average Relative Representation Error – ARRE). Στη συγκεκριμένη εργασία ακολουθείται μια λιγότερο δεσμευτική προσέγγιση, ορίζοντας την ισοδυναμία δύο αναπαραστάσεων με βάση την τελική απόδοση του συστήματος OFDM όσον αφορά το ρυθμό λαθών στο δέκτη (Bit Error Rate - BER). Με βάση αυτή τη λογική, αποδεικνύεται ότι μπορούν να χρησιμοποιηθούν αναπαραστάσεις πολύ μικρού μεγέθους λέξης και οι προσεγγίσεις που χρειάζεται να γίνουν κατά τις μετατροπές μεταξύ των δύο συστημάτων δεν είναι ανάγκη να είναι ιδιαίτερα ακριβείς. Έτσι, τα σχετικά κυκλώματα μπορούν να υλοποιηθούν αποδοτικά και με μικρό κόστος. Η υλοποίηση δύο συστημάτων για τον FFT 64 σημείων, ένα βασισμένο αποκλειστικά σε γραμμική αναπαράσταση σταθερής υποδιαστολής και ένα υβριδικό που χρησιμοποιεί γραμμική και λογαριθμική αναπαράσταση, δείχνει ότι χωρίς διαφορές όσον αφορά το BER και την καθυστέρηση (delay), η υβριδική προσέγγιση απαιτεί μικρότερη επιφάνεια ολοκλήρωσης και παρουσιάζει σημαντικά χαμηλότερη κατανάλωση ισχύος. / Orthogonal Frequency Division Multiplexing (OFDM) has been established as one of the most prevalent methods for high data rate transmission through wireless channels. In an OFDM communication system, one of the fundamental and most computationally intensive parts is the computation of the Fast Fourier Transform (FFT). The subject of this thesis is to investigate the use of alternative arithmetic representation systems for the implementation of FFT circuits. Typically, these circuits are implemented using linear fixed-point representations. In literature, implementations of the FFT using the Logarithmic Numbering System (LNS) have been proposed and significant gains in quantization errors, chip area and power consumption have been reported. The effectiveness of these proposals in the case of the FFT for OFDM systems is investigated. Focusing on the case of the 64-point FFT for an OFDM modem for an 802.11a wireless network, one of the most efficient architectures proposed is based on the concept of row-column FFT and presents a way of implementing the computation without using any digital (non-fixed input) multiplier. The most important feature of the LNS representation is the fact that multiplication operations turn to mere additions, thus there are significant implementation gains. On the downside though, addition in LNS is very expensive. Combining the aforementioned, it is shown that the implementation of the whole FFT computation in LNS is not a preferable solution. Although the row-column FFT architecture may offer high performance and low implementation cost, it presents a number of deficiencies mainly due to the fact that special purpose circuits are used to perform the multiplications with the complex constants (twiddle factors) that appear in the computation. In order to alleviate these deficiencies, it is proposed to use the LNS representation in some parts of the FFT circuit, thus forming a hybrid-representation system. In hybrid-representation systems two major issues are raised. The first one is how to define equivalence between the arithmetic representation systems used and the second one is related to the cost of the circuits required to perform the conversions between the numbers of the different arithmetic systems. Typically, the equivalence criterion used is mathematically defined and metrics like the Signal-to-Noise Ratio (SNR) or Average Relative Representation Error (ARRE) are commonly used. In this report, a less restrictive metric is used: two arithmetic representations are defined to be equal if the Bit Error Rate (BER) performance of the overall OFDM system is equal. Using this approach, it is shown that short word-length representations may be used and the conversions between the linear and logarithmic systems need not be very accurate. This results in great simplification of the conversion process and the respected circuits can be implemented with low cost. For comparison, two 64-point FFT systems have been implemented, one using a linear fixed-point 2’s complement representation and one using both linear and LNS representation. Without any differences in BER performance and circuit delay, the hybrid-representation system requires less chip area and consumes significantly lower power.

No description available.

Secret key generation from reciprocal multi-antenna channels is an interesting alternative to cryptographic key management in wireless systems without infrastructure access. In this work, we study the secret key rate for the basic source model with a MIMO channel. First, we derive an expression for the secret key rate under spatial correlation modelled by the Kronecker model and with spatial precoding at both communication nodes. Next, we analyze the result for uncorrelated antennas to understand the optimal precoding for this special case, which is equal power allocation. Then, the impact of correlation is characterized using Majorization theory. Surprisingly for small SNR, spatial correlation increases the secret key rate. For high SNR, the maximum secret key rate is achieved for uncorrelated antennas. The results indicate that a solid system design for reciprocal MIMO key generation is required to establish the secret key rate gains.

Thesis (Master').

Ce travail de thèse a pour but de suivre le comportement du liant ionomère après vieillissement en condition réelle d'utilisation de la pile. Dans un premier temps, diverses techniques de caractérisation en phase solide ont été utilisées afin d'étudier le ionomère présent dans les électrodes. La majeure partie de ces techniques se sont avérées insatisfaisantes pour le suivi du vieillissement du ionomère. Seule l'analyse par diffraction rayon X (DRX) a mis en avant un changement d'organisation structurale du ionomère dans les électrodes. Par la suite, nous avons choisi de caractériser le ionomère en solution après extraction par l'eau. Préalablement, un protocole de quantification en 2 à 3 étapes, dans lequel intervient une quantification via le rapport signal-sur-bruit (S/N), a été mis en place afin de quantifier le Nafion® et autres petites molécules fluorés. L'extraction Soxhlet a ensuite été réalisée sur les électrodes permettant de révéler la présence d'un produit de dégradation hydrosoluble après fonctionnement en pile, à la fois en cathode et en anode. Enfin, la macération des électrodes dans le diméthylacétamide (DMAc) a permis d'extraire le polymère Nafion® ainsi que deux acides : l'acide trifluoroacétique (TFA) et l'acide triflique (TFI). En conclusion, la corrélation de l'ensemble des observations nous a permis de proposer un mécanisme de dégradation du liant ionomère présent dans les électrodes. / The purpose of this study was to follow the behavior of ionomer binder after fuel cell operation. First, a series of techniques were used to investigate to characterize ionomer in electrode at solid state. Most of them were inefficient to study ionomer degradation. Only X-Ray Diffraction (XRD) pointed out a structural change of the binder in electrodes. Second, it has been decided to characterize ionomer in liquid state after water extraction. But first of all, a quantitative 19F NMR protocol composed of two to three steps, with a first step using a quantification through signal-to-noise ratio (S/N), was establish in order to quantify Nafion® and degradation products. Soxhlet extraction performed on electrodes allowed to detect a degradation product water-soluble. Finally, extraction with organic solvent (Dimethylacetamide) allowed to extract Nafion® and two acid: trifluoroacetic acid (TFA) and triflic acid (TFI) from electrodes. In conclusion, correlation between observation and literature allowed us to propose a degradation mechanism of ionomer in electrodes.

Contact microphones, especially throat microphones, have been developed to be used in environments with high background noises to improve the speech intelligibility in communication. They pick up vibrations from the surface they are attached to and are therefore less sensitive to sound and noise from the air. Comparison of the speech intelligibility with other types of microphones have previously been done by letting test persons examine the communication devices in question. This study examines the possibility to make use of the STI-method instead and therefore make a comparison faster, more cost-efficient and customizable. The thought is that if the relationship between speech signals and vibrations were known, it could be used to transform the STI test signal to vibrations and then use the STImethod to estimate the speech intelligibility for the chosen contact microphone. This study, containing 22 men and women, evaluated the vibrations at the most suitable locations on the head for contact microphones and compared it with the speech signal of the same test person. Frequency responses were calculated for all locations of the head and a more detailed study showed that the frequency response of the neck may be approximated as a second order lowpass filter with a cut-off frequency of about 300 Hz that attenuates speech signals with higher frequencies. Experiments were also done to measure the STI value of a throat microphone with the known relationship. However, the results pointed out several problems that needs to be addressed before a STI method can be performed successfully. The results from this study may also be used to deeper study the relationships between different vibrations resulting from speech signals and suggestions on how the performance of contact microphones may be improved are given.

L'imagerie par résonance magnétique (IRM) appliquée au domaine vétérinaire exploite des systèmes à bas champ magnétostatique qui ont de nombreux avantages, notamment leur faible coût d'achat et d'entretien. Mais sur ces machines, les capteurs radiofréquence (RF) sont initialement dédiées à l'homme et ne permettent pas une qualité d'image optimale. Dans le cadre de cette thèse, des méthodes simples d'optimisation de capteurs à 0,2 T ont été développées, puis exploitées pour des applications de recherche et préclinique. Le travail d'optimisation a été partagé en deux axes. Dans un premier temps, un modèle analytique a été développé sous MATLAB pour l'estimation du rapport signal sur bruit intrinsèque à un capteur paramétré par ses dimensions et les propriétés de l'objet imagé. La validation du modèle a été obtenue par la comparaison entre mesures et simulations du facteur de qualité. Cette méthode d'optimisation a été appliquée pour deux études spécifiques qui ont fait l'objet d'une publication. Dans un second temps, un travail sur le découplage actif a été mené. En effet, sur l'IRM 0,2 T à notre disposition, le découplage passif est la méthode retenue par le constructeur. Mais pour certaines applications des artefacts d'imagerie sont inévitables et le facteur de qualité réduit. Des moyens de découplage actif ont donc été développés. Les performances des capteurs ainsi équipés se sont avérées meilleures qu'en découplage passif. Ce système de découplage associé à un dispositif de connexion par couplage inductif du signal de résonance magnétique a été également démontré à 3 T comme une preuve de concept d'un dispositif de connexion universelle. Ce dispositif a fait l'objet d'un article récemment soumis pour publication / Magnetic resonance imaging {MRl) in veterinary practice employs low magnetostatic field devices which have numerous advantages such as their low maintenance and initial cost. Yet, the radiofrequency {RF) coils commercially provided with these devices are dedicated to human morphology, therefore reducing image quality. ln this work, simple optimization methods for 0.2 T RF coils were developed for an implementation in research and preclinical studies. Optimization protocol was subdivided into two main steps. First, an analytical model was developed using MATLAB in order to estimate the intrinsic signal to noise ratio variations with coil and imaged sample characteristics. Validation of the model was assessed thanks to quality factor comparison between simulated and measured values. The use of the analytical model for two specific studies was described in a recently accepted publication. Second, active decoupling was investigated. lndeed, passive decoupling is the decoupling method implemented on the 0.2 T MR device at our disposal. However, this technique can lack of efficiency in some experiments, inducing imaging artifacts and reduced quality factor. Active decoupling method was therefore implemented. The electronic performances of the coils equipped this way were better than in passive decoupling. This active decoupling device combined with an inductive coupling connecting system was tested at 3 T to demonstrate the technical feasibility of a new universal connecting device, for which an article was recently submitted

Die Positionsbestimmung von Verkehrsteilnehmern basiert häufig auf der Auswertung von Signalen globaler Navigationssatellitensysteme (GNSS). Dabei werden die Laufzeiten der Signale und darauf beruhend die Entfernungen zwischen den Satelliten und dem Verkehrsteilnehmer ermittelt. Die Positionsbestimmung erfolgt dann nach dem Prinzip der Trilateration. Bei hohen Anforderungen an die Genauigkeit der Position wird hierfür insbesondere die Phase des Trägersignals ausgewertet. Eine besondere Herausforderung stellt dabei die Mehrwegeausbreitung der Signale dar. Hervorgerufen wird diese durch Objekte, wie Bäume, Gebäude oder Fahrzeuge in der Umgebung des Verkehrsteilnehmers. Durch die Mehrwegeausbreitung werden die Laufzeiten der Signale und somit die Position fehlerhaft bestimmt. Es ist daher von großer Bedeutung, die mehrwegebedingten Fehleranteile zu detektieren und sie zu reduzieren. In dieser Arbeit wird dafür der Zusammenhang zwischen dem Trägerphasenmessfehler und der Signalqualität genutzt. Durch Anwendung einer im Rahmen dieser Arbeit entwickelten adaptierten Hilbert-Huang-Transformation auf die aus dem Signalqualitätsparameter des Signal-zu-Rauschleistungsdichte-Verhältnis abgeleiteten Signalamplituden können Mehrwegesignale detektiert und der durch sie verursachte Trägerphasenmessfehler berechnet werden. Anhand der Auswertung eines Experimentaldatensatzes sowie Daten von GNSS-Referenzstationen des SAPOS-Netzes kann der Erfolg des Einsatzes der adaptierten Hilbert-Huang-Transformation nachgewiesen werden.

La réflectométrie GNSS (ou GNSS-R) est une technique de télédétection originale et pportuniste qui consiste à analyser les ondes électromagnétiques émises en continu par la soixantaine de satellites des systèmes de positionnement GNSS (GPS, GLONASS, etc.), qui sont captées par une antenne après réflexion sur la surface terrestre. Ces signaux interagissent avec la surface réfléchissante et contiennent donc des informations sur ses propriétés. Au niveau de l'antenne, les ondes réfléchies interfèrent avec celles arrivant directement des satellites. Ces interférences sont particulièrement visibles dans le rapport signal-sur-bruit (SNR, i.e., Signal-to-Noise Ratio), paramètre enregistré par une station GNSS classique. Il est ainsi possible d'inverser les séries temporelles du SNR pour estimer des caractéristiques du milieu réfléchissant. Si la faisabilité et l'intérêt de cette méthode ne sont plus à démontrer, la mise en oeuvre de cette technique pose un certain nombre de problèmes, à savoir quelles précisions et résolutions spatio-temporelles peuvent être atteintes, et par conséquent, quels sont les observables géophysiques accessibles. Mon travail de thèse a pour objectif d'apporter des éléments de réponse sur ce point, et est axé sur le développement méthodologique et l'exploitation géophysique des mesures de SNR réalisées par des stations GNSS classiques. Je me suis focalisé sur l'estimation des variations de hauteur de l'antenne par rapport à la surface réfléchissante (altimétrie) et de l'humidité du sol en domaine continental. La méthode d'inversion des mesures SNR que je propose a été appliquée avec succès pour déterminer les variations locales de : (1) la hauteur de la mer au voisinage du phare de Cordouan du 3 mars au 31 mai 2013 où les ondes de marées et la houle ont pu être parfaitement identifiées ; et (2) l'humidité du sol dans un champ agricole à proximité de Toulouse, du 5 février au 15 mars 2014. Ma méthode permet de s'affranchir de certaines restrictions imposées jusqu'à présent dans les travaux antérieurs, où la vitesse de variation verticale de la surface de réflexion était supposée négligeable. De plus, j'ai développé un simulateur qui m'a permis de tester l'influence de nombreux paramètres (troposphère, angle d'élévation du satellite, hauteur d'antenne, relief local, etc.) sur la trajectoire des ondes réfléchies et donc sur la position des points de réflexion. Mon travail de thèse montre que le GNSS-R est une alternative performante et un complément non négligeable aux techniques de mesure actuelles, en faisant le lien entre les différentes résolutions temporelles et spatiales actuellement atteintes par les outils classiques (sondes, radar, diffusiomètres, etc.). Cette technique offre l'avantage majeur d'être basé sur un réseau de satellites déjà en place et pérenne, et est applicable à n'importe quelle station GNSS géodésique, notamment celles des réseaux permanents (e.g., le RGP français). Ainsi, en installant une chaîne de traitement de ces acquisitions de SNR en domaine côtier, il serait possible d'utiliser les mesures continues des centaines de stations pré-existantes, et d'envisager de réaliser des mesures altimétriques à l'échelle locale, ou de mesurer l'humidité du sol pour les antennes situées à l'intérieur des terres. / GNSS reflectometry (or GNSS-R) is an original and opportunistic remote sensing technique based on the analysis of the electromagnetic waves continuously emitted by GNSS positioning systems satellites (GPS, GLONASS, etc.) that are captured by an antenna after reflection on the Earth's surface. These signals interact with the reflective surface and hence contain information about its properties. When they reach the antenna, the reflected waves interfere with those coming directly from the satellites. This interference is particularly visible in the signal-to-noise ratio (SNR) parameter recorded by conventional GNSS stations. It is thus possible to reverse the SNR time series to estimate the reflective surface characteristics. If the feasibility and usefulness of thismethod are well established, the implementation of this technique poses a number of issues. Namely the spatio-temporal accuracies and resolutions that can be achieved and thus what geophysical observables are accessible.The aim of my PhD research work is to provide some answers on this point, focusing on the methodological development and geophysical exploitation of the SNR measurements performed by conventional GNSS stations. I focused on the estimation of variations in the antenna height relative to the reflecting surface (altimetry) and on the soil moisture in continental areas. The SNR data inversion method that I propose has been successfully applied to determine local variations of : (1) the sea level near the Cordouan lighthouse (not far from Bordeaux, France) from March 3 to May 31, 2013, where the main tidal periods and waves have been clearly identified ; and (2) the soil moisture in an agricultural plot near Toulouse, France, from February 5 to March 15, 2014. My method eliminates some restrictions imposed in earlier work, where the velocity of the vertical variation of the reflective surface was assumed to be negligible. Furthermore, I developed a simulator that allowed me to assess the influence of several parameters (troposphere, satellite elevation angle, antenna height, local relief, etc.) on the path of the reflected waves and hence on the position of the reflection points. My work shows that GNSS-R is a powerful alternative and a significant complement to the current measurement techniques, establishing a link between the different temporal and spatial resolutions currently achieved by conventional tools (sensors, radar, scatterometer, etc.). This technique offers the major advantage of being based on already-developed and sustainable satellites networks, and can be applied to any GNSS geodetic station, including permanent networks (e.g., the French RGP). Therefore, by installing a processing chain of these SNR acquisitions, data from hundreds of pre-existing stations could be used to make local altimetry measurements in coastal areas or to estimate soil moisture for inland antennas.

Introdução: O ruído é um dos principais agentes de poluição ambiental, que em intensidades elevadas, poderá causar a perda auditiva induzida por níveis de pressão sonora elevadas (PAINPSE). A PAINPSE é o segundo tipo de perda auditiva neurossensorial mais frequente, e considerada uma das principais doenças ocupacionais. Para evitar os prejuízos da exposição ao ruído ocupacional, legislações determinam o monitoramento e o desenvolvimento de ações para reduzir os efeitos da exposição, sendo o uso de dispositivos de proteção auditiva (DPA) a mais empregada. Contudo é comum que o trabalhador relate dificuldades para se comunicar com o DPA, o que pode induzi-lo ao uso inadequado. Os estudos já realizados apresentam discordâncias sobre os efeitos do DPA para a inteligibilidade da fala, além de serem poucos os que compararam participantes com perdas auditivas com normo-ouvintes. Objetivos: Avaliar a inteligibilidade de fala em trabalhadores expostos a ruído durante o uso de DPA. Métodos: Trata-se de um estudo transversal, aprovado pelo Comitê de Ética em Pesquisa da instituição nº 130/14, com amostra de 36 participantes, todos do sexo masculino e expostos ao ruído. Foram divididos em grupo normo-ouvintes (G1), com 20 participantes e com perda auditiva (G2), com 16 participantes. Todos os participantes realizaram avaliação da inteligibilidade de fala em campo acústico, utilizando listas de monossílabas e sentenças, em situações de uso do DPA no silêncio e no ruído e sem o uso do DPA no ruído. Resultados: Na avaliação da inteligibilidade de fala com DPA utilizando as monossílabas, ambos os grupos apresentaram porcentagens de acertos superiores à 95% na situação de silêncio. Na presença do ruído com o DPA, não foi detectado efeito de interação envolvendo grupo (p>=0,106) e nem efeito de grupo (p=0,182), mas houve efeito de interação entre intensidade de fala e relação sinal/ruído (S/R) (p=0,006). Na avaliação da inteligibilidade de fala no ruído sem o DPA, foi detectado efeito de interação entre uso do DPA e relação S/R (p=0,010) sobre a média das porcentagens de acertos. Na avaliação com listas de sentenças, a porcentagem de acertos com DPA foram superiores a 98% no silêncio; contudo, no ruído houve piora, com efeito de interação entre intensidade de fala e relação S/R (p=0,010) e entre grupo e relação S/R (p=0,039). Na comparação do uso/não uso do DPA, houve efeito de interação entre uso do DPA e as relações S/R (p=0,009) sobre a média das porcentagens de acertos. Conclusões: Para ambos os grupos, houve piora da inteligibilidade de fala com o aumento da intensidade do ruído; contudo, a dificuldade foi maior para o G2. Na comparação das situações com/sem o uso do DPA, o uso dificultou a inteligibilidade de fala, mas não houve diferença significante entre as situações. Por fim, ressalta-se a importância de realizar mais estudos dos efeitos do DPA na comunicação, além de inserir nas avaliações ocupacionais testes que avaliam a fala no ruído, para determinar as dificuldades que podem surgir no ambiente ocupacional, e assim desenvolver medidas que beneficiem os trabalhadores que necessitam se comunicar durante as atividades laborais / Introduction: Noise is one of the main agents of environmental pollution, and at high intensities, it may provoke noise-induced hearing loss (NIHL). NIHL is the second most frequent type of sensorineural hearing loss and it is considered one of the main occupational diseases. In order to avoid the effects of the exposition to occupational noise, legislation determines the monitoring and development of actions to reduce the effects of such exposure and one of the most applied measure is the mandatory use of hearing protection devices (HPD). However, workers often report difficulties to communicate when using the HPD, what may induce them to use it improperly. Previous studies showed divergences between the effects of HPD on speech intelligibility; besides, few studies compared participants with hearing loss to those with normal hearing. Purpose: To assess speech intelligibility of workers exposed to noise during the use of HPD. Methods: This is a cross-sectional study, approved by the Institution\'s Ethics Committee (nº 130/14) and counted on a sample of 36 participants, all males and exposed to noise. They were distributed into a normal hearing group (G1), with 20 participants and hearing loss group (G2), with 16 participants. All participants were submitted to an assessment of speech intelligibility in free field, using monosyllable and sentence lists, when using the HPD in silence and noise situation, and also without the use of HPD in noise. Results: In the assessment of speech intelligibility with HPD using the monosyllables, both groups presented percentages of correct answers above 95%. In the presence of noise with HPD, there was no interaction effect involving group (p>=0.106) or group effect (p=0.182), but there was found an interaction effect between speech intensity and signal-to-noise ratio (SNR) (p=0.006). In the assessment of speech intelligibility in noise without HPD, an interaction effect between HPD use and SNR (p=0.010) was detected over the mean percentage of correct answers. In the evaluation with lists of sentences, the percentage of correct answers with HPD were superior to 98% in the silence; however, in noise, there was a decrease on the interaction effect between speech intensity and SNR (p = 0.010) and between group and SNR (p = 0.039). In the comparison with/without use of HPD, there was found an interaction effect between hearing protection use and SNR (p = 0.009) on the mean percentage of correct answers. Conclusions: For both groups, it was noticed that speech intelligibility decrease with the increasing of noise intensity; nevertheless, the difficulty was greater for the group with hearing loss. In the comparison of situations with and without the use of HPD, its use made speech intelligibility more challenging, but no significant difference between the situations was found. Finally, it is important to highlight the importance of further studies regarding the effects of HPD on communication and, besides, inserting speech-in-noise tests in occupational examination in order to determine the difficulties that may arise in the occupational environment and thus, develop measures that benefits the workers who need communicate during their labor activities