Passive HF geolocation using TDoA based receiver network

22/12/2018
OAI : oai:www.see.asso.fr:1301:2018-5:24917
DOI :
contenu protégé  Document accessible sous conditions - vous devez vous connecter ou vous enregistrer pour accéder à ou acquérir ce document.
Prix : 10,00 € TVA 20,0% comprise frais d'expédition hors UE non compris (8,33 € hors TVA) - Accès libre pour les ayants-droit
se connecter (si vous disposez déjà d'un compte sur notre site) ou créer un compte pour commander ou s'inscrire.
 

Résumé

Passive HF geolocation using TDoA based receiver network

Métriques

0
0
372.13 Ko
 application/pdf
bitcache://a60b8bfe5833acea163935a6571d3c516d90736a

Licence

Creative Commons Aucune (Tous droits réservés)
<resource  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                xmlns="http://datacite.org/schema/kernel-4"
                xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
        <identifier identifierType="DOI">10.23723/1301:2018-5/24917</identifier><creators><creator><creatorName>Pascal PAGANI</creatorName></creator><creator><creatorName>Patrice Pajusco</creatorName></creator><creator><creatorName>Michel Ney</creatorName></creator><creator><creatorName>Ankit Jain</creatorName></creator><creator><creatorName>Roland Fleury</creatorName></creator></creators><titles>
            <title>Passive HF geolocation using TDoA based receiver network </title></titles>
        <publisher>SEE</publisher>
        <publicationYear>2018</publicationYear>
        <resourceType resourceTypeGeneral="Text">Text</resourceType><dates>
	    <date dateType="Created">Sat 22 Dec 2018</date>
	    <date dateType="Updated">Sat 22 Dec 2018</date>
            <date dateType="Submitted">Thu 17 Jan 2019</date>
	</dates>
        <alternateIdentifiers>
	    <alternateIdentifier alternateIdentifierType="bitstream">a60b8bfe5833acea163935a6571d3c516d90736a</alternateIdentifier>
	</alternateIdentifiers>
        <formats>
	    <format>application/pdf</format>
	</formats>
	<version>41194</version>
        <descriptions>
            <description descriptionType="Abstract"></description>
        </descriptions>
    </resource>
.

98 ZREE N°5/2018 DOSSIER 2 JOURNÉES SCIENTIFIQUES URSI 2018 This paper presents the geolocation of a High Frequency (HF) transmitter (Tx) located at distances of about 700 to 1300 km from a network of receivers (Rx’s). HF signals propagate through the ionospheric medium using skywaves and are reflected back to the Earth, thereby allowing HF communications over long dis- tances. In our study, the location of the HF Tx is found using time difference of arrival (TDoA) method. Four remotely controllable Rx’s which are built to cap- ture HF signals synchronously are installed in four different cities in France. HF radio broadcast signals are captured simultaneously on the four deployed Rx’s from the HF Tx situated in Nauen, Germany. The TDoA’s between the signals received on all the four synchronized Rx’s are obtained using the cross-channel sounding method. Using the obtained TDoA’s, the Tx location is found using the geolocation algorithm based on TDoA. Experimental results show that it is possible to locate the HF Tx in Nauen with a geolocation error of about 5 km. Cet article présente la géolocalisation d’un émetteur HF situé à des distances d’environ 700 à 1300 km d’un réseau de récepteurs. Les signaux HF se pro- pagent à travers le milieu ionosphérique en utilisant des ondes de ciel et sont réfléchis vers la Terre, permettant ainsi des communications HF sur de longues distances. Dans notre étude, la position de l’émetteur HF est obtenue en uti- lisant la méthode “Time Difference of Arrival” (TDoA). Des récepteurs contrô- lables à distance, conçus pour capturer des signaux HF de manière synchrone, sont installés dans quatre villes différentes en France. Ce réseau de quatre récepteurs permet l’acquisition simultanée de signaux de radiodiffusion HF transmis par un émetteur situé à Nauen, en Allemagne. La différence de temps d’arrivée entre les signaux reçus par les quatre récepteurs est obtenue par une technique appelée sondage de canaux croisés. En utilisant les différences de temps d’arrivée déduites, la position de l’émetteur est estimée en utilisant un algorithme de géolocalisation basé sur la technique TDoA. Les résultats expé- rimentaux montrent qu’il est possible de localiser l’émetteur HF à Nauen avec une erreur de géolocalisation d’environ 5 km. Ankit Jain Pascal Pagani Rolland Fleury Michel Ney Patrice Pajusco IMT Atlantique, Lab-STICC UMR CNRS 6285, Brest, France. Passive HF Geolocation Using TDoA Based Receiver Network Principle and measurement results Introduction HF radio signals propagating using skywaves are reflected once or multiple times between the Earth and the ionos- pheric medium. Thus, long-range communication can be established through radio signals transmitted in the HF band. HF communication is extremely useful in defense and civil applications as a complement or fallback solution to satellite communication. Owing to the importance of HF communica- tion in defense and civil applications, geolocation of HF trans- mitters are of prime importance. In particular, in the context of signal intelligence, passive geolocation methods consist of detecting the source of an electromagnetic signal using a non-intrusive system i.e. without the transmitter being able to detect the localization process [1]. Numerous methods to measure the Direction of Arrival (DoA) of the incoming HF signals are explained in [2]. Most of these direction finding (DF) systems require deployment of large antenna arrays at multiple locations which can be quite expensive and requires large space. HF geolocation can also be achieved using the single site location (SSL) method which measures the DoA of the HF signal in terms of azimuth and elevation angles [3]. SSL method requires installation of a large antenna array only on a single site which is an advan- tage over traditional DF systems. But, the ionospheric reflec- tion profile along the propagation path must be known to estimate the ionospheric reflection height which is required to transform the elevation angle to ground range. In the case of passive geolocation, the propagation paths through which the radio signals travels are unknown. Thus, SSL method can- not be used for locating HF transmitters passively. HF geolocation can also be achieved using time-domain techniques. The location of a HF transmitter can be found from the time of arrivals (ToA’s) obtained from multiple single antenna receivers under some assumptions and wit- REE N°5/2018 Z 99 Passive HF Geolocation Using TDoA Based Receiver Network hout the knowledge of the ionospheric profile along the HF propagation path [4]. Also, the cost of the receiver system is much less compared to SSL or DF systems. In passive geo- location systems, the signal transmission time is not known. Therefore, geolocation cannot be performed using the ToA method. But, the location of the transmitter could be found by multilateration technique using the time difference of arri- vals (TDoA’s) obtained from different receiver pairs. Further, the TDoA technique requires a minimum of four receivers to perform localization [5]. This paper aims to explain that it is possible to perform HF geolocation using the TDoA method. To validate this concept, multiple remotely controllable receivers capable of synchro- nously capturing HF radio are developed using software de- fined radio (SDR) modules. The next section explains the principles of HF propagation using skywaves and HF geoloca- tion algorithm. Later the HF receiver design and the receiver network setup are presented. Next, the data captures from the HF transmitter in Nauen, Germany, are analyzed and the geolocation results are presented. Finally, conclusions are drawn in the last section. HF Propagation Principles The ionosphere lies from about 50 km to 2000 km above the surface of the Earth [6]. The neutral atoms present in the atmosphere are ionized from the sun’s radiation which results in the formation of free electrons which affects HF radio pro- pagation. The ionosphere is split into D, E and F layers in the order of increasing altitude, respectively. Moreover, the elec- tron density is also variable in each of the layers making the ionosphere a dispersive medium and capable of refraction of the HF radio signals. The D layer mainly accounts for the absorption of the HF radio signals and has the least electron density. The electron density is the maximum in the F layer which is present between 150 to 600 km and it accounts for large propagation paths. The ionosphere is also a highly dynamic medium which varies diurnally, seasonally, geogra- phically and with respect to the solar activity. This variability in the ionosphere is one of the major issues in HF geolocation, for signals propagating through skywaves, as it leads to time delay variations [7]. HF radio propagation is possible using ground waves and skywaves. Long distance HF radio transmissions occur via the skywaves where the radio signals are reflected back to the Earth through the ionosphere. HF radio signals which are reflected once from the ionosphere are referred to as one- hop mode whereas radio signals which are reflected multiple times are known as multi-hop modes. It is possible to com- municate up to distances in the range of 500-3000 km using one-hop propagation mode. Due to the presence of Earth’s magnetic field, an additional propagation mode due to ex- traordinary waves exist whereas ordinary waves are present even in the absence of the magnetic field. These propagation modes can be distinguished with respect to the direction of rotation of the waves. In our geolocation study, we assume that all the received HF radio signals follow a one-hop propagation mode consi- dering the distances at which the HF transmitter is located from the receivers. Figure 1 presents the geometry of HF geolocation assuming a uniform ionosphere. The transmit- ter is located at point Tx and the receivers are located at point Rx1 and Rx2. Due to the varying refractive index in the ionosphere, the HF radio signals propagate along a curved path and the real reflection height in the ionosphere is repre- sented by hm , as seen in Figure 1. As per the Breit and Tuve’s theorem [6], the time required for the HF radio signal to travel along the red curved path through the ionosphere is equal to the time required to travel along the black triangular path in vacuum. As the ionosphere is assumed to be uniform, the reflection height for both the triangular paths is equal and is represented by hv ; it is referred to as virtual reflection height. The Earth is assumed to be flat by neglecting its curvature. Lastly the effect of earth’s magnetic field is also ignored. It is known that the geolocation errors calculated using the angle of arrival (AoA) method by neglecting the earth’s magnetic field are generally smaller at mid-latitudes which is the region of interest for our study [8]. Due to electromagnetic image theory, the length of the propagation path between Rx1 -Tx is equal to Rx1 -I1 -Tx’; similarly the propagation paths between Rx2 -Tx is equal to Rx2 -I2 -Tx’.The geolocation process consists of finding the coordinates of Tx’ which is situated at a height of h vertically above the location of the Tx on a flat Earth. In the case of TDoA, the coordinates of Tx’ can be found by the principle of multilateration where the transmitter lies at the intersection of three hyperboloids. The geometrical problem of identifying the intersection of three hyperboloids requires solving systems of quadratic Figure 1: Geometry of HF geolocation assuming a one-hop propagation mode and uniform ionosphere. 100 Z REE N°5/2018 DOSSIER 2 JOURNÉES SCIENTIFIQUES URSI 2018 equations which cannot be solved analytically. The mathe- matical formulations to obtain the location of Tx’ using the TDoA method even without the information of the ionos- pheric reflection height is explained in [9]. The mathematical approach is recalled here for consistency. Mathematical approach: TDoA Let the receivers (Rxi ) be located at position Pi = (x ( ( i x , yi y y , zi z) i i where i = 1, 2, 3.. i n; let the transmitter (Tx) on the flat Earth surface be located at position P0 = (x ( ( 0 , y0 y y , z0 z ). The coordi- nates of point Tx’ is located at P'0 = (x ( ( 0 , y0 y y , z'0 ) where z'0 is given by the following equation: (1) The propagation path length Di between a Tx-Rx pair via a i one-hop propagation mode is calculated as follows: (2) where to is the signal transmit time, ti t is the signal arrival time i at the ith receiver, c is the speed of light, c Di is the propagation i path length between the transmitter and ith receiver and 冩 冩 represents the Euclidean norm. After squaring equation (2), it is given as follows: (3) In the case of TDoA, the transmission time t0 t is unknown. Computing the difference between equation (3) at i = 1 i (Rx1 ) and equation (3) at i = 2 (Rx i 2 ); after further simplifica- tion one obtains the following equation: (4) where T represents the transpose operator. As all receivers are situated on the Earth surface, one can consider z1 = z2 = z3 = z0 After simplification, equation (4) can be expressed in the form of a linear equation with 3 unknowns x0 , y0 y y , t0 t : (5) where Pi(x,y) represents the projection of vector Pi on the (x, y ( ( ) plane (i = 1,2). In the TDoA method, time delays y y between different receivers are calculated considering a refe- rence receiver. So with 3 receivers, one can obtain two linear equations with 3 unknowns. Thus, to solve the equation with 3 unknowns, a 4th receiver is used. Equation (5) is obtained using receivers 1-2. Similarly using Rx pairs 1-3 and 1-4, res- pectively, one can obtain 2 more equations. Thus, a linear system with 3 independent equations with 3 unknowns is obtained having the following form: (6) The constants of 1st row of equation (6) are obtained by comparing with it equation (5) and are given as follows: (7) Similarly from the other 2 equations obtained from Rx pairs 1-3 and 1-4, constants of other rows of equation (6) can be obtained. Finally, the transmitter coordinates (x ( ( 0 , y0 y y ) and the transmit time t0 t can be obtained by doing a simple matrix inversion operation. The value of z'0 can be obtained by substituting (x ( ( 0 , y0 y y ) in equation (3). Finally the equivalent height hv can be obtained by resolving equation (1). v Note: To obtain the transmitter coordinates, Equation (6) can be realized in the following form: (8) where A is a 3x3 matrix with elements A aij and j b = (b ( ( 1 , b2 , b3 )T . In the case of N receivers, where N > 4, A is a (N - 1) x A 3 matrix, b is a vector of (N - 1) elements and the system is solved using the pseudo inverse of matrix A as follows: (9) In this case, the solution is optimum in the least square sense. HF Receiver Network In order to study HF geolocation using the TDoA method, four remotely controllable and programmable receivers are designed to synchronously capture HF radio signals. Receiver design The receiver system design is illustrated in Figure 2. It is composed of both hardware and software. The main com- ponent of the receiver setup is the SDR Ettus USRP N200/ N210 box whose RF frontend is connected to an active HF antenna. The SDR box consists of a direct sampling analog- to-digital converter (ADC) capable of sampling up to 100 MHz, coupled with an FPGA that performs downsampling and frequency conversion. The LFRX daughterboard is ca- REE N°5/2018 Z 101 Passive HF Geolocation Using TDoA Based Receiver Network pable of accepting signals from DC to 30 MHz and ampli- fies the received signals. Using the GPS antenna, the GPS disciplined oscillator (GPSDO) can be locked to global GPS standards and the signal fluctuation is within ±50 ns [10]. The GPSDO allows multiple USRP’s which are located far away from each other to synchronize their captures with the same clock. The receiving setup communicates to their respective computers via the Ethernet connection. The computer consists of several programs that are used to schedule and capture data according to user requirement. The scheduling of the captures is done by a Java interpreter. It also manages the launch of the Python scripts which act as an interface between hardware and software. Python scripts are written to synchronize the capture time with the GPS time, initialize the capture and capture the data as per GNU Radio standards [11]. A capture generates two files: a DAT file and a HDR file. The DAT files contain complex samples of each capture whe- reas the HDR files consist of metadata corresponding to the respective DAT files. TDoA Receiver Network The receiver system explained in the previous section is deployed in four different cities in France. As illustrated in Figure 3, the HF antennas are deployed at some height from the ground level and far away from urban areas to minimize the effect of noise. Figure 4 presents the GPS locations for the receivers deployed in Brest, Bordeaux, Grenoble and Lille. All the receivers were controlled from a central machine in Brest. Figure 2: Receiver system design prototype. Figure 3 : HF antenna deployed in (a) Brest, (b) Bordeaux, (c) Lille, (d) Grenoble. (a) (b) (c) (d) 102 Z REE N°5/2018 DOSSIER 2 JOURNÉES SCIENTIFIQUES URSI 2018 HF radio signals were captured simultaneously on all the receivers from the HF transmitter located in Nauen, Germany. The geographic location of the HF transmitter in Nauen is presented in Figure 4. The ground ranges for the different HF links are summarized in Table 1 and it is within the theoretical range of possible propagation paths using one-hop propaga- tion mode. Measurement details and signal processing All the HF radio broadcast signals are captured for a span of 5 seconds and sampled at the rate of 200 kHz. The captu- red signal is filtered in the frequency domain to remove un- desired parts of the spectrum. As most of the HF broadcasts are transmitted over a 10 kHz band, only signals within this bandwidth around the carrier frequency of the transmitter are used, the rest is filtered out. After filtering, the signal is trans- formed back to the time domain and the received signal ave- rage power is calculated. Finally the message signal samples are obtained by demodulating the transformed signal. Reference receiver selection The geolocation algorithm used to estimate the geo- graphic location of the HF transmitter is dependent on the reference receiver [9]. The received signal average powers for all the receivers were compared and the receiver with the highest average power was considered as the reference receiver. Using the cross-channel sounding technique [12], the message signal samples from the reference receiver are cross-correlated with the message signal samples from the other three receivers to obtain the three TDoA estimates. The TDoA estimate 6t corresponds to the maximum value of the cross-correlation and is expressed as follows: (10) where is the cross-correlation between the message signal samples s1 (t) obtained from the reference receiver t and the message signal samples s1 (t) obtained from the t other three receivers (i ( ( = 1, 2, 3), respectively. i Experimental Results Multiple HF broadcast radio signals were captured si- multaneously by all the receivers in the network between 19:11–19:20 UTC on 18th July 2017 from the HF transmit- ter located in Nauen. The interval between the start times of each capture was equal to one minute. The transmitter was emitting at a frequency of 11.790 MHz. The GPS loca- tions of the broadcast transmitter in Nauen and all the four receivers were transformed to x and y coordinates using the azimuthal equidistant projection [13] and considering the selected reference receiver as the origin of the coordinate system. An analysis of a data capture is presented in the following section. Data capture analysis A HF radio broadcast signal captured from Nauen simulta- neously by all the four Rx’s on 18th July 2017 at 19:17 UTC is analyzed. The signal received in Grenoble was considered as Figure 4: GPS coordinates of the transmitter site and all the receivers in the network. HF radio link (Rx-Tx) Ground Range (km) Brest-Nauen 1320 Bordeaux-Nauen 1313 Grenoble-Nauen 976 Lille-Nauen 710 Table 1: Ground distance between the HF transmitter and all the receivers. REE N°5/2018 Z 103 Passive HF Geolocation Using TDoA Based Receiver Network the reference signal on the basis of the received signals power. The signal captured in Grenoble is presented in Figure 5. The message signal samples were obtained from the four receivers as explained earlier. The propagation duration dif- ferences for the signals received at different receiver pairs (i.e. Grenoble-Brest, Grenoble-Bordeaux and Grenoble- Lille) were obtained using the cross-correlation method. An example of the cross-correlation output for the signals captured in Grenoble and Lille is shown in Figure 6. Using the obtained TDoA estimates, the Tx coordinates were esti- mated using the TDoA geolocation algorithm. The estimated xy coordinates were transformed to geographical coordinates and the geolocation error was found to be 5.7 km. Table 2 provides the summary of the geolocation error for the 10 different captured signals from the HF transmitter in Nauen on 18th July between 19:11-19:20 UTC. For all ten captures, the received signal in Grenoble was selected as the reference receiver on the basis of received signal intensity. The signal-to-noise ratio (SNR) for all ten signals captured in Grenoble was at least 12 dB. The minimum geolocation error of 5.7 km corresponds to a relative error of about 0.43% computed with respect to the maximum ground range among the four possible HF paths whereas the maximum geolocation error of 112.6 km corresponds to a relative error of about 8.51%. The minimum and maximum errors were obtained within an interval of 1 minute. This rapid fluctua- Figure 5: HF signal received in Grenoble from the HF transmitter in Nauen on 18th July at 19:17 UTC. Figure 6: Cross-correlation of radio signal captured in Grenoble and Lille from the HF transmitter in Nauen. 104 ZREE N°5/2018 DOSSIER 2 JOURNÉES SCIENTIFIQUES URSI 2018 tion in the geolocation accuracy is due to the fast variations of the ionosphere which leads to variable results in a very short period. In order to reduce this effect, one can consider ave- raging the obtained results within a certain time interval. The geolocation errors for all the 10 captures are provided in Table 2. Figure 7 presents the geographical coordinates of the esti- mated and true Tx locations. It is observed that the accuracy in terms of direction of localization is very high when compared to the geolocation errors. The obtained azimuth errors for all the captured signals are also listed in Table 2. Conclusion In this paper, we presented the results of geolocation of the HF transmitter in Nauen, performed using the TDoA method. The TDoA method is well suited for passive geolocation and can be performed using light and portable equipment. Four remotely receivers capable of synchronously capturing dis- tant HF signals have been built and deployed in four different cities in France. We have also presented the receiver design in detail. Multiple signals are captured simultaneously using the receiver network from the HF transmitter in Nauen. The measurement data are analyzed and the geolocation errors are presented. The analysis of the geolocation output allows us to believe that HF geolocation can be performed using the TDoA method. Further, the obtained geolocation errors can be reduced if we use more receivers [9]. After the preliminary measurement presented in this paper, we collected a large database of measurements obtained from different radio broadcast transmitters across Europe. These data are currently being statistically analyzed to evaluate the geolocation performance with respect to different system pa- rameters. In particular, the impact of the received SNR, TDoA accuracy, Tx-Rx distance and frequency will be investigated, with the aim of improving the geolocation accuracy. Figure 7: Estimated Tx location and the real Tx location. Data capture time (UTC) Error (km) Relative error (%) Azimuth error (°) 19:11 13.75 1.04 -0.45 19:12 24.90 1.88 -0.56 19:13 7.98 0.60 0.17 19:14 78.41 5.93 1.31 19:15 30.70 2.32 -0.72 19:16 112.59 (maximum) 8.51 -2.19 19:17 5.71 (minimum) 0.43 -0.22 19:18 30.70 2.32 -0.72 19:19 30.70 2.32 -0.72 19:20 30.70 2.32 -0.72 Table 2: Summary of geolocation and azimuth errors for all HF signals captured from Nauen. REE N°5/2018 Z 105 Passive HF Geolocation Using TDoA Based Receiver Network Acknowledgements This work is supported in part by the Carnot Institute and the Brittany region in France. The authors thank the Atmospheric Optics Laboratory in University of Lille 1 and the Electrical Engineering Department in IUT1-University Grenoble Alps for hosting our receiver setup. References [1] D. Heurguier, “La localisation d’émetteurs de radiocom- munication en zone urbaine,” Journées Scientifiques URSI France, Mar. 2014. [2] P. J. D. Gething, “HF direction finding,” Proceedings of the Institution of Electrical Engineers, vol. 113, no. 1, pp. 49–61, Jan. 1966. [3] G. Fabrizio and A. Heitmann, “A multipath-driven approach to HF geolocation,” Signal Processing, vol. 93, no. 12, pp. 3487–3503, Dec. 2013. [4] P. Pagani, R. Fleury, Y. L. Roux, and D. L. Jeune, “A study of HF transmitter geolocation through single-hop ionospheric propagation,” The 8th European Conference on Antennas and Propagation (EuCAP), pp. 2689–2693, Apr. 2014. [5] D. J. Torrieri, “Statistical theory of passive location systems,” IEEE Trans. on Aerospace and Electronic Systems, vol. AES- 20, no. 2, pp. 183–198, Mar. 1984. [6] K. Davies, “Ionospheric radio,” IEE Electromagnetic Waves Series 31, pp. 124-158, n° ISBN: 0 86341 186 X, May 1989. [7] R. W. LaBahn and R. B. Rose, “Time delay variations in HF propagation,” Radio Science, vol. 17, no. 5, pp. 1285–1299, Sep. 1982. [8] E. V. Dao, L. F. McNamara, and J. J. Colman, “Magnetic field effects on the accuracy of ionospheric mirror models for geolocation,” Radio Science, vol. 51, no. 4, pp. 284–300, Apr. 2016. [9] A. Jain, P. Pagani, R. Fleury, M. M. Ney, and P. Pajusco, “Efficient time domain HF geolocation using multiple distributed receivers,” The 11th European Conference on Antennas and Propagation (EuCAP), pp. 1852–1856, Mar. 2017. [10] “GPSDO Selection Guide Ettus.” [Online]. Available: https://kb.ettus.com/GPSDO. [11] “USRP-GNURadio.” [Online]. Available: https://wiki.gnuradio. org/index.php/Main_Page [12] A. Jain, P. Pagani, R. Fleury, M. M. Ney, and P. Pajusco, “Cross-Channel Sounding for HF Geolocation: Concepts and Experimental Results,” The 12th European Conference on Antennas and Propagation (EuCAP), Apr. 2018. [13] J. P. Snyder, “Map projections: A working manual,” U.S. Government Printing Office, Washington, D.C., Series 1395, pp. 191–202, 1987. LES AUTEURS Ankit Jain : Diplômé de l’Université de Mumbai (2012) et de l’Université technique de Chemnitz (2014), il a occupé en 2015 le poste d’ingénieur système au sein de L&T Infotech, Inde. Il effectue sa thèse de doctorat à IMT Atlantique (ex - Telecom Bretagne) depuis janvier 2016. Ses travaux de recherches portent sur la propagation HF en milieu ionos- phérique, la détection des émetteurs HF et le traitement du signal. Pascal Pagani : Diplômé de Telecom Bretagne et de l’Université de Bristol, il a obtenu un doctorat en électronique de l’INSA de Rennes en 2005, et une HDR de l’Université de Bretagne Occidentale en 2016. De 2002 à 2012, il a mené des travaux de recherche au sein d’Orange Labs dans les domaines de la transmission pour les réseaux résidentiels, qui ont été récompensés par le Grand Prix de l’Electronique Général Ferrié 2013. De 2012 à 2016, il a été Maître de Conférences dans le département Micro-ondes de Telecom Bretagne, où il a travaillé en particulier sur la propagation HF en milieu ionosphérique. Rolland Fleury : Rolland Fleury est docteur 3° cycle de l’Université de Rennes en 1972. Il rejoint le CNET à Lannion où il travaille sur la propagation des ondes VLF et HF. Il est copropriétaire du logiciel SATIS de calcul des prévisions de la propagation ionosphérique. Depuis 2001, il est Enseignant-Chercheur à l’ENST de Bretagne (IMT-Atlantique, site de Brest). Ses travaux portent sur les relations Soleil-Terre et la morphologie de l’ionosphère à l’aide de mesures GNSS. Dans le cadre du GIRGEA, il participe à la formation de nombreux étudiants africains. Michel Ney : Diplômé de l’Ecole Polytechnique Fédérale de Lausanne, il obtient un doctorat de l’université d’Ottawa, Canada en 1983. Il devient professeur dans la même institution jusqu’en 1993 et ensuite rejoint le Département Micro- ondes de l’ENST-Bretagne. Il reprend la direction du Laboratoire d’Electronique et des Systèmes de Télécommunica- tions (LEST UMR CNRS 6165) en 1998 jusqu’en décembre 2007. Il est “Life Fellow” de l’IEEE, membre représentant d’Institut Mines-Télécom (IMT) au bureau de l’URSI-France, membre du bureau des éditeurs du “International Journal 106 ZREE N°5/2018 DOSSIER 2 JOURNÉES SCIENTIFIQUES URSI 2018 of Numerical Modelling, Electronic Networks, Devices and Fields”, du comité scientifique des Annales des Télécom- munications et membre du comité scientifique permanent du Colloque International et Exposition sur la Compatibilité Electromagnétique. Professeur émérite à l’IMT Atlantique sur le campus de Brest dès 2018, il continue ses activités de recherche portant sur la modélisation électromagnétique appliquée aux dispositifs passifs, la propagation et les milieux complexes. Patrice Pajusco : Patrice Pajusco a obtenu son diplôme d’ingénieur SUPELEC en 1992. Il rejoint France Télécom R&D en 1993 où il mène des études sur les canaux de propagation large bande puis MIMO. En1999, il prend la direction de l’équipe chargée de la modélisation de la propagation pour l’étude et le déploiement de systèmes radio (UMTS, Wi-Fi, WIMAX...). En 2008, il rejoint IMT Atlantique (anciennement Télécom Bretagne) en tant que responsable du départe- ment Micro-ondes. Il y poursuit ses travaux sur le retournement temporel dans le cadre de son doctorat qu’il défend en 2011. Ses activités de recherche actuelles portent sur la caractérisation et modélisation multi-capteurs pour des applications telles que la transmission massive MIMO, la localisation ou la modulation spatiale.