Positioning in urban or indoor environment is a hot topic, either due to regulations such as the E911 requiring US mobile telecommunication operators to be able to locate their subscribers in case of emergency, or due to the market development, with the extension of location-based services targeting the mass market concentrated in metropolitan areas. In urban or indoor areas, it is generally recognized that satellite-based positioning systems are not suitable (alone) to provide a continuous, reliable and accurate position to the user. Therefore, alternative positioning techniques may be useful to complement or replace satellite positioning in these environments. This PhD study has studied the possibility of using a mobile TV system based on the DVB-SH standard as system of opportunity for positioning. The advantages of using a DVB-SH system for positioning are multiple. First, such system has a good availability in metropolitan areas, including indoor. Secondly, the emitters are synchronized and their density should be sufficient to track signals from several emitters simultaneously. This opens the possibility of using timing measurements from several emitters to find a position by tri-lateration. Also, the large bandwidth of the TV signal, required for the transmission of video content, should be beneficial for the accuracy of the timing measurements and for the robustness against multipaths. Therefore, DVB-SH system seems to be an interesting candidate as system of opportunity for positioning. However, several challenges are to be solved for such a solution to be relevant. First, the signals propagate in the urban environment, which creates challenging conditions for positioning such as strong power fading, blockage of the desired line-of-sight signal or large echoes. Secondly, the DVB-SH standard uses an OFDM modulation, which has not been studied for positioning purpose. Therefore, techniques for fine tracking of the first received signal replica will have to be developed. Finally, a particularity of modern broadcast system is the use of a Single Frequency Network, in which all emitters send the same signal on the same carrier frequency. Therefore emitter identification in a Single Frequency will be another issue to be solved. This PhD study has proved the feasibility of positioning using DVB-SH signals. The main contributions of this work are the propositions of (1) an OFDM signal delay tracking method working in urban propagation channels, and (2) a modification to the network deployment permitting emitter identification and (3) a first assessment of the position accuracy using the proposed algorithms. These two methods have very low impact on the initial TV broadcasting service if the right set of signal parameters is chosen: no signal modification is required and the network deployment modification uses a feature already present in the DVB-SH standard. The positioning method was simulated using real urban propagation channel measurements. The obtained position has root mean square error of 40m. The main error contribution comes from tracking a non-line-of-sight signal. Further work would be required to deal with this issue, which would lower the position root mean square error to 7m, which has been locally observed in the simulation for good tracking conditions.

In this paper, we present the implementation of the acquisition algorithm of a versatile Global Navigation Satellite System (GNSS) receiver for satellite applications. For versatility purpose, the choice of the receiver algorithms has been motivated by 1) their capability to fulfill the application requirements with a moderate complexity, 2) their capability of being factorized in a small set of elementary modules that can be configured and combined in various ways in order to process both GPS and Galileo current and future signals. These algorithms have been specified using SystemC, a modeling language that can be common to hardware and software flow. The use of a virtual platform for simulation allows us to identify bottleneck of the architecture and to propose algorithm modification to solve them.

A simple parametric model was recently introduced to model peaky altimetric waveforms [1] [2]. This model assumes that the received altimetric waveform is the sum of a Brown echo and Gaussian peaks. A maximum likelihood estimator for the parameters of this Brown + peak model was studied in [2] in the simple case where altimetric signals are corrupted by a single peak. However, an analysis conducted on real altimetric waveforms from the PISTACH project [3] shows it is also interesting to consider multi-peak models. This paper studies a generalization of the algorithm presented in [2] to estimate the parameters of multi-peak altimetric signals. The main contribution of this paper is a method allowing one to estimate the number of peaks which are present the Brown + peak model. The effects of model order mismatch will also be studied. Simulation results conducted on synthetic and real altimetric waveforms allow one to appreciate the performance of the proposed multi-peak model and its interest related to the single-peak version. Note that comparisons between the different proposed algorithms for altimetric waveform parameter estimation is done based on a 3 parameter Brown model estimating the amplitude, the epoch and the significant wave height of the echo. When dealing with peaky waveforms, the classical algorithm (MLE3) can fail to fit the altimetric signal, as shown in Fig. 1 (black curve). The single-peak model provides interesting results (left figure - red curve) but cannot model accurately the presence of multiple peaks in the observed signal. The multi-peak algorithm proposed in this paper clearly shows significant improved performance

The demodulation performance achieved by any of the existing GNSS signals, such as GPS L1 C/A, GPS L2C or GPS L5, is satisfactory in open environments where the available C/N0 is quite high. However, in indoor and in urban environments, several characteristics degrade the demodulation performance. In particular, in these environments, the C/N0 level of the received signal is often very low. Also, when the receiver is in motion, the C/N0 suffers additional fast variations due to changing diffraction conditions which can further affect the GNSS messages demodulation. Therefore, since the mass-market applications being conceived nowadays are aimed at indoor and urban environments, it is necessary to study and to search alternative demodulation/decoding methods which improve the GNSS messages demodulation performance in these environments. It is also needed to consider new GNSS signals, such as GPS L1C and GALILEO E1, which were developed recently. These signals aim at providing satellite navigation positioning service in any kind of environment, giving special attention to indoor and urban environments. Therefore, this dissertation also analyses the demodulation performances of the new GNSS signals as they are defined in the current public documents. Moreover, new GALILEO E1 message structures are proposed and analysed in order to optimize the demodulation performance as well as the quantity of broadcasted information. Therefore, the main goal of this dissertation is to analyse and to improve the demodulation performance of the GPS L1 C/A, GPS L2C, GPS L5, GPS L1C and GALILEO E1 signals, specifically in indoor and urban environments, and to propose new navigation message structures for GALILEO E1. A detailed structure of the sections of this dissertation is given next. First, the subject of this thesis is introduced, original contributions are highlighted, and the outline of the report is presented. Second, this dissertation begins by a description of the current structure of the different analysed GNSS signals, paying special attention to the navigation message structure and the implemented channel code. Also, the different channel code decoding techniques used in subsequent sections of this dissertation are fully described. In the third section, two types of transmission channel models are presented. These models represent the correlator outputs as used for carrier tracking and demodulation in the two environments, assuming ideal code tracking. On one hand, the open environments, and even some indoor environments, are modelled by an AWGN channel. Therefore, this report describes the AWGN channel mathematical model and presents its effects on the signal carrier phase tracking process, since the phase tracking performance altogether with the code tracking performance condition the demodulation performance. On the other hand, the urban environments and some indoor environments are modelled as a mobile channel. In this case, there exists several possible mathematical mobile channel models, and thus one of them is selected and this choice is justified. Moreover, the problems affecting the signal carrier phase tracking process are also presented, altogether with two techniques employed during the simulations in order to track the signal carrier phase. These techniques are the PLL and the channel estimation. Additionally, for both transmission channels, the structure implemented during the simulations in order to reproduce the channel characteristics is presented. In the fourth and fifth sections, efforts to improve the demodulation performance of the existing GNSS signals are presented. In the fourth section, a tentative to make a binary prediction of a part of the GPS L1 C/A navigation message is presented. Note that the binary prediction introduced is a prediction of the bits forming the navigation message, not a prediction of the physical magnitude of different message fields. More specifically, the prediction of the broadcasted satellite ephemeris is tried using the GPS L1 C/A almanacs data, a long term orbital prediction program provided by TAS-F, and some signal processing methods. These signal processing methods can be separated in two types. The first type searches the ephemeris data prediction by using the past ephemeris data values, or, in other words, the history of the ephemeris data. The methods tried are the ephemeris data spectral estimation, the PRONY method, and a neural network. The second type of methods tries to exploit the correlation between the different GPS satellite orbits. More precisely, these methods try to exploit the fact that some GPS satellites share the same orbit with their position delayed in time. Then, in the fifth section, improvements to the GPS L2C and GPS L5 navigation message demodulation performance are brought by using their channel codes in a non-traditional way. The proposed method consists in combining the navigation message inner and outer channel codes in order to correct more received words. The technique is such that the receiver accepts as the transmitted word the most probable word provided by the Viterbi decoding, or inner channel code, which meets the outer channel code verification. In fact, this technique follows the same principle as the list Viterbi decoding method although the proposed algorithm is completely different. Also, the technique solves some additional problems due to the GPS navigation messages. The proposed algorithm is described, its advantages and drawbacks presented, some possible modifications are given and its performance is compared with the performance obtained by traditional GPS L2C and GPS L5 decoding methods. Additionally, another method used to improve this performance is presented. This method consists in using the ephemeris data probabilities in order to improve the traditional Viterbi decoding. In the sixth section, the GPS L1C and GALILEO E1 Open Service demodulation performance is analysed in different environments. More specifically, this section presents a brief study of the structure of both signals to determine the received C/N0 in an AWGN channel. Then, the demodulation performance of these signals is analysed through simulations in different environments. The environments analysed are the open, indoor and urban environments modelled by the AWGN channel and the mobile channel. Therefore, this section presents the demodulation performance obtained when the GNSS signals are transmitted through an AWGN channel assuming perfect tracking, an AWGN channel with thermal noise affecting the PLL tracking, an AWGN channel with thermal noise and dynamic stress error affecting the PLL tracking, and a mobile channel with a GNSS receiver travelling at 5km/h, 30 km/h and 50 km/h. In this last case, the carrier phase tracking is achieved either by using a PLL or by applying a channel estimation technique. In the seventh section, efforts to improve the different GALILEO E1 signal performances are presented. Specifically, this dissertation presents a new navigation message structure which improves the demodulation performance and increases the signal information transmission rate. The new proposed navigation message structure consists in adopting a message structure similar to the GPS L1C message structure but also in including a signalling technique known as Code Shift Keying, or CSK, which increases the information transmission rate. In fact, the CSK technique consists in shifting the PRN code of each transmitted symbol in order to map for each code shift a fixed quantity of bits. This results into an increased quantity of bits transmitted during a symbol length, from one to the number of mapped bits. The CSK implementation into a navigation signal analysis is detailed next. First, the drawbacks introduced by the CSK implementation are analysed, and mainly the problems due to the fact that the data channel PRN code is no longer synchronized with the pilot channel PRN code. This induces that the data channel can no longer be used to acquire or to track the signal in the signal parts where the CSK is implemented. Therefore, the new acquisition performance is analysed. The optimal source mapping code shifts which reduce the acquisition false alarm arte when the data channel is employed are also presented. Initial conclusions concerning the impact of CSK on the tracking performance is also presented. Second, this dissertation searches for the demodulation performance of different CSK source mapping configurations. The CSK source mapping configuration refers to the bits and their distribution into packets to be transmitted by the same CSK symbol. Therefore, in order to analyse the different CSK source mapping options, the following points are studied. First, the interest in using the CSK polarity in order to encode an extra bit is analysed. Second, the study determines whether it is best to transmit bits belonging to several packets inside the same CSK symbol, or if it is best to transmit only bits belonging to the same packet. Third, the theoretical BER curves for different numbers of bits forming a CSK are shown. Fourth and last, figures depicting the BER and WER of different CSK source mapping configurations are presented. Additionally, since some of the packets transmitted by the CSK symbols implement channel codes with soft inputs, the theoretical expressions of the likelihood ratios of the bits coded by a CSK symbol have been calculated and verified. Finally, we conclude this study.

The Maseng-Bakken model has shown interesting properties to model rain attenuation for Ka and Q/V broadband satellite systems. This paper derives the optimal rain attenuation predictor based on the Maseng-Bakken model. The optimal predictor is obtained by minimizing the mean square error between the rain attenuation and its estimate. We show that this predictor reduces to a bank of filters whose parameters depend on the rain attenuation power spectral density. Simulation results allow us to appreciate the performance of the resulting rain attenuation prediction that is compared with more traditional strategies.

In this paper, we propose to extend the WAVE Architecture with satellite, WiFi andcellular technologies. Many applications can be considered in vehicular networks whichcan be provided through one or several communication systems. In this paper, the mapupdate application has been considered over different technologies. Simulations were runusing NS3 simulator to compare the behavior of files downloads with 802.11p and with othertechnologies. The aim is to optimize the necessary amount of resource, and to determinethe best combination of technologies to download maps.

In acoustic, ultrasonic or electromagnetic propagation, crossed media are often modelled by linear filters with complex gains in accordance with the Beer-Lambert law. This paper addresses the problem of propagation in media where polarization has to be taken into account. Because waves are now bi-dimensional, an unique filter is not sufficient to represent the effects of the medium. We propose a model which uses four linear invariant filters, which allows to take into account exchanges between components of the field. We call it bi-filter because it has two inputs and two outputs. Such a circuit can be fitted to light devices like polarizers, rotators and compensators and to propagation in free space. We give a generalization of the Beer-Lambert law which can be reduced to the usual one in some cases and which justifies the proposed model for propagation of electromagnic beams in continuous media.

This paper addresses the problem of classifying altimetric signals according to their shapes. The proposed classifier is divided into three steps. A one-class support vector machine method is first used to isolate the large amount of Brown-like echoes from others signals which are considered as outliers. The second step extracts pertinent features from the the remaining echoes (which cannot be well described by the Brown model). These features are projected onto discriminant axes using linear discriminant analysis. The final step classifies the projected feature vectors using a standard Bayesian classifier. The proposed three step classification strategy is evaluated on supervised real altimetric echoes.

In this paper, we investigate the application of multiuser detection techniques to a Low Polar Orbit (LPO) mobile satellite used in the ARGOS system. These techniques are used to mitigate the multiple access interference in the uplink transmission of the system. Unlike CDMA, due to the Doppler Effect, each signal has a different received carrier frequency and a different propagation delay. Multiuser detection techniques are proposed for asynchronous transmission in ARGOS system: the maximum likelihood detector, the conventional detector, and the sequential interference cancellation detector, as solutions to tackle the interference effects. Bit Error Rate performance graphs are shown for these techniques.

This paper studies a new Bayesian unmixing algorithm for hyperspectral images. Each pixel of the image is modeled as a linear combination of so-called endmembers. These endmembers are supposed to be random in order to model uncertainties regarding their knowledge. More precisely, we model endmembers as Gaussian vectors whose means have been determined using an endmember extraction algorithm such as the famous N-finder (N-FINDR) or Vertex Component Analysis (VCA) algorithms. This paper proposes to estimate the mixture coefficients (referred to as abundances) using a Bayesian algorithm. Suitable priors are assigned to the abundances in order to satisfy positivity and additivity constraints whereas conjugate priors are chosen for the remaining parameters. A hybrid Gibbs sampler is then constructed to generate abundance and variance samples distributed according to the joint posterior of the abundances and noise variances. The performance of the proposed methodology is evaluated by comparison with other unmixing algorithms on synthetic and real images.