Frequency domain techniques have been widely used in indoor radio propagation measurements and modeling for telecommunication applications. This work addresses measurement of the time of arrival (TOA) of the first path for geolocation applications using results of frequency domain channel measurements. First, we analyze the effect upon TOA measurement accuracy due to: sampling period of the radio channel in the frequency domain, sampling period in the time domain used for detection of the TOA and the windowing filter used before transformation to the time domain. Then, we provide some results of measurement made in line of sight (LOS) and Obstructed LOS (OLOS) indoor environments in order to compare the characteristics of the measured TOA in these two important scenarios for indoor geolocation applications. Finally, we compare the measurement results with the ray tracing based model that had been developed previously for indoor geolocation applications.

Networks with hundreds of ad hoc nodes equipped with communication and position finding abilities are conceivable with recent advancements in technology. Methods are presented in this thesis to assess the communicative capabilities and node position estimation of mobile ad hoc networks. Specifically, we investigate techniques for providing communication and geolocation with specific characteristics in wireless ad hoc networks. The material presented in this thesis, communication and geolocation, may initially seem a collection of disconnected topics related only distantly under the banner of ad hoc networks. However, systems currently in development combining these techniques into single integrated systems. In this thesis first, we investigate the effect of multilayer interaction, including fading and path loss, on ad hoc routing protocol performance, and present a procedure for deploying an ad hoc network based on extensive simulations. Our first goal is to test the routing protocols with parameters that can be used to characterize the environment in which they might be deployed. Second, we analyze the location discovery problem in ad hoc networks and propose a fully distributed, infrastructure-free positioning algorithm that does not rely on the Global Positioning System (GPS). The algorithm uses the approximate distances between the nodes to build a relative coordinate system in which the node positions are computed in three-dimensions. However, in reconstructing three-dimensional positions from approximate distances, we need to consider error threshold, graph connectivity, and graph rigidity. We also statistically evaluate the location discovery procedure with respect to a number of parameters, such as error propagation and the relative positions of the nodes.

La préparation d'interventions de maintenance dans les installations industrielles a dorénavant recours à des outils d'étude, de modélisation et de simulation basés sur l'exploitation de modèles virtuels 3D des installations. L'acquisition de ces modèles tridimensionnels s'effectue à partir de nuages de points mesurés, depuis plusieurs points de vue, par balayage angulaire horizontal et vertical d'un faisceau laser par scanner laser terrestre. L'expression dans un repère commun de l'ensemble des données acquises est appelée consolidation, au cours de laquelle les paramètres de changement de repères entre les stations sont calculés. L'objectif de cette thèse est d'améliorer la méthode d'acquisition de données laser en environnements industriels. Celle-ci doit, au final, garantir la précision et l'exactitude nécessaires des données tout en optimisant le temps et les protocoles d'acquisition sur site, en libérant l'opérateur d'un certain nombre de contraintes inhérentes au relevé topographique classique. Nous examinons, dans un premier temps, l'état de l'art des moyens et méthodes mis en œuvre lors de l'acquisition de nuages de points denses de scènes d'intérieurs complexes (Partie I). Dans un deuxième temps, nous étudions et évaluons les données utilisables pour la consolidation: données laser terrestres, algorithmes de reconstruction de primitives et systèmes de géolocalisation d'intérieur (Partie II). Dans une troisième partie, nous formalisons et expérimentons ensuite un algorithme de recalage basé sur l'utilisation de primitives appariées, reconstruites dans les nuages de points (Partie~III). Nous proposons finalement une approche probabiliste de l'appariement de primitives permettant l'intégration des informations et incertitudes a priori dans le système de contraintes utilisé pour le calcul des poses (Partie IV). / Many pre-maintenance operations of industrial facilities currently resort on to three-dimensional CAD models. The acquisition of these models is performed from point clouds measured by Terrestrial Laser Scanning (TLS). When the scenes are complex, several viewpoints for scanning, also known as stations, are necessary to ensure the completeness and the density of the survey data. The generation of a global point cloud, i.e. the expression of all the acquired data in a common reference frame, is a crucial step called registration. During this process, the pose parameters are estimated. If the GNSS systems are now a solution for many outdoor scenes, the registration of indoor TLS data still remains a challenge. The objective of this thesis is to improve the acquisition process of TLS data in industrial environments. The aim is to guarantee the precision and accuracy of acquired data, while optimizing on-site acquisition time and protocols by, as often as possible, freeing the operator from the constraints inherent to conventional topographic surveys. In a first part, we consider the state of the art of the means and methods used during the acquisition of dense point clouds of complex interior scenes (Part I). In a second part, we study and evaluate the data available for the registration: terrestrial laser scanner data, primitive reconstruction algorithms in point clouds and indoor geolocation systems (Part II). In the third part, we then formalize and experiment a registration algorithm based on the use of matched primitives, reconstructed from per station point clouds (Part III). We finally propose a probabilistic approach for matching primitives, allowing the integration of a priori information and uncertainty in the constraints system used for calculating poses (Part IV).