Study Of A Localization System For Wi-Fi Networks Using HomeRF, Based On Signal Levels And Spread Referenced Model

Location based services and sensor networks using IEEE-802.11, applied in the context of wireless local area networks are being studied in recent researches, once they open new perspectives for new applications and aggregate value to such networks. In this sort of application some service must be provided, where input and also output parameters are directly influenced by the user’s physical location. In this thesis we will present a system for locating wireless local area network devices, based on the radio frequency characteristics of such networks using HomeRF. Thus, we addressed issues related to some aspects of localization systems through, an architecture based on wireless sniffers and by constructing a localization model based on signal propagation models, in which its parameters are calculated in real time. This will guarantee a considerable self-sufficiency and adaptation capacity to this proposed system, once it does not need human intervention to work, neither from the network administrator or the wireless user being located. Moreover, a probabilistic method was used to help us estimating wireless devices positions, based on the previous constructed model. We later demonstrate the feasibility of our approach by reporting results of field tests in which the proposed technique was implemented and validated in a real world indoor environment.Various methods of localization models have been proposed in the literature in order to improve the performance of locator’s that determine the physical location of a wireless user’s network, making it possible to develop applications that make use of this information. Some ideas for localization based applications are discussed, showing how the actual position information can be useful within the context of applications and services for wireless networks.This thesis provides a brief introduction to HomeRF wireless communications and applications for LANs localization based. In this type of application, services may be provided so that the user input parameters and/or output are directly influenced by their physical position. Various methods and localization models have been proposed in the literature in order to improve the performance of locators that determine the physical location of a wireless user’s network, making it possible to develop applications that make use of this information. This work was carried out on networks that operate using the HomeRF standard2.0, so the first section presents an overview of these networks together with the motivation for this work. Some ideas for localization based applications are discussed, showing how the actual information about position can be useful within the context of applications and services for wireless networks.The objectives of this work are summarized in Chapter1, followed by a brief description of its main contributions and innovations. Finally, the work structure is described in the last section of Chapter1.In Chapter2, some references related to the topic addressed here will be discussed. The classes of systems will be presented, showing that the ideal applications for each class, pointing out advantages and disadvantages of their use. Always will be taken into consideration, cost of implementation and applicability in a WLAN environment.The chapter3, first introduces the standard used in wireless networks, including aspects of the physical layer and basic operation of the devices that make up these networks. It is shown here also, through experiments conducted in the laboratory, that it is possible by using the RSSI number to assist in tracking devices. Finally, are discussed propagation models and how they apply to this job.Chapter4describes in detail the proposed system. The localization mechanism is provided, including a definition of its components and how they interact. The whole step-by-step system is shown, since the measurements of signal’s quality through construction of propagation map until the position estimation performed by the localization server. Also in this chapter is found the results obtained after the implementation of the proposed localization mechanism. First results will be presented regarding the accuracy of the system described in this work. Some system parameters were changed so that it was possible to evaluate the impact of these errors in the localization estimation. Then, the proposed system is compared with other methods of finding, already established and frequently referenced in the literature. Also Methods for improving the accuracy of the estimator are proposed. To finalize the chapter the results regarding the accuracy of the system will be presented. Quantitative and qualitative comparisons between the proposed system and other studies in the literature will be presented, along with observations, justifications and explanations.Chapter5concludes the work done by consolidating the results presented and drawing relevant conclusions. The main contributions made during the research are presented. Finally, some recommendations are also made for future research. Figures17, Tables4, References56.