Novel Methodologies to Mitigate Multipath for Indoor Localization

Author(s): Kabiri, Saman | Advisor(s): De Flaviis, Franco | Abstract: The primary purpose of this dissertation is to address the critical issues in the development and design of an accurate direction finding system with an emphasis on indoor applications. One of the main limitations of current direction finding methods is the lack of utilizing a precise data model in estimation algorithms. To achieve an accurate data model as the basis of any estimation algorithm, the antenna element or an array of antennas, as the main block of any direction finders, must be appropriately modeled by taking into account any practical considerations including the interaction between elements and non-uniformity of the gain and phase pattern response. In this dissertation, for the first time, the phase center (PC) of the receiving antenna is employed to construct the data model accurately. The theoretical model is validated by both full-wave simulation and measurement results. As investigated thoroughly in this dissertation, the estimation accuracy of the direction finding system based on the phase center model remains acceptable even in the case of placing the array elements very close to each other, i.e., when the array components are tightly coupled. This outcome is of vital importance since the total direction finder size, which is limited by the antenna array size can be ultra-miniaturized without compromising the detection accuracy of the system. Although utilizing the novel phase center model will provide accurate results for outdoor direction findings, a novel estimation algorithm designed based on this newly introduced model is imperative to resolve the multipath dilemma for indoor applications completely. As a significant part of this dissertation, an advanced algorithm is proposed and tested in both simulated and real-world environments. Several highly miniaturized arrays so-called tightly coupled arrays are designed, fabricated, and measured as viable candidates for indoor applications including Real Time Locating Systems (RTLS), Indoor Positioning Systems (IPS), and Internet of Things (IoT). This dissertation also presented detailed analytical and simulation-based studies of the angle-of-arrival (AoA) estimation by a single antenna for indoor direction finding applications in three-dimensional space. Using a single radiator instead of an array of antennas can broaden the scope of indoor applications by providing a low-cost, ultra-miniaturized, and low-complexity solution. The proposed solution is realized by using the phase center model as the central apparatus of the single element direction finder. The theory of the single antenna in three-dimensional space is developed under the assumption of knowing the radiator location and orientation at any given time. In the era of smartphones and wearables, an inertial measurement unit (IMU) is a universal built-in module in most devices which is employed to estimate the displacement and rotation of the single radiator. As a result, a hybrid sensor consisting of a single radiator and an IMU is designed and tested to estimate the AoA in real-world situations. The accuracy of the proposed method is validated by conducting extensive measurements for several real-world scenarios.