Abstract
In recent decades, localization using wireless technologies has gained significant attention, and different location-related measurements, such as time-of-arrival (TOA), time-difference-of-arrival (TDOA), time-sum-of-arrival (TSOA), received signal strength (RSS), and angle-of-arrival (AOA), can be employed to determine the target position. The advancements in the fifth-generation beyond (5GB) network enable high-precision distance and angle measurements, leading to centimeter-level localization accuracy. This improvement has facilitated the development and expansion of various location-based services (LBSs), including advanced driver assistance systems (ADAS), simultaneous localization and mapping (SLAM), surveillance, object tracking, etc.In general, two crucial metrics can be used to assess the performance limits of the localization system: 1) localizability, and 2) Cramér-Rao lower bound (CRLB). In practice, the inadequate availability of anchors leads to the failure of localization, and localizability can be utilized to quantify whether a sufficient number of anchors is available to determine the target position without ambiguity. When an adequate number of anchors is detected, the mean-square error (MSE) can be employed to assess the localization performance. The resulting MSE will be compared against the CRLB, which represents the minimum achievable error for unbiased estimators.
The thesis aims to establish a comprehensive theoretical framework for assessing the performance of diverse localization strategies in wireless networks, with a particular focus on the metrics of localizability and CRLB. Different from the conventional approaches that rely on fixed, deterministic anchor positions, the devised framework explores the evaluation metrics with arbitrary anchor geometries, where their positions are modeled as random parameters with probability density functions (PDFs) specified. Considering the variability in anchor placements, the devised framework provides a more realistic representation of localization performance in real-world scenarios. Furthermore, this framework integrates various network parameters, such as the availability of anchors, path-loss exponent, types of wireless channels, etc. Therefore, the system designers can gain insights into optimizing the network configurations to fulfill the desired localization requirements.
Lastly, the thesis focuses on characterizing the localization performance within millimeter-wave (mmWave) and terahertz (THz) heterogeneous networks (HetNets), where mmWave anchors serve as macro anchors and THz anchors function as micro anchors, thereby improving network coverage. The average achievable rate of the HetNet is analytically derived to quantify the capability of serving anchor to deliver sufficient data transmission rates to the target for high-precision localization, and the single-anchor CRLB distribution is presented to portray the overall localization performance with arbitrary mmWave and THz anchor placements.
To summarize, the thesis studies the performance limits of various localization systems by assessing their localizability and CRLB with random network geometries. The tractable expressions for these metrics are analytically derived, allowing system designers to evaluate the localization schemes without time-consuming and complex simulations, enabling more efficient evaluation and design of localization systems.
| Date of Award | 17 Jun 2024 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Young Jin CHUN (Supervisor) |