Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Land Surveying and Geo-Informatics | en_US |
| dc.contributor.advisor | Shi, Wenzhong (LSGI) | en_US |
| dc.creator | Yu, Yue | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/12281 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Autonomous localization by integrating Wi-Fi and MEMS sensors in large-scale indoor spaces | en_US |
| dcterms.abstract | Location-based services (LBS) have become more and more important with the development of Internet of Things (IoT) technology and increasing popularity of IoT terminals in recent years. Global Navigation Satellite System (GNSS) is widely used for positioning outdoors while it is still challenging to realize autonomous, precise and universal indoor localization based on the existing devices. Among most indoor positioning technologies, the Wireless Fidelity (Wi-Fi) based positioning is regarded as an effective way for realizing ubiquitous and high-precision indoor navigation, especially the presentation of next generation Wi-Fi access point which supports the state-of-art Wi-Fi Fine Time Measurement (FTM) protocol. Micro-Electro-Mechanical System (MEMS) sensors can provide an accurate short-term navigation solution, which also provides a potential way for autonomously generating the crowdsourced Wi-Fi received signal strength indication (RSSI) based fingerprinting database, by collecting and mining the users' daily-life trajectories and corresponding signals of opportunity. | en_US |
| dcterms.abstract | This thesis proposes an automatic and precision-controllable algorithm for multi-source fusion based wireless positioning using the combination of Wi-Fi FTM, crowdsourced Wi-Fi RSSI fingerprinting, and IoT terminals integrated MEMS sensors, by which the realized ubiquitous positioning accuracy can reach 1.5~4.5m (within 75th percentile), and meter-level accuracy can be achieved under Wi-Fi FTM covered indoor scenes. Compared with previous hybrid navigation algorithms or structures, the main innovation points of this research are: | en_US |
| dcterms.abstract | 1)This research presents an autonomous three-dimensional (3D) positioning algorithm for low-cost MEMS sensors. This algorithm is based on the inertial navigation system (INS) mechanization and comprehensively utilizes multi-level constraints and observables (including: pseudo observations, gravity vector, altitude increment, pedestrian dead reckoning (PDR), zero velocity update (ZUPT), zero angular rate update (ZARU), quasi-static magnetic field (QSMF), non-holonomic constraint (NHC)). The proposed algorithm can be used without any external equipment and user intervention, and the autonomous 3D indoor positioning performance can be realized under changeable motion and handheld modes and environmental interference. | en_US |
| dcterms.abstract | 2)This research proposes and compares three different Wi-Fi FTM bias estimation algorithms to solve the problem of Wi-Fi FTM based ranging biases between changing terminals and Wi-Fi access points (APs). In which the polynomial-based (PB) approach can provide the best performance of ranging bias estimation, but requires the priori information; The Gradient Descent (GD) based calibration algorithm does not need the priori information, but needs to extract the initial quasistatic status information; The tightly-coupled bias estimation algorithm integrates multiple location sources and pedestrian's motion information, and calculates and feedbacks the ranging bias estimation result in real-time to obtain the optimal convergence value. In this point, corresponding error and iterative models are designed, which can realize adaptive ranging bias estimation towards different scenarios and improve the accuracy and universality at the signal source level. | en_US |
| dcterms.abstract | 3)This research develops an autonomous 3D indoor localization and trajectory reconstruction framework based on MEMS sensors and sparsely deployed Wi-Fi FTM stations, Bluetooth Low Energy (BLE) nodes, and Quick Response (QR) codes based landmarks, and proposes and testes two corresponding trajectory error optimization algorithms, including the two-sided filtering and smoothing algorithm based on the adaptive unscented Kalman filter (AUKF) and Rauch-Tung-Striebel (RTS), and the GD based global optimization algorithm. The proposed trajectory optimization algorithms can effectively eliminate the cumulative error caused by the MEMS/landmarks integration framework and maintain the calculation efficiency, and more accurate smoothed navigation results can be acquired compared with one-sided filtering. | en_US |
| dcterms.abstract | 4)This research proposes a deep-learning based crowdsourced Wi-Fi fingerprinting database generation and updating framework based on the daily-life trajectories of public users. The influencing factors and time correlation of the optimized crowdsourced trajectory error are modeled and predicted by the multi-layer perception (MLP) network. In addition, the results of trajectories error prediction are further applied for crowdsourced trajectories classification, segmentation, merging, and the final Wi-Fi RSSI fingerprinting database construction and updating, which can effectively reduce the redundancy of the generated database and improve the accuracy and the stability of database matching. | en_US |
| dcterms.abstract | 5)This research proposes a 3D navigation architecture based on the integration of MEMS sensors, Wi-Fi FTM, and crowdsourced RSSI fingerprinting, and makes the comprehensively experimental analysis. In which the MEMS sensors/Wi-Fi FTM tightly-coupled integration model can realize meter-level positioning accuracy in Wi-Fi FTM covered indoor environments, and has strong anti-interference ability; the MEMS sensors/RSSI fingerprint loosely-coupled integration model can provide a more universal and wide-coverage positioning solution and compensate for the limited deployment of Wi-Fi FTM stations; and the final hybrid MEMS sensors/Wi-Fi FTM/RSSI fingerprint integration model can effectively achieve automatic, and precision-controllable positioning in large-scale 3D indoor spaces. In addition, this research designs corresponding signal quality evaluation strategies for all three integration models to achieve adaptive weight adjustment of each observation. Therefore, by taking better advantage of the merits of low-cost sensors, Wi-Fi FTM, and crowdsourced RSSI fingerprinting, the proposed algorithm has the following advantages: | en_US |
| dcterms.abstract | 1)The algorithm can significantly improve the performance of attitude estimation and 3D dead reckoning by self-calibrating the navigation parameters without the need for any external equipment or user intervention, which can be applied in case of complex indoor environments and changeable handheld modes of smartphones. | en_US |
| dcterms.abstract | 2)The algorithm can provide accurate and reliable 3D indoor navigation results in large-scale indoor spaces using smartphone integrated MEMS sensors and sparsely deployed landmarks such as Wi-Fi stations, BLE nodes, and QR codes; In addition, different error optimization methods are further applied for decreasing the cumulative error of forward navigation. | en_US |
| dcterms.abstract | 3)The algorithm can realize the automatic construction of Wi-Fi fingerprinting database using the collected crowdsourced trajectories, and develops a comprehensive deep-learning based trajectories evaluation, selection, partition, and merging framework to improve the robustness and efficiency of final generated database. | en_US |
| dcterms.abstract | 4)The algorithm can provide universal and precision-controllable positioning performance by integrating both Wi-Fi FTM and RSSI fingerprinting based absolute location sources and MEMS sensors based DR approach. Autonomous 3D localization can be realized in large-scale indoor spaces and meter-level positioning accuracy can be realized in Wi-Fi FTM covered indoor scenes. | en_US |
| dcterms.abstract | There are various potential applications for the outcomes of this research, for example: | en_US |
| dcterms.abstract | • Precise location based services that use IoT terminals; | en_US |
| dcterms.abstract | • Mobile mapping and crowd-sensing; | en_US |
| dcterms.abstract | • Crowdsourced navigation by using daily-life data provided by public users; | en_US |
| dcterms.abstract | • Crowdsourced data mining and geo-spatial big data analysis; | en_US |
| dcterms.abstract | • Multi-source fusion based seamless localization towards pedestrians and vehicles; | en_US |
| dcterms.extent | xxiii, 177 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2023 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.LCSH | Wireless localization | en_US |
| dcterms.LCSH | Microelectromechanical systems | en_US |
| dcterms.LCSH | Detectors | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12281