Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2013. / This electronic version was submitted and approved by the author's academic department as part of an electronic thesis pilot project. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from department-submitted PDF version of thesis. / Includes bibliographical references (p. 141-153). / Most current methods for 802.11-based indoor localization depend on either simple radio propagation models or exhaustive, costly surveys conducted by skilled technicians. These methods are not satisfactory for long-term, large-scale positioning of mobile devices in practice. This thesis describes two approaches to the indoor localization problem, which we formulate as discovering user locations using place and motion signatures. The first approach, organic indoor localization, combines the idea of crowd-sourcing, encouraging end-users to contribute place signatures (location RF fingerprints) in an organic fashion. Based on prior work on organic localization systems, we study algorithmic challenges associated with structuring such organic location systems: the design of localization algorithms suitable for organic localization systems, qualitative and quantitative control of user inputs to "grow" an organic system from the very beginning, and handling the device heterogeneity problem, in which different devices have different RF characteristics. In the second approach, motion compatibility-based indoor localization, we formulate the localization problem as trajectory matching of a user motion sequence onto a prior map. Our method estimates indoor location with respect to a prior map consisting of a set of 2D floor plans linked through horizontal and vertical adjacencies. To enable the localization system, we present a motion classification algorithm that estimates user motions from the sensors available in commodity mobile devices. We also present a route network generation method, which constructs a graph representation of all user routes from legacy floor plans. Given these inputs, our HMM-based trajectory matching algorithm recovers user trajectories. The main contribution is the notion of path compatibility, in which the sequential output of a classifier of inertial data producing low-level motion estimates (standing still, walking straight, going upstairs, turning left etc.) is examined for metric/topological/semantic agreement with the prior map. We show that, using only proprioceptive data of the quality typically available on a modern smartphone, our method can recover the user's location to within several meters in one to two minutes after a "cold start." / by Jun-geun Park. / Ph.D.
| Identifer | oai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/82475 |
| Date | January 2013 |
| Creators | Park, Jun-geun |
| Contributors | Seth Teller., Massachusetts Institute of Technology. Department of Aeronautics and Astronautics., Massachusetts Institute of Technology. Department of Aeronautics and Astronautics. |
| Publisher | Massachusetts Institute of Technology |
| Source Sets | M.I.T. Theses and Dissertation |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | 153 p., application/pdf |
| Rights | M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission., http://dspace.mit.edu/handle/1721.1/7582 |
Page generated in 0.0012 seconds