Indoor positioning is an enabling technology primed to impact the indoor application space as Global Navigation Satellite Systems (GNSS) did for the outdoor space. Amongst the competing positioning technologies are methods of different mediums: light, radio frequency and ultra-wideband, ultrasonic, and imaging; methods of different modalities: received signal strength, angle-of-arrival, time-of-flight; and methods of different mathematics: trilateration, triangulation, machine learning, and signal processing.
Light-based positioning compared to other positioning schemes exploits fixed-location directional luminaires placed regularly throughout a space as anchor points -- there is an efficiency argument for multi-purpose lighting and a cost-share argument for infrastructure-based positioning. Similar to the satellite infrastructure with GNSS, with anchor points and models for light propagation and construction, position is estimated based on received signals at active photodiode-equipped target devices. Received signal strength, a common first order attribute, alone is not noise resilient enough for centimeter-level 3D positioning. Methods using angle diversity produce better results particularly in 3D but with more complex hardware.
For this dissertation, we exploit angle diversity and modulated optical sources in light-based positioning systems to estimate position to centimeter-level accuracy in 3D. We propose, analyze, and contribute two novel positioning schemes that use these concepts. One of the proposed schemes is a new hybrid 3D indoor positioning technique, Ray-Surface Positioning (RSP), which incorporates a narrow field-of-view (FOV) optical source (Ray) with wide diffuse optical sources (Surfaces) to position active devices in 3D. The second scheme, a Zone-based Positioning Service (ZPS), is a positioning scheme and architecture that incorporates an angle diverse narrow FOV optical source at the positioned device. This unique design decision allows the active device to position itself directly with respect to photovoltaic anchor points but also to position other devices in its FOV called transitive positioning. Along with these contributions, we also investigate several other related topics.
Concisely, as part of the dissertation, we contribute (a) review of the state-of-the-art, (b) analysis for steering Lambertian sources, (c) method of creating angle diversity from a narrow FOV optical source, (d) novel positioning approaches in (1) RSP and (2) ZPS, (e) proof of concept prototypes for (1) RSP and (2) ZPS, and (f) architectures for indoor positioning applications.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/41040 |
| Date | 19 May 2020 |
| Creators | Lam, Emily |
| Contributors | Little, Thomas D. C. |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Rights | Attribution-NonCommercial-ShareAlike 4.0 International, http://creativecommons.org/licenses/by-nc-sa/4.0/ |
Page generated in 0.0021 seconds