eprintid: 1507935 rev_number: 25 eprint_status: archive userid: 608 dir: disk0/01/50/79/35 datestamp: 2016-11-10 10:13:55 lastmod: 2020-02-12 16:15:31 status_changed: 2016-11-10 10:13:55 type: thesis metadata_visibility: show creators_name: Atkins, CLE title: Observation-domain sidereal filtering for high-rate GPS precise point positioning ispublished: unpub divisions: UCL divisions: A01 divisions: B04 divisions: C05 divisions: F44 keywords: sidereal filtering, precise point positioning, GPS, carrier phase, seismology, multipath abstract: Accurate and timely determination of coseismic displacement (the quasi-permanent displacement of the ground caused by an earthquake) is crucial for tsunami warning and situational awareness systems. Seismometers measure acceleration or velocity very precisely. However, the process of integrating such data to yield displacement data is difficult and error-prone. In contrast, The Global Positioning System (GPS) can measure displacement directly, but is also subject to errors, one of which is multipath. Multipath errors can lead to errors in the measurement of small displacements or obscure the displacement entirely. Errors due to multipath are highly dependent on the geometry of GPS satellite orbits and surrounding reflectors relative to the receiving antenna itself. Each satellite has a ground track repeat period of approximately one sidereal day. Hence, this relative satellite-reflector geometry will repeat with the same period. Using this fact, it is possible to identify and remove the error signature induced by multipath by analysing data from adjacent days, yielding an improved time series of displacements and hence a more discernible coseismic offset. This process is commonly known as 'sidereal filtering'. This thesis describes a sidereal filter algorithm that attempts to remove the multipath error signature from the GPS measurements themselves before processing them rather than from the resulting position time-series. It is shown to generally produce a more stable position time series over periods from a few tens to a few hundred seconds, remove high-frequency multipath error more effectively, yield better stability during satellite outages and measure small centimetre-level displacements more accurately than a position-domain sidereal filter. However, results are inconclusive when applying the algorithm to the measurement of the coseismic displacements of a real earthquake, but it is demonstrated that an observation-domain sidereal filter is better at enabling one to distinguish certain types of seismic wave from a position time series. date: 2016-08-28 date_type: published oa_status: green full_text_type: other thesis_class: doctoral_open language: eng thesis_view: UCL_Thesis primo: open primo_central: open_green verified: verified_manual elements_id: 1146341 lyricists_name: Atkins, Christopher lyricists_name: Ziebart, Marek lyricists_id: CLEAT95 lyricists_id: MZIEB32 actors_name: Atkins, Christopher actors_id: CLEAT95 actors_role: owner full_text_status: public pagerange: 1-363 pages: 363 event_title: UCL (University College London) institution: UCL (University College London) department: Civil, Environmental & Geomatic Engineering thesis_type: Doctoral editors_name: Ziebart, M citation: Atkins, CLE; (2016) Observation-domain sidereal filtering for high-rate GPS precise point positioning. Doctoral thesis , UCL (University College London). Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/1507935/1/Thesis_final.pdf