Long-term ephemeris generation for mobile devices
GPS receivers in challenging signal reception areas can find it difficult to maintain continuous signal lock for the up-to 30 seconds needed to obtain the broadcast ephemeris from the satellites. It can take many minutes for receivers to obtain the ephemeris parameters from enough satellites to obta...
Main Author: | |
---|---|
Published: |
University College London (University of London)
2013
|
Subjects: | |
Online Access: | http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.626137 |
id |
ndltd-bl.uk-oai-ethos.bl.uk-626137 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-bl.uk-oai-ethos.bl.uk-6261372017-02-17T03:19:47ZLong-term ephemeris generation for mobile devicesStacey, P.2013GPS receivers in challenging signal reception areas can find it difficult to maintain continuous signal lock for the up-to 30 seconds needed to obtain the broadcast ephemeris from the satellites. It can take many minutes for receivers to obtain the ephemeris parameters from enough satellites to obtain a position fix. As well as users expecting a rapid time-to-first-fix (TTFF) when performing a warm start, the advent of the E-911 legislature in the USA also provides motivation for rapid TTFF with accurate positions. In this thesis, a scalable orbit prediction algorithm is developed to calculate an extended ephemeris of GPS satellites of up to 7 days on a mobile device to within an accuracy of 20 m rms (radial). The algorithm numerically integrates force models from a set of initial conditions, position and velocity, determined solely from the broadcast ephemeris. Forces used include Earth gravity models of varying resolution, third-body gravity effects, solar radiation pressure, relativity, and solid Earth tides. Adding these forces improves the ephemeris accuracy at the cost of the execution time, so research is undertaken to produce ephemeris data quickly whilst maintaining similar accuracy. An orbit determination algorithm is developed that uses a least-squares technique to optimise the computed orbit over multiple sets of previously observed broadcast ephemeris data. This modifies the estimate of the initial conditions made available to the orbit predictor, which then provides a modified ephemeris which has a substantially lower root-mean-square error. The algorithm again uses a suite of force models, and only broadcast ephemeris orbit data was utilised, no other satellite ephemeris is required. Studies determining the effect of modifying the a priori weights and the number of sets of ephemeris data used for observations are undertaken and results demonstrated on a HTC TyTN II mobile device. The orbit determination technique presented in this thesis is novel, in that it utilises only broadcast ephemeris parameters previously acquired by a single GPS receiver to modify the initial position and velocity for the orbit prediction algorithm. Previous techniques used either empirical modelling or least squares techniques to modify empirical parameters of the orbit prediction algorithm.621.384University College London (University of London)http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.626137http://discovery.ucl.ac.uk/1389940/Electronic Thesis or Dissertation |
collection |
NDLTD |
sources |
NDLTD |
topic |
621.384 |
spellingShingle |
621.384 Stacey, P. Long-term ephemeris generation for mobile devices |
description |
GPS receivers in challenging signal reception areas can find it difficult to maintain continuous signal lock for the up-to 30 seconds needed to obtain the broadcast ephemeris from the satellites. It can take many minutes for receivers to obtain the ephemeris parameters from enough satellites to obtain a position fix. As well as users expecting a rapid time-to-first-fix (TTFF) when performing a warm start, the advent of the E-911 legislature in the USA also provides motivation for rapid TTFF with accurate positions. In this thesis, a scalable orbit prediction algorithm is developed to calculate an extended ephemeris of GPS satellites of up to 7 days on a mobile device to within an accuracy of 20 m rms (radial). The algorithm numerically integrates force models from a set of initial conditions, position and velocity, determined solely from the broadcast ephemeris. Forces used include Earth gravity models of varying resolution, third-body gravity effects, solar radiation pressure, relativity, and solid Earth tides. Adding these forces improves the ephemeris accuracy at the cost of the execution time, so research is undertaken to produce ephemeris data quickly whilst maintaining similar accuracy. An orbit determination algorithm is developed that uses a least-squares technique to optimise the computed orbit over multiple sets of previously observed broadcast ephemeris data. This modifies the estimate of the initial conditions made available to the orbit predictor, which then provides a modified ephemeris which has a substantially lower root-mean-square error. The algorithm again uses a suite of force models, and only broadcast ephemeris orbit data was utilised, no other satellite ephemeris is required. Studies determining the effect of modifying the a priori weights and the number of sets of ephemeris data used for observations are undertaken and results demonstrated on a HTC TyTN II mobile device. The orbit determination technique presented in this thesis is novel, in that it utilises only broadcast ephemeris parameters previously acquired by a single GPS receiver to modify the initial position and velocity for the orbit prediction algorithm. Previous techniques used either empirical modelling or least squares techniques to modify empirical parameters of the orbit prediction algorithm. |
author |
Stacey, P. |
author_facet |
Stacey, P. |
author_sort |
Stacey, P. |
title |
Long-term ephemeris generation for mobile devices |
title_short |
Long-term ephemeris generation for mobile devices |
title_full |
Long-term ephemeris generation for mobile devices |
title_fullStr |
Long-term ephemeris generation for mobile devices |
title_full_unstemmed |
Long-term ephemeris generation for mobile devices |
title_sort |
long-term ephemeris generation for mobile devices |
publisher |
University College London (University of London) |
publishDate |
2013 |
url |
http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.626137 |
work_keys_str_mv |
AT staceyp longtermephemerisgenerationformobiledevices |
_version_ |
1718414019289677824 |