Please use this identifier to cite or link to this item:
|Title:||Improved GNSS Carrier Phase Tracking for Hand Held Receivers|
|Keywords:||Geodesy;Engineering--Electronics and Electrical|
|Abstract:||Despite constituting a prime portion of GNSS applications, pedestrian hand held users still cannot achieve centimetre-level carrier phase based solutions; rather, they currently provide code phase based metre-level solutions. The limiting factors that inhibit the provision of kinematic solutions include poor antenna quality, harsh operating environments, poor receiver components, especially low-cost oscillators, to name a few. Many of these factors needs to be addressed to improve the performance of kinematic solutions to hand held users. To this end, the first phase of this research is aimed at understanding the pedestrian GNSS signal propagation channel. A study of the effects of pedestrian dynamics on GNSS signal parameters, namely carrier frequency and phase, is conducted. A theoretical relationship is established between the dynamics measured by inertial sensors and GNSS signal parameters, also considering the oscillator g-sensitivity, and verified empirically using live GPS L1 C/A signals. Considering this relationship, pedestrian dynamics is characterized using appropriately processed inertial sensors data to understand the effects on GNSS signal parameters. The oscillator g-sensitivity and line-of-sight dynamics are identified to be two major sources of carrier phase disruptions. Pedestrian dynamics are characterized in the frequency domain in terms of power spectral densities and cumulative spectral densities. One of the important results of this research phase is that the pedestrian dynamics exhibit quasi-periodicity; this is proven to be useful in developing improved tracking architectures that reduce the impact of pedestrian dynamics in subsequent research phases. A study of signal attenuation due to the presence of human body is also conducted and impact of such attenuation on the carrier phase tracking performance is empirically measured. The second phase of this research addresses oscillator g-sensitivity issues. Existing theoretical models are used to quantify the carrier frequency perturbations due to changes in acceleration. A method of laboratory calibration of g-sensitivity using GPS signals is employed. A novel tracking architecture that compensates for g-sensitivity induced carrier frequency perturbations using accelerometers measurements via feed-forward correction is proposed. The method is tested using live GPS L1 C/A signals in a pedestrian environment with various accelerometer specifications ranging from tactical to MEMS grade. Performance is also evaluated for vehicular navigation. The third phase of this research results in a novel tracking architecture that measures pedestrian dynamics and implements a feed-forward compensation technique. A simple short term Fourier transform based technique is used to isolate harmonics inherent to pedestrian dynamics and predict the dynamics ahead of time. The prediction is proven useful in delivering improved carrier phase tracking performance, again, evaluated with live GPS signals. The final phase of the research performs a comparative analysis of carrier phase continuity in pedestrian environment between a software-defined receiver and a survey-grade receiver. It is also demonstrated, by comparing kinematic solutions performance, that a software-defined receiver can also deliver comparable performance to a commercial survey-grade receiver when properly tuned. Overall, this thesis is aimed at addressing some of the key issues arising in pedestrian environments that are detrimental to carrier phase tracking, so that GNSS receiver design progresses one step closer to the possibility of ubiquitous kinematic solutions.|
|Appears in Collections:||Electronic Theses|
Files in This Item:
|ucalgary_2015_bhaskar_srinivas.pdf||5.13 MB||Adobe PDF||View/Open|
Items in The Vault are protected by copyright, with all rights reserved, unless otherwise indicated.