High data rate global positioning system receiver performance analysis for ionospheric monitoring within the Canadian high arctic region

Abstract

The Canadian High Arctic Ionospheric Network (CHAIN) has recently begun an expansion of Global Positioning System (GPS) receiver stations within the Canadian auroral and high latitude regions which will be used to monitor the ionosphere. The stations included in this expansion will utilize the Septentrio PolaRxS Pro Global Navigation Satellite System (GNSS) receiver. This receiver is new to the CHAIN network and work regarding its performance as well as newly obtainable data thanks to its improved specifications, specifically the increased sampling rate of 100 Hz, must be analyzed. One method of high arctic ionospheric monitoring in which the receiver will take part involves the calculation of G PS-derived total electron content (TEC). The accuracy of GPS-derived TEC relies heavily on the accurate estimation of the instrumental biases, known as the differential code bias (DCB). The most appropriate methods of estimating the DCBs for the Septentrio PolaRxS Pro receivers are determined and tested. These include two variations of the minimization of standard deviations method as well as two variations of bias estimation through the comparison of stationderived TEC and TEC maps provided by the International GNSS Service (IGS). Biases obtained using the minimization of standard deviations methods range from -9.81 TECU to 9.36 TECU. Methods involving the comparison of station-derived TEC and TEC maps return bias values ranging from -4.01 TECU to 18.05 TECU and -8.84 TECU to 11.57 TECU for a least squares comparison and a direct, per epoch, comparison method, respectively. Another method of ionospheric monitoring in which the receiver will be used involves the logging and analysis of signal intensity and phase scintillation. The PolaRxS Pro is capable of sampling GPS carrier signal intensity and phase at a maximum rate of 100 Hz, double that of previous receivers. The characteristics of the amplitude and phase scintillation spectra at 100 Hz sampling rates are described. Results are obtained specifically within the auroral region during May 24th through 31st 2013. Wavelet and Fourier transform methods of analysis are described for a qualitative and quantitative comparison of the higher frequency spectral range with the current theoretical predictions. Higher frequency amplitude spectra shows an abrupt deviation from theoretical predictions. Temporal analysis shows no dominant characteristics during the scintillation event in the higher frequency region where static analysis shows a near zero spectral slope before, during and after the event. This constant spectrum in the high frequency amplitude alludes to noise in the region. Phase spectral analysis shows a more subtle deviation from theoretical predictions in the higher frequency regime. In the lower frequency portion, up to about 20 Hz, the expected behavior based on previous work is observed, a power law behavior with a mean spectral slope around -2. In the higher frequencies the mean spectral slope becomes increasingly more positive up to a value of -0.4394 seen in the 40 Hz to 50 Hz range.