This article gives a brief overview of the acquisition and tracking of Galileo IOV signals received from the GSAT0101 satellite on the morning of December 15. Researchers in the PLAN Group successfully recorded E1 and E5 data using a single dual-channel front-end and subsequently acquired and tracked E1 B/C, E5a and E5b signals using the PLAN Group GSNRx software GNSS receiver.
A little over seven weeks after launch, one of the two Galileo IOV satellites began to transmit on the E1 band. To the delight of eagerly waiting researchers worldwide, Galileo-PFM (GSAT0101) broke radio silence on December 10, 2011. Within hours the community was alive with reports of successful acquisition and tracking of the E1 B/C signals. Four days later the E5 signal was also activated. In the early hours of the morning of the 15th of December researchers gathered in the PLAN Group at the University of Calgary and observed the sky filled with broadcasting satellites from three GNSS. Using a dual channel front-end designed in-house, a Novatel GPS-703-GGG antenna and a laptop computer, IF data was collected to examine these new signals. This data was processed by GSNRx, a reconfigurable a multi-system, multi-frequency software receiver developed by the PLAN Group [1]. The equipment used to acquire and process the data is shown in Figure 1.
Figure 1 The equipment used to acquire and process the Galileo-PFM signals included an in-house dual frequency front-end, a 10 MHz OCXO, a Novatel GPS-703-GGG antenna and a standard laptop computer running the GSNRx software receiver.
At approximately 03:20 MST (UTC – 7:00) more than 20 GNSS satellites were visible from a rooftop mounted antenna. Having reconfigured the front-end to accommodate the E5 band, IF data was collected which included Galileo E1 B/C and E5 A/B, GIOVE-B E1 B/C and E5a, GPS L1 C/A and L5, and GLONASS L1 C/A. Following some last minute modifications to GSNRx to include the Galileo E5b signals, the samples were processed, simultaneously tracking GPS and Galileo on both the L1/E1 and L5/E5 frequencies and GLONASS on L1. A screenshot of the receiver in operation is shown in Figure 2.
Figure 2 Screenshot of GSNRx while processing the Galileo PFM signals
The versatility of GSNRx had been exploited in the past when new signals were brought online. In particular, the modular design adapted for PLAN’s software receiver had been utilized to quickly add new signals and new signal processing techniques. Once again this flexibility was drawn upon to facilitate the last-minute addition of the E5b I/Q signals (that very night) and to enable the stand-alone tracking of each signal component. By the same means, of course, this structure could be easily manipulated to enable composite tracking of data/pilot signal pairs or even facilitate vector tracking of all signals in view.
A subset of the raw correlator values for the E1 B, E1 C, E5a I and E5a Q signals are shown in Figure 3, (note that the E1 C values have been offset by -2.0×105 for clarity). A data-rate of 250 symbol/s is clearly visible on the E1 B and E5b signals while a 50 symbol/s stream can be observed on the E5a I signal. The 25 chip secondary code is also evident on E1 C at a rate of 250 chip/s.
Figure 3 Raw Correlator Values for the E1 B/C, E5aI/Q and E5bI/Q signals. The bit periods can be clearly seen on E1B, E5aI and E5bI. The secondary code can be observed on E1C while the pilot signal can be seen on singals E5aQ and E5bQ.
All six components of the Galileo-PFM signals shown above (transmitted on PRN 11) were tracked independently and their signal modulations were found to agree with the Galileo Open Service ICD [2]. A trace of the measured carrier-to-noise floor ratios for the Galileo signals is shown in Figure 4. As indicated by the ICD, the E5b signals were observed at 2 dB lower power than the E1 B and C signals. The E5a signals, however, were expected to be received at the same power as E5b and yet were observed at approximately 4 dB lower power. This is believed to be a combination of the antenna and IF filtering within the front-end as the E5a center frequency is located relatively near the pass-band edge of both. This front-end was initially designed for 40 MHz bandwidth, but used in this experiment at 50 MHz, as will be discussed later.
Figure 4 Measured C/N0 for Galileo-PFM Signals
The software receiver was once again reconfigured, this time to produce signal correlator values spaced along a delay of approximately 700 m and 70 m for the E1 A/B and E5 A/B signals, respectively, such that the cross-correlation of the received and local-replica PRN sequences could be examined. The signals were tracked for 10 seconds and the 1 ms correlator values averaged, to produce estimates of the code cross-correlation function. The characteristic ripple of the CBOC modulation on E1 B/C can be seen in Figure 5 (left), particularly on the right-most ascending feature of the envelope. Likewise, the alt-BOC cross-correlation of E5a Q in Figure 5 (right) is as expected. It is noted that the E5a I signal has suffered some distortion due to the filtering effects mentioned above.
Figure 5 Measured cross-correlation functions for the Galileo PFM E1 B and C signals (left) and E5a I and E5b I signals (right).
The PLAN group’s front-end is a highly flexible GNSS signal capture tool ideally suited for use with the GSNRx software receiver. The front-end, photographed in Figure 6, allows software reconfiguration of oscillator source (onboard, or external), antenna bias voltage, sampling rate, and IF bandwidth in addition to other low level control options making it highly adaptable. Furthermore, the center frequency, and filter bandwidth of each of the two hardware channels is independently configurable between 1150 – 2000 MHz, and between 4—40 MHz bandwidth (single sided) respectively.
Figure 6: PLAN group two-channel reconfigurable front-end with main system blocks labeled. The external clock and GNSS antenna SMA connectors are along the right edge, while the data interface is via mini-USB on the opposite side of the front-end.
Typically the front-end is configured to collect dual bands of 40 MHz two-sided bandwidth in order to cover the L1 and L2 transmission bands of both GPS and GLONASS as is shown in the right and central blocks within Figure 7. To allow the capture of E5a/E5b, the front-end configuration software was used to move the center frequency of channel B from 1237 MHz to 1192 MHz, the bandwidth of channel B from 33 MHz to 50 MHz, and to increase the sampling rate of both channels from 40 to 50 Ms/s.
Figure 7: Front-end channel A and channel B typically configured to capture GPS and GLONASS L1+L2, but reconfigured here to allow capture of Galileo IOV E5a+E5b signal in lieu of L2 band.
While each of the E5a and E5b signals have main lobe widths of 20.46 MHz (two sided), the composite E5 signal covers 50 MHz of spectrum, overlaying both the current GPS L5 signal at 1176, and the future GLONASS L3 signal near 1207 MHz. In order to demonstrate the capabilities of the GSNRx software receiver as an L5/E5 + L1/E1 system, it was desirable to capture the new IOV signals in their entirety.
The Galileo PFM satellite was observed from the Calgary Laboratory on the E1 link since the 12th of December at approximately 08:00 hrs and on the E5 link since the 14th of December at approximately 18:00 hrs. The last successful acquisition of the satellite on either E1 or E5 was at 03:20 hrs on the 15th of December and indicated a Doppler of approximately +2.3 kHz at E1. This figure is compatible with a reported elevation of approximately 40 degrees and rising, as reported by a number of software packages operating on a TLE [3]. Researchers recorded IF data once again at 03:55 on the 15th of December but failed to acquire any of the Galileo-PFM signals, suggesting the satellite may temporarily have ceased transmission.
References
Petovello, M. G., and C. O’Driscoll, G. Lachapelle, D. Borio and H. Murtaza (2008), “Architecture and Benefits of an Advanced GNSS Software Receiver,” Journal of Global Positioning Systems, vol. 7, no. 2, pp. 156-168.
Galileo Project Office. Galileo OS SIS ICD. http://ec.europa.eu/…/galileo/files/galileo-os-sis-icd-issue1-revision1_en [Accessed: 15 December 2011].
NORAD Two-Line Element Sets. http://celestrak.com/NORAD/elements/, [Accessed: 15 December 2011].
