This article shows actual examples of how the information provided by VeEX's SkyView™ GNSS Antenna Installation Verification feature has been used to troubleshoot modern multi-band GNSS antenna installations.
Tabe of Content
- Case 1: Inconsistent C/No Traces - Santiago, Chile
- Case 2: Improved Performance of Existing GNSS Antenna Distribution
Case 1: Inconsistent C/No Traces - Santiago, Chile
While writing a short article to explain the "No Satellites" zones in GPS/GNSS antenna reception coverage polar graphs, we reached out to one of our partners in the southern hemisphere, to run some tests and capture GNSS antenna reception heat maps with SkyView™. About a year ago, they had upgraded their sync lab with new antenna, equipment and a GNSS Assurance (monitoring) system, making them a good choice for the task. They had installed a new dual-band quad-constellation GNSS antenna on their roof, to feed an Adtran-Adva-Oscilloquartz OSA5422HQ++ PRTC and PTP Grandmaster (T-GM) as well as other precision timing equipment, as part of their synchronization lab improvements. Which they felt had been working fine.
After performing a preliminary 24-hour test with MTTplus' SkyView™, we noticed that there were some degraded (yellow, orange, red) satellite traces on an otherwise good satellite reception area of the coverage map. That seemed inconsistent, even for cloudy autumn days (no rain). What could have caused those streaks?
1.1 Check Individual Reception Quality Patterns
First, we used the captured SkyView GNSS antenna reception quality (C/No) data to identify which satellite system (constellation) was suffering from poor signal quality reception issues. After reviewing the coverage map for each individual constellation, it is quite clear that GLONASS was the culprit.
Traditionally, users of dual-constellation GNSS receivers have selected GPS+GLONASS settings on their precision timing equipment, because they were the most mature and reliable PNT (position, navigation and time) systems. However, Galileo and BeiDou systems have matured, therefore, depending on location, Galileo and BeiDou are becoming better choices to complement GPS. Also, since 2023, we have received anecdotical reports about some degradation in the GLONASS signal quality (C/No density) and we can only speculate that it could be related to the current conflicts (2024). Therefore, based on test results, we may no longer recommend the use of GLONASS in certain situations.
Recommendation: Change the GNSS receiver settings in their dual-constellation equipment, from GPS+GLONASS (left) to GPS+Galileo (right), to improve overall reception performance, as seen below. You can get a good idea of the expected final antenna reception pattern improvements, based on data from the previous test, by selecting (combining) GPS and Galileo heat maps on the SkyView polar graph page.
For equipment supporting more than two constellations, we recommend the use of GPS, Galileo and BeiDou (add GLONASS if the signals are consistently of good quality in your region).
1.2 Check C/No Density Values (Signals' Quality)
It is also important to look at the C/No density table, as it contains more details about the whole antenna installation performance.
Although we were told that the new antenna supports 1.2 and 1.5 GHz bands, we mainly see satellite signals from "band 1" (1.5 GHz) and only a few from "band 2" (1.2 GHz). However, those few L2 signals seem to have poor reception quality (C/No < 18 dB-Hz). It indicated that we had more troubleshooting to do.
Identified their GNSS antenna as an ADVA 1047020173-01 and checked its datasheet to confirm it is indeed a multi-band device (supports 1.16 to 1.25 GHz and 1.52 to 1.6 GHz).- Identified a primary GEMS MGS12 1x2 active splitter and checked its datasheet to confirm it has the bandwidth required to cover the two sets of bands (supports 1.15 to 1.65 GHz).
Identified a secondary Microchip-Microsemi-Symmetricom 58536A 1x4 active splitter and checked its datasheet. Unfortunately, it turns out it only has bandwidth for the GPS L1 band (supports 1.555 to 1.595 GHz). Therefore, the performance of the whole GNSS antenna installation was being limited by this element, which most likely was a leftover from their original GPS distribution system. Also, since GLONASS L1 frequency range (1.598 to 1.604 GHz) is outside of the splitter's specs, this may have contributed to the marginal signal quality streaks reported earlier.
Recommendation: Replace the secondary 1x4 active splitter with a more modern multi-band version, supporting 1.15 to 1.61 GHz, to take advantage of the new antenna's and equipment's capabilities, improving overall system reliability.
It is also important to confirm that any unused ports on the splitters are properly terminated (50Ω).
1.3 Final Results (72-hour test)
As expected, after replacing the limiting 1x4 active splitter, all GNSS receivers now benefit from improved redundancy, with many more good satellite signals to lock-on to (70 vs. 41), on different frequency bands and with better C/No quality, resulting on a much stronger reception pattern, improved accuracy, as well as more resilient multi-path and interference/jamming mitigation.
The test set's dual-band quad-constellation GNSS receiver can now see satellites from GPS (L1C/A, L2C), Galileo (E1B/C, E5B), GLONASS (L1OF, L2OF), and BeiDou (B1I, B2I) bands.
Here are the improved GNSS reception quality patterns, represented by the C/No density heat maps, following a comprehensive 72-hour SkyView test:
Although some of GLONASS satellite trails still show some signs of degradation (yellow), there are significant improvements from their original "Poor-Marginal" to "Fair-Good", compared to the initial test results.
Regarding the degraded areas to the North-East and East, the influence of major topographic features and tall buildings is significant and unavoidable. For context, here is the full SkyView polar graph, superimposed over the map of Santiago.
Modern mapping services are very useful tools for getting a good initial idea of the environment around an antenna under test, without having to go up to the roof or be physically present on site. Then you can decide if it may be necessary to climb on to the roof for a visual inspection of the antenna, connections and surroundings.
Later, a physical visual inspection on the roof revealed a few nearby brick walls and roof lines blocking the low elevation path towards the north-east direction. Elevating the antenna by approximately 2 to 3 meters (6 to 9 feet) could further enhance reception quality for low elevation angles.
1.4 Closing Comments
Why didn't their GNSS Assurance (monitoring) system warn the end user about their problem or limitation? We are not sure, since we are not familiar with the monitoring tool being used. From the GNSS monitoring system's "before" heat map screenshots we saw, the issue wasn't obvious, perhaps giving end users a false sense of security.
In our assessment, the GNSS monitoring (assurance) system used by them may have a too wide (very generous) GOOD range (green, 21 to 50 dB-Hz), which might overlook potential issues. While we agree that the vendor's own GNSS receivers may operate effectively within those ranges, VeEX's SkyView test focuses on validating the optimal quality of an antenna installation. Therefore, we provide a more detailed color grading scale, in which 21 dB-Hz is not considered "good" for a new antenna or distribution system. It's not just about whether our test set's GNSS receiver can work with the antenna under test or not; it's about ensuring the signal quality and coverage are the best possible for the specific antenna system being tested. While 21 dB-Hz could be viewed as "Marginal" or "Fair" for lower elevation angles (towards the horizon), they should be viewed as "Poor" for higher angles (towards the zenith). This highlights the importance of having a more detailed color scale, which can help uncover issues like the one addressed in this case study.
This case in Santiago Chile was remotely troubleshot and diagnosed from Fremont California, using the test set's built-in EZ-Remote access functionality and the assistance from our local partner, MIDEX Chile.
Sites with Internet access can be remotely monitored during the test. In this particular case, being able to check it after 24 hours (instead of waiting until the first 72-hour test finished) saved time, money and resources.
Case 2: Improved Performance of Existing GNSS Antenna Distribution
2.1 What We Started With
We used the SkyView™ GNSS Reception Verification tool to qualify an existing roof antenna installation and distribution system. The antenna was a multi-band quad-constellation, which is a good starting point. However, the relatively short antenna feed (30.5m / 100ft) was all passive, with a 1x2 passive splitter to feed two receivers. Although the initial 72-hour results showed a decent performance, based on the color scale, it also shows plenty of room for improvements.
Signal Quality (C/No) Color Scale:
45 to 53+ dB-Hz, Excellent- 36 to 44 dB-Hz, Good
- 27 to 35 dB-Hz, Fair
- 18 to 26 dB-Hz, Marginal
- 9 to 17 dB-Hz, Poor
- 1 to 8 dB-Hz, Bad
2.2 Improvements
After replacing the 1x2 passive splitter with an active 1x4 splitter, with 10 dB total gain to compensate for the split loss and some cable loss, the overall signal quality improvements are significant, as seen in the second SkyView map results.
The new GNSS signal distribution not only offers better signal quality, stability and resiliency to the systems connected to it, it also provides two extra ports for new equipment. VeEX's SkyView GNSS reception coverage maps helped validate and document the improvements, as well as justify (bring into perspective) the small investment on the new 1x4 active splitter.
2.3 General Recommendations
- Do not use T connectors as RF splitters.
- Always terminate unused ports on active and passive GNSS splitters.
- Procure good quality active GNSS splitters with enough gain to compensate for the power split and cable losses. Make sure they are multiband.
We do recommend testing for 72 hours, to log more data and obtain better coverage, since the SkyView test runs unattended anyway. However, if time is not on your side, one can use a dual-band quad-constellation receiver to run tests for 24 hours, with decent coverage since there are a lot more satellite signals to log (assuming that the antenna under test is also multiband). In single or dual constellation receiver scenarios, 24 hours may not provide enough datapoints in the coverage map.
Related Test Solutions
- MTTplus - Modular Test Platform
- CX41 - Coaxial Time Domain Reflectometer (TDR)