QZSS Support [GPS, added in v1.170]

Is it possible to support QZSS reception on KiwiSDR's GPS receiver ?
QZSS (Quasi-Zenith Satellite System)  is a Japanese satellite positioning system and compatible with GPS L1C/A code. So the QZSS is sometimes called the “Japanese GPS.”
In Japan, there is not a sufficient number of GPS satellites to perform high-precision positioning constantly.
Currently, four satellites on QZSS will be visible at all times from locations in the Asia-Oceania regions. (PRN 193,194,195 & 199GEO)
If QZSS can be used in an integrated way with GPS, ensuring a sufficient number of satellites for stable, high-precision positioning in the Asia-Oceania regions.

Thanks for all experts.


  • jksjks
    edited February 8
    That's an interesting question. I read a little bit about QZSS, but I don't have a lot of time to look into it.

    Does anyone know if you just have to search for the QZSS PRN codes using the same techniques as for Navstar L1 C/A to get an acquisition?

  • jksjks
    edited February 6
    Yeah I've been looking at that. The section about the PRN code generation is ambiguous. What I've tried implementing so far hasn't worked.

    Update: got it now..

  • Well, that was easier than I thought.

    I'm tracking PRN 194 and 195 no problem from here in New Zealand (QZS-2 and -4 respectively). Subframes are decoding and the software is flagging them as good. I have a little work to do to make the az/el display work correctly for them.

  • jksjks
    Accepted Answer
    Okay, v1.170 is released with support for QZSS.

  • That's a great job!!
    Thank you for your quick work.
    Now PRN 193,194 & 195 can be received. (See attached image file)

    PRN199(GEO) seems to be unusable due to some obstacle now.

    Since GEO has been newly added to the QZSS specification, there is currently no commercial equipment that can be received.
    Therefore, if PRN199 can be received by Kiwi SDR, it is probably the world's first available GPS device in general.
    I'm paying considerable attention to that point now.
    1920 x 1234 - 218K
  • I have not seen any acquisitions of PRN199 yet. There is some test code in the file Beagle_SDR_GPS/gps/search.cpp to make sure I got the PRN 193-202 initialization correct and the first 10 chips match what is written in the specification.

  • Looks like PRN199 is being acquired now.

  • Yes, I also succeeded in receiving PRN199(QZS-3).

    1920 x 1234 - 231K
  • Dear jks, I am excited with the update. This is really a great achievement! I am running a kiwisdr in Japan, and now it almost always has more than four satellites. This should be the case in other part of Asia and Oceania.
  • I have recently moved to a new location where I have a limited view of the sky. Either blocked by other apartment buildings or a nearby (hopefully inactive) volcanic cone. The addition of the QZ sats really improves the amount of time I have enough sats in view (4) to compute location fixes.

  • jksjks
    edited February 10
    Today's v1.171 update adds a shadow map to the GPS tab of the admin page. This is a display built up over a long period of time showing all az/el areas of reception colored in a transparent aqua color. That way you can see areas of non-reception that are "shadowed" by buildings or terrain.

    If you're in the Asia/Oceania region a circle will be shown where the new geosynchronous QZS-3 sat should appear when it is transmitting (it seems to be off at the moment). Combined with the shadow map you should get an idea if QZS-3 will be receivable from your location. Because QZS-3 is geostationary its az/el position will not change. But of course the exact az/el depends on where the Kiwi is located.

    An alternate method to determine QZS-3 visibility: There are "dish pointing" apps on mobile devices that look through the device's camera while you point it at the sky. It then shows you a picture of what the camera sees, as usual, plus an overlay of the geosynchronous orbit belt with various longitude positions labeled. Just move the device until the cursor is over 127 deg E. If there is sky behind the cursor, and not an obstruction, then you'll probably be able to receive QZS-3.

    The image below shows the shadowing I get from the terrain to the west below about 20 degrees and buildings to the east and south. PRN195 (another QZSS sat) is passing through the QZS-3 circle, so I'll probably get reception when QZS-3 is back on.

    1024 x 640 - 202K
  • GPS alert flag: There was a late change that went out after the initial v1.171 release. Most of you will have picked up the change. But if not, force an update by clicking the "build now" button on the update tab of the admin page.

    As shown below if a satellite has the "alert" flag set in any subframe a red "A" will appear where the yellow "U" normally appears when the tracking PLL is unlocked. As of this writing PRN 195 has its alert flag set which is reflected in the outage status messages published here: http://sys.qzss.go.jp/dod/en/naqu.html

    When the alert flag is set the satellite data will not be used in computing a position fix even though subframe data is decoding properly.

    Screen Shot 2018-02-12 at 7.32.04 PM copy.jpg
    1024 x 640 - 197K
  • v1.172 adds a switch to the admin page GPS tab called "include alerted?" that overrides the exclusion of alerted sats from the position fix. The alert "A" symbol will appear green instead of red when the sat is actually being used. It will be interesting to see if some position fixes get worse when alerted sats are included.

  • It is an interesting attempt. 
    Although I tried to include the alerted sats(PRN193,194&195), it seems that there is no particular impact on the positioning result at the moment.
    By the way, about PRN199(QZS-3), the signal has been stopped for several days. 
    This may be quite a serious trouble. 
    1920 x 1234 - 248K
  • I also have not seen any problems including PRN193-5 in position fixes when they are in alert mode.

    But a few days ago when PRN199 was transmitting I noticed position fixes on your Kiwi often had altitude values that were very wrong. By thousands of meters sometimes. And the lat/lon positions were also wrong by some amount. Before I could add any code to ignore PRN199 it had stopped transmitting. So when it does return it will be very interesting to experiment with the new alert switch and see if there is any effect.

    On my own Kiwi I didn't have enough sats for any position fix during the time when I saw PRN199 transmitting. So I couldn't verify if it was also a problem here.

  • The attached image is the result of reception including PRN199 on February 9, 8:00 UTC.
    Certainly the value of altitude was quite strange.
    The contents of the navigation data may have been abnormal.

    According to some information, the last PRN199(QZS-3) signals were sent between 7:00 and 11:00 UTC on February 9.
    I can not receive it at all after that.

    1920 x 1234 - 231K
  • The alert flag of PRN193-195 has been released earlier.
    PRN199 is still stopped.
Sign In or Register to comment.