CLAS multi-stream test broadcast

category: gnss

Introduction

The Quasi-Zenith Satellite System (QZSS, or Michibiki), operated by the Cabinet Office, Govenment of Japan, broadcasts not only positioning signals like GPS, but also augmentation information signals to improve positioning accuracy. One type of augmentation information is the Centimeter-Level Augmentation Service (CLAS, pronounced as same as the Cirrus), which provides high-precision augmentation within Japan in a short period of time.

In the future, in order to expand the augmentation capabilities of CLAS, the augmentation information will be multi-streamed. Test broadcasts of this will begin on June 30, 2025, so we will review the details.

CLAS augmentation information

Currently, the available Michibiki satellites are QZS-2 (No. 2), QZS-4, and QZS-1R (successor to QZS-1), which are in quasi-zenith orbit, and QZS-3, which is in geostationary orbit. Currently, all Michibiki satellites transmit the same CLAS augmentation information.

CLAS augments the positioning results of the US Global Positioning System (GPS), the European Galileo (E), and Michibiki (J). CLAS has an information transmission rate of 2 kilobits per second. Because CLAS transmits information on ionospheric and tropospheric delays for representative locations within Japan, the number of augmenting satellites is limited to 17, including GEJ. Because Japan is long from north to south, available satellites may vary by region, and low-elevation-angle satellites may be unavailable due to building obstructions, so an increase in the number of augmenting satellites is desirable.

CLAS multi-streaming

In order to increase the number of satellites that can be augmented, it was announced that the Michibiki satellite constellation will be divided into two groups (Pattern 1 and Pattern 2), with different satellites being augmented in each group. The CLAS multi-stream test broadcast will be carried out from June 30 to July 14, 2025.

Update the User Interface Specification (IS-QZSS-L6) for CLAS and CLAS test library, CLASLIB, and the Test Transmission from QZS-3

By using CLAS multi-streaming, it is possible to simultaneously receive two augmentation pattern information, potentially enabling the use of more augmentation target satellites. The satellite signal specification IS-QZSS-L6-007-Draft (Interface Specification - Quasi-Zenith Satellite System - L6 band), which includes this information, has been published. The broadcast pattern information can be found in the header (Table 4.1.2-2 L6 message type ID, p.24).

According to IS-QZSS-L6-007-Draft 4.1.1.2 Compositions of Augmented Satellites (page 20), the Michibiki satellites are classified as follows:

pattern #1pattern #2
QZS-4, QZS-5, QZS-3QZS-2, QZS-1R (QZS-3)

QZS-5 is scheduled for launch at the end of 2025. QZS-3 normally broadcasts Pattern 1, but will broadcast Pattern 2 when other Pattern 1 satellites are unavailable. QZS-6, which will be in geostationary orbit and begin operation on July 18, 2025, will broadcast MADOCA-PPP ionospheric delay information using the L6D signal, so it is not listed here.

Receivers with only one L6D signal reception channel, or receivers that receive multiple L6D channels but select the single L6D signal with the strongest signal, require a firmware update. If the signal strength of the receiving satellite weakens and the receiver switches to another satellite with a different pattern, the information read position becomes unstable and decoding may fail. Such receivers must either change their behavior to select a satellite within the same pattern or wait up to 30 seconds for Subtype 1 to be broadcast.

This pattern information has reserved bits, so it may be possible to extend it. CLAS information is transmitted every 30 seconds, so in the worst case scenario, it could take 30 seconds from the start of reception until information acquisition can begin. However, if this backup pattern can be used to stagger the start of transmission, it is expected that the time until information acquisition can begin can be shortened.

Multi-stream test broadcast

For this multi-stream test broadcast, the augmentation information from QZS-3 will be changed to Pattern 2. Other satellites will broadcast the conventional Pattern 1. To prevent malfunction of conventional receivers as mentioned above, the Pattern 2 augmentation information from QZS-3 will be broadcast with the alert flag on (i.e. prohibited for use).

I tried receiving CLAS augmentation information from QZS-2 (pseudo random noise number PRN 194) and CLAS augmentation information from QZS-3 (PRN 199) side by side. I used RTKLIB 2.4.3b34, QZS L6 Tool ver.0.1.5, and the CLAS stream from a GNSS observation station.

str2str -in ntrip://ntrip.rnav.info.hiroshima-cu.ac.jp:80/CLAS 2>/dev/null | alstread.py -p 194 -l | qzsl6read.py
str2str -in ntrip://ntrip.rnav.info.hiroshima-cu.ac.jp:80/CLAS 2>/dev/null | alstread.py -p 199 -l | qzsl6read.py

QZS-2 (the left side of the screen) and QZS-3 (the right side) transmit almost identical subtype (ST) information at the same time. The alert flag (an asterisk before the CLAS display) is on for the augmentation information from QZS-3. Because the satellites augmented by the two are different, the required bit length for augmentation differs. In addition, with CLAS, information is transmitted closer to the beginning of the subframe (SF), so some of the ST information placement differs.

Although the ST transmission order is exemplified in IS-QZSS-L6 4.1.2.2.14 Sub Type Transmission Pattern, it does not appear to be exactly like this in reality. However, since the ST transmission order for both satellites was nearly identical in this test broadcast, it appears that the Michibiki Control Station is allocating one augmentation information stream into two patterns and uploading them to each Michibiki satellite.

Multi-stream augmentation target satellites

Next, I used qzsl6read.py with the -t 1 option to know the satellites and signals to be augmented described in ST1. The observation date and time is June 30, 2025, 19:20 UTC.

pattern #1 onlypattern #2 onlycommon
G01 L1C/A L2CM+L L2Z-tracking L5I+QE21 E1B+C E5aI+QG02 L1C/A L2Z-tracking
G10 L1C/A L2CM+L L2Z-tracking L5I+QE23 E1B+C E5aI+QG04 L1C/A L2CM+L L2Z-tracking L5I+Q
G23 L1C/A L2CM+L L2Z-tracking L5I+QE29 E1B+C E5aI+QG07 L1C/A L2CM+L L2Z-tracking
G08 L1C/A L2CM+L L2Z-tracking L5I+Q
G09 L1C/A L2CM+L L2Z-tracking L5I+Q
G16 L1C/A L2Z-tracking
G26 L1C/A L2CM+L L2Z-tracking L5I+Q
G27 L1C/A L2CM+L L2Z-tracking L5I+Q
E13 E1B+C E5aI+Q
E15 E1B+C E5aI+Q
E19 E1B+C E5aI+Q
E26 E1B+C E5aI+Q
E27 E1B+C E5aI+Q
J02 L1C/A L2CM+L L5I+Q
J03 L1C/A L2CM+L L5I+Q
J04 L1C/A L2CM+L L5I+Q

In the current test broadcast, there are three different satellites targeted for augmentation between Pattern 1 and Pattern 2. Pattern 1 can be said to be a setting that prioritizes GPS use, while Pattern 2 is a setting that prioritizes Galileo use.

I checked the satellite position for this date and time on GNSS View for Sapporo, Tokyo, and Naha.

GNSS View

The satellite constellations seen in the sky do not appear to change significantly in these regions.

I plotted Sapporo, Tokyo, and Naha on the Geospatial Information Authority (GSI) of Japan Vector Map.

GSI Vector Map

The Michibiki satellites are at an altitude of 20,000 to 40,000 kilometers above the Earth’s surface. Even if they are 1,000 kilometers away on a 6,370-kilometer radius of the Earth, the satellite position does not appear to change significantly.

In the future, the difference in the satellites targeted for augmentation between patterns may become larger. When this happens, it will be necessary to simultaneously receive multiple L6D signals and combine the streams.

Conclusion

Test broadcasts of multi-stream CLAS augmentation information for Michibiki have begun, so I tried receiving it. By receiving and combining these two streams, we can expect to increase the number of satellites that can be augmented. In the current test broadcasts, the difference in the number of satellites that can be augmented between these streams is about three satellites.

Making CLAS multi-stream is one of the challenges toward future advances in high-precision positioning. I will continue to closely monitor developments.


Related article(s):