Capacity analysis of CLAS satellite augmentation information using QZSS archive data

categories: gnss
tags: clas qzss

Introduction

The information contained in the centimeter-level augmentation service (CLAS) broadcast by the quasi-zenith satellite system (QZSS), petnamed Michibiki, is roughly divided into those related to satellites and those related to regions. Regarding the former, I examined the possibility of increasing the number of augmented satellites based on the number of transmission bits.

The content of the article will be presented at the research presentation of the 2022 General Conference of the Institute of Positioning, Navigation and Timing of Japan. The manuscript (in Japanese) is also published on the site.

Presentation material for IPNTJ Conference on June 2022

Modification of QZS L6 Tool

Here, 1 hour of CLAS augmentation information from 2022-01-01 00:00:00 UTC (2022-01-01 09:00:00 Japan time) is downloaded from the QZSS archive data service.

I have modified QZS L6 Tool for this capacity analysis to include the number of satellites (Nsat), the number of signals (Nsig), and the number of augmented information bits related to the satellites (Bsat), the number of supplementary information bits related to the signal (Bsig), the number of other information bits such as header and convection delay (Bother), the number of null information (Bnull) are recorded. You can download of this modified code by checking out this 202206ipntj branch of QZS L6 Tool. The modified code is qzsl62rtcm.py only.

git clone https://github.com/yoronneko/qzsl6tool -b 202206ipntj

QZS L6 Tool on GitHub

You can easily check this correction on GitHub.

Modification display of QZS L6 Tool on GitHub

If we set this modified code qzsl62rtcm.py to trace option level 3 -t 3, the above information will be printed with the IPNTJ header. When we execute the following command in the python directory, the CLAS information file in the sample of QZS L6 Tool will be analyzed and the number of satellites and required bits will be saved in 202206ipntj.txt.

cat ../samples/2022001A.l6 | ./qzsl62rtcm.py -t 3 | grep IPNTJ > 202206ipntj.txt

In this code, when it find a message of subtype (ST) 1, it displays the number of bits aggregated. Therefore, the first data is invalid and will be removed during aggregation. Also, since the last data will not be output, the result is forcibly output in the main routine.

It takes 30 seconds to transmit a frame (a series of augmented information blocks), and since this data is for 1 hour, 120 frames of data are output during this period. Here, this data was aggregated using Excel (202206ipntj.xlsx).

Current status of CLAS augmentation information

The CLAS message broadcast from this 2022-01-01 00:00:00 UTC in one hour is as follows.

STmessage nameitems to be augmented
1Mask(definition)
2GNSS Orbit Correctionsatellite
3GNSS Clock Correctionsatellite
4GNSS Satellite Code Biassignal
6GNSS Satellite Code and Phase Biassignal
7GNSS URAsatellite
11GNSS Combined Correctionsatellite
12Atmospheric Correctionregional, satellite

When ST1 was decoded, the number of satellites to be augmented was 17 to 19, and the average number was 18.3. With ST12, which began broadcasting on November 30, 2020, there was an announcement that the maximum number of satellites to be augmented will be expanded from 11 to 17. Now it seems to be augmented for more satellites.

For these CLAS messages, the bit number ratios for satellite augmentation, signal augmentation, and other non-information are as follows. It can be seen that the bit occupation ratio of satellite augmentation is higher than that of signal augmentation and tropospheric delay augmentation.

CLAS capacity ratio

The percentage of zero-padded no-information bits accounted for 22% of the total. The average of these numbers and the number of bits is summarized in the table.

average number per periodaverage bit per period
satellite18.326169.2
signal50.69584.2
other3509.6
null11587.0
total50850.0

From this table, it can be seen that as the number of satellites increases, it increases by 1340.7 bits, as the number of signals increases, it increases by 189.5 bits, and 2.8 signals are broadcast per satellite.

If the non-information margin bit is used to increase the number of satellites, it seems possible to add about 5 satellites by simple calculation. In addition, since the bit capacity depending on the number of signals is small, it is possible to augment the Galileo E5b signal, which is not currently being broadcast, in terms of line capacity.

Conclusion

I investigated the capacity of CLAS augmentation information regarding satellite augmentation. When subtype 12 was introduced, it was announced that the maximum number of satellites to be augmented would be 17, but now I have confirmed the augmentation of 17 to 19 satellites, which is more than that. In the current CLAS, the augmentation signal for Galileo’s E5b signal is not broadcast, but the E5b signal augmentation information can also be broadcast at the transmission capacity point. Regarding the number of satellites, it seems that the number of satellites to be augmented can be added a few more.

This result also depends on the satellite placement, so at least 24 hours of analysis had to be done. It was also necessary to analyze the active ionospheric activity around July. This is a topic for future study.