Logging Settings?

Can someone verify the correct settings to input for logging Raw Data (RINEX)? I’m using an RS2 to record data for drone mapping using a WingtraOne Gen II. I was receiving a poor PPK fix % after my first flight, and this was the response they gave me after reviewing the .SBF files: “L1CA, L2P on the BS end look fine. However, the signal is completely missing on L1P and L2C, which leads to low number of satellites and made it impossible for WiH to calculate the PPK solution.
I suggest the client checks BS parameters and whether the BS meets the requirements.”

I’ll attach a link to the Wingtra specs here: Wingtra PPK minimum specs

Also attached is a screenshot of the current RINEX settings that I’m using in ReachView3

I appreciate the help!

Hi Cole,

The signals you’re describing are determined by the hardware, not by the software. So in any case of logging, for example, the L2C GPS signal should be collected. Unlike the L2P signal which RS2 doesn’t track at all.

Can you please share your base file with me? Usually, the issues with processing happen because of a rough overall quality of the data. To make any informed conclusions, I’ll need to check it myself.

1 Like

Thank you, Polina. So to clarify, the Emlid RS2 doesn’t track L2P? Can you confirm if the RS2 meets the below requirements? I will send you my base file as well.

General base station requirements

  • Possibility of continuous logging with logging interval of 15s or faster (1s is recommended for highest accuracy)
  • Logging at least two frequencies L1 (1575.42MHz) and L2 (1227.60MHz)
  • Receiving constellations GPS + GLONASS
  • Possibility to save / convert logging data to RINEX files (receiver independent exchange format) with version 2.11 or newer

Cole,

If we’re talking about GPS L2 frequency, Reach receivers track L2C signal only. It’s the updated signal but there can still be satellites that don’t provide it yet.

Answering your questions, our receivers suit all of those requirements. If there are processing issues, then we need to assess the settings you used during processing and the overall quality of the observational data. This quality, in turn, is dependent on the hardware setup and the environmental conditions that are important to know for further assessment.

I can’t help with the settings of the Wingra software, but I can check the data from Reach to see what’s there. I can see that you’ve sent it in an email. I’ll check it and get back to you with the results. Meanwhile, please share your environmental conditions and the base placement setup.

2 Likes

Thanks for getting back to me, Polina. I will attach a screenshot of the base location from the day of the flight. Environmental conditions this day were optimal - 75 degrees Fahrenheit and partly cloudy. No obstructions were in view of the base station.

I will email you the SBF file from the drone and the JSON file for your reference.

2 Likes

It looks like the base station has obstructions to the southwest, west: a shed and house to the east (assuming north is at top of imagery). This is not an optimal place for a base station.

If the shed (?) is built with any kind of metal siding, you will have severe multi-path for the base. Also, the trees and house probably obstructs at least 30° of the sky view which will lower the number of visible sats and increase the PDOP ( >2).

This is probably why you are having problems PP the data. The base needs to have a clear view of the sky in order to have good data to PP with the rover.

Hi everyone,

I’ve checked the data and I’m ready to share my findings with you.

Let’s see what the data from RS2 looks like. Below are the L1 recordings:

And this is L2:

There are rather few satellites with a good SNR (over 45) and they happen to be mostly for GPS and Galileo satellites. As Wingtra drone doesn’t track Galileo, it’s not possible to use this data for processing.

It’s okay for L2 to be generally lower than L1. Still, you can see that it stays generally at the 40 level. This is fine, even though not all of the visible satellites transmit the necessary frequency.

Overall, I can say that the data is okay. The only satellites that have cycle slips are the ones closer to the horizon. There are really few GLONASS satellites visible, too, so in the end, it may affect the solution.

The situation is a bit different for Wingtra. This is the L1 frequency:

You can see that the SNR stays around 40 and even drops to 35 sometimes. This is probably explained by the different type of antenna used on the drone.

This is L2 frequency:

It looks like the info on the GPS frequency is either indeed as low as 20 in SNR or it’s not recorded in the log at all. Either way, it’s hard to assess its quality as there’s not much data to go on.

So in the final processing, we can use only GPS and GLONASS data as there are no more recordings on the rover. Also, we need to tweak the settings during post-processing to include the L2 frequency data from the drone.

I used our Emlid Studio to post-process the data. I’ve changed the default settings a bit:

  • SNR mask to 20 to include the data from the drone
  • Elevation mask to 10 for the same reason
  • Filer type to Combined for a thorough run-through

I’ve managed to get the 98.5% of Fix. Here’s a screenshot:

The default settings are working a bit worse as they completely cut off the L2 data from Wigntra.

So what I can suggest here. You can place the base a bit more in accordance with our Placing the base guide. This will help to ensure that all of the GPS satellites have a good SNR. It may be a good idea to choose the day of the survey that allows you to see more GLONASS satellites. More GPS sats that can provide L2C can be helpful as well. You can track them via various tools, like GNSS radar.

Also, it may be possible to play with the settings in your post-processing software. I don’t have much experience in working with Wingtra products so I can’t really guide you there. But I believe their support will be able to help you out with it.

3 Likes

Thank you, Polina. I appreciate you working so diligently on this. This is the reply I got from Wingtra Support on this issue.

”The EMLID meets 2nd bullet as visible from the tech specs - it tracks L1CA and L2C → for reliable results and 100% PPK solutions, L1P and L2P should be tracked

The problem with EMLID is it does not track L1P and L2P, so there is no redundancy in frequencies in cases the signal on L1CA and L2C is bad or the number of satellites is low what can cause failure while geotagging."

It appears that the issue is in WingtraHub due to the fact that it relies heavily on L2P signals… I’ve asked them if this can be adjusted and will let you know what they reply with.

Thanks for the reply, Bryan. I should have been more clear. The “shed” that you’re seeing is actually the tent we had set up over our work station. I agree after looking at the location of the house in relation to the base that it could have been an issue. I’ll keep that in mind for future missions.

2 Likes

Hi Cole,

You’re very welcome :slight_smile: Yes, please let us know so that we know it for future reference. Each software uses its own configurations, it’s hard to match everything.

For now, I can see that you can maximize the number of satellites transmitting the L2C frequency by planning the day and time for your survey. It takes more time but can actually help with the amount of data the software will be able to post-process.

2 Likes

This topic was automatically closed 100 days after the last reply. New replies are no longer allowed.