About the calculated position from gps measurement tools matlab source code

120 views
Skip to first unread message

mohamed fathi

unread,
Aug 25, 2022, 1:42:23 AM8/25/22
to gpstest...@googlegroups.com
Why always the calculated position (lat long alt )from Weighted least square method by the matlab source  in gpsPvt.allLlaDegDegM different from the values from the txt file from GNSSLOGGER or GPSTEST 
From Fix message whether the provider is Fused location or GPS ?

mohamed fathi

unread,
Aug 26, 2022, 4:57:55 AM8/26/22
to GPSTest
in other words, anyone works with the GPS measurement tools Matlab source code

Sean Barbeau

unread,
Aug 26, 2022, 9:59:50 AM8/26/22
to mohamed fathi, GPSTest
Each of those positions are calculated via different methods/sources:
* WLS is calculated using weighted least squares algorithm implemented in GNSSLogger or GNSS Analysis Tool
* Fused location comes from Google's fused location provider implemented in Google Play services
* GPS location comes from the GNSS chipset on the phone

Sean

--
You received this message because you are subscribed to the Google Groups "GPSTest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gpstest_andro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/73afa21a-3420-442a-83e4-7a8ba0b2acb3n%40googlegroups.com.

mohamed fathi

unread,
Aug 26, 2022, 10:52:09 AM8/26/22
to Sean Barbeau, GPSTest
Dr sean thanks for your reply
But the wls is not accurate enough to compare with others  difference  is large sometimes reach hundred of meters 
Is it because the pseudorange and pseudorange rate calculated in Gnss logger without the ionosphere and troposphere delay ? 
And if yes or no
How to reduce the  difference between the calculated from wls and other from Gps and Fused location because the Gps and Fused location already have small difference between each other 


From: Sean Barbeau <sjba...@gmail.com>
Sent: Friday, August 26, 2022 3:59:37 PM
To: mohamed fathi <mohammedf...@gmail.com>
Cc: GPSTest <gpstest...@googlegroups.com>
Subject: Re: About the calculated position from gps measurement tools MATLAB source code
 

Sean Barbeau

unread,
Aug 26, 2022, 6:38:53 PM8/26/22
to mohamed fathi, GPSTest
The fused provider and the GPS from the chipset are both post processed in proprietary ways, evidently with better accuracy than WLS alone. Here's the latest open source implementation of WLS for an older version of GnssLogger:

mohamed fathi

unread,
Aug 26, 2022, 7:03:50 PM8/26/22
to Sean Barbeau, GPSTest
Thanks Dr sean for your help 🙏
From: Sean Barbeau <sjba...@gmail.com>
Sent: Saturday, August 27, 2022 12:38:39 AM
Reply all
Reply to author
Forward
0 new messages