Accurate gps module

i don’t know if it is open source and if is affordable, it seems to use a new kind of technology who make it more accurate and use less power

http://phx.corporate-ir.net/phoenix.zhtml?c=203541&p=RssLanding&cat=news&id=2302120

1 Like

Thank you very much for the hint! We will surely have a closer look at it.

Though I doubt we can use it. First of all Broadcom is traditionally bad at giving specs to smaller projects - and we are small, we will not order tens of thousands or millions of these chips. Broadcom is typically only interested in high volume customers.

The other problem may be, and here we have to start to speculate because of missing specs (see above), that this GNSS solution uses an approach which Broadcom already used in the predecessor chip BCM4752:

The problem with these chips is that they do not include a processor to calculate the solution, i.e. these are only smart receivers outputting the GPS/Galileo/Glonass/… digital signal data. This is in general a cool thing because then one could develop own cool (and free) algorithms on top of it, calculate the solution on your own etc. - there is even existing free code for some other receivers that does this. The big but is that Broadcom does not provide any information on the protocol to talk to the BCM4752 and I would suspect that the same will be true for this new chip.

So yes, good idea, but I have only very little hope that it will be useful for us if we ever want to be able to use it. Currently we are more focusing on existing, working and also very good solutions like from uBlox. These provide a simple UART interface and an NMEA data stream.

Cheers
nicole

1 Like

Oh, I just read the product brief a little further, it seems that the BCM47758 might be usable for us since it does the PVT calculation on chip. I have contacted Broadcom for tech specs, availability and prices.

Thanks for the hint!

Cheers
nicole

5 Likes

you’re welcome, i’m glad to help :blush:

that page is already gone it seems :frowning:

I did some test for GNSS Sensitivity or GPS Reception for librem 5 revisions rv1.0.6.1 and rv1.02-1.03
I got better result on rv1.03 than rv1.0.6.1. I do not know why rv1.0.6.1 which has improvement for g.n.s.s sensitivity is not working here. :sneezing_face:

Can you please share (with a photo) how you got to know the board revision?

1 Like

I going to do a retest by just GPSd tools i am tired of geoclue for sure.

I did one more test for Librem 5 revision 1.02-1.03 on GNSS - GPS Reception, Looks like that aGPS(wlan-gsm) it has priority than the main module gnss, so putting HKS OFF for modem and wlan-bt to shutdown aGPS we get better signal and true gps position, seems a bug on software side.

2 Likes

/etc/geoclue/geoclue.conf ?

I’ve disabled WiFi as a source.

2 Likes

Aaa thank you @irvinewade I did not know that it could be disabled from there. I going to “false” wlan and “gsm” to give the gnss-module a top priority. I retest.

Yes, definitely. If you are testing the GNSS then you should disable all other sources so that you know where the information is coming from. You probably want to disable WiFi as a source permanently, for privacy reasons.

1 Like

Thank you @irvinewade i falsed wlan and gsm position, and now my Librem 5 rv 1.02-1.03 which it come without gnss-sensivity-improvement the GPS Reception it working good. :smiley:

3 Likes