Cheap reference clock using a cell phone

GNU Radio
1

I found something interesting on the internet today, while browsing for
rtl_sdr related information. It seems like a guy called steve-m has fed
his
rtl_sdr dongle a clock from his mobile phone. Here is a picture:

Now mobile phone networks can be used to estimate the frequency error of
your oscillator, because base stations will have clocks that are in sync
to
a global reference. I think this is what steve-m is aiming for. Has
anyone
else attempted to do this? I’d be interested to hear about your
experiences.

It would be neat if a $8 dongle and a $20 cell phone could be combined
to
create a SDR locked to a global reference.

juha

2

Neat idea! Presumably he just tapped the (buffered?) output of the
VCTCXO on the phone which has is being actively steered by the phones
operation, and I’m guessing the breadboard is a fractional-N PLL to get
it to the frequency needed to replace the existing RTL_SDR clock.
Not exactly durable though. Plus you now have an active cellphone next
to your SDR

3

Isn’t that what Kalibrate does? http://thre.at/kalibrate/

Very Respectfully,

Dan CaJacob

– This email contains sensitive proprietary and confidential
information.
The technical data contained herein is/may be controlled under the U.S.
International Traffic in Arms Regulations (ITAR) and may not be exported
to
a Foreign Person, either in the U.S. or abroad, without the proper
authorization by the U.S. Department of State. –

4

Hi,

Neat idea! Presumably he just tapped the (buffered?) output of the VCTCXO on
the phone which has is being actively steered by the phones operation, and

Exactly.

I’m guessing the breadboard is a fractional-N PLL to get it to the frequency
needed to replace the existing RTL_SDR clock.

Yes. Although it’s actually possible to feed the 26 MHz to the E4000
only (and leave the 28.8MHz xtal for the RTL).
This way, no need for PLL.

Not exactly durable though. Plus you now have an active cellphone next to
your SDR…

The phone was running a RX only firmware in that picture.

Cheers,

Sylvain
5

Not exactly new, but nice anyway :slight_smile: I have been using the leakage of 13
MHz
and its harmonics for more than a decade to calibrate other RF stuff on
zero
beat. It is relatively accurate, but could be better. The GSM base
station
usually do not have a GPS receiver here in Germany, I guess they still
derive their clock from the backbone, with all inaccuracies such a
system
brings.

Atomic clocks (rubidium systems) are down in the two digit US$ range
when
buying a used one, they do a fine job and need nothing more but a power
supply. I installed one into my old spectrum analyzer, this solves
everything when a frequency is in question, and from that I calibrated
the
52 and 64 MHz clocks for my USRP. More than sufficient :slight_smile:

Ralph.

From: discuss-gnuradio-bounces+ralph=removed_email_address@domain.invalid
[mailto:discuss-gnuradio-bounces+ralph=removed_email_address@domain.invalid] On Behalf Of
Juha
Vierinen
Sent: Friday, 05 April, 2013 21:13
To: gnuradio mailing list
Subject: [Discuss-gnuradio] cheap reference clock using a cell phone

I found something interesting on the internet today, while browsing for
rtl_sdr related information. It seems like a guy called steve-m has fed
his
rtl_sdr dongle a clock from his mobile phone. Here is a picture:

Now mobile phone networks can be used to estimate the frequency error of
your oscillator, because base stations will have clocks that are in sync
to
a global reference. I think this is what steve-m is aiming for. Has
anyone
else attempted to do this? I’d be interested to hear about your
experiences.

It would be neat if a $8 dongle and a $20 cell phone could be combined
to
create a SDR locked to a global reference.

juha