Re: Power line communications

GNU Radio
1

On Sun, Feb 19, 2012 at 17:33, George N. [email protected] wrote:

This was a super informative post, it’s really interesting to hear about
experience of actually building PLC in practice.

Thanks.

I haven’t seen very many
studies on it, but it seems like different buildings have greatly varying
profiles. I didn’t think of things like elevator motors in office
buildings, but that’s got to introduce a lot of interference.

Cheap switching power supplies (wall warts) are issues as well, if
they are plugged in close to the power coupler.

If you ever do this again, I would love to see some of the channels over
time. I ordered some basic PLC equipment, but all I really have access to
are packets :stuck_out_tongue:

I’ll try to grab a spectrum analyzer plot from my lab. If I find time
I’ll see if I can get a channel sounding.

How do you actually get echoes in a PL? I never thought about this.
Any time there is a discontinuity in impedance on a transmission line,
there will be a reflection, including at the end of an open circuit.
Power lines have lots of open branches that signals can go down, get
reflected back from, and form a delayed sum with the original signal.
This can include, for example, traveling out to pole transformers and
back.

One idea would be to measure the impulse response between two power
line endpoints, then use that as a baseline to detect changes which
could mean damage or unauthorized access.

It would be really cool to be able to hook up the USRP to a PL and
understand various channels in different types of buildings. Totally not
conducive to my dissertation, but I find it interesting.

Happy to distract you :slight_smile:

Johnathan

2

On Mon, Feb 20, 2012 at 07:32, Johnathan C.
[email protected] wrote:

If you ever do this again, I would love to see some of the channels over
time. I ordered some basic PLC equipment, but all I really have access to
are packets :stuck_out_tongue:

I’ll try to grab a spectrum analyzer plot from my lab.

Below are two images of the RF spectrum as received through a power
coupler plugged into a wall socket in my office building.

The first is DC-500MHz:

http://gnuradio.org/data/power-line-500MHz.jpg

The green line is the thermal noise floor in the RF bandwidth used
(the measurement setup has a NF of about 9dB).

There are no quiet parts of the spectrum, and many narrowband signals.
FM radio stations are obvious around 100 MHz.

The 125MHz spectral line and 100 MHz harmonics go away when I unplug a
USRP N210 on the same power strip :slight_smile:

A zoom of the first 50 MHz:

http://gnuradio.org/data/power-line-50MHz.jpg

It’s pretty noisy all over.

Johnathan

3

On Mon, Feb 20, 2012 at 6:32 PM, Johnathan C. <
[email protected]> wrote:

Below are two images of the RF spectrum as received through a power
FM radio stations are obvious around 100 MHz.

Wow… so let me make sure I get this straight. That’s just a passive
reading of the RF spectrum on your wall socket… you are not
introducing
any form of transmission on the line? The USRP N210 was simply just
plugged in to the power strip, right?

If so, that’s unbelievably noisy. I didn’t expect it to be that bad.
The FM stations are definitely clear.

It’s very hard for me to find specs of commercial power line equipment,
but
it typically operates between 2-28MHz? Even within that short range,
given
your zoomed in snapshot, there is almost a 40dB difference in the amount
of
noise from across that band from valley to peak.

4

On Mon, Feb 20, 2012 at 15:32, Johnathan C.
[email protected] wrote:

The 125MHz spectral line and 100 MHz harmonics go away when I unplug a
USRP N210 on the same power strip :slight_smile:

To be fair, these also went away when I unplugged my logic analyzer
Mictor connector from the N210. So it was the probe cables with the
100MHz and 125MHz clock signals that were radiating, not the USRP
motherboard.

Sorry, Matt :slight_smile:

Johnathan

5

On Tue, Feb 21, 2012 at 10:18, George N. [email protected] wrote:

Wow… so let me make sure I get this straight. That’s just a passive
reading of the RF spectrum on your wall socket… you are not introducing
any form of transmission on the line? The USRP N210 was simply just plugged
in to the power strip, right?

As I mentioned in another email–the loud signals were my logic
analyzer probe radiating the 125MHz and 100MHz clocks from the USRP
N210 and being received by the power line. Once I unplugged the
cable, those spectral lines went completely away. Actually, the USRP
N210 was pretty quiet considering the case was off.

But yes, both spectrum analyzer plots were passive measurements of the
building power infrastructure coupled to 50 ohms.

If so, that’s unbelievably noisy. I didn’t expect it to be that bad. The
FM stations are definitely clear.

Yep. It’s why we researched the DSSS technique, it works very well to
increase the SNR (at the expense of bit rate.)

It’s very hard for me to find specs of commercial power line equipment, but
it typically operates between 2-28MHz? Even within that short range, given
your zoomed in snapshot, there is almost a 40dB difference in the amount of
noise from across that band from valley to peak.

Yes, 2-28 MHz is typical. Power lines have increasingly higher
attenuation above that; below that, transformer high-pass
characteristics come into play. The zoomed in plot shows the
approximately 1 MHz corner of the coupler used.

I was, however, able to load up my house wiring through the coupler
with a few watts at 146MHz and successfully contact a local ham radio
repeater :slight_smile:

Johnathan

6

On Tue, Feb 21, 2012 at 1:55 PM, Johnathan C.
[email protected] wrote:

I was, however, able to load up my house wiring through the coupler
with a few watts at 146MHz and successfully contact a local ham radio
repeater :slight_smile:

Johnathan

Ah - that sounds like one of them fancy ‘house-tennas’ used to get
around local housing association rules against ham antennas…

–
Doug G.
[email protected]

7

Yup. When you think about it, Im surprised the signals levels werent higher.

Doug: one of the benefits about living on my own
37-acre spread that I miss is the ability to erect arbitrarily large and
ugly antennas subject only to:

o budget

o wifely approval

On
Tue, 21 Feb 2012 14:01:26 -0500, Douglas G. wrote:

On Tue, Feb
21, 2012 at 1:55 PM, Johnathan C.
wrote:

I was, however,
able to load up my house wiring through the coupler with a few watts at
146MHz and successfully contact a local ham radio repeater :slight_smile:
Johnathan
Ah - that sounds like one of them fancy ‘house-tennas’ used
to get around local housing association rules against ham antennas…

Links:

8

Hi Johnathan,

The 125MHz interferece (AC) sounded familiar to me so I dug in the
archives.
The image below is received at 137MHz @ Los A. and they have the
AC lines dangling in the air like a spiderweb. When it gets rainy we see
those 125MHz in our images.
The antenna was ~20m avay from the AC lines

The format is APT (Automatic Picture Transmission) from polar orbiting
NOAA
satellites.
One pixel line is 2080px wide and transmitted every 0.5 second.

So, from the image
http://www.poes-weather.com/~jerry/apt/logs/daily/images/img_001.jpg it
is
easy to calculate the interference

image width = 2080px = 0.5 seconds
interference width ~34px = X seconds

Patrik

----- Original Message -----
From: “Johnathan C.” [email protected]
To: “George N.” [email protected]
Cc: “GNURadio D.ion List” [email protected]
Sent: Tuesday, February 21, 2012 20:55
Subject: Re: [Discuss-gnuradio] Power line communications

On Tue, Feb 21, 2012 at 10:18, George N. [email protected] wrote:

Wow… so let me make sure I get this straight. That’s just a passive
reading of the RF spectrum on your wall socket… you are not introducing
any form of transmission on the line? The USRP N210 was simply just
plugged
in to the power strip, right?

As I mentioned in another email–the loud signals were my logic
analyzer probe radiating the 125MHz and 100MHz clocks from the USRP
N210 and being received by the power line. Once I unplugged the
cable, those spectral lines went completely away. Actually, the USRP
N210 was pretty quiet considering the case was off.

But yes, both spectrum analyzer plots were passive measurements of the
building power infrastructure coupled to 50 ohms.

If so, that’s unbelievably noisy. I didn’t expect it to be that bad. The
FM stations are definitely clear.

Yep. It’s why we researched the DSSS technique, it works very well to
increase the SNR (at the expense of bit rate.)

It’s very hard for me to find specs of commercial power line equipment,
but
it typically operates between 2-28MHz? Even within that short range, given
your zoomed in snapshot, there is almost a 40dB difference in the amount
of
noise from across that band from valley to peak.

Yes, 2-28 MHz is typical. Power lines have increasingly higher
attenuation above that; below that, transformer high-pass
characteristics come into play. The zoomed in plot shows the
approximately 1 MHz corner of the coupler used.

I was, however, able to load up my house wiring through the coupler
with a few watts at 146MHz and successfully contact a local ham radio
repeater :slight_smile:

Johnathan

Discuss-gnuradio mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio