toggle quoted messageShow quoted text
Sounds like a very interesting research project and I imagine a
LOT of people could learn from your findings!
One thought though: you mentioned this research was going to be
"EMF noise that these devices email through the air" but I've
always been curious if they are also radiating noise via the AC
power lines as well. Much like the electric or magnetic
interference, this "conducted" interference on the AC lines could
also help you find and detect things potentially even farther
away. I have to believe there can be a way to couple an SDR to
the power line, cancel out the 50/60 cycle power, and view things
in the frequency and/or time domains.
Anyway.. good luck with the research!
On 07/09/2020 11:01 AM, Nathan R wrote:
I am a graduate student currently exploring a new research project
on activity recognition within a household setting by sensing when
a user turns on or interacts with a particular appliance or
electric device. The intent behind the project is to wirelessly
detect device use by passively measuring the EMF noise that these
devices emit through the air during regular use. Through past
literature searches, we have identified the very low frequency
range of 0-2 MHz as the target area where many of these devices
are emitting noise. I did not have a lot of knowledge on antenna
and radio design coming into this project, and only limited
experience with SDRs, but have made some headway in the past few
weeks. My setup thus far has consisted of using an RTL-SDR V.3 in
direct sampling mode paired with a variety of pre-made antenna
designs, of which I have achieved the best results with the
MegaLoop 30+ MLA. I have also just ordered an Airspy HF+ Discovery
and the Youloop antenna which I hope will help me to achieve even
better SNR in the HF range.
From my readings and my background knowledge in electromagnetics,
I have come to understand that these loop antennas perform better
in the very low frequency ranges, but also couple to the magnetic
component of radio waves. This appears to match with my
experimental results where although I have been able to detect a
variety of appliances and devices, such as TVs, washing machines,
and light bulbs, it is only devices that generate strong magnetic
fields, like power drills with DC motors, that I am able to detect
from more than a foot away.
Especially once we get the new Airspy SDR, I am fairly confident
in all of the areas of our pipeline except the antenna design.
Thus my question is if there is anyone who, either purposely or
accidentally, has had experience with detecting appliances from
longer distances before, suggestions on how we could improve our
antenna design for these lower frequencies, and thoughts on
whether detecting these waves through the air at a house or room
scale is even realistically possible at all.