Keep An Eye On The Neighborhood With This Passive Radar

If your neighborhood is anything like ours, walking across the street is like taking your life in your own hands. Drivers are increasingly unconcerned by such trivialities as speed limits or staying under control, and anything goes when they need to connect Point A to Point B in the least amount of time possible. Monitoring traffic with this passive radar will not do a thing to slow drivers down, but it’s a pretty cool hack that will at least yield some insights into traffic patterns.

The principle behind active radar – the kind police use to catch speeders in every neighborhood but yours – is simple: send a microwave signal towards a moving object, measure the frequency shift in the reflected signal, and do a little math to calculate the relative velocity. A passive radar like the one described in the article linked above is quite different. Rather than painting a target with an RF signal, it relies on signals from other transmitters, such as terrestrial TV or radio outlets in the area. Two different receivers are used, both with directional antennas. One points to the area to be monitored, while the other points directly to the transmitter. By comparing signals reflected off moving objects received by the former against the reference signal from the latter, information about the distance and velocity of objects in the target area can be obtained.

The RTL-SDR test used a pair of cheap Yagi antennas for a nearby DVB-T channel to feed their KerberosSDR four-channel coherent SDR, a device we last looked at when it was still in beta. Essentially four SDR dongles on a common board, it’s available now for $149. Using it to build a passive radar might not save the neighborhood, but it could be a lot of fun to try.

25 thoughts on “Keep An Eye On The Neighborhood With This Passive Radar

  1. It is like a 1D radar.
    You have a number that can be considered an approximation of the distance from the delay between the two signals.
    You have a number that can be considered an approximation of the speed (which would change based on relative angles) based on the doppler shift between the directly received reference signal and the reflected signal.

    This begs the question if you used all four ports on the KerberosSDR, one for the reference and three extremely well spaced antennas (using equal lengths of cable, the object to be observed being at least ten times the size of the wavelength lightening up the object and the spacing of the antennas at least ten times the size of the object to be observed) at known relative positions in 3D space, would you have a crude radar by solving the position location of the intersecting hyperboloids.

      1. I’d wager most of that capability has to do with the spectrum used. Stealth isn’t an invisibility cloak, It lowers your apparent cross-section to a given frequency band. You can (and they do) make a missile (or target drone) look like an B-52 by reversing the same principles that make a bomber look like a party balloon.

    1. I wonder if you pointed the four antennas skyward and tuned them to 1090 MHz could you use ADS-B Mode S Airborne Positions transponder squawks ( ) which contain Altitude, Latitude, Longitude and time to calculate the positions of the antennas. I suppose that it would be a kind of poor/smart persons GPS. One squawk would not provide an accurate position, but multiple squawks from multiple aircraft in multiple locations over time, could possibly average out to extremely accurate antenna locations.

      Probably not useful for a handful of antennas, that you could walk about with a handheld GPS (phone) and tag them individually, but if you had a very large array.

      1. Actually that message only has 1 bit allocated for the time, there must be another message type that contains the time, because 1 bit for time has me scratching my head. Time to dig up and start reading a very long and boring document.

    1. Given the response of some people to being caught speeding I don’t think it would. For them the blame lies with the police, the speed camera, etc. not with their own driving.

    2. Particularly on freeways without many lanes, sometimes it can be dangerous to go as slow as the speed limit. “Add 5-10mph when it’s >55mph” is the American way.

      …Sorry, that’s “add ~10kph when it’s >=100kph”? When you want to take things slow without annoying people, you can just follow a big truck.

      1. Not too far from here at the border of the next US state south of us the speed limit transitions from 65 mph to 55 mph. At the same place, the average traffic speed jumps from 75 mph to 85 mph for reasons that are beyond me. Anyone doing the speed limit there is basically a rolling road block causing all kinds of near misses and traffic snarles.

    3. No, you merely need to increase the radar power output as well as increase the gain directionally targeted at the location of the excess of speed. Just fry whatever is going to fast… call it the ping of death.

  2. I was thinking more along the lines of using a Hot Wheels Radar Gun strategically placed along the street and attached to a small solar panel. A 555 would trigger the radar at intervals.
    It could be enough to slow down traffic in the neighborhood as radar detectors go off.
    Relocate the radar gun after a week or so to keep the repeat offenders from thinking, “Oh, the detector lights up at this location all the time, so I can just ignore it.”

      1. Bill gets it.

        Fun story. My wife is a cop, we have a patrol car in the driveway most of the time. She works a few towns over, so she can’t do anything about the traffic in our neighborhood. People have become accustomed to the patrol car that can’t pull them over and the near permanent Waze report, and they speed by our house and run the stop sign in front of our house. I offered the local cops the use of my driveway for a day. The footage from my driveway camera is priceless.

    1. And you get reported to the FCC for deliberately causing radio interference to legal radio receivers and a nice fat fine along with it… Let the real police do their jobs. Not that many people even run radar detectors to be worth your trouble.

      But yeah, most people that are serious about radar detectors know that X-band is pretty much obsolete as enforcement radar and therefore have it locked out anyway rather than waste precious dwell time on it rather than scanning the much more wide and dangerous Ka band. And Escort has been making nice line of radar detectors with crowdsourcing built in for years now, which is effectively automatically tying their detectors to Waze and giving a heads up without even needing a sniff of RF… Cat and mouse.

      Or just make this fake sign which uses an Arduino to make up random numbers around 30-40 mph and just throws up any number when the Selenium photo cell detects shadow or headlights from an approaching car.It doesn’t matter that its telling a complete lie. They still slow down not knowing the truth. Looking for a cop to fly out after them.

      Or hook up a Hot Wheels Radar Gun to a solar cell or external power, aim a TV camera at the rear display, have a illumination light for the LCD, and project it on a portable TV screen also powered by the external supply. The HW Gun button would be jammed into permanent trigger mode with electrical tape. However, someone may mailbox vandalize it or steal it.

  3. Cool to read about, thanks for sharing! Amazing what can be done with passive systems and ambient conditions or environmental installation emission(s).

    I was surprised what I could determine using Spectrum Lab software’s waterfall and a single microphone with traffic in the sonic range passively, though didn’t perform any calibrations for the range of frequencies and conditions effect on the graph/diagram. Definitely looks cool to see the Doppler effect and direction correlation.

    I’ve since picked up two USB microphones to experiment with regarding more direction info to process… though haven’t started working on yet.

  4. I am interested in Kraken SDR primarily because of the claim of being able to measure speed of moving vehicles. I am concerned because I have never seen anyone actually do it especially for measuring vehicle traffic in on a residential street. I have implemented a speedcamera setup that essentially measure time for vehicles to pass between two regions of interest in a camera’s field of view. I was hoping that I could see something with Kracken SDR that would help me verify the speed I am measuring with the speedcamera. Isn’t it true that doppler shifts are most easily measured using high frequency signals (i.e., x-band)? I am concern because TV and Radio signals are generally in the MHz region. This video show how the use a 24 GHz source to measure the doppler shift. Thoughts?

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.