I have been following some passive radar tutorials and they all mention a reference channel (for a signal directly from a transmitter) and an observation channel (for a signal reflected off of some object).

I have a setup that includes a few dipoles. The data collected are separated into channels, such as CHANNEL A, CHANNEL B, CHANNEL Z...

I'm totally new to passive radar (and all things radio, for that matter), so these might be a silly/noobie questions:

  • How can I tell which of my channels represents the reference vs observation channel?
  • How do I specify that a particular channel is the reference channel rather than the observation channel? Are they tuned to different frequencies?
  • Can I do this by looking at the data stream?
  • What information do I need to know in order to determine which channel is the observation vs reference?



1 Answer 1


I have yet to play with passive radar myself, but my impression is that you will need to deliberately set up one of your antennas so that it receives only the "signal directly from a transmitter" and another of your antennas so that it receives "a signal reflected off of some object".

Once you've done that, it should be clear which channel you have plugged which antenna into, no?

My guess here is somewhat corroborated by this diagram:

Diagram of "non-cooperative illuminator" and "target" and two passive radar receivers — one "reference channel" aimed on a direct path to the illuminator and the other "surveillance channel" aimed away on the target

This particular diagram was found at https://www.researchgate.net/figure/A-passive-radar-system-with-a-reference-channel-and-surveillance-channel_fig1_283186198 from a published paper "Joint Delay and Doppler Estimation for Passive Sensing with Direct-Path Interference" by Xin Zhang and co-authors (DOI:10.1109/TSP.2015.2488584) but I'm seeing similar other examples in an image search for "passive radar reference channel".

Also corroborated by https://dopplerfish.com/passive-radar-hardware/ which says:

The observation channel antenna is a 5-element Yagi-Uda antenna […] The gain and directivity of the reference channel antenna is less critical for the radar’s performance, so we used a simple half-wave dipole.

This is somewhat opposite from what I first imagined but makes some sense especially since in this context: "The key was to select a location that provided an uninterrupted line of sight in the direction of the targets but provided some shielding from the direct path signal emitted by the [close and high-powered] radio tower."

Also corroborated by https://www.rtl-sdr.com/krakensdr-passive-radar-demonstration/ which uses two separate Yagi antennas, one pointed at the reference and one pointed in the direction of surveillance. (See https://github.com/krakenrf/krakensdr_docs/wiki/08.-Passive-Radar for more notes/tips, including "the reference signal should not be received directly by the surveillance antenna as much as possible".)

Based on https://en.wikipedia.org/w/index.php?title=Passive_radar&oldid=1112632179#Digital_beamforming and many of the pictures there of commercial systems, it looks like it's also fairly common (at least when cost is less of a concern) to use arrays of many separately-sampled antennas, which can be directionally focused electronically and/or digitally (rather than physically like a Yagi or dish).

  • $\begingroup$ Thanks! I also thought that it might be possible to set up and receive a specific reference/observation signal, but how is it done? Tune the reference receiver to the exact freq of the transmitter and have the observation receiver slightly off freq? For the "cross-talk", we can apply adaptive filtering. $\endgroup$
    – pookie
    Commented Oct 4, 2022 at 21:41
  • $\begingroup$ @pookie I don't think "channel" means a frequency in this context, I think it just means a distinct input signal, separate from another input signal. See e.g. dopplerfish.com/passive-radar. You might be right that cross-talk is pretty much expected and dealt with in the processing. But regardless you need a way to tell which energy comes from the "original" (illuminator) and which came by way of reflection off the target(s) — and it looks like that's usually directional in nature, using a dish/yagi for at least one of the inputs in simple, or array beamforming in fancier setups. $\endgroup$ Commented Oct 4, 2022 at 23:01
  • $\begingroup$ Updated my answer to get rid of my note about cross-talk and add more references supporting that the "how to separate the reference/observation channels" part is typically done by using directional antennas. $\endgroup$ Commented Oct 4, 2022 at 23:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .