A Screenshot based Meteor Scatter Detector for HDSDR

Over on our forums Andy (M0CYP) has posted about his new meteor scatter detection program which works with HDSDR and any supported SDR like an RTL-SDR. It works in an interesting way, as instead of analyzing sound files for blips of meteor scatter activity it analyzes screenshots of the HDSDR waterfall. The software automatically grabs the screenshots and determines if a signal is present on any given frequency. You can set a preconfigured detection frequency for a far away transmitter, and if the waterfall shows a reflection it will record that as a meteor.

Meteor scatter works by receiving a distant but powerful transmitter via reflections off the trails of ionized air that meteors leave behind when they enter the atmosphere. Normally the transmitter would be too far away to receive, but if its able to bounce off the ionized trail in the sky it can reach far over the horizon to your receiver. Typically powerful broadcast FM radio stations, analog TV, and radar signals at around 140 MHz are used. Some amateur radio enthusiasts also use this phenomena as a long range VHF communications tool with their own transmitted signals. See the website www.livemeteors.com for a livestream of a permanently set up RTL-SDR meteor detector (although that site does not use Andy’s software).

Andy writes that his meteor scatter detection software is still in beta so there might be some bugs. You can write feedback on the forum post, in the comments here, or contact Andy directly via the link on his website.

Andy's screenshot based meteor detection software
Andy’s screenshot based meteor detection software
Notify of

Inline Feedbacks
View all comments

Thanks that’s food for thought.


Jim Menefee - W4JWM

One other item, even the size of the amplitude and time length of the signal. You would need to consult an astrophysicist for that information.

Jim Menefee - W4JWM

Excellent project. I did it on 50 MHz. Train tracking project via meteor scatter for CSX in 2000.,Really long range . If you could find a method of scanning 5 maybe 10 Frez. Near real time, is the trick, show the ‘program’ source transmitter and print, it woul
d add direction as well. Using the same theory, sporadic E could also be log and the direction based on the amount of time the signals heard. Together they make a very strong & power tool .


I see 2 problems with using wideband signals for meteor detection:
– When looking at wider band, the total noise level is higher. This reduces sensitivity. High power of FM transmitters mitigates this to some extent.
– Impossible to filter out reflections from planes. Narrowband signals allow doing this by looking at the Doppler signature of the reflected signal


this fucking interface…