that's a hint for the person contemplating homebrew ground-pentrating radar. the hint being: the path loss from transmitter to receiver is quite high. the transmit power has to be large in order for the return echo to be detectable. similarities to E-M-E (aka 'moonbounce') communications are present.About Ground Penetrating RADAR. Requires a high voltage (Deadly 1000 Volts) 400 MHz Pulsed Transmitter driving a bow-tie antenna inside a large box facing the ground
Discuss KerberosSDR - 4x Coherent RTL-SDR
Assuming "1000V" is a peak voltage (plus and minus), that is 354 Vrms and into a 50 ohm load would be 2.5 kW.
According to http://www.radartutorial.eu/02.basics/G ... ar.en.html, a page I found to be correct on PSR and SSR RADAR so far, Ground Probing RADAR that use a short pulse with peak voltages is between 20 to 200 V. I have not seen details of antenna designs and how they are fed by the pulsegenerator, but I do not assume that they are anywhere near to 50 Ohm due to the high bandwidth and very low frequencies involved.
There are also Wall Probling RADAR operating above 1 GHz using UWB signals, which have much lower EIRP.
Problem for all types of such wideband sensors is as always Electro Magnetic Compatibility and Radio Frequency Interference to existing primary services.