Doppler and FMCW Radar experiments at 10 GHz

Many many years ago I was one of the pioneers in the 10 GHz Ham band .

Among many other I still keep some old materials like horn and Gunnplexers ( described here ):

https://www.qsl.net/n9zia/ma87127.html

It was years that I had the desire to test them in two different Radar configurations in medium range :

- CW Doppler ( only detects objects in movement )
- CWFM ( also detects static objects )

The CW Doppler theory ( used for instance in burgler alarm sensors ) is very easy , the frequency of the moving object is different from the one of the transmitter thet is also used as LO in RX .
The IF will be the doppler shift due to the  movement of the object .

The CWFM radar theory is also simple :

The carrier of the transmitter ( and again  LO in RX ) is sweeped linearly during the time .
In this case there is a shift of frequency when the signal reflected from even a static an object reach the receiver because there is a delay due to the path .
The LO frequency will be different from the one returning back reflected from the object .
The difference will be linearly proportional to the distance .

https://www.radartutorial.eu/02.basics/Frequency%20Modulated%20Continuous%20Wave%20Radar.en.html

Today it was time in some of my last remaining neurons to make some tests .

I have made two video-audio with the sound of moving and static objects with this portable battery  setup on my roof :

CW Doppler RADAR : signals only from moving objects

CWFM RADAR : signals from statics and moving objects 

Some remarks for the CWFM RADAR:

-There is a need of Ying-Yang  to find the proper Sweep Modulation Frequency and Output Frequency Sweep Range .
- There are two parameters that are not ideal and can generate problems that could mask the signals to receive .
- Theoretically it is  obvious that is convenient to have both parameters quite high , but in practice :
With high Output Frequency Sweep Range every not flat amplitude response of the circuits in the sweeped band will generate signals at the Sweep Modulation Frequency ( that is not sinusoidal , so have also harmonics).
In this case it is convenient to try to keep both this parameters low .
The problem is that in this case , near object will generate very low frequency signals ( difficult to be eard ) .
I could ear with the earphones the signals reflected from the two near houses , but I am not quite sure that could be heard in the video , being cut from the smartphones .
- Non linearity in the frequency sweep will generate not pure beat tones ( wobbled , like you can ear in the tests ) .
Also in this case it is convenient to keep both parameters relatively low .
Anyway for this simple setup I am happy

I have used together with the Gunnplexer only :

- A monolitic 10V voltage regulator
- A TL084 as AF preamplifier after the Gunnplexer mixer
- A LM386 to drive the earphones
- A XR2206 to generate the sweeeping signal for the varactor of the gunnplexer 

The Output Frequency Sweep Range was measured on air to be +/- 14 MHz .