I would like to describe a radar signal processing issue my research team has been struggling through the last few weeks. We are attempting to obtain a high range profile of a simple scatterer (Sphere in the image below). We use a bistatic transmitter receiver that steps across an 8 GHz bandwidth from 71-79 GHz and our local oscillator operates at 500 kHz. Let me try to walk through as much of the program as possible. Apologies if I make mistakes, I am a relative novice when it comes to this.
The signal is resolved into its I and Q components that we digitize using the DAQ stream VI. The components are then placed into an I+iQ complex form. From there, a hanning window is applied before a Complex FFT across the signal. The results of the hanning window and the FFT adjusted for the frequency axis are attached.
I then take only one half of the results of the FFT and from what I understand, the peak of the FFT should be at 500 kHz (the difference frequency) which it does appear is the case. I then subset the array to focus primarily around the 500 kHz difference frequency before using a max/min function to grab the maximum of the FFT results. This forms complex numbers that are located at the difference frequency across the 8GHz bandwidth. The amplitudes of which are also attached.
My professor proclaims that performing an FFT or IFFT across that Bandwidth information is supposed to be a range profile (with a peak associated at the location of the scatterer in the scenery). However when I perform this FFT, I obtain a very noisy result with no evident range profile.
By the way, we perform 512 frequency steps across the 8 GHz bandwidth and take 125000 samples from the DAQ board at a 625000 sample rate.
There are many pieces that I am missing and any help/resources would be appreciated. I have also uploaded the VI if that helps.