I wrote the FDTD simulation code using MATLAB. The center frequency of the electromagnetic wave was 2.38GHz, the bandwidth was 50MHz, and the pulse width was 50ns. I set up two receivers for echo reception. However, only when the spatial step size was set to 0.0005, the arrival delay calculated by the cross-correlation method based on the echo data was just close to the theoretical delay. Is this reasonable? I know that in the FDTD method, the higher the frequency, the higher the requirement for the refinement of the grid. But in my Settings, it should be considered an extremely detailed grid. Is this because there is an error in my code writing or is there another reason?
Hello. You should be able to see how your pulse moves from transmitter to receiver, if your space and time step sizes are set properly. Is this a square pulse? Is your CFL factor = 1? Then you should have deltax=c * deltat. Set CFL=1 is my suggestion. Also, try to do 1D simulation first.
I used left-hand circularly polarized LFM signals and applied windowing. I visualized the echoes from the two receivers, and visually, there was a certain delay and amplitude difference between them. Moreover, their waveforms were consistent with the source signal I set. After I later reduced the center frequency, this error decreased significantly.
- Badges
- Science topic