Hi,
I am trying to simulate k-wave with a depth-20 cm, sampling frequency-40 Mhz, c0-1540 m/s and 10000 samples per scan line. While simulating it shows some errors. What could be the reason?
Hi,
I am trying to simulate k-wave with a depth-20 cm, sampling frequency-40 Mhz, c0-1540 m/s and 10000 samples per scan line. While simulating it shows some errors. What could be the reason?
Hi Fathima,
It sounds as though you are attempting to create a very large grid. For a maximum supported frequency of say, 10 MHz, the grid spacing must be dx = c_0/(2*f_max) = 75 um. In this case, 20 cm corresponds to 2666 grid points. A 3D simulation of this size would need more than 1 TB of memory.
I'm afraid you'll have to be more specific if you need more help than this. There is a lot of information in the k-Wave Manual that you might find helpful.
Brad.
Thank you very much for your reply.
Yes. I found errors while creating such a large grid. I was trying to create an eight channel ultrasound system with these specifications (c=1540 m/s,Fs=40 MHz, prf=4 KHz, signal_freq=2.5 MHz, depth= 20 cm, 40 scan lines, 100 frame/s ).Is it possible to use these specifications for k-wave simulation? What is the maximum number that we can assign for Nx ? Is there any problem if i reduce CFL value?
Hi Fathima,
There is no maximum value for Nx, per se. The largest value you can use in practice will depend on how much memory you have in your computer, and the precise setup of your simulation. If you're using the C++ code, there is a formula to estimate the memory usage in Sec. 4.9 of the k-Wave Manual.
As to how your specifications given above map to the k-Wave simulation parameters, it's probably worthwhile going through some of the examples in the toolbox and the descriptions given in the manual. For example, the value you choose for the CFL will affect both the stability and the accuracy of the simulation.
Good luck with your simulations,
Brad.
Thank you so much for your information.
You must log in to post.