k-Wave

1. Tieyuan
   

   Member
   Joined: May '12
   Posts: 5

   Dear Dr.Treeby and Dr.Cox,

   I have a question about your modeling of absorption. I noticed that in your modeling paper (2010) you explained that you calculated your attenuation factor and dispersion by simulating monopolar plane wave. In k-wave package, source code is 1D for modeling of absorption. It works in 1D for me. I am confused that I have troubles to stabilize my solution (alpha and phase velocity with frequency) when I want to use point source in 2D. Or say solution is deviating theoretical solution much. I am working on seismic wave (frequency limited in tens of Hz range, not greater than 500Hz)

   Could you give me some suggestions on that?

   Thank you for your release such a great software for research.

   Cheers,
   Tieyuan

   Posted 12 years ago #

2. bencox
   

   Developer
   Joined: Mar '10
   Posts: 292

   Hi Tieyuan,

   The reason we test the absorption in 1D is because the wave is not attenuated due to geometric spreading, only due to absorption, so it is easy to check that the absorption is being modelled correctly. It should still work in 2D and 3D. What values are you using for the absorption medium.alpha_coeff and medium.alpha_power?

   Kind regards,

   Ben

   Posted 12 years ago #

3. Tieyuan
   

   Member
   Joined: May '12
   Posts: 5

   Hi Ben,

   Thanks for quick reply. It makes sense. So I need to compensate this amplitude loss due to geometric spreading (factor for 2D: 1/sqrt(r); 3D: 1/r).

   medium.alpha_coeff=0.25
   medium.alpha_power=1.1

   I modified the original 1D modeling absorbing code to 2D. r should be centimeter?
   I choose factor 1/sqrt(r_cm*30) to match with theoretical data very well not 1/sqrt(r).

   I am unclear why it is.

   Best regards,
   Tieyuan

   Posted 12 years ago #

4. Tieyuan
   

   Member
   Joined: May '12
   Posts: 5

   Hi Ben,

   I figure out what happen in amplitude compensation. I should consider sqrt(r1/r2) factor for two sensors.

   Thank you,
   Tieyuan

   Posted 12 years ago #

5. bencox
   

   Developer
   Joined: Mar '10
   Posts: 292

   Ok, good. Glad you sorted it out!

   Ben

   Posted 12 years ago #

RSS feed for this topic

Reply

You must log in to post.