k-Wave

1. abode2100
   

   Member
   Joined: Jul '19
   Posts: 1

   Hi,

   I have been trying to add a vacuum layer in my simulation to block the direct arrival signals from the receivers. I've successfully blocked them when I set the medium.density parameter to a small values (e.g., one) for this vacuum layer. However, the signals always get reflected, which is not what I want. I want to have an absorptive layer that completely absorbs the signal and reflects nothing (does not matter which material this is). Is this doable in K-wave? Note that I've been tuning medium.alpha_power and medium.alpha_coeff to have high attenuation coefficient in this layer but didn't get what I want.

   Any suggestion for such a layer with the corresponding parameters will be really helpful.

   Thanks in advance.

   Posted 5 years ago #

2. vich
   

   Member
   Joined: Jul '19
   Posts: 19

   As far as reflection goes, I would suspect it is only to be expected. For plane waves at normal incidence, the pressure reflection coefficient R is given by:

   
R = P_reflect / P_incid = (r_2 - r_1) / (r_2 + r_1)


   where the P are pressures, the lower-case r are characteristic acoustic impedances of the involved media, the direction of wave transmission is from medium 1 to 2, and:

   
r_n = rho_n * c_n


   with rho,c being the equilibrium density and speed of sound.

   Thus, if you set rho_2 too low in your simulations, R approaches -1; the reflected waves are strong and 180° out of phase with the incident waves.

   As for the attenuating layer, I guess you would need it to be many voxels thick in order to prevent the same reflection problem that occurs when using a PML with very high attenuations per voxel. And then your attenuation would still not be strictly complete, although probably sufficient.

   Maybe you could use 2 Np/voxel (the default for the PML) as a starting point (after conversion to dB/MHz^y/cm)?

   Posted 5 years ago #

RSS feed for this topic

Reply

You must log in to post.