k-Wave

1. benji
   

   Member
   Joined: Dec '15
   Posts: 14

   Hi everyone,
   I am trying to simulate propagation of an acoustic wave in a humen body phantom, emerging from and measure by piezoelectric devices (transmitter and reciever).
   I dont know much about such PZT devices, but i assume:
   1. The transimitting device must have some kind of hard mounting on it's "back" surface, preventing backword radiation and widening its frequency response.
   2. The recieving device must have a matced layer on the recieving surface, perfectly absorbing the recieved signal.

   how can these charcteristics be implemented in the simulation? regarding the reciever - is it possible and would it be correct to define a PML on the surface of the reciever sensor?
   regarding the reciever - would it be correct to define a layer of very high impedance behind it?

   Thank u very much,
   Benji

   Posted 8 years ago #

2. Bradley Treeby
   

   Developer
   Joined: Mar '10
   Posts: 1,640

   Hi benji,

   k-Wave is not the best tool to simulate the electro-mechanical response of a PZT device (a finite element model would be much better suited to that). It is possible to define a layer of high impedance behind the transducer surface, however, the normal way in k-Wave to account for the frequency response of a transmitting device is to use the appropriate time domain driving signal (this could be measured, or predicted based on transducer centre frequency and bandwidth).

   Similarly for modelling a receiving device. Within k-Wave the sensor positions are just positions in the domain where the pressure field is sampled at each time step (they don't affect the field in anyway). Probably the most straightforward approach would be to position the receiver near the edge of the grid, where there is already a PML implemented.

   Hope that helps,

   Brad.

   Posted 8 years ago #

3. benji
   

   Member
   Joined: Dec '15
   Posts: 14

   Hi Brad,
   thank u very much for your response. I am realy greatful for all that amazing work u guyz are doing!
   regarding the transmitter - I was not clear enough. i dont intend simulating mechanical response or frequancy response of the PZT. What i am interested in is to eliminate the backword projection. would a high impedance layer behind the tranmitter do that? would it have other effects i am not aware of?

   regarding the reciever - I will be more specific. I have a sensor on the skin surface. i have great interest in simulating reflection from the skin surface anywhere otherwise the surface underlying the sensor (right now i am sensing also reflected waves from the surface under the sensor). positioning the skin surface in the PML, without entring also neighboring skin surface into the PML, is possible only in very specific sensor locations and size.
   could there be any other more robust way?

   Thank u again
   benji

   Posted 8 years ago #

4. Bradley Treeby
   

   Developer
   Joined: Mar '10
   Posts: 1,640

   Hi Benji,

   For the transmitter, the easiest way is to position this close to one edge of the domain. The backward travelling wave will then be absorbed by the existing PML. If you can't do this you could try adding a reflecting layer. However, I don't have any experience of doing this. Keep in mind that using a material with a high sound speed will affect the stability and accuracy of the model (see the manual for more details on stability and the k-space correction term).

   I'm not sure I understand your second question. The receiver won't introduce any reflections, thus the only reflections will come from an impedance mismatch. If you want to avoid reflection from the skin surface in your simulation, you could match the sound speed and density to the surrounding medium.

   Brad.

   Posted 8 years ago #

RSS feed for this topic

Reply

You must log in to post.