http://www.k-wave.org/forum/topic/modelling-directivity-of-a-linear-transducer-array Support for the k-Wave MATLAB toolbox en-US Wed, 13 May 2026 01:02:25 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/modelling-directivity-of-a-linear-transducer-array#post-7102 Wed, 23 Oct 2019 00:01:01 +0000 bencox 7102@http://www.k-wave.org/forum/ <p>Hi Francis, </p> <p>If you have a single point detector (a single 1 in a binary mask) then it will be equally sensitive to waves arriving at it (passing over it) from all directions. You can endow it with a directional sensitivity by defining the direction in which it will be most sensitive to (by giving a single value to <code>sensor.directivity_angle</code>. The directional pattern (the sensitivity at the other angles) can be either set to <code>sensor.directivity_pattern = &#39;gradient&#39;</code>, which gives a cosine dependence or <code>sensor.directivity_pattern = &#39;pressure&#39;</code> which gives the kind of dependence you would get from spatial averaging (hence this also requires <code>sensor.directivity_size</code> to be set, as that is the averaging area that would give the same directional response). Of course, you could just do the spatial averaging by hand yourself but I wouldn't do that by setting the angle to 0 but by not setting any directivity at all, ie. average the measurements from a set of omnidirectional point detectors.</p> <p>You might also want to look at the Transducer class. See the 'Defining An Ultrasound Transducer' Example.</p> <p>Hope that helps!</p> <p>Best wishes,<br /> Ben </p> http://www.k-wave.org/forum/topic/modelling-directivity-of-a-linear-transducer-array#post-7068 Mon, 07 Oct 2019 13:05:20 +0000 K_J_Francis 7068@http://www.k-wave.org/forum/ <p>Hi</p> <p>What is the best way to model the directivity of a linear transducer array using k-Wave? We are trying to simulate a 128 element, 7MHz transducer. The goal is to mimic directivity as close as possible to the measured one. As per k-Wave examples, there are two ways (i) using spatial averaging and (ii) by modifying sensor.directivity_angle. </p> <p>We are performing 2D simulations and to be more realistic, we considered the second option of assigning directivity angle to each sensor point. How should we assign the sensor.directivity_angle matrix? We thought of the following implementations and we would like to know how to make it more realistic?</p> <p>(i) Assume that there are 5 sensor points in a transducer element. Should we assign [-θ2 -θ1 0 θ1 θ2], corresponding to 5 sensor points of an element and repeat it for all 128 elements to obtain sensor.directivity_angle? θi are equispaced angles within the opening angle of the single element.</p> <p>(ii) Consider the whole transducer array and assigning equispaced angles as in the k-Wave example.<br /> dir_angles = (-1:1/15:1).' * pi/2;<br /> sensor.directivity_angle = zeros(Nx, Ny);<br /> sensor.directivity_angle(sensor.mask == 1) = dir_angles;<br /> In this case, how to change it to the directivity of the transducer, considering each element has an opening angle and transducer has the combined effect?</p> <p>(iii) Simply make all sensor element forward-looking (sensor.directivity_angle(sensor.mask == 1) = 0) and spatial averaging of sensor points within an element. Is this an option?</p> <p>Is there a better way to implement the same?</p> <p>Thank you,<br /> Francis </p>