This kernel is used to compute the delay in reception for every transducer. It also sums the delay in transmission to obtain the valid samples for the interpolation.
- Template params:
- T: type of the operation;
- NUM_LINE_t: number of scanlines;
- NUM_ELEMENT_t: number of transducer’s elements;
- VECDIM_sample_t: dimension of the SIMD to be performed. Addressed in the Xilinx UG1076, it depends on the type chosen;
- LEN_IN_sample_t: number of inputdata per invoking;
- LEN_OUT_sample_t: number of outputdata per invoking;
- LEN32b_PARA_sample_t: number of inputdata per invoking;
- Kernel Inputs:
- para_const: kernel_samples self-used structural parameter include sampling frequency, inverse speed of sound, and so on. Its definition can be seen in L1/include/kernel_sample.hpp.
- para_rfdim: Input array, used to filter whether the input rf-data of each scanline is in the region of interest;
- para_elem: X-Y-Z vector which represents the positions of our transducers in the probe;
- in_img_x: The result of kernel_imagepoints in x-dimension;
- in_img_z: The result of kernel_imagepoints in z-dimension;
- in_delay: Result of kernel_delay;
- Kernel Outputs:
- out_sample: A vector that represents our valid entries in the rf-data vector;
- out_inside: A vector with only 0 and 1 values, which represents whether each element of our rf-data vector is in the region of interest;