Transducer Setup

Transducer modeling

PRESTUS does not provide default transducer calibrations. Instead, it encourages users to set up transducers and optimize settings to match their empirical setup.

See the full parameter documentation.

Static parameters

The transducer supports two alternative configurations:

• Annular array (type = annular)
• Matrix array (type = matrix)

Each type is configured in its own sub-struct (annular or matrix). Shared fields (trans_pos, focus_pos, focal_distance_ep/focal_distance_bowl, freq_hz) sit at the transducer level.

For the setup of annular transducers see the annular definition.

For the setup of matrix transducers see the matrix definition.

Dynamic parameters: amplitude and focus

These parameters are traditionally calibrated in free-water simulations (see below).

Note: Calibration is currently implemented only for annular arrays.

transducer.annular.elem_amp
Amplitude of the acoustic source (in Pa). [CALIBRATED]

transducer.annular.elem_phase_deg
Phase of the acoustic source per element (in degrees). [CALIBRATED]

For matrix arrays, phase delays are currently not provided as input.
Instead, steering is performed automatically by computing the phase for each element based on transducer.trans_pos and transducer.focus_pos.

See transducer calibration.

Transducer placement

See also transducer placement.

In water simulations in combination with annular arrays, the placement will be automatically determined based on the edge of the PML layer.

transducer.trans_pos
Position of transducer bowl (XYZ, T1 grid voxel space). Used for placement and phase calculation in matrix arrays.

transducer.focus_pos
Position of stimulation target (XYZ, T1 grid voxel space). Defines the focal point and is used for automatic steering in matrix arrays.

Transducer-target distance specification

The figure shows PRESTUS definitions for annular transducer distances.

PRESTUS expects one either of the following to be specified (or it will attempt a geometric distance calculation based on trans_pos and focus_pos, if specified).

focal_distance_ep
Expected distance from the transducer exit plane to the stimulation focus (mm). This often corresponds to the focal depth setting of the driving system / transducer calibration.

focal_distance_bowl
Expected distance from the transducer bowl to the stimulation focus (mm). Will be internally calculated based on focal_distance_ep, dist_geom_ep_mm, and curv_radius_mm.

---

Transducer geometry: annular vs. matrix array

Annular array

An annular array consists of concentric ring-shaped elements arranged on a spherically curved surface. Because all elements share the same acoustic axis, focusing is inherently one-dimensional (axial). Lateral steering is not possible; the transducer must be physically repositioned to change the target location.

Modeling in PRESTUS:
Element geometry is defined by inner/outer diameters (Elements_ID_mm, Elements_OD_mm) and bowl curvature (curv_radius_mm). Each element is assigned an independent pressure amplitude (elem_amp) and phase (elem_phase_deg), which must be calibrated empirically. PRESTUS includes a calibration pipeline for annular arrays to match simulated pressure to free-water measurements.

When to use:
Use annular arrays when your hardware is a single-focus or multi-ring bowl transducer with pre-calibrated per-element amplitudes and phases. This is the simpler and better-validated configuration.

---

Matrix array

A matrix array consists of many small elements distributed across a curved or flat surface. Because each element can be driven with an independent phase delay, the acoustic focus can be steered electronically in three dimensions without moving the transducer.

Modeling in PRESTUS:
Element positions are either defined analytically (Fibonacci, Fermat spiral, or rectangular grid) or loaded from a coordinate file. Element shape can be rectangular (rect), circular (disc), or curved (bowl). Phase delays are computed automatically from the geometric relationship between each element, trans_pos, and focus_pos — no manual phase input is required. For curved arrays, curv_radius_mm and dist_geom_ep_mm define the bowl geometry used for visualization and offset calculations. An optional Clover configuration replicates the array into a multi-aperture (up to three-leaf) arrangement.

When to use:
Use matrix arrays when your hardware supports electronic steering (e.g., a phased array system) or when you want to simulate volumetric targeting without repositioning the transducer. Note that amplitude calibration is not yet implemented for matrix arrays.

---

Transducer modeling: kWaveArray vs. simple elements

The setup_source function creates the acoustic source for k-Wave simulations. Two implementations are supported:

1. Realistic kWaveArray elements (use_kWaveArray = 1, recommended)
   Uses the k-Wave kWaveArray class to define transducer elements via physical parameters (position, radius, apodization) rather than discrete grid voxels. Elements render smoothly at any grid resolution. Independent per-element delays and weights enable dynamic focusing. Element overlap is handled automatically through weighted contributions. Supports annular rings, axisymmetric 2D simulations, matrix arrays, and GPU acceleration.

2. Simple custom elements (use_kWaveArray = 0)
   Creates bowl-shaped (3D) or arc-shaped (2D) pressure source regions per element using the k-Wave makeBowl function. Each element receives an identical continuous wave (CW) signal. Best for uniform single-element setups where simplicity is preferred. Grid discretization can introduce staircasing artifacts in the source mask.