Field II Correspondence¶

eSDIva deliberately mirrors Field II's conventions so a Field II user can transition (and cross-validate) with minimal friction. This page collects the correspondences in one place.

Transducer classes ↔ `xdc_*` functions¶

Field II	eSDIva	z-datum
`xdc_linear_array`	`LinearArrayTransducer`	flat face at z = 0
`xdc_focused_array`	`LinearArrayTransducer(elevation_focus_mm=...)`	element face (rim) at z = 0, lens dished back to −sag
`xdc_convex_array`	`ConvexArrayTransducer`	centre element at z = 0, arc centre at z = −R
`xdc_convex_focused_array`	`ConvexArrayTransducer(elevation_focus_mm=...)`	as above + rim-referenced lens
`xdc_2d_array` / `xdc_rectangles`	`MatrixArrayTransducer`	flat face at z = 0
`xdc_piston`	`FlatCircularTransducer`	face at z = 0
`xdc_concave`	`ConcaveCircularTransducer`	apex at z = 0, rim at z = +sag
— (convex bowl)	`ConvexCircularTransducer`	apex at z = 0, rim at z = −sag
`xdc_focused_array` (1 element, cylindrical lens)	`FocusedCircularTransducer`	face (curved-axis rim) at z = 0, centre line at −sag
any `Th` via `xdc_get(Th, 'all')`	`from_fieldii_xdc_data` → `FieldIITransducer`	as exported
any `Th` via `xdc_get(Th, 'rect')`	`from_fieldii_rect_data` → `FieldIITransducer`	as exported

sag = R − √(R² − (D/2)²) in every case.

Lens tiles of the native lensed arrays are sampled equal-arc in the lens angle θ (nodes at y = R·sin θ), exactly matching Field II's tiling of xdc_focused_array.

Simulation calls¶

Field II	eSDIva
`calc_h`	`Emission(tx)` — pulsed mode (returns `ρ₀·h`; identical arrays at `rho=1`)
`calc_hp`	`Emission(tx, fs=..., excitation=e)` with `xdc_impulse` ↔ `tx.set_impulse_response`
`calc_scat`	`Reception(tx, rx).scan_focusline(...)` (focused, apodized, summed on receive)
`calc_hhp` / `calc_scat` (unit point)	`Reception(tx, rx).pulse_echo_rf(...)` per-element RF
`calc_scat_all`	`Reception(tx, rx).synthetic_aperture_rf(...)` (FMC)

The pulse-echo derivative convention is shared: the physical ∂³v/∂t³ is carried by the band-limited excitation and TX/RX impulse responses — neither simulator applies an explicit derivative, so calc_scat for a unit point equals calc_hhp and eSDIva's RF coincides with both (correlation ≈ 1.0000).

Time origin (`t0`)¶

Field II reports absolute time from the excitation start. eSDIva returns each result with coords["t0"], referenced to the beam axis: the TX (and RX) focusing bulk delays.max() is subtracted so downstream beamforming needs no per-line correction.

For a lens-focused aperture, eSDIva's time grid is referenced to the first-arriving rim, but the focused elevation echo peaks one lens transit later; reception therefore adds elevation_lens_sag / c once per aperture (TX and RX). With this, a native elevation-focused linear array matches Field II pulse-echo RF at lag 0.

Imported probes: the lens curvature is already in the imported patches (the RF shape matches without help), but the t0 transit correction needs the lens focal length — pass elevation_focus_mm= (the Field II Rfocus) to from_fieldii_xdc_data / FieldIITransducer, or set tx.elevation_lens_sag (metres) directly.

Amplitude scale¶

eSDIva scales the pulse-echo RF by ρ₀ / 2c₀² (the physical scattering prefactor); Field II uses approximately ρ₀ / 2. For a unit-amplitude scatterer the raw RF amplitudes therefore differ by a factor c₀² (≈ 2.37 × 10⁶ at 1540 m/s). Normalised comparisons (envelope / peak, PSF, correlation) are unaffected.

Units¶

Field II is SI everywhere (metres, seconds). eSDIva's user-facing API is mm (_mm suffixes) with SI internals — the one deliberate departure. FieldIITransducer takes its patch geometry in metres, exactly as exported by xdc_get.