A synthetic aperture sonar draws the seabed on a chosen image surface, and anything standing off that surface is displaced. Whenever the vehicle's path is not straight it is blurred as well: a point target 5 m above the surface is drawn about 2.5 m nearer the track and, under chop, loses about 8 dB of peak.
Last updated .
Tour
Across the surface heights
Peak loss (dB) against surface height (m)
Shift in range (m) against surface height (m)
Dots: measured on the images for this point height, sound speed and aperture. Line under the shift: the predicted layover; below zero is nearer the track. Pick a dot to move the surface.
Controls
SAS image of the patch
8 m × 8 m, along-track up, range 12 to 20 m to the right · × where the point stands · ○ where it is drawn
Point target response 1.2 m across
Along-track cut dB, cm
Range cut dB, cm
Method
The sonar is the array described throughout this site: 36 receivers 30 mm long at 300 kHz with 60 kHz of bandwidth, on a track 10 m above a flat sand bed. It advances 34 phase-center spacings per ping, so that consecutive pings share two phase centers; a phase center is the midpoint between the transmitter and a receiver. The seabed is simulated with ApertureLab's point-scatterer engine at 20,000 scatterers per square meter, with the objects as meshes and their acoustic shadows cast by the simulator. The vehicle sways and heaves sinusoidally, the heave about half the sway at 1.3 times its frequency, and the beamformer is always given the true path. The vehicle's attitude is path-aligned: its heading and pitch follow the path it flies, so that it yaws with the sway and pitches with the heave, and the 3D scene draws it the same way. In the 3D scene the track's sway and heave are drawn five times larger than they are, so that they can be seen at the scale of the patch, and the vehicle's turns are larger by the same factor.
Every image is a time-domain back-projection on a 1.25 cm pixel grid with no glint suppression, formed on a flat surface at the height and with the sound speed that the controls set. Images are shown through the dataset's display chain with one fixed brightness scale per aperture, so that a blurred point is drawn dimmer than a focused one. The point target's own response is isolated by simulating each scene a second time without it and subtracting the two images; the subtraction is exact because the imaging chain is linear. The point-target panel and the numbers in the readout are measured on that response.
The steps that lead up to a synthetic aperture are described on Why synthetic aperture?, and the full imaging chain, including micronavigation, on Image formation. On a circular track the image surface becomes a true plane of focus, a case described on The plane of focus in circular SAS. How the length and position of the synthetic aperture set the image is shown on Aperture length and position. The effect of faults in the sonar's own hardware on the same scene is shown on Hardware faults.
The point-target panel is drawn at the image's 1.25 cm pixels, along-track up and range to the right, and shows the smallest of 25 cm, 50 cm and 1.2 m that holds the response, centered on its brightest pixel. It is in decibels against the same point perfectly focused under the same motion, with the surface at its height. The reference is taken per motion because a vehicle that yaws with its path sweeps its beam along the track, which changes the stretch of track that sees a given point: focused under chop, this point target holds 80 percent of its energy within 1.2 cm along-track, against 2.6 cm in still water. The panel's two cuts run through that pixel over the same width, with the dashed line at −3 dB. The charts under the readout plot, for every surface height, the loss of peak under each of the three motions and the shift in range of the point as drawn, all measured on these images, with the layover the geometry predicts. Holding the compare button shows the image formed with the surface on the bed, in still water, with the true sound speed and the full aperture. Slow swell is 5 cm of sway over a 20 m wavelength and chop 5 cm over 6 m, which turn the vehicle's heading 0.9 and 3.0 degrees each way. The images were computed ahead of time for every combination of the controls, 792 in all.