James T. Christoff- Christoff_jim@ccmail.ncsc.navy.mil
Code R21, Acoustical Sensing
Coastal Systems Station, Dahlgren Division
Naval Surface Warfare Center
Panama City, Florida, 32407
Popular version of paper 3pSP5
Presented Wednesday afternoon, June 24, 1998
ICA/ASA '98, Seattle, WA
The synthetic aperture sonar(SAS) concept uses modern signal processing technology to produce an acoustic image of the sea bottom. This acoustic imaging sonar operates on principles similar to that of a camera used to take pictures. But this camera uses underwater acoustic energy, a computer to store acoustic echoes and focussing algorithms to produce very sharp acoustic images of the sea bottom. The SAS camera has one big advantage over an underwater camera that uses light. Light does not travel very far underwater, especially if the water is cloudy or turbid. On the other hand, some forms of acoustic energy can travel many miles underwater. The SAS camera makes use of this fact and produces underwater images of much larger areas than could be achieved with an underwater light camera. The SAS concept is able to do this by towing a very small underwater transducer in a straight line and collect the echoes from each transmission. It then compensates for the underwa! ! ter environmental effects and the motion of the vehicle and focuses the acoustic energy much like a camera does and produces the underwater image upon a computer display. This paper describes this concept and the method called motion compensation that reduces the defocusing effects due to unexpected motion of the underwater transducer. This defocusing problem is very similar to the problem encountered by a camera that has a slow shutter and the person taking the picture shakes the camera just as he snaps the shutter, producing a blurred image. The SAS concept suffers the same degradation when the vehicle moves in an unexpected direction while the SAS is producing an acoustic picture.
In the mid seventies the Coastal Systems Station (CSS) began to seriously consider the SAS concept as a practical sonar system for underwater imaging and mine classification. Measurements by CSS and others indicated that the environment was stable enough to support SAS and CSS decided to proceed with development of SAS technology. In the late 80's and early 90's modern digital signal processors(DSP) made SAS a practical and affordable technology. The latest DSP technology makes processing, modeling, simulation and implementation of complex SAS algorithms an affordable reality.
SW/VSW SAS
A modern underwater SAS camera was built by Northrop Grumman( formerly Westinghouse), under contract to the Navy to make very sharp underwater acoustic images. The sonar is a dual frequency SAS, operating simultaneously at 20 and 180 kHz, producing 7.5 and 2.5 centimeter resolution respectively from a 0.5 meter physical aperture. These two images taken at two different acoustic frequencies provides more information to the operator about objects on the bottom and about the bottom type, mud, rock or sand than a single frequency sonar could. The blurring effects of the unwanted transducer motions are cancelled with a motion compensation algorithm. This algorithm is called the displaced phase center (DPC) algorithm first described by Sheriff(1) for motion compensation (MOCOMP).
RESULTS
The target field consisted of two target spheres, 15 cm and a 20 cm, and a resolution panel. The resolution panel is an aluminum 1.2 by 2.4 meters plate with gravel rings ranging from 30 cm to 5 cm inside diameter. Figure 1a and 1b are the SAS point response function (PRF) of the 15 cm sphere without and with MOCOMP. Image 1a is blurred by unknown motion, this can be easily recognized by comparing 1b with 1a, which is corrected for motion. The second plot shows the response with motions corrected and has much sharper peak than the first. The sharper the peak the sharper the image will be focused. The resolution of the SAS is near theoretical when the MOCOMP algorithm is used. Figure 2 is a bottom image, the first image, 2a, is without any motion compensation and is slightly blurred. Compare it with image 2b that has been corrected with the DPC algorithm. The second image is sharper and the undulations and pot holes in the sea bottom are darker.
CONCLUSIONS
The evidence presented in this paper proves the SAS concept can produce very good underwater images from very small and simple hardware. We expect that as new MOCOMP algorithms are developed and the as the DSP technology becomes faster and more powerful, that the image area will grow larger and the image quality will improve substantially
ACKNOWLEDGEMENTS
This work was sponsored by ONR Code 321TS. I thank Jose Fernandez for his assistance and suggestions in preparing this manuscript.
REFERENCES
1. [Robert Sheriff, Symposium On Autonomous Underwater Vehicle Technology, pp. 236-245, July 1992]