Optimization And Implementation Of Multi-beam Imaging Sonar System Based On Experimental Data

Imaging sonar, which can provide high-resolution pattern information of detection areas, has been more and more widely applied in ocean development. Based on the analysis on existing problems of the principled sample machine of multi-beam imaging sonar developed by the laboratory, optimizations of algorithm and soft and hard wares were conducted in this paper, thus further improving the imaging qualities of the sonar.Specifically, the aspects were included as follows:Firstly, according to requirements listed in the outline of imaging sonar system design, large amount of tests and verifications were conducted on the various performance indexes and imaging effects of the principled sample machine in different settings and water conditions. With the analysis on the test results, directions of improvement and optimization of imaging sonar were pointed out.Secondly, aiming at the impact of the inconformity of amplitudes and phases among sonar transducer channels on the system’s imaging qualities, correction methods for amplitude and phase error of channel were researched and range and phase corrections for the system were made in the two divided parts of conditioning channel and array-element channel. The self-correcting method for system array amplitude-phase errors was proposed based on the schema space theory and TSI arithmetic. Through this method, it became possible to estimate the direction of the correction source and the array amplitude-phase errors in three independent steps with only one far-field correction source in an unknown direction.Thirdly, several improved algorithm designs for beamforming was designed. The robustness design method based on the Least Squares increased the steady performance of the system against noise interferences and channel errors; the improved method of weighting of Chebyshev perfected the characters of the main sidelobe in beam response function; based on convex optimization, the beam forming method which could focus dynamically realized the elaboration imaging of targets; the tap rarefaction design on the strength of signal sparse representation theory enabled the possibility for the reduced complexity of transducer structure.Finally, improvement was made for problems in the system such as small display dynamic range, scarce storage capacity and low starting speed and so on, increased the stability of the system and the research on the low-power design of the system provided a sound base for the commercialization of the system.As shown by the measured results, the sonar system with optimized design features better imaging and satisfactory starting speed and video compression performance, with the expected goals of optimization design achieved. Some optimization methods which could hardly be tested were also verified through theory simulation.