Phased subarray processing for underwater 3D acoustic imaging

3D sonar imaging using a fully-populated rectangular 2D array has many promising applications for underwater imaging. A primary limitation of such systems is the large number of parallel front-end hardware channels needed to process the signals in transmit and receive when using conventional full phased array imaging. A subaperture beam acquisition and image formation process is presented that significantly reduces the number of front-end hardware channels while achieving image quality approaching that of full phased array imaging. Rather than transmitting and receiving on all NXN transducer elements to form each beam, an MXM subset of elements-called a subarray-is used for each firing. The limited number of front-end processing channels are used to acquire data from each subarray. Switching hardware allows the subarray to be multiplexed across the full array. Due to the Nyquist sampling criteria in beamspace, the number of beams acquired by each subarray can be significantly reduced compared to the number required for the full array. The phased subarray processing includes beam upsampling, lateral interpolation with a subarray dependent filter, and coherent weighting and summation of all subarray images to form a high resolution image. The phased array method achieves an image quality nearing that of full phased array imaging with significantly fewer processing channels, slightly reduced SNR, and roughly three times the number of firings for reasonable configurations.