Abstract
The mapping swath and along-track resolution can be simultaneously enhanced by using multireceiver synthetic aperture sonar (SAS). Unfortunately, the SAS parameters, including the pulse repetition frequency, SAS moving velocity, and the along-track aperture of the receiver array, are rigorously limited. Any deviation from the strict requirements could result in sampling that is not uniform. Based on the matrix inversion (MI) method, the uniform signal is recovered. The MI technique would not be able to recreate the SAS image if the redundant receivers within two subsequent pulses were extremely close or overlapped. This study discusses an innovative approach to tackle this issue. Our approach first identifies the receivers that the SAS image reconstruction algorithms exploit. Then, the approximation error caused by the range approximation is compensated for these receiver datasets. Thereafter, all receiver datasets corresponding to each pulse are stored pulse by pulse. After this operation, the omega-k algorithm is further used to reconstruct the SAS image. When the MI method works well, the proposed method can achieve high-quality images that closely match those of the matrix inversion method. More importantly, the proposed method can still work in cases where the MI method fails. The experiments further show the major superiority of our method over traditional methods.