Wei Bo, He Chuanlin, Xing Siyu, Zheng Yi
Institute of Oceanographic Instrumentation, Qilu University of Technology (Shandong Academy of Sciences), Qingdao 266061, China.
School of Ocean Technology Sciences, Qilu University of Technology (Shandong Academy of Sciences), Qingdao 266061, China.
Sensors (Basel). 2022 Aug 12;22(16):6016. doi: 10.3390/s22166016.
High-accuracy level underwater acoustical surveying plays an important role in ocean engineering applications, such as subaqueous tunnel construction, oil and gas exploration, and resources prospecting. This novel imaging method is eager to break through the existing theory to achieve a higher accuracy level of surveying. Multibeam Synthetic Aperture Sonar (MBSAS) is a kind of underwater acoustical imaging theory that can achieve 3D high-resolution detecting and overcome the disadvantages of traditional imaging methods, such as Multibeam Echo Sounder (MBES) and Synthetic Aperture Sonar (SAS). However, the resolution in the across-track direction inevitably decreases with increasing range, limited by the beamwidth of the transducer array of MBES. Furthermore, the sidelobe problem is also a significant interference of imaging sonar that introduces image noise and false peaks, which reduces the accuracy of the underwater images. Therefore, we proposed an accelerated deconvolved MBSAS beamforming method that introduces exponential acceleration and vector extrapolation to improve the convergence velocity of the classical Richardson-Lucy (R-L) iteration. The method proposed achieves a narrow beamwidth with a high sidelobe ratio in a few iterations. It can be applied to actual engineering applications, which breaks through the limitation of the actual transducer array scale. Simulations, tank, and field experiments also demonstrate the feasibility and advantages of the method proposed. 3D high-accuracy level underwater acoustical surveying can be achieved through this 2D MBES transducer array system, which can be widely promoted in the field of underwater acoustical remote sensing.
高精度水下声学测量在海洋工程应用中发挥着重要作用，如水下隧道建设、油气勘探和资源勘查等。这种新颖的成像方法迫切需要突破现有理论，以实现更高精度的测量。多波束合成孔径声纳（MBSAS）是一种水下声学成像理论，它可以实现三维高分辨率探测，并克服传统成像方法（如多波束回声测深仪（MBES）和合成孔径声纳（SAS））的缺点。然而，受MBES换能器阵列波束宽度的限制，沿航迹方向的分辨率不可避免地会随着距离的增加而降低。此外，旁瓣问题也是成像声纳的一个重大干扰因素，它会引入图像噪声和虚假峰值，从而降低水下图像的精度。因此，我们提出了一种加速去卷积MBSAS波束形成方法，该方法引入指数加速和向量外推来提高经典理查森-露西（R-L）迭代的收敛速度。所提出的方法在几次迭代中就能实现具有高旁瓣比的窄波束宽度。它可以应用于实际工程应用中，突破了实际换能器阵列规模的限制。仿真、水池和现场实验也证明了所提方法的可行性和优势。通过这种二维MBES换能器阵列系统可以实现三维高精度水下声学测量，该系统在水下声学遥感领域具有广泛的推广前景。
