浅海环境下稳健的最小二乘匹配场定位算法
Robust Matched Field Localization Algorithm Based on Least Squares in Shallow Water

匹配场处理在海洋环境参数估计领域具有优于常规算法的性能，但其对噪声及环境扰动十分敏感，制约了该方法的应用。为了提高匹配场处理的稳健性，提出了浅海环境下稳健的最小二乘匹配场定位算法。该方法用简正波模型把浅海环境的声场分解为模态函数矩阵和模态系数向量，把环境扰动构建为模态函数的扰动，并在最小范数意义下用最小二乘算法对噪声和环境扰动进行约束，从而提高了匹配场处理的稳健性和定位性能。最后，使用海试实验数据对最小二乘匹配场处理器进行了验证。结果表明：和常规匹配场处理器相比，该算法的峰值背景比提高了约0.6~1.7 dB，而且对声源位置参数的估计更准确。
The matched field processor(MFP) is prior to the nomal algorithms in the estimation of ocean environmental parameters, but the method is sensitive to the noise and the environmental mismatch, and this disadvantage hinders its application seriously. To overcome this shortcoming and improve the performance of MFP, an approach called robust matched field localization algorithm based on least squares(LSMFP) in shallow water is proposed, which decomposes the received fields in the shallow water into depth function matrixes and amplitudes of normal mode at the beginning. Then the environmental mismatch is considered as some noise, and combined with the other noise using the least squares in the sense of minimum norm, therefore the performance of MFP is improved. In the end, the vertical array data from the sea trials is used to evaluate the algorithm. The results show that the power to background ratio(PBR) of LSMFP is lager about 0.6~1.7dB than CMFP, and the localization is more accurate