超短基线定位系统融合分类解模糊技术研究

摘要

针对超短基线水声定位系统面临的相位差模糊问题,该文提出一种基于多分类器融合的定位解模糊算法.首先构建多分类器系统,将解模糊问题转化为对模糊数进行分类识别的问题,并采用Choquet积分对各分类器结果进行融合,得到模糊数所属类别,进而对目标进行定位.给出了目标位置的无模糊观测条件.该算法优点是无需构造小于半波长间距的阵列,有效扩大了无模糊阵列孔径,而且由于充分利用了相位差观测数据的统计特性,定位精度可接近克拉美-罗界.仿真结果验证了该算法的有效性.%To solve the phase-difference ambiguity problem in Ultra-Short BaseLine (USBL) underwater acoustic positioning systems, an ambiguity resolution and localization method based on Multiple Classifier Fusion (MCF) is proposed. Firstly, the multiple classifier system is built. Then, ambiguity resolution problem is formulated as classifying and recognizing the ambiguity integer, and Choquet integral is utilized for fusing the results of multiple classifiers. Finally, the category of ambiguity integer is obtained and the target is located. The unambiguous observation condition of the target position is derived. Without constructing an inter-sensor spacing less than half the wavelength, unambiguous aperture of the array is effectively enlarged. Moreover, as statistical characteristics of the observation data are fully utilized, the positioning accuracy approaches the Cramer-Rao bound. Simulation results verify the effectiveness of the proposed method.