General concept of reduction process for big data obtained by interferometric methods

Abstract

Interferometric sonar systems apply the phase content of the sonar signal to measure the angle of a wave front returned from the seafloor or from a target. It collect a big data - datasets that are so large or complex that traditional data processing application software is inadequate to deal with them. The recording a large number of data is associated with the difficulty of their efficient use. So data have to be reduced. The main goal of new reduction method developed by the authors is that, the data after reduction will not be an interpolated value. The proposed method is consists of two main stage: the grouping of data and the generalization of data. The first stage consists of two steps: initial division and clustering. In the first step, the area will be divided into a grid of squares. The maximum level of generalization of the grid will be founded and its size will be defined. In the second step of data grouping, namely clustering artificial neural networks will be used. Artificial neural networks are good alternative to traditional methods of clustering data. The authors decided to use artificial intelligence methods during the processing of data obtained by interferometric methods because it is novel approach to such issues and provides satisfactory results. The author's goal is to represent each group by a single sample depending on the compilation scale of final product. The article contains a detailed description of the proposed method.