Research Of Multibeam Interferometric Bathymetry Technology And Its Algorithm Implementation On DSP

Multibeam echosounder is an important bathymetric tool which is widely used in underwater engineering and sea exploration. Most multibeam echo- sounders use amplitude detection technique or phase detection technique which can get single depth value in each beam. These traditional algorithms can not achieve high resolution because the number of depth value is determined by beam number; the beam footprints in the edge of swath are larger than the beam footprints in the middle of swath, so the depth measurements in entire swath are not uniform. Moreover, small objects in beam footprint can not be detected.In order to solve these problems in traditional algorithms we use interferometric method in multibeam bathymetric snoar. The interferometric processing is performed in each beam footprint, using this method a large number of depth value can be acquired without increasing beam number, so the resolution of multibeam echosounder can be largely improved. In this paper we firstly introduce the principle of multibeam interferometric bathymetry using analogism. Then, the measurement error of this new algorithm is analyzed and three type of noise which impacts the depth measurement are particularly discussed. After error analyses, the signal process methods are proposed which include the variable bandwidth filter, phase unwrapping method based on multisource fusion, DOA estimation based on multilook approach, multi baseline detection and so on. A great many of emulational and experimental data is used to illustrate this new algorithm which are recorded by a 300-kHz multibeam echosounder in a lake experiment. Finally, we discuss the problems about DSP implements of this new algorithm.The research indicates that compared with traditional algorithm multibeam interferometric bathymety can improve the resolution of multibeam system and gets much number of depth values. Moreover, this new algorithm has a high ability to detect the detail of sea floor depth information. Using the signal processing methods proposed in this paper can effectively decrease the impact of noise, improves signal to noise ratio and largely increases the precision of depth measurements.