Research On Data Fusion Algorithm Of Passive Underwater Acoustic Detection

Due to the deepening of ocean exploration,the requirements for underwater target detection capabilities are becoming more strict.Therefore,in-depth research on underwater target detection technology is an urgent matter.Accurate detection of underwater targets is also the hot issue and current field of marine technology research.In response to the needs of underwater target detection,this subject explores the buoy's attitude measurement technology,the principle and algorithm of vector hydrophones for the orientation estimation of underwater targets.On this basis,the data fusion technology of multi-vector hydrophones is used to further improve the positioning.Finally,a Passive Underwater Detection Simulation System that can interact with researchers and simulate research content is designed,effectively reducing research costs.Firstly,a buoy attitude detection system based on a three-axis accelerometer and a three-axis gyroscope is designed and verified.The theoretical knowledge of the attitude detection system is analyzed in detail.Then,the hardware circuit and software algorithm of the system are designed,and the system is tested,including the analysis of the sensor modules,static performance and dynamic performance results.The results show that the whole system has good static and dynamic performance,which can meet the requirements of buoy attitude measurement under actual using conditions.Secondly,it mainly studies the orientation estimation principle of single vector hydrophone and vector hydrophone array.The single vector hydrophone is based on the average sound intensity method to estimate the target azimuth,and the Kalman filter algorithm is used to improve the accuracy of the target orientation.Later,based on its signal processing model,the vector hydrophone array uses the TLS-ESPRIT algorithm and the Root-MUSIC algorithm to estimate the target azimuth.The results show that improving the signal-to-noise ratio and increasing the number of snapshots can reduce the error of the target azimuth estimation and improve the positioning accuracy of the target.Then,the parameter-level and feature-level fusion positioning algorithms are studied.The parameter-level fusion positioning algorithm mainly includes the Chan positioning algorithm and the Taylor positioning algorithm.The principle is analyzed and the positioning accuracy of the two algorithms under different errors is compared.The performance of the Taylor algorithm is better than the Chan algorithm as the measurement distance error increases.The feature-level fusion positioning algorithm mainly studies the GDOP weight method.Based on the high-precision positioning of target by a single vector hydrophone,the dual-base station and four-element measurement array system model is further established,and the measured data is integrated.The comparison shows that the positioning accuracy of the four-array element is higher.Finally,according to the research content and modular design ideas,a Passive Underwater Detection Simulation System(PUDSS)that can combine the attitude measurement module,target detection module,three vector hydrophone data fusion algorithms,and the recording and monitoring of sea state information is built,while providing an intuitive interface for the simulation results display.