Key Technologies And Software Implementation Of Distributed Sonar Multi-target Passive Positioning

The distributed sonar multi-target passive positioning system not only needs to process the target original signal,but also needs to process the feature data.Aiming at the problems encountered in data processing at the target feature level,this paper studies two key technologies in multi-target passive positioning,namely target trajectory enhancement technology and target correlation technology.In addition,the distributed sonar multi-target passive positioning display and control software is designed to realize the real-time monitoring and display of the target's movement state,which helps the commander to make reasonable command decisions.In the process of multi-target passive positioning,weak targets are susceptible to interference from strong targets,which leads to the trajectory of the weak targets in the azimuth history map of the sonar output drowned in the noise background.To solve this problem,image processing technology can be used to improve the clarity of the weak target trajectory,which is convenient for subsequent zoom-in tracking of the target.This paper studies three methods of image processing: one is the histogram equalization method,which uses non-linear transformation to make the image gray uniform,so that the picture details are clear;the second is the histogram piecewise linear transformation method,which designs a linear mapping function to suppress the background section And stretch the target segment to improve image contrast;the third is the combination of image edge detection and histogram segmentation equalization(NIDO-ODSHE),which performs histogram segmentation equalization after convolution with the image edge detection operator,so that the edge of the target trajectory clearer.The simulation results show that the histogram equalization method can increase the intensity of the weak target trajectory,but cannot suppress the background noise well.The histogram piecewise linear transformation method can effectively suppress the background noise and improve the clarity of the target trajectory when the target intensity is small.Its processing effect is affected by the target intensity.The NIDO-ODSHE method performs better when the target trajectory changes at a slower rate.In the process of multi-target passive positioning,the correct division of target batches affects the accuracy of positioning solution results.Aiming at this problem,this paper studies two multi-objective data association algorithms.The first is the association of target position data.In this paper,the statistical form of the coarse association and the decision threshold are given,and the uncertain set after the coarse association is finely associated again to improve the data association speed.The second is target feature correlation.This paper uses the gray system theory and systematic cluster analysis to determine the similarity between targets.And implement the target adaptive entropy weight gray correlation degree algorithm.Compared with artificial weighting,the algorithm can reflect the importance of different features and is not affected by subjective factors.Simulation proves that both algorithms can achieve correct target correlation.Finally,through functional requirements analysis,distributed sonar multi-target passive positioning software is designed and implemented.The software implements the target trajectory enhancement algorithm and data association positioning algorithm studied in this paper,and completes functions such as data transmission,file management,and human-computer interaction.The laboratory joint debugging test combined with simulation data proved the correctness and stability of the software.