Multiple Target Sources DOA Estimation Algorithm Research In Shallow Ocean Environment

Underwater target location technology is an important component of marine science research,and it has wide application prospect in underwater communication,sonar and national security.DOA estimation based on the signal source can achieve high resolution direction estimation of multiple target sources in the space environment,and then realize the accurate positioning of the target.The application of this technique to the underwater environment has a good effect.In this paper,the DOA estimation algorithm of underwater target source is studied with mathematical analysis and simulation experiment,which is mainly realized by the following aspects:In this paper,the Fractional Fourier transform method is used to transform the output of the sensor array into the fractional domain to deal with the defects of traditional algorithms which cannot be measured accurately when the number of signal sources is more than the number of sensors.The multi-objective source separation is achieved successfully.Then,the signal output of the single vector sensor is used and a reasonable delay is given.The two-dimensional DOA parameters are estimated by FrFT-ESPRIT algorithm respectively.In this paper,we first describe the phenomenon of severe heavy-tailed noise in the underwater noise environment and simulate the underwater noise distribution by using the synthetic circularly symmetric generalized Gaussian distribution through the simulation modeling method.At the same time,for the noise output of the sensor array,the higher order moments of the variance matrix do not exist.In this paper,the method of fractional lower order matrix is used.And the simulation results are verified by MATLAB simulation software.In addition,due to the complexity of the underwater environment,the direction matrix of the sensor array may have non-ideal gain value.In this paper,the least squares method is used to optimize the instantaneous gain value of the sensor array,and the vector output method is used to increase the array output signal Then the sparse matrix is reconstructed and the DOA parameter of the signal source is estimated by the sparse representation.The simulation results show that the method can not only improve the parameters,but also improve the performance of the system.Measurement accuracy and detection probability,but also reduces the environmental signal to noise ratio requirements,which in the actual measurement of the marine environment is very useful.