Research On Eigen Space Spatial Spectrum Estimation And Its Improved Algorithm Based On Acoustic Vector Sensor Array

Compared with traditional sound pressure arrays,the acoustic vector sensor arrays can obtain both the sound pressure information in the sound field and the velocity information.As a result,the acoustic vector sensor arrays have higher gain and better DOA estimation performance,and therefore have attracted the attention and research of a large number of scholars.Based on this characteristic,this paper uses the acoustic vector sensor array to study and simulate the DOA estimation algorithm with the eigen space method.Compared with traditional DOA estimation algorithm,the performance of this method is more precise and the band width of this method is sharper.The simulation is used in this paper to verification this conclusion.Subspace algorithms usually require the number of sources for DOA estimation.Therefore,how to estimate the number of sources accurately is also one of the research focus in this paper.The recent research on the number of sources in array signal processing has achieved many results,but there are also many shortcomings.First,for the traditional method of source number estimation based on information theory,when the number of snapshots or array elements is small,its performance will decrease.Secondly,for the method of source number estimation based on the Gerschgorin's Disk Estimation method,twice feature decomposition operations of the covariance matrix are performed when estimating the number of sources and azimuth estimation,that makes the calculation amount larger.Finally,although the method of source number estimation based on the eigenvalue threshold method does not require multiple feature decomposition operations,each estimation requires manually setting the threshold,which makes this method practical in practice.In view of the shortcomings of traditional source number estimation methods,this paper proposes a source number estimation method based on Newton interpolation.This method can maintain good estimation performance even under the condition of the small number of snapshots and the small number of array elements.The estimation performance of this method does not change with the number of sources to be estimated,and the source number and DOA can both be estimated by only one feature decomposition.Meanwhile,there is no need to manually set the threshold.After that,this paper combines the compressive sampling array with the eigen space method,and uses the eigen space DOA estimation method based on the compressive sampling array.This algorithm greatly reduces the amount of calculation and the complexity of the hardware circuit structure.Through simulation analysis,the feasibility of the algorithm is verified.At the same time,this paper also researches the estimation of the number of sources based on the compressive sampling array,which avoids the phenomenon that the calculation volume increases due to multiple eigen decomposition when using the compressive sampling array.