Research On DOA Estimation Method Based On Higher Order Acoustic Vector Sensor Array

As a new type of sound signal detection device,acoustic vector sensor has played an increasingly important role in the signal processing field.Its characteristic is that it can not only measure the scalar sound pressure information,but also measure the vector velocity information.In the field of communication,sonar,and other aspects of the direction of the direction of the relevant engineering applications,a lot of practice has proved the superior performance of acoustic vector sensors.When estimating the arrival direction of a signal,the existing array structure model can only measure the first-order information in the sound field.The actual sound field contains not only first-order information but also high-order sound field information.Aiming at the problem of large measurement error and low estimation accuracy of the existing models,this paper proposes a high-order acoustic vector sensor array model to study the DOA estimation of high-order acoustic vector sensor arrays.The research work of this article is as follows:For high-order information contained in the sound field,a higher order acoustic vector sensor model containing high-order information of the sound field is introduced.In order to improve the DOA estimation accuracy of spatial signals,a HO-MUSIC algorithm based on DOA estimation of high-order acoustic vector sensor arrays was designed.By decomposing the covariance matrix,the signal subspace and the noise subspace are estimated,and the spectrum peak search is used to effectively estimate the pitch and azimuth angles of the sound source signal.Experiments show that the designed HO-MUSIC algorithm is effective,and because it contains higher-order information of the sound field,it better describes the sound field and estimates the angle of the sound source signal is more accurate.The optimal value analysis is given for the order value of each parameter in the model.The MUSIC algorithm combines the ESPRIT algorithm with high-order acoustic vector sensors to solve the disadvantages of large computations caused by peak searching.The HO-ESPRIT algorithm based on DOA estimation of high-order acoustic vector sensor arrays is designed.Compared with the HO-MUSIC algorithm,The amount of computation is reduced,and experiments have proved the effectiveness of the algorithm.The order of the algorithm model is discussed and the choice of the best order is given.In order to solve the problem of orthogonality between output components in high-order acoustic vector sensor array signal processing methods,the quaternion theory is introduced into higher-order acoustic vector sensor array models,and superior performance of high-order acoustic vector sensors is exploited to derive the corresponding quaternion model.Based on this,combined with the MUSIC and ESPRIT algorithms,the quaternion HOQ-MUSIC algorithm based on higher order acoustic vector sensor array DOA estimation and the quaternion HOQ-ESPRIT algorithm based on higher order acoustic vector sensor array DOA estimation are designed.The orthogonality of the quaternion improves the estimation accuracy,and the simulation experiment proves the effectiveness of the algorithm.