Research On Sound Source Tracking Technology Based On Small Aperture Microphone Array

With the rapid development of science and information technology,modern passive acoustic detection systems pay more and more attention to the characteristics of small size,low cost,and strong adaptability,whether it is military demand or civilian use.Compared with the traditional large-aperture acoustic array,the array aperture of the small-aperture array is only a few signal wavelengths.It has the advantages of high concealment,strong maneuverability,and convenient deployment,and can track the target sound source in all directions,all-weather,and multiple scenarios.However,the miniaturized acoustic array hardware system brings new challenges to passive acoustic detection and tracking.In practice,when the traditional direction finding tracking algorithm is applied to small aperture arrays,especially conventional beamforming algorithms,it is difficult to meet the requirements of tracking accuracy.Therefore,in the context of small-aperture arrays,this paper simulates and analyzes the influence of the array aperture on the estimation accuracy of the Direction Of Arrival(DOA)of the target signal.The DOA tracking algorithm for arrays and coprime linear acoustic arrays has been exploratively studied from the perspectives of theoretical derivation,simulation analysis and experimental verification.The specific work is as follows:Firstly,the research on DOA tracking algorithm based on uniform linear acoustic array is carried out.For algorithms such as MUSIC(Multiple Signal Classification),which are directly extended from static systems to non-static systems,the dynamic target angle of arrival changes rapidly,resulting in poor real-time performance.In this paper,the combination of the projection approximation subspace(PASTd)algorithm and the MUSIC spatial spectral function is used to avoid the eigendecomposition of the signal covariance matrix when obtaining the subspace,and the computational complexity is reduced,thereby realizing real-time tracking of moving targets.The impact of forgetting factor as an important parameter of the algorithm on the tracking performance is analyzed by simulation.The simulation results show that when the forgetting factor is set to 0.7,the tracking error of the algorithm is the lowest.Aiming at the problem that the subspace update class tracking algorithm and the traditional particle filter(Particle Filter,PF)algorithm are used for long-term tracking of moving targets and the accuracy of DOA estimation decreases in a low signal-to-noise ratio environment,an improved particle filter DOA tracking algorithm is proposed,which uses the MUSIC spatial spectral function as the likelihood function of the PF algorithm,and reshapes the likelihood function waveform through exponential weighting.Theoretical analysis and simulation results show that the proposed method has lower tracking error when tracking the target for a long time,and in the low signal-to-noise ratio environment,the proposed algorithm has the fastest convergence speed and the lowest tracking error.Then,aiming at the problem that the physical aperture of the traditional uniform linear array is limited by the Nyquist sampling theorem,the DOA tracking algorithm based on the co-prime acoustic array is researched,and the tracking algorithm based on the particle filter under the co-prime array is proposed.The algorithm applies the MUSIC spatial spectral function of spatial smoothing to the improvement of the likelihood function of PF,and aiming at the problem of the loss of array aperture caused by the introduction of the idea of spatial smoothing,an exponential r is used to weight the improved likelihood function.Theoretical derivation and simulation results show that the algorithm has higher estimation accuracy,and the proposed algorithm has faster convergence speed and better robustness in the low signal-to-noise ratio environment.Finally,in order to prove the effectiveness of DOA estimation and tracking algorithm under different array models from the perspective of experimental verification,this paper designs and builds a set of target tracking test platform based on small-aperture microphone array.By comparing the performance of several classical DOA estimation algorithms in processing measured data,the tracking performance of each algorithm is analyzed,and the feasibility and practicability of the experimental platform are verified.Based on the above research,the DOA estimation and tracking experiments of moving targets under different microphone array structures are carried out to verify the effectiveness of the proposed algorithm.