Research On Wideband Signal DOA Estimation Algorithm And Application Of Regional Positioning Method

In recent decades,with the development of array signal processing technology,the less cost of sensors and the faster speed of processors,acoustic sensor network has been widely applied in life,medical,military and other fields.Acoustic array passive detection system utilizes microphone arrays to receive target signal source,and adopts the array signal processing algorithm model to signal processing,finally realizes the source target detection,direction finding and location.The thesis mainly includes the introduction of the typical DOA estimation algorithms for wideband signals,then this paper proposes a high resolution direction finding algorithm based on wavelet transform,and it designs a regional positioning system,and then the effectiveness of the system is proved through the simulation experiment and application experiment.The specific of the paper is as follows:Firstly,it introduces the wideband array signal model,and the beamforming algorithm,the multiple signal classification algorithm and the wideband focusing matrix algorithm based on the mentioned model,and it gives the algorithm principle and implementation steps,then it uses the simulation experiment to analyze the error performance of these algorithms in the influence factors such as different SNR,the number of fast,and so on.Secondly,it describes the basic theory of wavelet transform,and it applies wavelet transform to set up a time-frequency array signal model,the model is different from the previous wideband array signal model which evenly divided the frequency band,it uses the multi-resolution characteristic of wavelet transform to implement multi-resolution classification of the received wideband signal,which can extract the more comprehensive time-frequency information from the signal.And then the paper puts forward the high resolution direction finding algorithm based on this model,and it deduces the Cramer-Rao Bound of the algorithm,then simulation experiments is constructed to analysis the error performance and multi-resolution performance of the algorithm,and verifies the robustness and high resolution properties of it.Again,this paper proposes a regional positioning system design scheme,the system mainly realize the sound source target signal detection and location function.The first step is to carry out detection on the received signal to determine whether there is a target signal,we extracts frequency spectrum feature and implements the target source positioning if there is any target signal source,and for both the near-field and far-field,it proposes respectively the geometric positioning model and spectrum estimation model,then it gets the target source localization results by regional spectrum peak search.In order to ensure spectral resolution as well as reduce the computational cost,first using low resolution spectrum estimation get rough location estimation,then for high resolution spectrum estimation in the above estimated position of small area and achieves the final positioning.Finally,the paper uses the existing array device of research group,and it designs a positioning simulation experiment for simulation of the explosion and whistle positioning experiment which carried on the semi-anechoic laboratory.The experiment proved the practicability of the positioning system.