Research On Spatial Parameter Estimation Of Distributed Sources Based On Finite-rate-of-innovation

Due to the complexity of the environment in the signal propagation process,the openness of the channel,and the unfixed user location,the high-rank signal model corresponding to the multipath scattering phenomenon—distributed sources are widely used in the actual environment,and then distributed estimation of spatial parameters of the type source has also become a hot issue in the field of array signal processing.The airspace parameter estimation method of the traditional distributed signal source can only accurately estimate the airspace parameter under the conditions of high signal-to-noise ratio and multiple snapshots.However,when the amount of data transmission is large and the calculation complexity of the algorithm is high,it is difficult to quickly and accurately restore the airspace parameters.To solve this problem,a method for estimating spatial parameters of distributed sources based on sparse representation has been proposed.This type of method reduces the amount of data transmission and can restore the spatial parameters more accurately,but it also introduces the constraints of the spatial grid.Great limitations.The Finite Rate of Innovation(FRI)sampling theory proposed in recent years can use the FRI sampling structure to directly sample and reconstruct continuous signals at a rate lower than the Nyquist sampling frequency,thereby effectively reducing the pressure in the process of sampling,processing,storage and transmission,and get rid of the constraints of the spatial grid.However,the current DOA(Direction of-Arrival)estimation method based on FRI theory can only be applied to point source signals,and cannot be applied to spatial parameter estimation of distributed sources.The VPW-FRI(Variable Pulse Width-FRI)model has strong characterization capabilities.The model can accurately describe complex distributed sources by adjusting the pulse width parameters,amplitude and number of asymmetric pulses.This article uses VPW-FRI signal model combines the characteristics of the FRI signal,and proposes a distributed source spatial parameter estimation method based on the FRI theory.This article mainly focuses on the following three parts:(1)Research on DOA estimation of point source signal based on FRI theory.First study the basic theory of FRI sampling,the basic sampling framework and reconstruction algorithm of common FRI signals;secondly,study the most classic point source signal estimation method based on FRI theory—FRIDA method,study the practical application background of this method,and construct the point source signal estimation method.The model of the source signal arrival angle estimation and the reconstruction model of the point source signal have completed the theoretical derivation of the FRIDA algorithm and verified the feasibility of the method through software simulation;finally analyzed the number of shots,noise(signal-to-noise ratio)the influence of other factors on the reconstruction accuracy of FRIDA estimation method.(2)To study the establishment of the distributed source model,firstly,the reasons for the generation of distributed sources are studied,and the characteristics of the traditional distributed source models are analyzed;secondly,the VPW-FRI model,AP-FRI(Asymmetric Pulses-FRI)model,Fr HT-FRI(Fractional Hilbert Transform-FRI)model are three of the most classic non-ideal pulse signal models.The characteristics of the three models are compared and analyzed.Finally,VPW-FRI model and AP-FRI are used respectively.Model,Fr HT-FRI model to fit the distributed information source and compare with the original signal,choose the best fitting model to describe the distributed information source.(3)Combining the VPW-FRI model with the FRIDA algorithm,a method for estimating the spatial parameters of distributed sources based on the VPW-FRI model(VPW-FRIDA)is proposed to realize the estimation of the central direction of arrival and the spread angle of the distributed sources.The VPW-FRIDA algorithm is implemented through software simulation to verify the feasibility of the method.In order to further analyze the factors that affect the reconstruction of the VPW-FRIDA algorithm,the effects of factors such as the number of snapshots,the spread angle,and the signal-to-noise ratio on the spatial parameter estimation of the distributed signal are respectively analyzed.Finally,the comparative experiments of the VPW-FRIDA algorithm,the FRIDA algorithm,and the traditional distributed source airspace parameter estimation method are carried out to analyze the performance of the VPW-FRIDA algorithm.