Research On DOA Estimation Technology Of Cyclostationary Signal Based On Coprime Array

Directional-of-arrial(DOA)estimation technology is one of the key technologies for the integration of 5G and the Internet of Things,among which the super-resolution subspace class estimation algorithms achieve better estimation performance.However,the performance of uniform array still needs to be improved because of the disadvantages such as small aperture,large mutual coupling error between array elements and low resolution of array.As a new type of sparse array,the coprime array has the advantages of large array aperture when the number of elements is equal,less cost of the element at the same aperture,small mutual coupling error of the elements,and high resolution,and it can obtain more excellent array signal processing gain.Therefore,DOA estimation method combined with coprime array has attracted extensive attention from experts and scholars at home and abroad.A large number of signals in the communication system have timedomain cyclostationary characteristic.Combining this characteristic with the DOA estimation method can obtain signal screening and separation estimation ability,strong noise and interference suppression ability.At present,in the research field of DOA estimation based on coprime array,the research of coprime array combined with cyclostationary signal DOA estimation is still in the initial stage.There are few DOA solutions for specific signal scenarios,and the research on estimation methods combining array and signal characteristics needs further research.This paper takes the coprime array as a core,deeply explores the applicability of the coprime array in the DOA estimation of cyclostationary signals,and develops the research on the DOA estimation method of the cyclostationary signal based on the coprime array to improve the performance of the DOA estimation method.It mainly includes two aspects: the research on the calculation method of the fusion of formation characteristic and cyclostationary characteristic,and the research on the calculation method of combined cyclostationary characteristic and non-circular characteristic.The main research contents and innovations of this paper are as follows:1.Aiming at the problem that the accuracy of DOA estimation algorithm for cyclostationary signal is limited by the uniform array structure,the coprime array is combined into the DOA estimation of the cyclostationary signal,and a high-precision DOA estimation algorithm for the cyclostationary signal based on spectral peak fusion is proposed.Firstly,the receiving signal model of the reciprocal array is constructed,and the cyclic correlation function of the array elements is calculated based on the cyclostationary characteristics of the signal,and the array output data matrix is reconstructed.Secondly,the cyclic MUSIC algorithm is used to solve the DOA estimates of the sub-arrays respectively.Finally,the proposed algorithm uses the coprime characteristic between the solution sets combined with the idea of spectral peak fusion,the fuzzy solution problem of the sparse array element spacing is eliminated.Simulation results show that the proposed algorithm has higher estimation accuracy than the existing DOA estimation algorithms for cyclostationary signals based on uniform arrays.2.Aiming at the problem of the inability to achieve effective DOA estimation for cyclostationary signals under the condition of underdetermined estimation,the cyclic correlation matrix is vectorized and reconstructed to construct a virtual array,and an underdetermined DOA estimation algorithm for cyclostationary signals based on vectorized reconstruction is proposed.Firstly,the proposed algorithm constructs a coprime array receiving signal model,constructs a virtual array model through vectorization processing,and introduces the virtual array elements to achieve an improvement in the degree of freedom of the array;Secondly,the spatial smoothing technology is used to process the signal output matrix of the maximum continuous virtual array elements,and the cyclic MUSIC algorithm is used to solve the DOA;Finally,in order to reduce the computational complexity of the algorithm,the DOA value is solved by combining the root polynomial method without grid search.Theoretical analysis and simulation verification show that the algorithm can effectively improve the degree of freedom,and the computational complexity of the algorithm is low.The algorithm is based on a coprime array and uses vectorized reconstruction method to achieve effective estimation of useful signals under the condition of underdetermined estimation,breakthrough the limit that the number of estimable signals need to be smaller than the number of physical array elements,which improves the DOA estimation performance.3.Aiming at the problem of DOA estimation for cyclostationary signal with non-circular characteristic,combining the non-circular characteristic of the signal to construct a virtual extended array,a joint non-circular characteristic DOA estimation algorithm for cyclostationary signal is proposed.First of all,the algorithm establishes the array receiving signal model,using the non-circular characteristic of the incident signal on the statistical characteristics,and the array output signal vector and its conjugate are cascaded into a non-circular extended array output vector,which expands the equivalent aperture of the array model;Secondly,the cyclic correlation function of each array element are calculated according to the cyclostationary characteristic,and the receiving data matrix of the non-circular extended array is constructed.The DOA values of the subarray are obtained by using the subspace algorithm.Finally,the spectrum peak fusion method is used to eliminate the ambiguity values to obtain the true signal arrival angle value.Simulation experiments are set to evaluate the influence of the combined signal cyclostationary and noncircular characteristic of on DOA estimation performance.Simulation experiments show that the algorithm has higher estimation accuracy.At the same time,it is verified that compared with the existing algorithms,the proposed algorithm makes full use of the non-circular characteristic of the signal and effectively improves the DOA estimation accuracy of the algorithm.