The Research Of Source Enumeration Of ULA Based On Combination Hankel Matrix Of Spatial Cross-Correlation

Source Enumeration is an important research direction in array signal processing and has received extensive attention in recent years.In the actual environment,the number of signal sources is often unknown.If the estimated number of sources does not match the actual number,the performance of most of the direction of arrival estimation algorithms will deteriorate or even completely fail.Therefore,this article focuses on the issue of the number of signal sources.In the meantime,an improved source enumeration method based on Hankel matrix with high-dimensional combination is proposed.Firstly,the basic knowledge of array signal processing is introduced,including array model,signal model and noise model.Then,some classical methods for source enumeration were introduced,including the Eigen-decomposition algorithm,information-based method,smooth-ordered column method,the Gai's circle method for colored noise,etc.The performance of each algorithm is analyzed and compared with the literature.Secondly,on the basis of the uniform linear array model,the structure and properties of the Hankel matrix in the source enumeration method based on the singular value decomposition of Hankel matrix with array element received covariance function are analyzed and introduced in detail,after which,an improved method for constructing high-dimensional combined Hankel matrix is proposed and the combined matrix is analyzed.Analysis and simulation experiments show that when using the improved high-dimensional combinatorial Hankel matrix algorithm to deal with coherent signals,the improved algorithm can increase the maximum number of source enumeration,while retaining the good de-coherence capability of the original algorithm.The performance of the algorithm is effectively improved.Finally,this paper analyzes and summarizes the regularity of the singular value in the algorithm,then proposes the idea of the adaptive threshold,and gives the threshold formula which is suitable for a certain condition.Simulation experiments show that the performance of the algorithm based on singular value decomposition of high-dimensional combined Hankel matrix using adaptive threshold is further improved,and it has good gradual consistency.