Near-field source positioning method based on factor analysis

Abstract

The invention discloses a near-field source positioning method based on factor analysis, and aims to solve the problems that a traditional subspace method is complex in calculation, cannot perform real-time processing and is poor in parameter estimation performance under a low signal-to-noise ratio. According to a neural network method, upper triangular elements of a covariance matrix of trainingsample signals are generally used as features of the signals to perform network training, and in a large array with a large number of array elements, the upper triangular elements of the covariance matrix of the signals are used as input signal features, so that the complexity of a neural network is improved, and the network training time is prolonged. Therefore, the invention provides the near-field source signal positioning method for dimension reduction by using the factor analysis method. According to the method, a few reconstructed feature variables are used for replacing original featurevariables to research and analyze things, so that the feature dimension of the network input signals is reduced, the input signal features after dimension reduction are used for training of the neural network, the training speed is increased, the real-time performance of the algorithm is high, and the engineering application value of the method is enhanced.

Description

technical field [0001] The invention belongs to the technical field of array signal processing, and in particular relates to a factor analysis method for locating near-field sources, which simplifies the network structure and reduces the amount of calculation. Background technique [0002] Direction of Arrival (DOA) plays a very important role. Traditional DOA estimation mainly uses the Multiple Signal Classification (MUSIC) algorithm, the Rotation Invariant Subspace (Estimation of Signal Parameters via Rotational Invariance Techniques, ESPRIT ) algorithm and its derivative methods. Due to the large amount of calculation required by the traditional angle of arrival estimation algorithm, poor prediction performance under low signal-to-noise ratio, and poor estimation adaptability to the actual application environment, most of these estimation algorithms mainly stay in theory and simulation, and it is difficult to be widely used. applied in practical engineering. With the po...

AI Technical Summary

Problems solved by technology

Due to the large amount of calculation required by the traditional angle of arrival estimation algorithm, poor prediction performance under low signal-to-noise ratio, and poor estimation adaptability to the actual application environment, most of these estimation algorithms mainly stay in theory and simulation, and it is difficult to be widely used. applied in practical engineering

Images