Directions-of-arrival Estimation of Coherent Cyclostationary Signals Based on Deep Learning
In this paper,we focus on the problem of direction-of-arrival(DOA)estimation with applications of deep learning in the case of small signal-to-noise ratio and non-stationary signals.The paper adopts linear equidistant array to receive coherent cyclosta-tionary signals,calculates the cyclic autocorrelation function of cyclostationary signals and constructs the data vector matrix,then de-composes the obtained matrix by vector singular value method,and finally inputs the decomposed matrix into convolutional neural network to obtain the DOA estimation result.Compared with the traditional DOA estimation algorithm,the convolution neural net-work has less estimation time.The simulations show that the root mean square error of the algorithm is reduced by 1 degree com-pared with the existing optimal algorithm in the environment of non-stationary signal and low signal-to-noise ratio.
DOA estimationSVD algorithmMUSIC algorithmcoherent signalcyclic stationary signalconvolutional neural net-work
国家科技期刊平台
NSTL
万方数据
