Research On Sea Clutter Suppression Method Of HFSWR Based On RBF Neural Network

As a new type of ocean detection radar,the detection range of HFSWR is not affected by the curvature of the earth,and it can realize the over-the-horizon and all-weather monitoring of the ocean surface.At present,HFSWR is mainly used in the detection of marine targets and marine dynamic parameters,which plays an important role in the military and civil fields.When HFSWR is working,high-intensity sea clutter often submerges the target signals,which has a great impact on the detection accuracy of radar at sea.The effective suppression of sea clutter becomes the key to improve the detection accuracy of radar.In this paper,based on the chaotic characteristics of sea clutter,the prediction model of sea clutter is established by using RBF neural network.The sea clutter is predicted by using the prediction model.In time domain,the sea clutter is effectively suppressed by subtracting the predicted sea clutter data from the radar echo signal.In addition,the intelligent optimization algorithm is introduced and improved to optimize the RBF neural network,improve the accuracy of the prediction model,and achieve more effective suppression of sea clutter.Around the above content,the main work of this paper is as follows:Firstly,the formation mechanism of sea clutter is introduced.Based on the measured data,the chaos of sea clutter is proved by experiments;the theory of phase space reconstruction is introduced,the prediction equation of sea clutter is derived,and the phase space reconstruction parameters of sea clutter data are calculated by various classical algorithms,which can be used as the basis to determine the number of input nodes of neural network.Then,the RBF neural network is constructed to learn the sea clutter prediction equation through training;in order to improve the robustness and accuracy of the network,the classical particle swarm optimization algorithm is improved to optimize the initial parameters of the RBF neural network.The experimental result shows that the improved PSO algorithm has higher optimization accuracy,and the optimized network prediction model has faster convergence speed and higher prediction accuracy for sea clutter.Finally,in order to further improve the prediction accuracy of RBF neural network,the WOA and the GWO algorithm are introduced respectively,and the shortcomings of each algorithm are improved.The initial parameters of RBF neural network are optimized by the improved algorithms.The optimized network learns the real part and imaginary part of sea clutter,and establishes the prediction models of real part and imaginary part of sea clutter respectively;the simulation results show that the improved optimization algorithm has faster calculation speed,faster convergence speed and higher optimization precision.The prediction accuracy of the network model optimized by the improved algorithm is further improved compared with the network model optimized by the traditional algorithm,and it has a better performance in the prediction and suppression of sea clutter.In addition,we add analog target signal to radar echo,the simulation experiment shows that the algorithm in this paper can effectively suppress sea clutter and retain the target signal well.This study has important theoretical significance and application value for realizing sea clutter suppression and improving the detection accuracy of radar to sea targets.