The Prediction Model of High-Frequency Surface Wave Radar Sea Clutter with Improved PSO-RBF Neural Network

Abstract

The sea clutter of high-frequency surface wave radar (HFSWR) has chaotic characteristics. Using the phase space reconstruction method to extend the one-dimensional sea clutter time series to the multi-dimensional phase space to fully demonstrate the internal dynamics of the sea clutter, and then training the Radial basis function (RBF) neural network to learn the internal dynamics of sea clutter and establishing the prediction model. The initial parameters of the network affect the convergence speed and the accuracy of the network model, so the particle swarm optimization (PSO) algorithm is used to optimize the initial parameters of the RBF neural network. Aiming at the PSO algorithm problems of the slow convergence speed and easily getting into local optimum, this paper proposes an improved PSO algorithm based on stage optimization. The simulation results show that the improved PSO algorithm has higher convergence accuracy; the optimized RBF neural network prediction model has higher stability and accuracy and has a better prediction effect on sea clutter.