Font Size: a A A

Research And Application On Sea Clutter Suppression Algorithm Based On GA-WNN Neural Network

Posted on:2012-07-31Degree:MasterType:Thesis
Country:ChinaCandidate:C X LiuFull Text:PDF
GTID:2218330368482313Subject:System theory
Abstract/Summary:PDF Full Text Request
Sea clutter is a back-scattering echo coming from the ocean surface by radar transmitted signal irradiation, whose existence puts a bad effect for radar in the capability of checking and following target on the ocean surface, which results in the performance of radar for detecting targets in strong sea clutter background and low false alarm probability rate situation is so bad that it even can't detect the target. So if sea clutter could be suppressed or diminished to some extent, the capability of detecting targets for ocean defense radar will probably be upgraded. With modern technology development, what is the key point in current ocean surface target detecting research, what is concerned by most researchers is how to suppress sea clutter by modern technology.In this paper, the improved gradient algorithm for training WNN (Wavelet Neural Network) is presented. Combined with phase-space reconstruction theory, actual measured sea clutter data is peformed to get a reasonable embedding dimension and embedding delay by the antocorrelation function method, Mutual information method, G-P (Grassberger-Procaccia) and Cao methods respectively. The results will be used to determine the number of input node. In this paper, GA (Genetic Algorithm) is introduced to improved network. The GA is used to optimization the WNN of initial weights and threshold etc. A better search space is determined, the optimal solution or near optimal solution is finded quickly. In this paper, the inner dynamical characteristic of sea clutter is studied by the GA-WNN, the predicting equation entailing the inner dynamical characteristic is attained.The results of emulation experiment, which is based on the real measured datas from the IPIX (Intelligent Pixel-Processing) radar, show that the two algorithms proposed in this paper, small weak target signals could be detected easier in sea clutter background, Compensation and suppression of sea clutter is attained. Those algorithms both do well in suppressing sea clutter.
Keywords/Search Tags:Sea clutter, Phase space reconstruction, Genetic Algorithms, Wavelet neural network, Suppression
PDF Full Text Request
Related items