Research And Implmentation Of The Key Algorithms Of FMCW Radar Signal Prosessing For Ship

FMCW(frequency modulated continuous wave) radar has the advantages of simple structure, low peak power, high range resolution, no blind area, and so on, which can be applied to the volume, weight and power limited platform. However, in the field of marine navigation radar, due to the impact of sea clutter, its advantages are not fully reflected, there are also clutter suppression, clutter target detection and estimation and other issues need to be resolved.In order to increase the radar ranging accuracy effectively, the main content of this theses is around the process of FMCW radar difference signal processing, which mainly includes the following three aspects.1) As the sea clutter prediction algorithm based on GRNN in the traditional experimental method has the defects in the parts of estimating the parameters, this theses proposed a new sea clutter prediction method, which is based on PSO-GRNN algorithm. the proposed new algorithm based on PSO algorithm can more easily find group optimal solution, and simple and practical characteristics, use it to find GRNN prediction of sea clutter wave Gaussian function bandwidth coefficient, and then according to the theory of GRNN to forecast sea clutter. The simulation results show that the improved method can effectively improve the accuracy of sea clutter prediction and suppression.2) The theses proposed an improved CFAR detection algorithm STMVI-CFAR. The algorithm based on VI-CFAR detection algorithm has combined characters of TM-CFAR and S-CFAR detection algorithm, in different sea clutter environment adaptive selection of different detection methods. In this paper, the proposed algorithm is able to detect the target signal in a homogeneous environment, a multi- target interference environment and a clutter edge environment. Finally, in the MATLAB software platform, through simulation experiments, analysis the detection performance of the proposed improved algorithm in different background, and compared with other existing detection algorithms.3) In order to further improve the accuracy of radar measurement, firstly through the research on the principle of several existing spectrum analysis method, and the simulation is carried out based on the ideal differential frequency signal. And then proposes an improved spectral estimation algorithm, the algorithm combined the fitting method with Chirp-Z algorithm, verified by simulation experiments. The experiment results show that the proposed algorithm has higher ranging accuracy than other methods, and through extensive Monte Caro simulation experiments to verify the stability of the algorithm.4) A simple FMCW radar signal processing system is realized by simulation, which is based on the simulation of the signal source, the acquisition of the difference frequency signal, the detection of the target and the final analysis of the spectrum. The effectiveness of the proposed algorithm is further verified.