Study On Radar Dim Moving Target Detection Algorithm

This dissertation launches an in-depth study of single and multi-target detection based on the theory of long time pulse integration. The mono-and-bi-static radar echo signal model is established and the impact of long time integration on maneuvering targets is analyzed. Furthermore, an extended Keystone transform-based algorithm for dim maneuvering targets, a motion compensation-based algorithm for bistatic radar and a multi-target detection algorithm based on Radon transform and signal separation are proposed. This dissertation mainly focuses on the following aspects of work:1. A novel dim maneuvering target detection algorithm based on extended Keystone transform is proposed. Firstly, the echo signal model of dim maneuvering target is established, the cause of range cell migration and Doppler spread is analyzed. Secondly, the limitation of traditional keystone transform for maneuvering target is researched and the nonlinear phase estimation is performed by Taylor expand and bilinear regression. Then extended Keystone transform is applied to range migration and Doppler spread compensation. Finally, the signal’s coherent integration is realized by 2D-NUFFT. Simulation results present the effectiveness of the proposed algorithm.2. A target detection algorithm based on envelope peak dynamic clustering is studied for bi-static radar. The pulse compression and sliding-window method is used to maximize the peak of echo signal. The dynamic clustering algorithm is studied to construct range compensation function for RCM correction. The accumulation of signal power is performed by FFT along slow time direction. Simulation results show the effectiveness of the proposed method at high SNR case.3. A target detection algorithm based on modified Keystone transform is studied for bi-static radar. The range cell migration correction method based on Keystone transform is studied. As for the problem introduced by under-sampled data, an ambiguous factor searching method is proposed based on Radon transform. Simulation results show the proposed algorithm can have a good detection performance under low SNR.4. A novel dim maneuvering multi-target detection algorithm is proposed based on Radon transform and signal separation. The feature of multi-target’s echo signal is analyzed and pulse compression is realized by “STRETCH” technique. The range curve caused by target’s acceleration is corrected by second-order keystone transform. As the parameters of targets are different, a signal separation method is proposed based on Radon transform. Simulation results demonstrate the effectiveness of the proposed algorithm.