Research On Dim Target Detection Algorithm With Monopulse Doppler Radar

The flyers, such as stealth aircraft, Anti radiation Missile, and Cruise Missile and so on, are widely used in modern war. The radar echoes of them are much weaker than those of traditional flyers. The signal to noise ratio(SNR) is too low to detect target reliably with general algorithms. Although by increasing the transmitter power, antenna aperture would improve SNR, the concrete realization is always subject to many reality conditions, the project implementation is very difficult, even impossible. Therefore, digging more useful target information from Frequency domain, Time domain and Spatial domain, and through advanced signal processing to detect weak target become necessary technical meansThe dissertation focuses on the problem of weak target detection with Monopole Doppler radar. In this paper, we fully use helpful information of three channels of radar (such as the angle error, complex phase angle and track etc.) to distinguish target from noises. New algorithms effectively improve radar's detection capability. This paper mainly covers the following aspects:Chapter 1 introduces the background and significance of the problem to be studied, and briefly introduces the concept of Monopulse and PD radar besides their working elements. At last the corresponding techniques and research trends at home and abroad are described.Chapter 2 is the signal detection basic theory of Monopulse PD radar. Firstly, the signal echo receiving model of Monopulse PD radar is presented, then introduces the typical flow of signal detection under the system. Finally, analyzes and summarizes the characteristics of weak target problem, proposes several possible methods to solve the problem.Chapter 3 is the new algorithm designing which is based on difference channel information. Firstly, the analysis of angle error measurement under Monopulse system is proposed, and the possible way of increasing the difference of angle distribution between the target and noises is discussed. Secondly, the test statistic based on angle error is constructed, and in the traditional non phase coherent integration algorithm, the test statistic auxiliary detection branch is added. The simulation results show that the new algorithm improves performance of primary non coherent detection algorithm. In addition to angle error information, difference channel also have phase information, which is useful information for detection too. Therefore, dim target detection algorithm based on neural network of which inputs are normalized signal amplitude, angle error and phase of angle is proposed. New algorithm attempts to make fully use of three channel information, and combine the powerful classification ability of probabilistic neural network to better distinguish between the target and noise. Simulation results show that the algorithm has better performance compared with non coherent integration algorithm.Chapter 4 is Track Before Detect algorithm based on dynamic programming in high PRF PD radar. Firstly, target motion model and observation model are briefly introduced. Secondly, the principles and processes of TBD algorithm based on dynamic programming are described, and DP TBD algorithm in HPRF mode is proposed. At last, with the angle error information a new DP TBD algorithm which process in the Angle Doppler Time three dimensional space is proposed, Simulation results show that the new algorithm can further improve the detection capability.