Research On Detection And Angle Estimation Of Low-angle Target In Mmultipath

The detection and localization of targets are the most primitive and the most traditional function of radars, but with the continuous development and great progress of technology, modern warfare has become a three-dimensional war, which is comprehensive and has multi-levels and multiple directions. The invasion of the low-level aircraft or missile poses a great threat to ground targets. When detecting and tracking targets in low altitude areas, radar needs to launch beams with enough energy and the signal to clutter and noise ratio is reduced due to the existing strong sea clutters. In addition, the presence of multipath effect leads to a fact that the received echo signal is usually the superposition of the direct wave signal and the reflected wave signal(one or more) in strong clutter environment, which may be enhanced or declined and can affect the performance of target detection seriously. At the same time, the direct wave and the mirror echo are coherent in an antenna beam and the angle difference between them is very small. All these could make angle estimation difficult. Therefore, how to solve the problems of detection and estimation of Low-angle targets is very challenging.In this paper, the research on detection and angle estimation of low-angle targets in multipath is carried out. At First, the surface multipath model and the echo signal model of low-angle targets are established, and then the relationships between various parameters of the multipath model and echo amplitude are introduced in detail. After that this paper studies the detection methods of low-angle multipath targets, including M/N detection methods on multi-frequency echo and low-angle detection methods of netted radars. In the first one, detection probabilities of unfluctuating targets and fluctuating targets of Swerling III in the free space or multipath environment are derived, and we also compare the simulation results of M/N detection method with another one in the single pulse detection method. The experiment fully reflects the performance of the proposed approach. The study on the detection probabilities of the netted radars is mainly aimed at the two key techniques, including fusion rules and target detection methods of netted radars. First of all, three classical fusion rules are deduced, and the effects of detection are studied. Finally we analyze the detection of low-angle targets in multipath.In order to distinguish coherent signals and improve the angular resolution, our group proposes a low-angle estimation method based on compression sensing, which integrates the relationship between the multipath attenuation coefficient and echo angles. So we can apply this relationship to reconstruct the composite steering vector. This method can improve angular resolution in low signal-to-noise ratio compared with L1-SVD estimation algorithm and doesn’t need terrain information. Due to the large computation of the classical maximum likelihood, a new low-angle estimation algorithm is proposed, which is named multi-frequency maximum likelihood. This method adapts the comprehensive information on multi-frequency echo that signal phases of two echo change with the frequency points and can weaken the influence of some frequency points. It can also improve the resolution of the low elevation angle. Adopting the alternating projection algorithm to reduce the computational complexity is also worth mentioning. The alternating projection algorithm can transfer complex multidimensional search into two simple one-dimensional ones. Then the simulation based on the surface multipath model is verified. Netted radar system can improve the detection accuracy and reliability of the whole system through the fusion of the monostatic radar data. And this field has become a current research hot spot. At the end, this paper analyzes the key technologies of netted radar system: coordinate transformation and data fusion methods. It also gives some preliminary ideas on solving the key technologies and analyzes the application of the geographical coordinate transformation method. The positioning precision of the radar system is also illustrated. On the other hand, this paper studies angle measuring algorithm of netted radar system. In this algorithm, everyone in netted radar system can apperceive the observation space(or the target environment), and obtains target location information(such as distance, azimuth, pitching angle and so on). Because of the space diversity characteristics, netted radar system can improve the estimation performance and has a higher positioning accuracy by networking fusion. The more sparsely the radar stations distribute, the more obvious the netted radar system’s performance is.