Research On Stealthy Targets Detecting Based On Microwave Radiation And Spatial Spectrum Estimation

How to effectively detect stealthy targets has become a critical problem as far as national defense issues are concerned. The microwave radiometer detection approach, which is highly invisible and barely influenced by weather or battlefield environment, has proved itself a potent way to detect targets passively. The antenna array constructed of several small antennas has a very high space resolution which is equivalent to that of a big antenna. While the space resolution of traditional array signal processing algorithms, such as inversion algorithm of synthetic aperture microwave radiometer, are determined by the maximum baseline of antenna array.In this paper, spatial spectrum estimation algorithm is employed to calculate the directions of microwave radiation signals received by the microwave radiation array systems. A novel spatial spectrum estimation method for passive detection is proposed. By using this method a super resolution-beyond resolution of the hardware system- can be achieved. Furthermore, with the same resolution performance, system complexity and equipment expenditure can be reduced by sparse array arrangement.Based on the analysis of microwave radiation signals characteristic and the atmosphere channel, a transmission model of stealthy targets, atmosphere channel and microwave radiation detecting system is established, which can show the distribution of stealthy targets' and the background's brightness temperature. The validity of this model is proved by the apropos simulations and the experiments.The intrinsic relationships of those common spatial spectrum estimation algorithms are discussed. And the performances of some algorithms are analyzed and simulated. By applying spatial spectrum estimation algorithms to the sparse array system, the problem of DOA (direction of arrival) ambiguity which results from array manifold ambiguity is discussed. Based on that, a novel algorithm is proposed to calculate the manifold dimension of a linear array arranged by arbitrarily integral multiple of half-wavelength, which can easily find out the manifold dimension and all the possible ambiguous angles due to the impinging of multi-signals. This method can be used to resolve the problem of the DOA ambiguity by removing some sensors of certain positions in the array. Furthermore, a model based on spatial spectrum estimation and the radiation signals is proposed to detect stealthy targets. The performances of this model, including sensitivity and space resolution, are analyzed and comparisons are made between our model and others employing other signal processing methods. The results demonstrate that our model is superior to others in the aspect of resolution and even a super-resolution could be achieved.In the practical scenarios, there are many inevitable errors which would deteriorate the performance of spatial spectrum estimation algorithms. Therefore, error calibration algorithms must be investigated to model the practical environments. A brief review of various errors models and errors calibration algorithms is first presented. Considering the weakness of radiation signals' energy, a new array errors model for low SNR scenario is proposed, then two calibration methods based on this model are further presented. In the situation of low SNR and in the presence of practical errors, the proposed array errors model and the calibration methods can facilitate the application of spatial spectrum estimation algorithms in the microwave radiation array systems effectively, which is proved by the simulations and the experiments.