Yang Jun

Distributed adaptive OSWCA CFAR detector

In the sense of likelihood ratio test (LRT), a weighted adaptive OSWCA CFAR detector using distributed sensors is proposed in this paper. In the scheme, a local test statistic, which is the ratio of its test sample level and a designated order statistic (OS) of its reference samples, is calculated by each sensor, and then each sensor transmits its local test statistic to the fusion center. At the fusion center, the final decision is made based on weighted cell averaging algorithm (WCA). Since the weights are adjusted according to different signal-to-noise/clutter ratio (SNR), adaptively, the proposed detector can be available in the case of that the target echo and noise/clutter that have different level for each sensor. Unlike many current distributed CFAR detectors, which assuming that all noise/clutter samples satisfy independent identical distributed (IID) and have same SNR for each sensor. Meanwhile, for a Rayleigh fluctuating target in Gaussian noise of unknown level, we obtain its closed-form expressions for the false alarm probability and the detection probability. The numerical analysis results indicate that the proposed OSWCA scheme can be available.

A fast complex linearized Bregman iteration algorithm and its application in ISAR imaging

To implement fast reconstruction of the multiple measurement vectors (MMV) problem under arbitrary sparse structures, a fast complex linearized Bregman iteration (FCLBI) algorithm is proposed and applied to inverse synthetic aperture radar (ISAR) imaging. First, a signal model of an arbitrary sparse structure is established and its signature is analyzed. Second, an iterative formula for the FCLBI algorithm is deduced in a complex domain to recover the signals of the arbitrary sparse structure, thus extending its universality for complex-valued data. Third, by combining stagnation step estimation and sensing matrix optimization, the total iterative numbers are decreased to improve computational efficiency. Finally, the algorithm is applied to ISAR imaging, which reduces imaging time. Simulations and experiments with real-world data show the effectiveness and robustness of the proposed algorithm.

A Modified RBF Neural Network and Its Application in Radar

Aiming at the problem of parameter estimation in radar detection, a modified RBF neural network is proposed to estimate parameter accurately because of its good approximation ability to random nonlinear function and quick convergence speed. Two classical detection methods, which widely used in radar field, are listed in this paper, and their corresponding parameters are estimated with modified RBF neural network. Theoretical analysis and numerical results both show that the proposed method has good parameter estimation accuracy and quick convergence speed.

Performance of Distributed OSGOR CFAR Detector with Nonconherent Integration

A novel distributed detector, called OSGOR detector, is proposed for multiple pulse noncoherent integration. In this detector, each sensor transmits its test cell sample and a designated order statistic (OS) of its reference observations surrounding the test cell to the fusion center. At the fusion center, the global noise power level is obtained with greatest of (GO) processing. Without knowing the level of Gauss noise, the closed-form expressions of probabilities of false alarm (Pfa) and probabilities of detection (Pd) about the four types of Swerling target fluctuating models (I, II, III and IV) are derived. The proposed detectors detection performance is performed with numerical simulation

Detection Performance Analysis of Distributed OSCA CFAR with Nonconherent Integration

A novel distributed detector, called order statistic cell-averaging (OSCA) detector, are proposed in the case of noncoherent integration, In this detector, each sensor transmits its test cell sample and a designated order statistic (OS) of its reference observations surrounding test cell to the fusion center. At the fusion center, the global noise power level is obtained with cell-averaging (CA) processing. Under the assumption of chi-square family target model and exponential clutter for independent square-law detector, the exact formulas of probability of false alarm (Pfa) and probability of detection (Pd), are derived. Finally, the corresponding detection performance is calculated with numerical analysis

Tearing Properties Investigation of a New Airship Envelope Fabric Based on Single Edge Notched Film Experiment

In the present work, a new phenomenal model is proposed for modeling and simulation of tearing mechanical properties on an airship envelope fabric. The greatest advantage of the phenomenal model is that such model is capable of calculating the tear strength of a plain weave fabric by simply uniaxial tearing tests rather than the expensive and complex pressurized cylinder tests. The geometric progression technique is employed to calculate the stress distribution on the crack tip. The stress transfer coefficient Q is derived to be 0.9656. Three types of SEN specimens are researched during the uniaxial tearing tensile tests. The effects of specimen width and the crack length ratio on the tear strength of the airship envelope fabric is studied thoroughly. A good agreement between such model and experimental data are obtained, and the deviation is lower than 5%. Meanwhile, compared with the new phenomenal model and the Thiele’s formula, this new phenomenal model is verified to be more accurate while calculating the tear strength of a plain weaving airship envelope fabric in the uniaxial tearing tensile load.

A fast sequential orthogonal matching pursuit algorithm with threshold for rapidly spinning targets

Limited echoes and large Doppler bandwidth in a low PRF radar system is a challenging problem for rapidly spinning targets imaging. To simultaneously improve the efficiency and accuracy when imaging the spinning targets, this paper proposes a novel fast sequential orthogonal matching pursuit algorithm with threshold (FSOMP-T). Firstly, the complexity is reduced by decreasing total iteration numbers and the computational burden in each iteration. Secondly, a novel threshold is designed to eliminate the false reconstructed scatters, which can help to improve the image accuracy under low SNR. Simulations demonstrate the effectiveness of the proposed algorithm.

A novel CFNN model for designing complex FIR digital filters

A complex feedforward neural network (CFNN) model is proposed via energy function optimum theory and the complex frequency response of complex finite impulse response (FIR) digital filter. Then the convergence of the model is proved, and the universality of the model is studied. Simulation results show that the proposed CFNN model can achieve a good approach to an arbitrary digital filter and has universal performance.A complex feedforward neural network (CFNN) model is proposed via energy function optimum theory and the complex frequency response of complex finite impulse response (FIR) digital filter. Then the convergence of the model is proved, and the universality of the model is studied. Simulation results show that the proposed CFNN model can achieve a good approach to an arbitrary digital filter and has universal performance.

The method to inverse-simulate spaceborne MPC SAR data based on common airborne SAR data

This paper proposes a method to generate simulation data of spaceborne MPC SAR from real airborne complex image SAR data. This method can be used for testing the theoretical analysis result of spaceborne MPC SAR.