Meiguo Gao

Bistatic three-dimensional interferometric ISAR image reconstruction

A three-dimensional (3D) bistatic inverse synthetic aperture radar (ISAR) imaging method is proposed in this paper. The proposed method makes use of interferometry and technically speaking, produces a 3D target reconstruction by estimating a scattering center position in 3D Cartesian space. The proposed method makes use of a combined ISAR/interferometry technique that also allows the ISAR image plane to orientation to be estimated. Cross- or L-shaped antenna configurations are discussed, and the effects of the baseline length along the horizontal and vertical direction on the scatterers position estimation are analyzed in detail. Finally, numerical simulations are used to evaluate the proposed methods performance.

Robust adaptive beamforming based on the effectiveness of reconstruction

The beamformers by joint estimation of the steering vector (SV) and the reconstructed interference-plus-noise covariance (INC) matrix have been investigated. However, few of them concerned about the effectiveness of reconstruction. In this paper, a novel beamformer is proposed, which introduces the improve factor (IF) to evaluate the improvement scale of output signal-to-interference-plus-noise ratio (SINR) with and without reconstruction. By exploiting the relationship between the IF and input signal-to-noise ratio (SNR), it can be made a trade off between the extent of the performance improved and computational load that reconstruction required. Therefore, the SV is estimated at first, and then the level of SNR determines whether the INC matrix is required to reconstruct. During the implementation, a spherical uncertainty set and the sparsity of interference direction are used. Simulation results indicate that the proposed beamformer has better performance compared with existing adaptive beamforming algorithms. HighlightsThe beamformer based on joint estimation is investigated.The relationship between the improve factor (IF) and SNR is exploited.The novel beamformer based the effectiveness of reconstruction is proposed.

The hardware platform design for DRFM system

The hardware platform design for digital radio frequency memory (DRFM) is the analogue-digital combined circuit with high speed data transfer speed, and the design quantities of the hardware platform significantly affect the overall performance of the whole system. The hardware requirement for general purpose DRFM system is analyzed and the hardware architecture based on FPGA and DSP techniques is proposed. The general principle for high-performance hardware design is withdrawn based on the engineering implementation experience. The entire design example with high speed data sample rate and real-time modulation ability is illustrated, and several important circuits example is given. The ideal performance of the example board guarantee the correctness of the design idea, and the concept stated in the article is valuable for implementation reference.

Realization of high data rate DUC based on FPGA

This paper discusses the issues on “how to save multiplier resource” and “how to cope with high data rate”, when implementing Digital Up Converter (DUC) with Field Programmable Gate Array (FPGA). A novel implementation structure of poly-phase interpolation filter is proposed, which could reduce the consumption of multiplier by half. Parallel processing is adopted in designing Numerical Controlled Oscillator (NCO), which could generate high data rate LO required. By following these methods, a DUC module is designed for certain radar IF echo simulator, which could generate digital IF signal with data rate up to 1.2Gsps.

A Nearest Neighbor Fuzzy Classifier for Radar Target Recognition Using Combined Features

This paper presents a nearest neighbor fuzzy classifier (NNFC), which is very suitable to process the combined features with different data types and scales. The NNFC does not require the combined features with the same data types and scales, and it is not necessary to perform any pre-processing. It uses the fuzzy membership function to process each feature of the combined features. Experiments with real satellite data show that the NNFC can effectively perform radar target recognition of multiple features combination

Parameters selection of FBLMS algorithm for noise FM-CW radar

The direct wave and clutter wave cancellation is a key procedure for noise FM-CW (frequency-modulated continuous-wave) radar. Frequency-domain Block LMS (FBLMS) adaptive filters have better performance of clutter (include the direct wave) cancellation in low computational complex. In practical system, however, the chosen of step-size and the tap length of FBLMS are extremely important for the subsequent target detection. Although, the step-size bounds of FBLMS have been derived and presented by different expressions, the drawbacks of them as follows: on the one hand, the step-size bound are usually complex expressions including eigenvalues which are related with the covariance matrix of the input signal, it has high computational complexity to calculate the step-size; on the other hand, the relationship between step-size bound, the tap length of FBLMS and clutter attenuation (CA) has been not discussed comprehensively. To address above-mentions, in this paper, the step-size bound is re-derived based on Feuers work. Furthermore, the relationship between the step-size, the tap length and CA is elaborated, it shows that the CA is dominated by the tap length of FBLMS, once the tap length is not less than the number of clutter path, the CA will be decreased as the step-size increases. In addition, the relationship is intuitively shown by three-dimensional plot. Numerical results show that the parameters selection of FBLMS is beneficial to implement in the system of noise FM-CW radar.

Sequential Separation of Radar Target Micro-Motions Based on IRT

A micro-motion sequential separation method based on inverse Radon transform (IRT) is proposed. The method has an accumulation effect and, therefore can adapt to environment with relatively low signal-to-noise ratio (SNR). This method has the capability of antidiscontinuous tracking and thus is suitable in the multi-target tracking scenario. The method is a good candidate in large dynamic range scenes to detect relatively weak scatterers of micro-motions. It is more computationally efficient in comparison with the methods based on the Hough transform (HT).

Space object identification based on narrowband radar cross section

A space object identification method based on narrowband radar cross section (RCS) is proposed in this paper. With RCS sequences, statistical features are extracted, the two-dimensional sizes of the object are estimated by adopting an improved size estimation model (SEM) and the attitude stability of the object is determined by periodicity analysis. Feature templates are constructed in matrix form with those three kinds of features of the training sample set and especially when constructing the template for a large number of statistical features, dynamic fuzzy clustering technology is used. The k-nearest neighbor fuzzy classifier is adopted to accomplish the identification. At the end, experiments with the measured data are conducted to confirm the effectiveness of the method.

A phase compensation method for azimuth interpolation based on zero-padded DFT in interferometric ISAR imaging

In interferometirc inverse synthetic aperture radar (InISAR) imaging, the azimuth interpolation based on zero-padded DFT changes the phase of each scattering center. This paper studies the adverse effects of zero-padded DFT method on the phase information. A phase compensation method to eliminate the phase error caused by zero-padded DFT is proposed. Besides, the effects of adding a hamming window before azimuth DFT on this method is also investigated. Simulation results show that the theoretical analysis is correct and the proposed phase compensation method is effective.

Direction estimation for two steady targets in monopulse radar

Traditional monopulse radar cannot resolve two targets present in one range and Doppler cell by means of the monopulse ratio. A novel algorithm is proposed to estimate the directions of two steady targets with two pulses. The algorithm has a closed-form expression and its variance is derived at high signal-to-noise ratios (SNRs). Furthermore, the pulse pair selection criterion and the estimation method with multiple pulses are given. Finally, some numerical results are shown to validate the proposed algorithm and the effect of slight target fluctuations is tested.