Xiutan Wang

Maximum likelihood parameter estimation of multiple chirp signals by a new Markov chain Monte Carlo approach

In this paper, a novel method for estimating the parameters of multiple chirp signals in additive Gaussian white noise is proposed. The method combines a global optimization theorem with a new Markov chain Monte Carlo algorithm, called the simulated annealing one-variable-at-a-time random walk Metropolis-Hastings algorithm. It is a computationally modest implementation of maximum likelihood estimation and has no error propagation effect. Simulation results show that the proposed method can give good estimates for the unknown parameters, even when the parameters of the individual chirp signals are closely spaced and the Cramer-Rao lower bound can be attained even at low signal-to-noise ratio.

Frequency estimation of the sinusoidal signals with unknown lowpass envelopes based on the eigenanalysis

In this paper, the frequency estimation of the sinusoidal signals with unknown lowpass envelope is addressed. Due to mismodeling, the performance of the conventional subspace-based method degrades significantly in these cases. By developing the method applied to the parametric localization of distributed sources, an eigenanalysis-based method is proposed for the frequency estimate. The comparisons of the proposed method and the nonlinear least-squares (NLS) approach with each other as well as the Cramer-Rao bound (CRB), are presented. The simulations illustrate the good performance in the precision and super-resolution.

A relaxation-based sequential optimizing method in CLSAR

A relaxation-based sequential optimizing (RESO) method is proposed for three-dimensional (3-D) target, feature extraction via the curvilinear synthetic aperture radar (CLSAR) system. Compared with the similarly structured algorithm RELAX, RESO obtains the scatterers 3-D position estimate via one-dimensional (1-D) and two-dimensional (2-D) FFTs instead of 3-D FFT employed by the former. Due to the utilization of these lower dimensional FFTs, the new algorithm is more efficient. Simulation results show that RESO and RELAX have similar performance and the estimates obtained via both of them can reach the corresponding Cramer-Rao bounds (CRB).

Fast non-iterative imaging algorithm for CLSAR

Though curvilinear synthetic aperture radar (CLSAR) has the three-dimensional (3-D) imaging capability, the images obtained by non-parametric methods are of little practical use because the data collected by CLSAR is not full volume but sparse data in 3-D frequency space. In this paper, a fast non-iterative imaging (FNII) method is proposed for extracting the scatterers 3-D features in CLSAR. Based on the loose coupling between the range and cross-range parameters, the new algorithm first obtains the range estimates via subspace techniques, then estimates the cross-range parameters from the reformed data with the above range estimates, and finally calculates all the amplitudes from the original data. Without iterative produce, FNII algorithm is computational efficiency. Simulation results indicate that the target image obtained via FNII is similar to the original image and comparison with another efficient algorithm LODIPS, FNII has less computational cost and lower SNR threshold.

A time domain synthesized binary phase code sidelobe suppression filter based on genetic algorithm

THe binary phase coded waveform (BPCW) is widely used in tracking radars, while the sidelobe suppression filter (SSF) designing methods for BPCW are always limited to short codes such as the Barker code. A new kind of SSF based on time domain synthesis and genetic algorithm (GA) is presented. The filter can be used for various lengthened codes, including the m-sequence etc. Some simulation results of time domain synthesized SSF for the Barker codes and m-sequence codes are given, together with comments on the performance comparison of the new filter with some known filters.

Adaptive clutter suppression scheme with high performance

Based on the frequency domain processing, a new adaptive clutter suppression scheme named FFT processing method in frequency domain (FPMFD) is presented. The scheme integrates the moving target indication (MTI) with the MTD processing and has low computational burden. A simulation example shows that the FPMFD has a higher improvement factor and smaller signal-to-noise ratio loss than the classical adaptive MTI that is based on the maximum entropy method (MEM) spectral estimation. This scheme will be beneficial for detecting weak targets.

Non-iterative imaging algorithm for CLSAR

Curvilinear synthetic aperture radar (CLSAR), whose aperture is formed via a curvilinear trajectory, is considered as a practical three-dimensional (3D) imaging system. 3D images obtained by using non-parametric methods, however, have little practical use because the data collected by CLSAR is sparse in 3D frequency space. Some parametric methods have been successfully applied to CLSAR for imaging, but are computationally expensive since they are iteration methods. A non-iterative imaging (NII) algorithm is proposed. The new algorithm estimates the range parameters of all scatterers via a modern spectrum method. It then uses these range estimates and the received data to form two-dimensional (2D) data slices, from which the cross-range parameters are estimated. Once the position (range and cross-range) estimates are obtained, the radar cross section (RCS) can be calculated from the data. Simulation results show that the new algorithm can efficiently form the targets 3D image via CLSAR.

A novel airborne radar signal simulator

Based on analysis of the properties of airborne radar echoes, a parallel processing technology is adopted in the simulator research. The novel architecture of combining microcomputer and multi-DSP parallel computing architecture is presented in detail. The simulator shows a friendly GUI, strong computational capability and broad communication bandwidth. Not only can it simulate airborne radar echoes in real-time, but also it has excellent expansion capability.

An iterative approach of parameter estimation for LFM signals

This paper provides a novel method for estimating the parameters of linear frequency modulated (LFM) signals, or the so called chirp signals, based on a modification to the filter bank approach for the maximum likelihood estimation. This method is especially effective in the case of low SNR. By using numerical simulations, the advantages of the new approach are demonstrated.

Joint parameters estimation scheme for nonlinear frequency modulated radars

A novel pulse compression scheme for tracking radars is presented which achieves joint parameter estimation of velocity and range. The scheme includes two pulse compression channels matched at different Doppler frequency shift. Through a magnitude comparing procedure, Doppler-range coupling is overcome and joint parameter estimation is achieved.