Fengzhou Dai

Dr. Fengzhou Dai

orcid

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 July 2012 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
FDFengzhou Dai
Fengzhou Dai
Hongwei LiuHongwei Liu
Hongwei Liu
PSPeng-Lang Shui
Peng-Lang Shui
SWShunjun Wu
Shunjun Wu
In this paper, we address detection of range spread targets with range walking in partially homogeneous clutter, which is often encountered in wideband radar. By considering the target range walking effect, a target returns model is proposed that represents the relationship between Doppler frequency and range frequency. Based on this, a generalized matched subspace detector (GMSD) and a general adaptive subspace detector (GASD) are proposed in the frequency domain. Moreover, to eliminate excess integration loss introduced by the mismatched detection window size, two improved detectors are proposed in range domain. The theoretical analysis shows that the GMSDs are of a constant false alarm rate (CFAR) with respect to both the power and the covariance matrix of the clutter but that the GASDs are of a CFAR with respect to the clutter power and an approximate CFAR with respect to the clutter covariance matrix. Simulated experiments are carried out to verify the effectiveness of the proposed detectors.

26 July 2023 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
HFHaosheng Fu
Haosheng Fu
FDFengzhou Dai
Fengzhou Dai
LHLing Hong
Ling Hong
Compared with the traditional subspace-based methods, sparse signal recovery (SSR) based methods have obvious advantages in performing the direction of arrival (DOA) estimation of an array, such as working well on the limited number of snapshots, good noise robustness, and so on. However, the occurrence of grid mismatch limits the estimation accuracy of these methods. To solve the problem that the grid mismatch impacts DOA estimation accuracy, this article presents a new method for off-grid DOA estimation using weighted Sinc interpolation, referred to as OGWSISBL. Specifically, the off-grid error is represented as the parameter to be estimated in the observation model. Then, under the sparse Bayes learning (SBL) framework, the off-grid error is calculated using the variational Bayesian expectation maximization (VBEM) method and eliminated by updating the grids with weighted Sinc interpolation. The proposed method can work well in both single and multiple snapshots. Simulation and actual measurement results illustrate that the proposed method is superior to the state-of-the-art methods reported.

24 February 2022 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
FDFengzhou Dai
Fengzhou Dai
YWYuanyuan Wang
Yuanyuan Wang
LHLing Hong
Ling Hong
Sparse Bayesian learning (SBL) is a popular and robust algorithm for sparse signal reconstruction (SSR). Unfortunately, the SBL algorithm suffers from heavy computational complexity when it is implemented directly since the inversion and multiplying operations of the estimation of the covariance matrix are involved in each iteration, which is proportional to the cube of the observed data length and thus prevents it solving large-scale problems. In many applications, such as radar imaging and array signal processing, the signal to be recovered is sparse in the Fourier dictionary. In this article, we propose an efficient implementation method for the Fourier dictionary-based SBL (FD-SBL). In the case that the Fourier dictionary is adopted, the estimation of the covariance matrix is a Toeplitz matrix for 1-D data or a Toeplitz-block-Toeplitz (TBT) matrix for 2-D data during the FD-SBL iterations. By utilizing this property, we employ the Gohberg–Semencul (G-S)-type factorization to accelerate the implementation of FD-SBL. To be noted, there is no approximation in our proposed method, and the computational cost is reduced by several orders of magnitude compared with the direct implementation of SBL. Finally, the experimental results verify the effectiveness of the proposed method.

25 September 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HFHaosheng Fu
Haosheng Fu
FDFengzhou Dai
Fengzhou Dai
LHLing Hong
Ling Hong
Metasurface aperture-based systems (MABSs) have recently attracted more attention due to their simple hardware structure, easy fabrication, and low-cost measurement, and they have been developed for the direction of arrival (DOA) estimation. In this paper, we focus on the two-dimensional (2D) off-grid DOA estimation problem for MABSs. Unlike the traditional multi-channel array measurement mechanism, MABSs retrieve the directional of incident sources by generating multiple spatially incoherent quasi-random radiation patterns. Meanwhile, it will bring a new problem in that the measurement matrix of the MABS is a pseudo-random matrix and cannot be described by mathematical formulas. Besides, under the compressed sensing (CS) framework, the grid mismatch caused by discretization operation cannot be ignored. To handle these problems, we innovatively propose a 2D off-grid DOA estimation model for MABSs using a Gaussian interpolation kernel. Then, based on the proposed model, a 2D off-grid DOA estimation method, referred to as 2D-OGGISBL, is developed. Specifically, the off-grid errors are considered as parameters to be calibrated in the model, which are estimated by using the expectation maximization (EM) algorithm under the framework of multiple measurement vector (MMV) sparse Bayesian learning (SBL) and eliminated via updating the grids using the Gaussian interpolation kernel. As shown by the simulation and experimental results, compared with the existing DOA estimation methods used for MABSs, the proposed method can calibrate more than 95% of the off-grid errors and works well even at low SNR (5 dB) and few numbers of snapshots, which significantly improves the measurement performance of MABSs.

11 July 2023 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
HFHaosheng Fu
Haosheng Fu
FDFengzhou Dai
Fengzhou Dai
LHLing Hong
Ling Hong
Improving the resolution of metasurface apertures (MAs)-based computational microwave imaging (CMI) is of great significance for its practical application. Existing MAs-based CMI (MAs-CMI) has limited resolution due to the limited size of MAs, the so-called Rayleigh diffraction limit (RDL), and must rely on complex back-end algorithms to achieve far-field super-resolution imaging. In this article, a single-frequency (10 GHz) single-sensor double-layer transmitarray (DTA) is designed to improve the resolution of MAs-CMI from the hardware aspect, which consists of a horn as feed, a layer of passive MA (PMA), and a layer of dynamic MA (DMA). By carefully designing the structure of the DTA, a variety of far-field random radiation patterns beyond RDL can be obtained, and the imaging resolution of the MA-CMI can be effectively improved. Simulation and measurement results show that the resolution of the far-field radiation patterns of the proposed DTA is at least 1.56 times that of RDL at the same aperture. Finally, an imaging experiment based on a manufactured DTA is performed to further demonstrate its far-field super-resolution characteristic for CMI. This is the first time that the far-field super-resolution TA based on MAs for CMI is reported.

13 March 2023 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
FDFengzhou Dai
Fengzhou Dai
JLJiang Liu
Jiang Liu
LTLong Tian
Long Tian
HDHang Dong
Hang Dong
LHLing Hong
Ling Hong
Micro-Doppler analysis of rigid-body target is significant for attitude estimation and recognition of space objects. The traditional micro-Doppler analysis method for rigid-body targets includes two sequential steps. First, the time-frequency analysis is performed on the radar echo data of the target, and then the micro-Doppler curve of each scattering center is separated and extracted from the time-frequency map. The second step depends on the micro motion model of the target. In the micro-Doppler analysis of real rigid-body target, there are some problems such as the mismatch of the micro motion model, the incidence angle dependence of the scattering center position, and the partial occlusion of the scattering center. Therefore, it is very difficult to correctly extract the micro-Doppler curves of multiple scatterers. In this paper, an end-to-end micro-Doppler analysis method for rigid-body target based on deep learning network is proposed, which can directly separate and extract the micro-Doppler curves of multiple scatterers from the target echo data. Specifically, an Asymmetrical AutoEncoding (A2E) network equipped with a Data Pre-Processing (DP2) module is developed to extract Time-Frequency Curves (TFCs) from radar echos. Considering the sparseness of Time-Frequency Distribution (TFDs), we then develop a novel Energy-Concentration Objective (ECO) function based on Min-Max game to enhance curves energy while suppress the background energy. In practice, measured TFDs are rarely annotated, it restricts the generalization capability of the A2E from the simulation to the measurement. To bridge the gap, two-fold modifications are finally constructed: i) we insert a partial-shared branch of decoder to reconstruct the TFD from the DP2 module; ii) we regularize the ECO function with a knowledge preservation based reconstruction bound to further capture the characteristics of measured echos in a semi-supervised way at test time to relieve the domain-shift problem.