A5100413696- Blind Source Separation Techniques
- Matrix Theory and Algorithms
- Speech and Audio Processing
- Machine Fault Diagnosis Techniques
- Holomorphic and Operator Theory
- Structural Health Monitoring Techniques
- Analytic and geometric function theory
- Advanced Decision-Making Techniques
- Sparse and Compressive Sensing Techniques
- Image and Signal Denoising Methods
- Electromagnetic Scattering and Analysis
- Advanced Algorithms and Applications
- Geometry and complex manifolds
- Evaluation Methods in Various Fields
- Neural Networks and Applications
- Game Theory and Applications
- Wireless Communication Networks Research
- Advanced Mathematical Theories and Applications
- Algebraic Geometry and Number Theory
- Advanced Image and Video Retrieval Techniques
- Numerical methods in inverse problems
- Advanced Adaptive Filtering Techniques
- Mathematical Inequalities and Applications
- Advanced Sensor and Control Systems
- Advanced Computational Techniques and Applications
Beihang University
2015-2025
Fudan University
2021-2025
Neijiang Normal University
2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2023
Northwest Normal University
2020
Jishou University
2014
Shenyang Institute of Automation
2011-2012
Chinese Academy of Sciences
2009-2012
University of Chinese Academy of Sciences
2012
Vrije Universiteit Brussel
2005
Human activity recognition by using wearable sensors has gained tremendous interest in recent years among a range of health-related areas. To automatically recognize various human activities from sensor data, many classification methods have been tried prior studies, but most them lack the incremental learning abilities. In this study, an method is proposed for sensor-based recognition. The designed based on probabilistic neural networks and adjustable fuzzy clustering algorithm. may achieve...
In this paper, we propose a new geometric-shaping design for golden angle modulation (GAM) based on the complex geometric properties of open symmetrized bidisc, termed Bd-GAM, future generation wireless communication systems. Inspired from circular symmetric structure GAM, construct schemes, Bd-GAM1 and Bd-GAM2. Specifically, consider MI-optimized probabilistic scheme with geometrics bidisc. With minimum SNR entropy constraint, Bd-GAM2 can overcome shaping-loss. Compared existed introduced,...
The Kohn-Nirenberg domains are unbounded in $\mathbb{C}^{n}$. In this article, we modify the domain $\Omega_{K,L} =\left\{(z_{1},\ldots, z_{n}) \right.\in \mathbb{C}^{n} : Re z_{n} + g \mid z_{n}\mid^{2} \sum_{j = 1}^{n-1} (\mid z_{j}\mid^{p} +K_{j} z_{j}\mid^{p-q} z_{j}^{q} L_{j} z_{j}\mid^{p-2q} Im z_{j}^{2q}) < 0 \}$ and discuss existence of supporting surface peak function at origin.
Human activity recognition (HAR) by using wearable accelerometers has gained significant interest in recent years a range of healthcare areas, including inferring metabolic energy expenditure, predicting falls, measuring gait parameters and monitoring daily activities. The implementation HAR relies heavily on the correctness sensor fixation. installation errors may dramatically decrease accuracy HAR. In this paper, method is proposed to improve robustness accelerometers. first calculates...
A B-spline empirical mode decomposition (BEMD) method is proposed to improve the celebrated (EMD) method. The improvement of BEMD on EMD mainly concentrates sifting process. First, instead curve that resulted from computing average upper and lower envelopes, interpolated by midpoints local maximal minimal points used as mean curve, which can reduce cost computation. Second, cubic spline interpolation replaced with account advantages over polynomial spline. effectiveness compared validated...
Envelope modified versions of the empirical mode decomposition (EMD) method such as B-spline interpolation-based EMD (B-EMD) and cardinal spline (C-EMD) have been proposed recently for purpose improving its effectiveness. To shed further light on their performance, behaviours these EMD-type methods in presence white Gaussian noises are investigated this study based extensive numerical experiments. Similarly to method, it turns out that envelope also act filter banks essentially. However,...
With the growing demand for data sharing file formats in financial applications driven by open banking, use of OFD (open fixed-layout document) format has become widespread. However, ensuring security, traceability, and accountability poses significant challenges. To address these concerns, we propose E-SAWM, a dynamic watermarking service framework designed edge cloud scenarios. This incorporates watermark information at edge, allowing precise tracking leakage throughout data-sharing...
Abstract Background Cerebrovascular segmentation is a crucial step in the computer‐assisted diagnosis of cerebrovascular pathologies. However, accurate extraction cerebral vessels from time‐of‐flight magnetic resonance angiography (TOF‐MRA) data still challenging due to complex topology and slender shape. Purpose The existing deep learning‐based approaches pay more attention skeleton ignore contour, which limits performance structure. We aim weight contour brain shallow features when...
In this study, the authors propose a novel method, namely Toeplitz-based singular value decomposition (or TL-SVD for short) reconstruction and basis function construction of electromagnetic interference (EMI) source signals. Given specific EMI signal, they first construct Toeplitz type data matrix. By applying (SVD) to constructed matrix, obtain set values, which are further divided into two parts, corresponding clear noisy components input signals, respectively. The de-noised signal can...
In this paper, we shall prove for normal matrices A, B. particular, if B are Hermitian matrices, yielding the Golden-Thompson inequality. MSC:15A16, 47A63, 15A45.
In brain–computer interface (BCI)-based motor imagery, the symmetric positive definite (SPD) covariance matrices of electroencephalogram (EEG) signals with discriminative information features lie on a Riemannian manifold, which is currently attracting increasing attention. Under manifold perspective, we propose non-linear dimensionality reduction algorithm based neural networks to construct more low-dimensional SPD manifold. To this end, design novel shrinkage layer modify extreme...
In the field of electromagnetic compatibility, interference (EMI) has always been a difficult problem for researchers. order to achieve accurate separation single-channel EMI signal, this paper proposes one-stop method based on ICEEMDAN and FastICA. Firstly, algorithm is used extract several intrinsic mode functions (IMFs) from mixture. Secondly, multi-dimensional signals are reconstructed according IMFs original aliasing signal separated by FastICA algorithm. Finally, feasibility proposed...
In this paper, the dynamic niching particle swarm optimization (DNPSO) is proposed to solve linear blind source separation problem. The key point use DNPSO rather than (PSO) and fast-ICA as algorithm in Independent Component Analysis (ICA). By using DNPSO, which has global superiority, performance of ICA will be improved accuracy convergence rate. idea sub-population leads greater efficiency compared with other methods when solving high dimensional cost functions ICA. based on investigated...