Yuli Fu

• South China University of Technology

In our paper, we introduce the sparse-friendly distillation framework as an effective training strategy for knowledge distillation. While model sparsity techniques have been widely adopted to reduce training overhead, sparse student models often struggle to achieve good performance in knowledge distillation. To address this issue, our framework lev...

Reconfigurable intelligent surface (RIS) is a promising technique that smartly reshapes wireless propagation environment in the future wireless networks. In this paper, we apply RIS to an unmanned aerial vehicle (UAV)-assisted non-orthogonal multiple access (NOMA) network, in which the transmit signals from multiple UAVs to ground users are strengt...

Multi-scale architectures and attention modules have shown effectiveness in many deep learning-based image de-raining methods. However, manually designing and integrating these two components into a neural network requires a bulk of labor and extensive expertise. In this article, a high-performance multi-scale attentive neural architecture search (...

The industrial Internet of Things (IIoT) has been viewed as a typical application for the fifth generation (5G) mobile networks. This paper investigates the energy efficiency (EE) optimization problem for the device-to-device (D2D) communications underlaying unmanned aerial vehicles (UAVs)-assisted IIoT networks with simultaneous wireless informati...

Multi-scale architectures and attention modules have shown effectiveness in many deep learning-based image de-raining methods. However, manually designing and integrating these two components into a neural network requires a bulk of labor and extensive expertise. In this article, a high-performance multi-scale attentive neural architecture search...

Compressed sensing (CS) can recover an image from a few random measurements by exploiting the sparsity assumption on the structure of images. Some recent generative model-based CS recovery methods have removed the sparsity constraint, but their recovery process is slow and the recovered signal is constrained to be in the generator range. Here, we p...

Single image de-raining is an important and highly challenging problem. To address this problem, some depth or density guided single-image de-raining methods have been developed with encouraging performance. However, these methods individually use the depth or the density to guide the network to conduct image de-raining. In this paper, a novel joi...

Nonlocal self-similarity has been proven to be a useful tool for image denoising. For MR image denoising, the method combining the nonlocal self-similarity with the low-rank approximation has been recently attracting considerable attentions, due to its favorable performance. Since the original low-rank approximation problem is difficult to be solve...

Stereo matching, which is a key problem in computer vision, faces the challenge of radiometric distortions. Most of the existing stereo matching methods are based on simple matching cost algorithms and appear the problem of mismatch under radiometric distortions. It is necessary to improve the robustness and accuracy of matching cost algorithms. A...

As a prior knowledge, non‐local self‐similarity (NSS) has been widely utilised in ill‐posed problems. Actually, similar textures appear not only in a single scale, but also in different scales. Unlike most existing patch‐based methods that only explore NSS in the same scale, a multi‐scale patches based image denoising algorithm is proposed in this...

Theoretical guarantees for the ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0,∞</sub> -pseudo-norm based convolutional sparse coding have been established in a recent work. However, the stability analysis in the noisy case via the stripe coherence is absent. This coherence is a stronger characteri...

We propose a novel feature extraction method and a cost function for stereo matching, that is roubust and stable in matching images from different photographing conditions. Based on the Spearman rank correlation, for the pixel in the window centered around a certain pixel, the code is proposed as a rank sequence obtained from an ordered set of the...

Multipath matching pursuit (MMP) has been developed to solve the sparse signal recovery problems in compressed sensing, which is better than traditional orthogonal matching pursuit (OMP) type algorithms in empirical performance. However, the computational burden of MMP is seriously heavy, and limits itself in applications. It needs to be improved u...

Compressed sensing magnetic resonance imaging (CS-MRI) using &ell;1-norm minimization has been widely and successfully applied. However, &ell;1-norm minimization often leads to bias estimation and the solution is not as accurate as desired. In this paper, we propose a novel model for MR image reconstruction, which takes as a smoothed &ell;1-norm re...

In this paper, robust complex-valued sparse signal recovery is considered in the presence of impulse noise. A generalized Lorentzian norm is defined for complex-valued signals. A complex Lorentzian iterative hard thresholding algorithm is proposed to realize the signal recovery. Simulations are given to demonstrate the validity of our results.

The low-rank matrix reconstruction has been attracted significant interest in compressed sensing magnetic resonance imaging (CS-MRI). To the end of computability, rank is often modeled by nuclear norm. The singular value thresholding (SVT) algorithm is taken as a solver of this model, usually. However, this model with the solver may be insufficient...

In this paper, a received signal strength indicator (RSSI) based indoor localization system is implemented employing WiFi infrastructure. In the light of the feature-scaling based k-nearest neighbor (FS-kNN) algorithm, a new continuous-feature-scaling model is proposed, which uses continuous weights instead of the discrete weights used in the FS-kN...

The least mean p-power error criterion has been successfully used in adaptive filtering due to its strong robustness against large outliers. In this paper, we develop a new adaptive filtering algorithm, named the proportionate least mean p-power (PLMP) algorithm, which uses the mean p-power error as the adaptation cost function. Compared with the s...

The correlation based framework has recently been proposed for sparse support recovery in noiseless case. To solve this framework, the constrained least absolute shrinkage and selection operator (LASSO) was employed. The regularization parameter in the constrained LASSO was found to be a key to the recovery. This paper will discuss the sparse suppo...

This paper studies the stability problem of Yang–Chen system. By introducing different radial unbounded Lyapunov functions in different regions, global exponential attractive set of Yang–Chen chaotic system is constructed with geometrical and algebraic methods. Then, simple algebraic sufficient and necessary conditions of global exponential stabili...

Recently, based on restricted isometry property (RIP), some sufficient conditions for exact support recovery with simultaneous orthogonal matching pursuit (SOMP) algorithm have been proposed when measurement matrices are different. In this paper, in the noiseless case, one sufficient condition for exact support recovery with SOMP is presented to im...

A new algorithm is proposed for compressed sensingmagnetic resonance imaging (CS-MRI). The l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</sub> -norm (0 <; p ≤ 1) based adaptive regularization model is used for MRI. The algorithm is established by using a novel iterative shrinkage scheme. In the i...

Block-sparsity is an extension of the ordinary sparsity in the realm of the sparse signal representation. Exploiting the block structure of the sparsity pattern, recovery may be possible under more general conditions. In this study, a block version of the orthogonal matching pursuit with thresholding (block-OMPT) algorithm is proposed. Compared wit...

Occlusion is a common yet challenging problem in face recognition. Most of the existing approaches cannot achieve the accuracy of the recognition with high efficiency in the occlusion case. To address this problem, this paper proposes a novel algorithm, called Efficient Locality-constrained Occlusion Coding (ELOC), improving the previous Sparse Err...

In this letter, direction of arrival (DOA) estimation is considered for impulsive noise case. The original complex-valued measurement for DOA problem is expressed as an augmented real-valued measurement using the real and imaginary parts. To deal with the impulsive noises, a function approximation technique is used. For the real-valued model, a rob...

In this letter, robust sparse signal recovery is considered in the presence of the symmetric -stable (SS) distributed noise. An M-estimate type model is constructed by approximating the location score function of the noise. A reweighed iterative hard thresholding algorithm is proposed to recover the sparse signal. The basis functions for the approx...

In this brief, a robust and sparse recursive adaptive filtering algorithm, called convex regularized recursive maximum correntropy (CR-RMC), is derived by adding a general convex regularization penalty term to the maximum correntropy criterion (MCC). An approximate expression for automatically selecting the regularization parameter is also introduc...

Spectrum access strategy plays a critical role in multichannel cognitive radio networks (CRNs). However, the CRNs cannot obtain the maximal throughput, when the existing access strategies, including overlay, underlay, and hybrid access strategies, are applied to multichannel CRNs. In this paper, we present a generalized access strategy in a multich...

A spectrum handoff model and optimal channel decision method based on Extenics have been proposed in order to resolve the optimal channel decision problem of spectrum handoff in Cognitive Radio Sensor Networks. The method of matter-element Extenics is used to analyses the spectrum handoff process. The channel state through the spectrum sensing and...

In order to maximize throughput and minimize interference of the wideband spectrum sensing problem in OFDM cognitive radio sensor networks, a linear weighted sum multi-objective algorithm based on the Particle Swarm Optimization is proposed. The multi-objective optimization advantages of Particle Swarm Optimization are utilized to solve the optimal...

Cognitive radio (CR)-based smart grid (SG) networks have been widely recognised as emerging communication paradigms in power grids. However, a sufficient spectrum resource and reliability are two major challenges for real-time applications in CR-based SG networks. In this article, we study the traffic data collection problem. Based on the two-stage...

In this paper, the Positive constrained Least Absolute Shrinkage and Selection Operator (P-LASSO) is studied for sparse support recovery using the correlation information in Compressive sensing (CS). A structural constraint is obtained for selecting the regularization parameter in the case of additive Gaussian noise. Since the measurements are fini...

Misaligned face recognition has been studied in the past decades, but still remains an open challenge. To address this problem, we propose a highly efficient misalignment-robust locality-constrained representation (MRLR) algorithm. Specifically, MRLR first aligns the query face via the L2-norm locality-constrained representation, and then recognize...

The cognitive radio technology can provide dynamic spectrum access and improve the efficiency of spectrum utilization. Spectrum sensing is one of the key technologies of cognitive radio networks. The spectrum sensing performance of cognitive radio networks will be greatly reduced in the low SNR environment, especially when using energy detection. D...

The cognitive radio technology can improve the efficiency of spectrum utilization byproviding dynamic spectrum access to unoccupied frequency bands. Spectrum sensing is one of the key technologies of cognitive radio networks. The spectrum sensing performance of cognitive radio networks will be greatly reduced in the low SNR environment, especially...

A more relaxed condition means that fewer of measurements are needed to ensure the exact sparse recovery from the theoretical aspect. The sufficient condition for the greedy block coordinate descent (GBCD) algorithm is relaxed using the near-orthogonality property. It is also shown that the GBCD algorithm fails when (1/(√K+1)≥δK+1<1).

Occlusion in face recognition is a common yet challenging problem. While sparse representation based classification (SRC) has been shown promising performance in laboratory conditions (i.e. noiseless or random pixel corrupted), it performs much worse in practical scenarios. In this paper, we consider the practical face recognition problem, where th...

We consider the problem of automatically recognizing human faces in which sparse representation-based classification (SRC) offers a key. SRC includes two steps: seeking sparest solution and making decision by dictionary classifier (DC). Aiming at improving the performance of face recognition, this paper proposes a joint classification approach base...

This paper presents a selective spectrum sensing and access strategy in a cognitive radio sensor network (CRSN), in order to maximize the throughput of secondary user (SU) system. An SU senses multiple channels simultaneously via wideband spectrum sensing. To maximize the throughput and reduce the sensing energy consumption, not all of the channels...

Sparse group lasso, concerning with group-wise and within-group sparsity, is generally considered difficult to solve due to the mixed-norm structure. In this paper, we propose efficient algorithms based on split Bregman iteration to solve sparse group lasso problems, including a synthesis prior form and an analysis prior form. These algorithms have...

Recoverability of block-sparse signals by convex relaxation methods is considered for the underdetermined linear model. In previous works, some explicit but pessimistic recoverability results which were associated with the dictionary were presented. This paper shows the recoverability of block-sparse signals are associated with the block structure...

Most of current biped robots are active walking platforms. Though they have strong locomotion ability and good adaptability to environments, they have a lot of degrees of freedom (DoFs) and hence result in complex control and high energy consumption. On the other hand, passive or semi-passive walking robots require less DoFs and energy, but their w...

Practically, in the underdetermined model ${bf Y} = {bf AX}$, where ${bf X}$ is a $K$-group sparse matrix (i.e., it has no more than $K$ nonzero rows), both ${bf Y}$ and ${bf A}$ could be totally perturbed. In this paper, based on restricted isometry property, for the greedy block coordinate descent algorithm, a sufficient condition of exact recove...

Compressed sensing ensures the accurate reconstruction of sparse signals from far fewer samples than required in the classical Shannon–Nyquist theorem. In this paper, a generalized hard thresholding pursuit (GHTP) algorithm is presented that can recover unknown vectors without the sparsity level information. We also analyze the convergence of the p...

Recently, it has been found that the redundant blocks problem existed in many fields, such as face recognition and motion segmentation. In this paper, taking the redundant blocks into account, we propose some greedy type algorithms that exploit the subspace information of the redundant blocks to solve the redundant blocks problem. The exact recover...

Different signals from the various sensors of the same scene form an ensemble. Distributed compressed sensing (DCS) rests on a new concept called the joint sparsity of the ensemble. JSM-1 is a model that describes the joint sparsity by one dictionary. Previously, the generalisation of JSM-1 was proposed where the signal ensemble depends on two dict...

In cognitive radio networks, wideband spectrum sensing is a promising technology which allows a secondary user (SU) to detect the signals of primary users (PUs) over multiple channels, the sensing overhead is reduced effectively. Together with spectrum sensing, spectrum access strategy affects the system performance. In this paper, we propose an ef...

Compressed sensing (CS)-based cross-and-bouquet (CAB) model was proposed by J. Wright et al. to reduce the complexity of sparse error correcting. For the sake of leading to better performance of CS-based decoding for the CAB model, an algorithm is proposed in this paper for constructing a well-designed projection matrix to minimize the average meas...

Recently, sparse representation (SR) and joint sparse representation (JSR) have attracted a lot of interest in image fusion. The SR models signals by sparse linear combinations of prototype signal atoms that make a dictionary. The JSR indicates that different signals from the various sensors of the same scene form an ensemble. These signals have a...

Sparse signals can be reconstructed from far fewer samples than those that were required by the Shannon sampling theorem, if compressed sensing (CS) is employed. Traditionally, a random Gaussian (rGauss) matrix is used as a projection matrix in CS. Alternatively, optimization of the projection matrix is considered in this paper to enhance the quali...

This letter focuses on an energy-efficient hybrid spectrum access scheme in a vehicle-to-infrastructure uplink communication scenario in cognitive vehicular ad hoc networks (cognitive VANETs). Considering path loss exponent of the channel between a vehicle and an access point, a constrained optimization problem is formulated to minimize the overall...

The standard sparse representation aims to reconstruct sparse signal from single measurement vector which is known as SMV model. In some applications, the SMV model extend to the multiple measurement vector (MMV) model, in which the signal consists of a set of jointly sparse vectors. In this paper, efficient algorithms based on split Bregman iterat...

Block-sparse reconstruction, which arises from the reconstruction of block-sparse signals in structured compressed sensing, is generally considered to be difficult due to the mixed-norm structure. In this paper, we propose efficient algorithms based on split Bregman iteration to solve the block-sparse reconstruction problems, including the constrai...

Block-sparse reconstruction, which arises from the reconstruction of block-sparse signals in structured compressed sensing, is generally considered difficult to solve due to the mixed-norm structure. In this letter, we propose an algorithm for reconstructing block-sparse signals, that is an extension of fixed point continuation in block-wise case b...

In cognitive radio networks, spectrum sensing and access scheme affects the system performance. In this paper, a new wideband mixed access scheme is proposed, in which the Secondary Users (SUs) sense the channels via wideband spectrum sensing, and access them with a mixed access strategy. In order to maximize the ergodic throughput of SUs, we find...

This paper presents an improved algorithm for detecting the SYN flooding attacks. The algorithm is based on the characteristics of the network processor IXP2850's hardware and software framework. It improves the typical method that is based on checking the received SYN segments twice from the same source. The improved algorithm will label the segme...

This letter discusses blind separability based on temporal predictability (Stone, 2001; Xie, He, & Fu, 2005). Our results show that the sources are separable using the temporal predictability method if and only if they have different temporal structures (i.e., autocorrelations). Consequently, the applicability and limitations of the temporal predic...

It is well known that stability of Hopfield type neural networks plays a very impor- tant role in both theoretical research and applications. So, it has been kept on study- ing in two decades. Stochastic eectiveness to this kind of neural networks has also re- ceived a lot of attention (ref. (Liao et al, 1996 A), (Liao et al, 1996 B), (Blythe,S. et...

In this paper, a mobile payment scheme is proposed based on Radio Frequency Identification (RFID) for both micro-payment and high value payment. The payment processes are designed under the structure of mobile communication and RFID systems. The potential risks are discussed. Also, a mutual authentication protocol is presented using secure certific...

The speed of interconnection has grown continually in the fast developing Internet and other networks. Routing lookup has become the bottleneck of high-speed packet forwarding. Obviously, high-speed packet forwarding depends on high-speed routing lookup and update algorithms. This paper discusses Longest Prefix Match algorithm (LPM) based on the ha...

Nonnegative matrix factorization (NMF) is widely used in signal separation and image compression. Motivated by its successful applications, we propose a new cryptosystem based on NMF, where the nonlinear mixing (NLM) model with a strong noise is introduced for encryption and NMF is used for decryption. The security of the cryptosystem relies on fol...

Using IXP2XXX network processor (NP), a fast processing of Internet Control Message Protocol (ICMP) is embedded in the NP based firewall. By interpolating the module before processing the information of network interface layer and setting a special jump identifier for the exceptional process microblock on NP, simulation results show that the optimi...

The lack of inherent security in the GPRS tunneling protocol (GTP) leaves security vulnerabilities in GPRS networks. A novel scheme to defend the network against overbilling is proposed. The scheme only requires a dynamic memory in a GTP firewall to store some information flow such as tunnel ID. It needs not work jointly with an extra firewall at t...

This paper discusses the source recovery step in two-stage blind separation algorithm of underdetermined mixtures. A statistically non-sparse decomposition principle of two mixtures (2d-SNSDP), which is an extension of the SSDP algorithm about two mixtures, is proposed. It overcomes the disadvantage of the SSDP algorithm and sparse representation b...

The independence priori is very often used in the conventional blind source separation (BSS). Naturally, independent component analysis (ICA) is also employed to perform BSS very often. However, ICA is difficult to use in some challenging cases, such as underdetermined BSS or blind separation of dependent sources. Recently, sparse component analysi...

In this paper, the concept of globally exponentially attractive set is proposed and used to consider the ultimate bounds of the family of Lorenz systems with varying parameters. Explicit estimations of the ultimate bounds are derived. The results presented in this paper contain all the existing results as special cases. In particular, the critical...

Underdetermined blind signal separation (BSS) (with fewer observed mixtures than sources) is discussed. A novel searching-and-averaging method in time domain (SAMTD) is proposed. It can solve a kind of problems that are very hard to solve by using sparse representation in frequency domain. Bypassing the disadvantages of traditional clustering (e.g...

Based on clustering method on planes, blind signal separation (BSS) of underdetermined mixtures with three observed signals is discussed. The condition of sufficient sparsity of the source signals is not necessary when clustering method on planes is used. In other words, it needs not that only one source signal plays the main role among others at o...

In light of radio frequency Identification (RFID) and universal serial bus (USB) Key techniques, an RFID-USB key is designed. With both advantages of RFID and USB key, this RFID-USB key can be used to establish some safe environments of authentication, identification and other applications feasibly. The idea of this design is detailed and an implem...

Through discussing the security problems of a key network protocol - the GPRS tunneling protocol (GTP), this paper points out 3rd generation network is at risk from both its own subscribers and its partner networks. So the 3G firewall should encrypt the GTP. State-of-the-art high-speed network processor-IXA2850 is described. And its features of cry...

Sparse representation of complex valued signals is addressed in this paper. Considering the statistical dependence between real part and imaginary part of a complex valued signal (e.g., the discrete-time Fourier transform of a real valued signal), a special probability density function (PDF) is introduced to describe the complex random variable in...

A penalty function based algorithm of blind separation is proposed in this paper. The nonlinearity in this kind of algorithms is discussed. Some stable regions of the nonlinearity are pointed to demonstrate the properties of it. A simulation is given to illustrate availability of the algorithm

Underdetermined blind source separation and sparse component analysis aim at to recover the unknown source signals under the assumption that the observations are less than the source signals and the source signals can be sparse expressed. Many methods to deal with this problem related to clustering. For underdetermined blind source separation model...

In this paper, we first give a constructive proof for the existence of globally exponential attractive set of Chua's system with a smooth nonlinear function. Then, we derive a series of simple algebraic sufficient conditions under which two same type of smooth Chua's systems are globally exponentially synchronized using simple linear feedback contr...

Bofill et al. discussed blind source separation (BSS) of sparse signals in the case of two sensors. However, as Bofill et al. pointed out, this method has some limitation. The potential function they introduced is lack of theoretical basis. Also the method could not be extended to solve the problem in the case of more than three sensors. In this pa...

Conventionally, multi-channel acoustic echo cancellation (AEC) achieves the goal by estimating the impulse responses of the local room. However, generally, conventional AEC methods have no unique solutions. Due to the strong correlation of the input signals, conventional methods are with many disadvantages. To overcome this problem, a new framework...

The signals with generalized Gaussian distribution are considered. A mathematical formula is given to illustrate the sparsity of the signals. According to this formula, the measure of the Laplacian signal is 1, and Gaussian signal is 2. Given a signal, compared with Laplacian signal and Gaussian signal, we can intuitively know how sparse the signal...

The underdetermined case in blind source separation, that is, separation of n sources from m (m<n) linearly mixtures, is probed in this paper. First, we analyzed the disadvantage of l¹ -norm solution. Second, we present a new sparse representation based on second order statistic, which is called statistically sparse decomposition princip...

In this paper generalized Gaussian distribution is employed to discuss sparseness measure for signals. At first we established a mathematical formula to calculate the sparseness measure of signals. According to this measure formula, the sparseness measure value of the Laplacian signal is 1, and Gaussian signal is 2. Given a signal, from its sparsen...

Constructing a family of generalized Lyapunov functions, a new method is proposed to obtain new global attractive set and positive invariant set of the Lorenz chaotic system. The method we proposed greatly simplifies the complex proofs of the two famous estimations presented by the Russian scholar Leonov. Our uniform formula can derive a series of...

Based on sparse representation, this paper discusses convolutive BSS of sparse sources and presents a FIR convolutive BSS algorithm that works in the frequency domain. This algorithm does not require that source signals be i.i.d or stationary, but require that source signals be sufficiently sparse in frequency domain. Furthermore, our algorithm can...

Stone's method is one of the novel approaches to the blind source separation (BSS) problem and is based on Stone's conjecture. However, this conjecture has not been proved. We present a simple simulation to demonstrate that Stone's conjecture is incorrect. We then modify Stone's conjecture and prove this modified conjecture as a theorem, which can...

In this paper, a new algorithm of iterative learning control with forgetting factor is proposed by using a new norm and a new analysis method. The new method applies the whole information of systems to transfer the iterative learning control problem into a stability problem of a discrete system with parameters. This algorithm improves the shortage...

Through an analysis and comparison of the algorithm proposed by Hyvarinen-Oja (1996), we present an approach of blind separation based on the penalty functions with multipliers. The approach gives the method to select the penalty function and speeds up the convergence of the algorithm. It avoids the ill-posed problem that may be caused by the pure...