Guangdong Polytechnic Normal University

Electroencephalography (EEG)-based brain-computer interfaces (BCIs), particularly those using motor-imagery (MI) data, have the potential to become groundbreaking technologies in both clinical and entertainment settings. MI data is generated when a subject imagines the movement of a limb. This paper reviews state-of-the-art signal processing techniques for MI EEG-based BCIs, with a particular focus on the feature extraction, feature selection and classification techniques used. It also summarizes the main applications of EEG-based BCIs, particularly those based on MI data, and finally presents a detailed discussion of the most prevalent challenges impeding the development and commercialization of EEG-based BCIs.

This research introduces a path planning method based on the geometric A-star algorithm. The whole approach is applied to an Automated Guided Vehicle (AGV) in order to avoid the problems of many nodes, long-distance and large turning angle, and these problems usually exist in the sawtooth and cross paths produced by the traditional A-star algorithm. First, a grid method models a port environment. Second, the nodes in the close-list are filtered by the functions P(x,y ) and W(x,y ) and the nodes that do not meet the requirements are removed to avoid the generation of irregular paths. Simultaneously, to enhance the stability of the AGV regarding turning paths, the polyline at the turning path is replaced by a cubic B-spline curve. The path planning experimental results applied to different scenarios and different specifications showed that compared with other seven different algorithms, the geometric A-star algorithm reduces the number of nodes by 10% ~ 40%, while the number of turns is reduced by 25%, the turning angle is reduced by 33.3%, and the total distance is reduced by 25.5%. Overall, the simulation results of the path planning confirmed the effectiveness of the geometric A-star algorithm.

As one of the most promising fluorescent nanomaterials, the fluorescence of carbon dots (CDs) in solution is extensively studied. Nevertheless, the synthesis of multicolor solid-state fluorescence (SSF) CDs is rarely reported. Herein, CDs with multicolor aggregation-induced emission are prepared using amine molecules, all of them exhibiting dual fluorescence emission at 480 nm (Em-1) and 580-620 nm (Em-2), which is related to the SS bonds of dithiosalicylic acid and the conjugated structure attached to CO/CN bonds, respectively. As a strong electron-withdrawing group, the increase of CN content makes dual-fluorescent groups on the surface of CDs produce push and pull electrons, which determines intramolecular charge transfer (ICT) between the double emission. With the increase in CN content from 35.6% to 58.4%, the ICT efficiency increases from 8.71% to 45.94%, changing the fluorescence of CDs from green to red. The increase of ICT efficiency causes fluorescence quantum yield enhancement by nearly five times and redshift of the fluorescence peak. Finally, based on the multicolor luminescence properties induced by the aggregation of CDs, pattern encryption and white-LED devices are realized. Based on the fat solubility and strong ultraviolet absorption characteristics of CDs, fingerprint detection and leaf anti-UV hazards are applied.

In imbalanced network traffic, malicious cyber-attacks can often hide in large amounts of normal data. It exhibits a high degree of stealth and obfuscation in cyberspace, making it difficult for Network Intrusion Detection System(NIDS) to ensure the accuracy and timeliness of detection. This paper researches machine learning and deep learning for intrusion detection in imbalanced network traffic. It proposes a novel Difficult Set Sampling Technique(DSSTE) algorithm to tackle the class imbalance problem. First, use the Edited Nearest Neighbor(ENN) algorithm to divide the imbalanced training set into the difficult set and the easy set. Next, use the KMeans algorithm to compress the majority samples in the difficult set to reduce the majority. Zoom in and out the minority samples' continuous attributes in the difficult set synthesize new samples to increase the minority number. Finally, the easy set, the compressed set of majority in the difficult, and the minority in the difficult set are combined with its augmentation samples to make up a new training set. The algorithm reduces the imbalance of the original training set and provides targeted data augment for the minority class that needs to learn. It enables the classifier to learn the differences in the training stage better and improve classification performance. To verify the proposed method, we conduct experiments on the classic intrusion dataset NSL-KDD and the newer and comprehensive intrusion dataset CSE-CIC-IDS2018. We use classical classification models: random forest(RF), Support Vector Machine(SVM), XGBoost, Long and Short-term Memory(LSTM), AlexNet, Mini-VGGNet. We compare the other 24 methods; the experimental results demonstrate that our proposed DSSTE algorithm outperforms the other methods.

Fog computing is an end-to-end horizontal architecture that distributes computing, storage, control, and networking functions closer to users along the cloud-to-thing continuum. The word “edge” may carry different meanings. A common usage of the term refers to the edge network as opposed to the core network, with equipment such as edge routers, base stations, and home gateways. In that sense, there are several differences between fog and edge. First, fog is inclusive of cloud, core, metro, edge, clients, and things. The fog architecture will further enable pooling, orchestrating, managing, and securing the resources and functions distributed in the cloud, anywhere along the cloud-to-thing continuum, and on the things to support end-to-end services and applications. Second, fog seeks to realize a seamless continuum of computing services from the cloud to the things rather than treating the network edges as isolated computing platforms. Third, fog envisions a horizontal platform that will support the common fog computing functions for multiple industries and application domains, including but not limited to traditional telco services. Fourth, a dominant part of edge is mobile edge, whereas the fog computing architecture will be flexible enough to work over wireline as well as wireless networks.

The resolution limit of far-field optical microscopy is reexamined with a full vectorial theoretical analysis. A highly symmetric excitation optical field and optimized detection scheme are proposed to harness the total point-spread function for a microscopic system. Spatial resolution of better than 1/6λ is shown to be obtainable, giving rise to a resolution better than 100 nm with visible light excitation. The experimental measurement is applied to examine nonfluorescent samples. A lateral resolution of 1/5λ is obtained in truly far-field optical microscopy with a working distance greater than ∼500λ. Comparison is made for the far-field microscopic measurement with that of a nearfield scanning optical microscopy, showing that the proposed scheme provides a better image quality.

Cognitive radio networks (CRNs) have emerged as advanced and promising paradigm to exploit the existing wireless spectrum opportunistically. It is crucial for users in CRNs to search for neighbors via rendezvous process and thereby establish the communication links to exchange the information necessary for spectrum management and channel contention etc. This paper focuses on the design of algorithms for blind rendezvous, i.e., rendezvous without using any central controller and common control channel (CCC). We propose a jump-stay based channel-hopping (CH) algorithm for blind rendezvous. The basic idea is to generate CH sequence in rounds and each round consists of a jump-pattern and a stay-pattern. Users “jump” on available channels in the jump-pattern while “stay” on a specific channel in the stay-pattern. Compared with the existing CH algorithms, our algorithm achieves the following advances: i) guaranteed rendezvous without the need of time-synchronization; ii) applicability to rendezvous of multi-user and multi-hop scenarios. We derive the maximum time-to-rendezvous (TTR) and the upper-bound of expected TTR of our algorithm for both 2-user and multi-user scenarios (shown in Table I). Extensive simulations are further conducted to evaluate performance of our algorithm.

A single-fed miniaturized wide-beamwidth circularly polarized implantable antenna, operating in the Industrial, Scientific, and Medical band (2.40-2.48 GHz), is designed and experimentally verified for subcutaneous real-time glucose monitoring applications. The proposed antenna features a very good miniaturization with the dimensions of 8.5 × 8.5 × 1.27 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> byemploying four C-shaped slots and a complementary split-ring resonator (CSRR). Meanwhile, by adjusting the slits of CSRR, circular polarization is realized. The simulation results in a three-layer phantom demonstrate that the impedance bandwidth is 12.2% (2.32- 2.62 GHz) with a peak gain of -17 dBi, and the 3-dB axial-ratio bandwidth is 2.4% (2.42-2.48 GHz) with a wide beamwidth of around 140°. An in vitro test was carried out in a pork slab, and the measured impedance bandwidth is 13% (2.31-2.63 GHz) with S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">11</sub> <; -10 dB, confirming the simulated results. The specific absorption rate distribution has been evaluated for the consideration of health safety, and the calculated link margin shows that the reliable communication can be guaranteed within 10 m in free space.

Locality and label information of training samples play an important role in image classification. However, previous dictionary learning algorithms do not take the locality and label information of atoms into account together in the learning process, and thus their performance is limited. In this paper, a discriminative dictionary learning algorithm, called the locality-constrained and label embedding dictionary learning (LCLE-DL) algorithm, was proposed for image classification. First, the locality information was preserved using the graph Laplacian matrix of the learned dictionary instead of the conventional one derived from the training samples. Then, the label embedding term was constructed using the label information of atoms instead of the classification error term, which contained discriminating information of the learned dictionary. The optimal coding coefficients derived by the locality-based and label-based reconstruction were effective for image classification. Experimental results demonstrated that the LCLE-DL algorithm can achieve better performance than some state-of-the-art algorithms.

With the increasing scale and complexity of the network, the network attack technology is also changing, such as malicious program attack, Trojan horse, distributed denial of service attack, worm, virus, web code injection, botnet, and other new network attack tools emerge in large numbers. As the core hotspot of network information security, network security situational awareness has received more and more attention. The traditional way of network security situational awareness prediction is relatively single. Usually, only one algorithm is used for perception and prediction, and its prediction accuracy is limited. To explore the application effect of intelligent learning algorithm, this study takes radial basis function (RBF) neural network as the main research object, optimizes RBF by simulated annealing (SA) algorithm and hybrid hierarchy genetic algorithm (HHGA), constructs RBF neural network prediction model based on SA–HHGA optimization, and carries out relevant experiments. The results show that the predicted situation value of the optimized RBF neural network in 15 samples is very close to the actual situation value. The neural network has good prediction effect and can provide assistance for the maintenance of network security. Received: 9 November 2021 | Revised: 7 March 2022 | Accepted: 22 March 2022 Conflicts of Interest The author declares that he has no conflicts of interest to this work.

We report the existence and stability of lattice solitons in parity-time ($\mathcal{PT}$)-symmetric mixed linear-nonlinear optical lattices in Kerr media. We focus on studying the characteristic effects on soliton propagation in the semi-infinite gap if we consider different amplitudes of real and imaginary parts of both the linear refractive index modulation profile and of periodic nonlinearity-modulation spatial distribution. It was found that the combination of $\mathcal{PT}$-symmetric linear and nonlinear lattices can stabilize lattice solitons and can provide unique soliton properties. It is revealed that the parameters of the linear lattice periodic potential play a significant role in controlling the extent of the stability domains and that the lattice solitons can stably propagate only in the low-power regime.

Cognitive radio networks (CRNs) have emerged as advanced and promising paradigm to exploit the existing wireless spectrum opportunistically. It is crucial for users in CRNs to search for neighbors via rendezvous process and thereby establish the communication links to exchange the information necessary for spectrum management and channel contention, etc. This paper focuses on the design of algorithms for blind rendezvous, i.e., rendezvous without using any centralized controller and common control channel (CCC). We propose a jump-stay channel-hopping (CH) algorithm for blind rendezvous. The basic idea is to generate CH sequence in rounds and each round consists of a jump-pattern and a stay-pattern. Users “jump” on available channels in the jump-pattern while “stay” on a specific channel in the stay-pattern. We prove that two users can achieve rendezvous in one of four possible pattern combinations: jump-stay, stay-jump, jump-jump, and stay-stay. Compared with the existing CH algorithms, our algorithm has the overall best performance in various scenarios and is applicable to rendezvous of multiuser and multihop scenarios. We derive upper bounds on the maximum time-to-rendezvous (TTR) and the expected TTR of our algorithm for both 2-user and multiuser scenarios (shown in Table 1). Extensive simulations are conducted to evaluate the performance of our algorithm.

Host-pathogen interactions are fundamental to our understanding of infectious diseases. Protein glycosylation is one kind of common post-translational modification, forming glycoproteins and modulating numerous important biological processes. It also occurs in host-pathogen interaction, affecting host resistance or pathogen virulence often because glycans regulate protein conformation, activity, and stability, etc. This review summarizes various roles of different glycoproteins during the interaction, which include: host glycoproteins prevent pathogens as barriers; pathogen glycoproteins promote pathogens to attack host proteins as weapons; pathogens glycosylate proteins of the host to enhance virulence; and hosts sense pathogen glycoproteins to induce resistance. In addition, this review also intends to summarize the roles of lectin (a class of protein entangled with glycoprotein) in host-pathogen interactions, including bacterial adhesins, viral lectins or host lectins. Although these studies show the importance of protein glycosylation in host-pathogen interaction, much remains to be discovered about the interaction mechanism.

Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (2D-SSA), which is a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trends, oscillations, and noise. Using the trend and the selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD that requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the execution time in feature extraction can be also dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, i.e., the k-nearest neighbor, is used for data classification. In addition, the combination of 2D-SSA with 1-D principal component analysis (2D-SSA-PCA) has generated the best results among several other approaches, demonstrating the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.

Skin cancer is one of the most dangerous diseases in the world. Correctly classifying skin lesions at an early stage could aid clinical decision-making by providing an accurate disease diagnosis, potentially increasing the chances of cure before cancer spreads. However, achieving automatic skin cancer classification is difficult because the majority of skin disease images used for training are imbalanced and in short supply; meanwhile, the model's cross-domain adaptability and robustness are also critical challenges. Recently, many deep learning-based methods have been widely used in skin cancer classification to solve the above issues and achieve satisfactory results. Nonetheless, reviews that include the abovementioned frontier problems in skin cancer classification are still scarce. Therefore, in this article, we provide a comprehensive overview of the latest deep learning-based algorithms for skin cancer classification. We begin with an overview of three types of dermatological images, followed by a list of publicly available datasets relating to skin cancers. After that, we review the successful applications of typical convolutional neural networks for skin cancer classification. As a highlight of this paper, we next summarize several frontier problems, including data imbalance, data limitation, domain adaptation, model robustness, and model efficiency, followed by corresponding solutions in the skin cancer classification task. Finally, by summarizing different deep learning-based methods to solve the frontier challenges in skin cancer classification, we can conclude that the general development direction of these approaches is structured, lightweight, and multimodal. Besides, for readers' convenience, we have summarized our findings in figures and tables. Considering the growing popularity of deep learning, there are still many issues to overcome as well as chances to pursue in the future.

In information-centric networking, accurately predicting content popularity can improve the performance of caching. Therefore, based on software defined network (SDN), this paper proposes Deep-Learning-based Content Popularity Prediction (DLCPP) to achieve the popularity prediction. DLCPP adopts the switch's computing resources and links in the SDN to build a distributed and reconfigurable deep learning network. For DLCPP, we initially determine the metrics that can reflect changes in content popularity. Second, each network node collects the spatial-temporal joint distribution data of these metrics. Then, the data are used as input to stacked auto-encoders (SAE) in DLCPP to extract the spatiotemporal features of popularity. Finally, we transform the popularity prediction into a multi-classification problem through discretizing the content popularity into multiple classifications. The Softmax classifier is used to achieve the content popularity prediction. Some challenges for DLCPP are also addressed, such as determining the structure of SAE, realizing the neuron function on an SDN switch, and deploying DLCPP on an OpenFlow-based SDN. At the same time, we propose a lightweight caching scheme that integrates cache placement and cache replacement-caching based on popularity prediction and cache capacity (CPC). Abundant experiments demonstrate good performance of DLCPP, and it achieves close to 2.1%~15% and 5.2%~40% accuracy improvements over neural networks and auto regressive, respectively. Benefitting from DLCPP's better prediction accuracy, CPC can yield a steady improvement of caching performance over other dominant cache management frameworks.

Plant phenotypic image recognition (PPIR) is an important branch of smart agriculture. In recent years, deep learning has achieved significant breakthroughs in image recognition. Consequently, PPIR technology that is based on deep learning is becoming increasingly popular. First, this paper introduces the development and application of PPIR technology, followed by its classification and analysis. Second, it presents the theory of four types of deep learning methods and their applications in PPIR. These methods include the convolutional neural network, deep belief network, recurrent neural network, and stacked autoencoder, and they are applied to identify plant species, diagnose plant diseases, etc. Finally, the difficulties and challenges of deep learning in PPIR are discussed.

Metaverse is a living space and cyberspace that realizes the process of virtualizing and digitizing the real world. It integrates a plethora of existing technologies with the goal of being able to map the real world, even beyond the real world. Metaverse has a bright future and is expected to have many applications in various scenarios. The support of the Metaverse is based on numerous related technologies becoming mature. Hence, there is no doubt that the security risks of the development of the Metaverse may be more prominent and more complex. We present some Metaverse-related technologies and some potential security and privacy issues in the Metaverse. We present current solutions for Metaverse security and privacy derived from these technologies. In addition, we also raise some unresolved questions about the potential Metaverse. To summarize, this survey provides an in-depth review of the security and privacy issues raised by key technologies in Metaverse applications. We hope that this survey will provide insightful research directions and prospects for the Metaverse's development, particularly in terms of security and privacy protection in the Metaverse.

A compact dual-band and dual-circular polarization (CP) stacked patch antenna is investigated for BeiDou navigation satellite system. The proposed antenna employs a single coaxial probe to feed two layers of patches simultaneously, realizing dual-band operation. The bottom patch with symmetrical slant cornercuts and the top one with two rectangular stubs on the diagonal produce a pair of degenerated modes, achieving CP radiation. Furthermore, two L-shaped stubs are loaded on the top patch to widen impedance bandwidth. A prototype operating at the bands of 1.615/2.492 GHz was fabricated and measured. Both the left- and right-handed CP radiations are obtained concurrently in the dual bands. The proposed antenna has a profile of 4.6 mm (0.024λ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> ) and occupies an area of 70 mm × 70 mm (0.38λ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> × 0.38λ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> ). The measured impedance bandwidth with |S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">11</sub> | <; -10 dB is 9.1% and 5.1% for the lower and upper bands, respectively. The measured 3 dB axial-ratio bandwidth is more than 1% for both bands.

Both interclass variances and intraclass similarities are crucial for improving the classification performance of discriminative dictionary learning (DDL) algorithms. However, existing DDL methods often ignore the combination between the interclass and intraclass properties of dictionary atoms and coding coefficients. To address this problem, in this paper, we propose a discriminative Fisher embedding dictionary learning (DFEDL) algorithm that simultaneously establishes Fisher embedding models on learned atoms and coefficients. Specifically, we first construct a discriminative Fisher atom embedding model by exploring the Fisher criterion of the atoms, which encourages the atoms of the same class to reconstruct the corresponding training samples as much as possible. At the same time, a discriminative Fisher coefficient embedding model is formulated by imposing the Fisher criterion on the profiles (row vectors of the coding coefficient matrix) and coding coefficients, which forces the coding coefficient matrix to become a block-diagonal matrix. Since the profiles can indicate which training samples are represented by the corresponding atoms, the proposed two discriminative Fisher embedding models can alternatively and interactively promote the discriminative capabilities of the learned dictionary and coding coefficients. The extensive experimental results demonstrate that the proposed DFEDL algorithm achieves superior performance in comparison with some state-of-the-art dictionary learning algorithms on both hand-crafted and deep learning-based features.