Search

The detection of zero-day attacks remains one of the most critical challenges in cybersecurity. This systematic literature review focuses on the various AI-based methods employed for detecting zero-day attacks, identifying both the strengths and weaknesses of these approaches. By critically evaluating existing literature, this review provides new insights and highlights the gaps that future research must address. The findings suggest that while artificial intelligence, particularly machine learning, offers promising solutions, there are significant challenges related to data availability, algorithmic complexity, and real-time application. This review contributes to the field by providing a comprehensive analysis of current AI-driven methods and proposing future research directions to enhance zero-day attack detection.

Effective content recommendation in new media relies heavily on algorithmic approaches to enhance user engagement and satisfaction. This abstract explores the current landscape of content recommendation systems in new media platforms, focusing on how algorithms are leveraged to deliver personalized and relevant content to users. The success of these systems hinges on their ability to analyze vast amounts of user data, such as browsing history, preferences, and social interactions, to predict content that aligns with individual interests. Key algorithmic techniques include collaborative filtering, content-based filtering, and hybrid methods, each serving distinct purposes in enhancing recommendation accuracy. The abstract examines the challenges and opportunities in algorithmic content recommendation, including issues of privacy, algorithm bias, and the need for continuous algorithm refinement. Effective algorithms not only increase user engagement but also drive business objectives such as increased user retention and monetization through targeted advertising. Ultimately, the abstract concludes by highlighting the importance of ongoing research and development in algorithmic approaches to keep pace with the evolving demands and complexities of new media content recommendation systems.

It is essential to achieve the high-accuracy recognition of low probability of intercept (LPI) radar signals in modern electronic warfare. However, under low signal-to-noise ratio (SNR), the recognition accuracy of the LPI radar signals is relatively low. In this paper, a novel radar signal recognition method based on Convolutional Stacked Recurrent Deep Neural Network (CSRDNN) is proposed. Firstly, we design a Convolutional Neural Network (CNN) to expand the feature space of input time domain signals, the features extracted by CNN were then used as inputs of the Stacked Recurrent Neural Networks (SRNN) module. In the SRNN module, we sequentially stack GRU, LSTM, and BGRU, enabling the model to better handle the short-term and long-term dependence of signal features and effectively solve asynchronous problems in unidirectional RNN networks. Subsequently, a Fully Connected Deep Neural Network (FCDNN) was employed to accomplish the recognition task. In addition, we design a training algorithm composed of the Nesterov-Adaptive Moment Estimation (Nadam) algorithm and the CosineAnnealing Learning Rate (LR) adjustment strategy to improve the training efficiency of the model. The experimental results demonstrate that the proposed model has higher recognition accuracy at low SNR compared to other models, with an overall recognition accuracy of 92.96% at −4 dB.

This research proposes a music-adaptive IoT exercise system integrating edge computing and blockchain technology to enhance middle-distance running exercise. The system personalizes music tempo based on real-time biosignals to provide a tailored experience. Experiments demonstrate that adaptive music significantly augments users’ physiological, psychological, and behavioral responses compared to no music, improving engagement and effectiveness. The system overcomes limitations in existing solutions regarding personalization, real-time adaptation, and data privacy. Edge computing enables on-device processing, and blockchain encryption guarantees security. Results highlight the benefits of HR adaptive running music for arousal, pleasure, and performance. This study offers insights into designing personalized, rewarding exercise systems using emerging technologies. It has promising implications for exercise, rehabilitation, and recreational fitness. Further work should investigate specific mechanisms and long-term impacts, integrate additional signals and music parameters, and enhance security. In conclusion, this innovative system combines IoT, adaptive music, edge computing, and blockchain to unlock new possibilities in exercise experiences.

Edge computing plays a crucial role in the field of the Internet of Vehicles (IoV), meeting the resource and latency requirements of time-sensitive vehicle applications. However, the emergence of numerous compute-intensive and latency-sensitive applications, such as augmented reality and autonomous driving, has led to a situation where traditional edge computing architectures cannot meet the increasing application demands of the IoV. This paper extends the paradigm of vehicular edge computing to a collaborative cloud-edge cluster resource provisioning framework. Integrating compute resources from multiple Edge Service Providers (ESP) and the cloud enables horizontal and vertical collaborative computation offloading among service nodes. To facilitate resource sharing among different ESPs, we introduce a dynamic pricing model and utilize software-defined networking (SDN) to tackle this scenario’s complex resource management challenges. Furthermore, with the optimization objectives of minimizing task computation latency and maximizing the profits of ESPs, we establish a mathematical model. Before resource allocation, we employ a clustering algorithm to determine initial offloading decisions, reducing the dimensionality of the action space. Subsequently, we employ the Double Deep Q-Network (DDQN) algorithm to achieve a rational allocation of compute resources. Simulation results demonstrate that compared to the Deep Q-Network (DQN) algorithm and greedy strategy, the proposed approach reduces latency by 18.18% and 34.85%, respectively, while increasing the profits of edge service providers by 16.25% and 33.33%, respectively.

The use of video surveillance to monitor interrogation behavior can effectively maintain judicial civility in the context of law enforcement cases. However, analyzing and reviewing law enforcement videos can be a time-consuming and resource-intensive process, particularly in the manual identification of interrogation violations. This work is dedicated to the development of an intelligent recognition system for interrogation violations by using a spatio-temporal attention fusion SlowFast Network. To address the issue of feature information underutilization in the slow path of the traditional SlowFast, a slow-to-fast path is incorporated into the original SlowFast to enhance learning. The model fuses the attention of spatial and temporal channels, replacing the traditional convolution module with this new approach. The proposed model was evaluated using the publicly available UCF101 action recognition dataset, resulting in a 1.52% improvement in Top-1 recognition accuracy compared to the traditional SlowFast. Based on two custom interrogation misconduct datasets, the proposed model was evaluated, achieving a recognition rate of 99.16% for interrogation misconduct. This demonstrates its effectiveness in identifying misconduct behaviors inside interrogation rooms. Compared to some advanced behavior recognition models, the proposed model demonstrates strong competitiveness in identifying misconduct during interrogations.

Owing to the nature of wireless channels, wireless transmission is vulnerable to attacks by adversaries; therefore, security has always been a critical issue in wireless networks. In this context, intelligent reflecting surfaces (IRS), as an emerging and promising technology, synergize with physical layer security (PLS), offering novel avenues to enhance privacy and resistance against interference in wireless communication. This paper investigates a cooperative jamming communication model assisted by IRS. Under the constraints of minimum safe rate and inaccurate channel state information (CSI), a deep reinforcement learning (DRL)-based framework is proposed to jointly optimize the BS transmitting beamforming power distribution and IRS phase shift matrix to maximize the system energy efficiency. We first formulate an anti-jamming communication optimization problem as a Markov decision process (MDP) framework and then design a DRL-based algorithm, in which the joint design is obtained through trial-and-error interactions with the environment by observing predefined rewards in the context of continuous state and action to generate an optimal policy. The simulation results show that when the number of IRS components is increased from 20 to 100, the proposed scheme can improve energy efficiency by 40.1%, which is better than other schemes.

Pedestrian detection is the use of computer vision technology to identify and accurately locate pedestrians in image or video data, which has a strong use value. This technology can be used as the research basis for visual tasks such as person re-identification, human pose estimation and behavior analysis, and can also be applied to industrial fields such as intelligent security, automatic driving and human-computer interaction. However, the problems of low image resolution, blurred appearance, large scale difference of pedestrians, occluded pedestrians and complex background still bring great challenges to the detection performance. To solve these problems, this paper proposes a high-performance pedestrian detection network dedicated to difficult conditions: DCPDN. Firstly, we design an optimized super-resolution reconstruction network to preprocess the image to alleviate the performance damage caused by low-resolution and blurred images. Then, to solve the multi-scale problem in pedestrian detection, we propose a weighted cross-scale feature fusion module, which adopts a hierarchical detection strategy to deal with pedestrian objects of different scales while fully fusing feature maps of different levels. Finally, to solve the occlusion problem that has plagued pedestrian detection for a long time, we design an occlusion processing module based on graph convolutional network, which can effectively use the correlation information between different parts of the human body and promote the feature expression of occluded objects. On the CityPersons dataset, the ${MR}^{-2}$ of the detector is reduced by 6.9%, 19.2%, 8.9%, 1.9%, 3.6% and 14.2%, respectively, corresponding to different partition subsets of R, HO, A, L, M and S. On the Caltech dataset, corresponding to different divisions of R, HO, A, L and S, the ${MR}^{-2}$ of the detector is reduced by 9.9%, 15.8%, 16.3%, 6.8% and 25.8%, respectively. The experimental results show that the performance improvement of the detector is significant on both severe occlusion (HO) and small scale (S) subsets. After testing, the algorithm has strong robustness to occluded pedestrians, and can be easily embedded in other detection frameworks. Our DCPDN is able to compete with the state of the art methods and is especially effective when dealing with the pedestrian detection problem under difficult conditions.

The increasing demand for wireless Internet of Things (IoT) calls for power efficient RF energy harvesting approach. The current dominant single-band RF energy harvesting front-end system restricts itself to a single frequency, which is at the risk of ineffective operation when the harvesting frequency is unavailable. This paper reviews and explores the alternative approach of a multiband RF energy-harvesting front-end system. It covers all essential circuitry of a multiband RF energy harvesting front-end system, starting from the recent RF surveys which investigate the typical, usable, and high-strength RF input, to the overview of the state-of-the-art antenna, impedance matching network (IMN), and RF-DC rectifier. The recent advancement of the multiple RF input harvesting abilities, reflection loss minimization, and performance improvement are comprehensively reviewed for different operating conditions, and this review also presents the advantages and disadvantages of the different circuit architecture combinations for multiband RF energy harvesting front-end systems. In summary, this review aims to fill the research gap in the further advancement of multiband RF energy harvesting towards enhancing its performance through optimal circuit integration of the front-end system.

In the electroencephalography (EEG) based cross-subject motor imagery (MI) classification task, the device and subject problems can cause the time-related data distribution shift problem. In a single-source to single-target (STS) MI classification task, such a shift problem will certainly provoke an increase in the overall data distribution difference between the source and target domains, giving rise to poor classification accuracy. In this paper, a novel multi-subdomain adaptation method (MSDAN) is proposed to solve the shift problem and improve the classification accuracy of the traditional approaches. In the proposed MSDAN, the adaptation losses in both class-related and time-related subdomains (that are divided by different data labels and session labels) are obtained by measuring the distribution differences between the source and target subdomains. Then, the adaptation and classification losses in the loss function of MSDAN are minimized concurrently. To illustrate the application value of the proposed method, our method is applied to solve the STS MI classification task about data analysis with respect to the brain-computer interface (BCI) competition III-IVa dataset. The resultant experiment results demonstrate that compared with other well-known domain adaptation and deep learning methods, the proposed method is capable of solving the time-related data distribution problem at higher classification accuracy.

With the development of multimedia editing technologies, the copyright protection has attacked more attentions. Reversible data hiding (RDH), in which the cover can be recovered losslessly, is an effect method to eliminate embedding distortions. As a typical RDH method, histogram shifting (HS) is used widely. Most existing RDH schemes based on HS usually build sharp histograms by predicting and sorting techniques. To make use of spatial correlations of multimedia, several RDH schemes based on multiple HS (MHS) are proposed to protect copyright, in which some rigid rules are used to build multiple histograms. Against images, videos have more spatial and temple correlations and it is easier to acquire sharper histograms. In this paper, a video MHS scheme based on compression sensing (CS) is proposed. As a linear sensing algorithm, CS can measure macroblock residuals by reducing corrections among pixels to acquire distinguishable macroblock features, while keeping their statistical characteristics immutable. By employing CS, macroblocks with similar characteristics cluster together to formulate multiple histograms. For each of these histograms, data embedding is implemented to reduce shifting distortions by expanding the outermost bins while other bins are unchanged. Experimental results show that the quality of most test videos in our scheme are higher than that in the state-of-art schemes.

Surface deformation poses a great threat to the safety of Jinchuan mining area production activities. At present, the spatio-temporal evolution law and mechanism of surface deformation in the Jinchuan mining area are unclear, and it is difficult to obtain reliable prediction results using the existing spatio-temporal prediction methods due to the lack of model parameters or relevant data. To solve these problems, this study proposes a new unified convolutional neural network with peephole long short-term memory (CNN-PhLSTM). Small baseline subset interferometric synthetic aperture radar (SBAS-InSAR) technology was used to obtain the spatio-temporal evolution characteristics of surface deformation in the period of 2014–2021. Time series InSAR deformation data are merged into a unified network model in series with a time-distributed CNN segmentation and stacked PhLSTM. The InSAR measurement results are shown to be reliable by comparison and verification with the benchmark and InSAR results of different orbits. The proposed CNN-PhLSTM model was evaluated by mean absolute error and structural similarity (SSIM) evaluation indexes, and was compared with support vector regression (SVR), multilayer perceptron (MLP) and CNN-LSTM models. The results show three continuous subsidence areas, namely the Longshou, second western and third eastern mining areas. The cumulative surface deformation continued to increase from 2014 to 2021. Faults and lithology control the spatial distribution of surface deformation in the Jinchuan mining area. The prediction results demonstrate that the surface deformation range will continue to expand and that time-series surface deformation will show a slow deceleration trend in the next two years.

Interferometric phase filtering is a crucial step in the interferometric synthetic aperture radar (InSAR) data processing, which is also important for improving the accuracy of topography mapping and deformation monitoring. Most of the commonly used phase filtering methods perform windowing computations based on the statistical characteristics of a single interferogram in the spatial or frequency domain. However, the difficulty in taking into account the diversity and complexity of the phase image results in filtering methods with weak denoising, limited detail preservation, and poor generalization ability. At the same time, regardless of the spatial or frequency domain, improved phase filtering performance inevitably leads to the problem of declining effectiveness. This article proposes a phase filtering method based on the deep convolution neural network with multiscale feature dynamic fusion (MSFF). Unlike the traditional feedforward neural networks, the proposed method adopts a strategy of multiscale feature dynamic fusion that accounts for the deep and shallow features of the interferometric phase while also taking into account image detail preservation and noise suppression during phase filtering. Based on both subjective and objective evaluations, the experimental results using the simulated data prove that the proposed method has better noise suppression and detail preservation than the commonly used methods and that the filtering performance is less dependent on noise level. Experiments using the real data confirm that the proposed method has better generalization ability and can meet the precision requirements of practical applications. The method presented in this article can provide a new approach for research in high-precision InSAR data processing technology while also offering technical support for practical InSAR applications.

For real-time industrial wireless edge-enabled cyber-physical systems, this paper proposes an adaptive secure transmission mechanism under the large scale path loss to fight against randomly distributed illegal sensing devices. According to changes in the communication environment, the proposed scheme adaptively obtains the corresponding optimal power allocations and wiretap code rates. Specifically, by injecting artificial noise, an edge controller provides confidential control information for a legitimate wireless sensing device and achieves non-confidential data service for a service subscriber, simultaneously. Considering whether the secrecy and non-confidential service rate constraints are met, a dynamic secrecy transmission mechanism is introduced to minimize secrecy outage probability (SOP) based on random channel realizations. The optimal power allocation for minimizing the SOP is derived through the efficient alternating optimization algorithm with numerical methods. Numerical results show that the proposed adaptive secure transmission strategy can efficiently utilize transmit power to achieve non-confidential information traffic without compromising the secrecy performance.

Diabetic Retinopathy (DR) may lead to blindness in diabetic patients, which is one of the most severe eye diseases. Therefore, using automatical technology to detect DR at the early phase has very vital clinical significance. In order to detect the microaneurysms (MAs) and hard exudates (HEs) of DR, a novel detection method based on deep symmetric convolutional neural network is proposed in this paper. The symmetric convolutional structure is used to improve the effectiveness of feature extraction. The proposed method also can overcome the imbalance of positive and negative samples to avoid overfitting by increasing the width and depth of the network. Furthermore, different network structures (convolution, pooling) are used to achieve different feature filtering in the stage of feature extractions. According to the experimental results, the proposed method is superior to the state-of-the-art approach on the public dataset DIARETDB1 (DB1). The detection accuracy of the objects is 92.0%, 93.2%, 93.6%, when using different filtering structures (convolution, max-pooling, ave-pooling) respectively. The detection of microaneurysms is much improved by using ave-pooling layer for feature filtering, and the max-pooling layer can improve the detection of hard exudates.

Loss reduction to improve the power efficiency in active integrated antenna (AIA) is a key design drive. This paper first analyzes the loss mechanism in a convention AIA structure. A new integration scheme of a GaN power amplifier (PA) transistor with an antenna without using any output matching network (OMN) and harmonic tuning network (HTN) is then proposed to construct a seamlessly integrated AIA. This is achieved by a novel design of a slot antenna with optimized input impedance at its fundamental frequency as well as for harmonic tuning, which essentially absorbs the OMN and HTN functions in the conventional Class-F power amplifier design. By eliminating these passive networks between the transistor and the antenna, the associated insertion and mismatch losses as well as the overall circuit size are reduced. For verification, two prototypes are designed, fabricated and measured, one with the integrated design and the other with a conventional design for comparison. Both AIAs operate between 3.4 and 3.6 GHz. Experimental results show that the power-added efficiency (PAE) of the seamlessly integrated AIA is over 52% within the operating band. Compared with the conventional cascaded design of a PA and an antenna, the PAE is improved by 14.2%.

The stability of flying of a hummingbird-like flapping-wing micro air vehicle (MAV) has been challenging. In this paper, experimental studies are reported on the tail shapes of hummingbird-like flapping-wing MAVs, since tails play an important role in-flight stability. Dynamics parameters of hummingbird tails are firstly studied and evaluated. Then man-made tails inspired by the natural hummingbirds are designed, manufactured and optimized for experimental tests. The results show that lift generated by the tail is independent of a fan angle, whereas the pitch moment is related to the fan angle. Further, the tail can be applied to stabilising hovering twin-wing flapping wing MAVs.

Retinopathy of prematurity (ROP) is one of the main causes of childhood blindness. However, insufficient ophthalmologists are qualified for ROP screening. The objective of this paper is to evaluate the performance of a deep neural network (DNN) for the automated screening of ROP. The training and test sets came from 420 365 wide-angle retina images from ROP screening. A transfer learning scheme was designed to train the DNN classifier. First, a pre-processing classifier separated unqualified images. Then, pediatric ophthalmologists labeled each image as either ROP or negative. The labeled training set (8090 positive images and 9711 negative ones) was used to fine-tune three candidate DNN classifiers (AlexNet, VGG-16, and GoogLeNet) with the transfer learning approach. The resultant classifiers were evaluated on a test dataset of 1742 samples and compared with five independent pediatric retinal ophthalmologists. The receiver operating characteristic (ROC) curve, ROC area under the curve, and precision-recall (P-R) curve on the test dataset were analyzed. Accuracy, precision, sensitivity (recall), specificity, F1 score, the Youden index, and the Matthews correlation coefficient were evaluated at different sensitivity cutoffs. The data from the five pediatric ophthalmologists were plotted in the ROC and P-R curves to visualize their performances. VGG-16 achieved the best performance. At the cutoff point that maximized F1 score in the P-R curve, the final DNN model achieved 98.8% accuracy, 94.1% sensitivity, 99.3% specificity, and 93.0% precision. This was comparable to the pediatric ophthalmologists (98.8% accuracy, 93.5% sensitivity, 99.5% specificity, and 96.7% precision). In the screening of ROP using the evaluation of wide-angle retinal images, DNNs had high accuracy, sensitivity, specificity, and precision, comparable to that of pediatric ophthalmologists.

K-pass pixel value ordering (PVO) is an effective reversible data hiding (RDH) technique. In k-pass PVO, the complexity measurement may lead to a weak estimation result because the unaltered pixels in a block are excluded to estimate block complexity. In addition, the prediction-error is computed without considering the location relationship of the second largest and largest pixels or the second smallest and smallest pixels. To this end, an improved RDH technique is proposed in this paper to enhance the embedding performance. The improvement mainly lies in the following two aspects. First, some pixels in a block, which are excluded from data hiding in some existing RDH methods, are exploited together with the neighborhood surrounding this block to increase the estimation accuracy of local complexity. Second, the remaining pixels in a block, i.e., three largest and three smallest pixels are involved in data embedding. Taking three largest pixels for example, when the difference between the largest and third largest pixels is relatively large (e.g., > 1), we improve k-pass PVO by considering the location relationship of the second largest and largest pixels. The advantage of doing this is that the difference valued 3 between the maximum and the second largest pixel which is shifted in k-pass PVO, is able to carry 1 bit data in our method. In other words, a larger amount of pixels are able to carry data bits in our scheme compared with k-pass PVO. Abundant experimental results reveal that the proposed method achieves preferable embedding performance compared with the previous work, especially when a larger payload is required.

In the reliability literature, reliability allocation is an important and widely studied topic. The existing reliability allocation methods, however, have limitations, including imprecise system decomposition, single-factor consideration, and poor practicability. To overcome those limitations, we propose an integrated fuzzy reliability allocation method based on micro-motion decomposition, cost function, and multi-factor analysis. The problems in the existing methods caused by equally weighted factors and influences of failures were overcome by correcting the traditional risk priority number method and evaluating the uncertainties and subjective factors during allocation using fuzzy language and triangular fuzzy number. A cost model was established based on the state of the art, subsystem intricacy, and environmental conditions, with which the issues of difficulty applying cost statistics and computational complexity in the current allocation methods were solved. The contribution of this paper is as follows. A fast approach of Pareto-optimal solution recommendation using the Pareto reliability index has been utilized to recommend a list of optimal ranking for decision-makers. Besides, the moving mean of the average precision and the moving mean standard deviation are utilized to demonstrate the trend of the evolutionary process. A multi-objective swarm bat algorithm has been developed to handle the multi-objective problems and its feasibility has been verified in a case study comparing the performance of the proposed method with that of the existing methods.

Food security is one of the societal challenge topics. As one-third of all food consumed by humans relies on animal pollination currently, this research provides an emerging solution to food supply reduction caused by population shrinking of natural pollinators, so as to reduce its impact on ecological relationships, ecosystem conservation and stability, genetic variation in the crop plant community, floral diversity, specialisation and evolution. This paper develops a conceptual technical roadmap of autonomous pollination for future farming using robotic micro air vehicle pollinators (MPrs). The research provides new insights into autonomous design and manufacture and into possible ways to increase the production efficiency which shortens the time from lab to market. The autonomous MPrs are realized using artificial intelligence and human expertise in the loop for smart agricultural industry. Further, this work identifies scientific and technological advances that are expected to translate, within proposed regulatory frameworks, into the pervasive use of MPrs for agricultural applications and beyond.

Interactive axis extraction for high-dimensional data visualization has been demonstrated to be powerful in high-dimensional data exploring and understanding. The extracted axes help to yield new 2-D arrangements of data points, providing new insights into the data. However, the existing interfaces for extraction only support linear axes or non-linear axes without specific semantics. When the data points lie in a manifold, it is hard to capture intrinsic features of the manifold by either linear axes or non-linear axes without specific semantics. Furthermore, a dataset with complicated topology would contain holes and branches. While a branch often indicates a local trend, it may not make sense to project data points to an axis in a different branch. In this paper, we propose an interactive visual interface to identify and extract intrinsic axes in high-dimensional data. The system contains four major views. The topology view presents the skeleton-based topology of the dataset. The detail view provides a force-directed layout of a high-dimensional data and allows interactive extracting intrinsic axes. The characteristics of extracted axes are visualized in the intrinsic axes view. The projection view layouts data points aligning with extracted intrinsic axes. Case studies and comparative experiments demonstrate the usefulness of our visual analytics system.

MATLAB® has built in five derivative-free optimizers (DFOs), including two direct search algorithms (simplex search, pattern search) and three heuristic algorithms (simulated annealing, particle swarm optimization, and genetic algorithm), plus a few in the official user repository, such as Powell's conjugate (PC) direct search recommended by MathWorks®. To help a practicing engineer or scientist to choose a MATLAB DFO most suitable for their application at hand, this paper presents a set of five benchmarking criteria for optimization algorithms and then uses four widely adopted benchmark problems to evaluate the DFOs systematically. Comprehensive tests recommend that the PC be most suitable for a unimodal or relatively simple problem, whilst the genetic algorithm (with elitism in MATLAB, GAe) for a relatively complex, multimodal or unknown problem. This paper also provides an amalgamated scoring system and a decision tree for specific objectives, in addition to recommending the GAe for optimizing structures and categories as well as for offline global search together with PC for local parameter tuning or online adaptation. To verify these recommendations, all the six DFOs are further tested in a case study optimizing a popular nonlinear filter. The results corroborate the benchmarking results. It is expected that the benchmarking system would help select optimizers for practical applications.

In the era of artificial intelligence (AI), many industry sectors, including space exploration, have experienced a shift in the way business is conducted due to the widespread use of AI technologies. In the past few years, AI has become a key tool used to explore the universe in space missions. In this paper, a multi-objective optimal design for payload orbital transfer involving space tethers is proposed based on a computational intelligence-assisted design framework with the artificial wolf pack algorithm (AWPA). Enlightened by the social behaviors of a wolf pack and its swarm intelligence, the AWPA is utilized for optimization problems in which a logsig function randomly obtains assignments for parents and offspring. $Swarmwolf$ , a simulation toolbox with given initial conditions. The proposed method effectively performs optimization tasks based on index of evolutionary pathway trends, has been defined to demonstrate the optimizing process. The results show that the proposed approach works expeditiously for the optimization of space tether model and its application.

In this paper, the interpretation of hovering flight for hummingbirds is studied from a hummingbird morphology perspective (muscle and skeleton) including weight distribution, followed by a discussion of hovering aerodynamics. Next, by studying the scale laws, geometry similarity, and statistical analysis on wing parameters, the parametric relation between wing performances and weight is studied, followed by flapping wing micro autonomous drones (FWMADs) design. The efficiency of the designed wings based on the scaling law is verified by flying test. Material difference and methods of design are summarized. Last, the morphology of bird's tails is presented, and then the designs of tails are introduced, followed by discussion of tail performances. The results show that the tail could be predicted to apply to the stability of hovering twin-wing FWMADs. The current studies provide a simple but powerful guideline for biologists and engineers who study the morphology of hummingbirds and design FWMADs.

Retinopathy of prematurity (ROP) has been one of the worldwide causes of blindness among children. Grading and treatment guidelines of ROP are mainly based on zone, stage, and plus disease. For serious ROP, identifying zone is more important than staging. However, identifying zone I from RetCam fundus images is not accurate and subjective by ophthalmologists. To address it, we develop a new deep learning framework to automatically identify zone I from RetCam images. Specifically, we train a deep convolutional neural network (DCNN) algorithm based on the RetCam images. The disc and macular center in terms of the threshold of intersection over union (IOU) were identified automatically. The algorithm is validated on fundus images and results show that zone I identification accuracy of 91% is achieved when the IOU threshold is 0.8. The obtained promising identification accuracy of zone I from the RetCam images indicates the potential applications in ROP grading, monitoring, and prognosis for infants.

As a novel financing method, peer-to-peer (P2P) lending has drawn extensive attention as it provides those financers who cannot participate in the traditional financial market with funds. In P2P lending marketplaces, one of the crucial challenges that P2P online lending platforms are facing is to accurately predict the default risk of each loan by tapping into default prediction models, thus effectively helping P2P lending companies avoid credit risks. That traditional credit risk prediction models fail to meet the demand of P2P lending companies for default risk prediction, which is because of the uneven distribution of credit data samples in the P2P lending marketplaces (i.e., the default sampled data are scarce). In this paper, we designed a multi-round ensemble learning model based on heterogeneous ensemble frameworks to predict default risk. In this model, an extreme gradient boosting (XGBoost) is initially used for ensemble learning, and the XGBoost, deep neural network, and logistic regression are then regarded as heterogeneous individual learners to undergo a linear weighted fusion. To verify the designed default risk prediction model, real credit data from a famous P2P online lending marketplace in China were used in a test. The results of the experiment indicate that this model can effectively increase the predictive accuracy compared with traditional machine learning models and ensemble learning models.

To detect multiple sclerosis (MS) diseases early, we proposed a novel method on the hardware of magnetic resonance imaging, and on the software of three successful methods: biorthogonal wavelet transform, kernel principal component analysis, and logistic regression. The materials were 676 MR slices containing plaques from 38 MS patients, and 880 MR slices from 34 healthy controls. The statistical analysis showed our method achieved a sensitivity of 97.12±.14%, a specificity of 98.25±0.16%, and an accuracy of 97.76±0.10%. Our method is superior to five state-of-the-art approaches in MS detection.

Recently, more and more traditional services are being migrated into a cloud computing environment that makes the quality of service (QoS) becomes an important factor for service selection and optimal service composition while forming cross-cloud service applications. Considering the nonlinear and dynamic property of QoS data, it is so difficult to achieve dynamic prediction while designing a QoS prediction method with unsatisfactory prediction accuracy. It is thus desirable to explore how to design an effective approach by incorporating some intelligent techniques into the QoS prediction method to improve prediction performance. In this paper, motivated by the adaptive critic design and Q-learning technique, we propose a novel QoS prediction approach to serve this purpose through the combination of fuzzy neural networks and adaptive dynamic programming (ADP), i.e., an online learning scheme. This approach extracts fuzzy rules from QoS data and employs the ADP method to parameter learning of the fuzzy rules. Moreover, we provide a convergence boundedness result for our proposed approach to guarantee the stability. Experimental results on a large-scale QoS service data set verify the prediction accuracy of our proposed approach.