Search

Thanks to the increasing number and massive coverage, delivery drones, equipped with various sensors, have demonstrated significant but unexplored potentials for large-scale and low-cost urban sensing during package delivery. In this paper, we propose novel studies on the reutilization of such delivery drone resources to fill this void in urban crowdsensing. Accounting for interdependency between flying/sensing and drone delivery weight, we jointly optimize route selection, sensing time, and delivery weight allocation, to maximize delivery and sensing utility under drones energy constraints. This problem is formulated as a non-convex mixed integer non-Linear programming problem, which is proved to be NP-hard. To address this intricate problem, we propose near-optimal algorithms that leverage the equivalent objective function construction, the local search scheme, and the alternating iteration technique. Theoretical analysis indicates that our algorithms can achieve the 1/(4+ -approximation ratio (where is an arbitrarily small positive parameter and the convergence guarantee in polynomial time, for the scenarios of fixed and adjustable delivery weights, respectively. Extensive trace-based simulations and field experiments demonstrate that ours can significantly improve the delivery & sensing utility by 124.7% and the energy utilization rate by 72.2% on average, compared with the drone delivery without reusing.

A crucial privacy-driven issue nowadays is re-identifying anonymized social networks by mapping them to correlated crossdomain auxiliary networks. Prior works are typically based on modeling social networks as random graphs representing users and their relations, and subsequently quantify the quality of mappings through cost functions that are proposed with insufficient theoretical support. Also, it remains unknown how to algorithmically meet the demand of such quantifications, i.e., to find the minimizer of the cost functions. We address those concerns in a social network modeling parameterized by community structures that can be leveraged as side information for de-anonymization. By Maximum A Posteriori (MAP) estimation, our first contribution is a series of MAP-based cost functions, which, when minimized, enjoy superiority to previous ones in finding the correct mapping with the highest probability. The feasibility of the cost functions is then for the first time algorithmically characterized.We prove the general multiplicative inapproximability, and thus propose two heuristics, which, respectively, enjoy an -additive approximation and a conditional optimality in carrying out successful user re-identification. Our theoretical findings are also empirically validated, with one of the datasets extracted from rare true cross-domain networks that reproduce genuine social network de-anonymization. Both theoretical and empirical observations manifest the importance of community information in enhancing privacy inferencing.

A major shortcoming of the conventional car-following models is that these models only consider the current spacing and speeds of the target vehicle and its immediate leading vehicle, without taking into account prior driving actions, even for those from the same driver. In other words, the numerous prior experiences have no influence in predicting vehicular movements for the next time step. In this research, we propose a machine-learning-based data-driven methodology that is able to take advantage of the high-resolution historical traffic data in the current data-rich era, to predict vehicular movements in an accurate manner with high computational efficiency. The proposed car-following model has a simple model structure based on a fixed-radius near neighbors (FRNN) search algorithm and it can be applied to high-resolution, real-time vehicle movement prediction, modeling, and control. A comprehensive performance comparison is also conducted among the proposed car-following model, another similar data-driven model, and two conventional formula-based models. The results indicate that the FRNN algorithm-based car-following model is superior to all other three models in terms of prediction accuracy and is more computationally efficient compared to its data-driven-based counterpart. Some extensive applications of the proposed car-following model are also discussed at the end of this article.

The state of health (SOH) estimation is of great importance in the battery management system. However, the accurate real time state of health estimation is still a challenge because the capacity of the battery cannot be precisely monitored and measured in real-time. In response to this problem, this paper proposes an online SOH monitoring method for lithium-ion battery (LIB) based on the charge-discharge feature of battery in real time. Firstly, the relationship between the charge-discharge feature and SOH through principal component analysis (PCA) is analyzed, and the appropriate feature which can precisely describe the battery aging process is then selected. Secondly, as to decompose and denoise the collected charge-discharge feature data, the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) is applied. The major trend obtained is predicted by logistic regression based on the sliding time window (LR-STW), and the minor fluctuation is predicted by Kalman Filter (KF). Finally, an online SOH estimation method based on RBF neural network is proposed to establish the mapping relationship between feature and SOH. The proposed method was verified by NASA experimental data. The experimental results show that the proposed method can accurately predict the LIB SOH in real time, and the robustness of the proposed method is strengthened and promised.

With the rapid growth in the demand for location-based services in indoor environments, wireless fingerprint localization has attracted increasing attention because of its high precision and easy implementation. However, an effective method does not exist owing to the problems of data loss, noise interference in the fingerprint database, and being time-consuming during the offline training phase. Therefore, this paper presents a novel indoor wireless fingerprint localization algorithm, termed BLS-Location, based on a broad learning system (BLS) that utilizes channel state information (CSI) to overcome the aforementioned problems. It includes an offline training phase and an online localization phase. In the offline training phase, the Kalman filter and the expectation-maximization (EM) algorithm are utilized for completing and denoising the data. Moreover, principal component analysis (PCA) is used to reconstruct the CSI data to reduce complexity and train the weights by BLS. In the online localization phase, we employ a novel probabilistic method based on the regression results of BLS to obtain the estimated location. The experimental results show that BLS-Location can significantly reduce the training time with a high accuracy, compared to several machine learning algorithms and four existing methods in two representative indoor environments.

Functional nanomaterials possess exceptional multi-physical (e.g., mechanical, electrical and optical) properties compared with their bulk counterparts. To facilitate both synthesis and device applications of these nanomaterials, it is highly desired to characterize their multi-physical properties with high accuracy and efficiency. The nanomanipulation techniques under scanning electron microscopy (SEM) has enabled the testing of mechanical and electrical properties of various nanomaterials. However, the seamless integration of mechanical, electrical, and optical testing techniques into an SEM for triple-field-coupled characterization of single nanostructures is still unexplored. In this work, we report the first SEM-based nanomanipulation system for high-resolution mechano-optoelectronic testing of single semiconductor InGaN/GaN nanowires (NWs). A custom-made optical measurement setup was integrated onto a four-probe nanomanipulator inside an SEM, with two optical microfibers actuated by the nanoma-nipulator for NW excitation and emission measurement. A conductive tungsten nanoprobe and a conductive atomic force microscopy (AFM) cantilever probe were integrated onto the nanomanipulator for electrical nanoprobing of single NWs for electroluminescence (EL) measurement. The AFM probe also served as a force sensor for quantifying the contact force applied to the NW during nanoprobing. Using this unique system, we examined, for the first time, the effect of mechanical compression applied to an InGaN/GaN NW on its optoelectronic properties.

In Named Data Networking, cache pollution, cache poisoning and interest flooding are three popular types of attacks that can drastically degrade the network performance. However, previous methods for mitigating these attacks are not sufficiently effective or efficient. Also, they cannot simultaneously handle the three attacks, or the case that core routers or edge routers are compromised. To handle these problems, we propose an effective and lightweight countermeasure scheme. It consists of token-based router monitoring policy (TRM), hierarchical consensus-based trust management (HCT), and popularity-based probabilistic caching and caching replacement policy (PPC). In TRM, each edge router monitors and evaluates each data requester's probability of launching the cache pollution attack and each data provider's probability of launching the cache poisoning attack, and accordingly assigns, rewards and penalizes tokens to them to control their data request and data provision activities. Thus, the interest flooding attack can also be mitigated by limiting the consumption of tokens. In HCT, each core router manages its directly connected edge routers using TRM, and the core routers trust each other through adopting the concept of consensus in Blockchain. Thus, the edge and core routers executing monitoring and evaluation are trustable. PPC uses probabilistic caching and caching replacement based on the popularity of received content to further mitigate the attacks and reduce caching and data verification overhead. Results from simulation experiments demonstrate that our proposed scheme has better performance, in terms of interest satisfaction ratio and average end-to-end delay than current mechanisms.

This paper presents the field programmable gate array (FPGA) implementation of a particle swarm optimization (PSO) based nonlinear model predictive control (NMPC) for applications with millisecond timescales and resource-constrained embedded systems. A custom hardware circuit architecture of NMPC controller, which mainly includes phase locking loop (PLL) clock, universal asynchronous receiver transmitter (UART) interface, trigonometric function, random number generator, objective function, PSO solver modules, is implemented on FPGA based on a hardware description language (HDL) specification. It employs parallelism, pipelining, and specialized numerical formats to enhance the online computation performance of NMPC. And the partial parallel expansion calculations of fitness (objective) function and particle position update are adopted for trading off the computational performance against resource usage. The hardware-in-the-loop experimental results through controlling a wheeled mobile robot model demonstrate that the design is capable of computing the required control action in real-time by using a midrange FPGA, while still offering good control performance.

Deep reinforcement learning (DRL)-based recommender systems have recently come into the limelight due to their ability to optimize long-term user engagement. A significant challenge in DRL-based recommender systems is the large action space required to represent a variety of items. The large action space weakens the sampling efficiency and thereby, affects the recommendation accuracy. In this article, we propose a DRL-based method called deep hierarchical category-based recommender system (DHCRS) to handle the large action space problem. In DHCRS, categories of items are used to reconstruct the original flat action space into a two-level category-item hierarchy. DHCRS uses two deep Q -networks (DQNs): 1) a high-level DQN for selecting a category and 2) a low-level DQN to choose an item in this category for the recommendation. Hence, the action space of each DQN is significantly reduced. Furthermore, the categorization of items helps capture the users' preferences more effectively. We also propose a bidirectional category selection (BCS) technique, which explicitly considers the category-item relationships. The experiments show that DHCRS can significantly outperform state-of-the-art methods in terms of hit rate and normalized discounted cumulative gain for long-term recommendations.

The advanced metering infrastructure (AMI) in modern networked smart homes brings various advantages. However, smart homes are vulnerable to many cyberattacks, and the most striking one is energy theft. Researchers have developed many countermeasures, fostered by advanced machine learning (ML) techniques. Nevertheless, recent advances are not robust enough in practice, partially due to the vulnerabilities of ML algorithms. In this paper, we present a covert electricity theft strategy through mimicking normal consumption patterns. Such attack is almost impossible to be detected by existing solutions as the manipulated data have little deviation against honest usage records. To address this threat, we initially identify and define two levels of consumption deviations: home-level and interpersonal-level, respectively. Then, we propose a feature extraction method and develop a novel detection model based on deep learning. Extensive experiments show that the presented attack could evade existing mainstream detectors and the proposed countermeasure outperforms existing leading methods.

Mobile social networks include a large number of social members who forward messages cooperatively. However, spammers post links to viruses and advertisements, or follow a large number of users, which produces many misleading messages in mobile social networks. In this paper, we propose an adaptive social spammer detection (ASSD) model. We build a spammer classifier by using a small number of labeled patterns and some unlabeled patterns. The prediction accuracy is high compared with some conventional supervised learning methods. Moreover, the time and energy required to label the identity of social members are reduced by applying ASSD. Because social spammers frequently change their behavior to deceive the spammer detection model, an incremental learning method is designed to update the spammer detection model adaptively, without retraining. We evaluate ASSD by comparing it with other supervised and semi-supervised machine learning methods using the Social Honeypot Dataset. Experimental results show that the proposed model outperforms the baseline methods in terms of recall and precision. Additionally, ASSD maintains a high detection accuracy by adaptively updating the model with newly generated social media data.

Distracted driving has been recognized as a major challenge to traffic safety improvement. This article presents a novel driving distraction detection method that is based on a new deep network. Unlike traditional methods, the proposed method uses both temporal information and spatial information of electroencephalography (EEG) signals as model inputs. Convolutional techniques and gated recurrent units were adopted to map the relationship between drivers' distraction status and EEG signals in the time domain. A driving simulation experiment was conducted to examine the effectiveness of the proposed method. Twenty-four healthy volunteers participated and three types of secondary tasks (i.e., cellphone operation task, clock task, and 2-back task) were used to induce distraction during driving. Drivers' EEG responses were measured using a 32-channel electrode cap, and the EEG signals were preprocessed to remove artifacts and then split into short EEG sequences. The proposed deep-network-based distraction detection method was trained and tested on the collected EEG data. To evaluate its effectiveness, it was also compared with the networks using temporal or spatial information alone. The results showed that our proposed distraction detection method achieved an overall binary (distraction versus nondistraction) classification accuracy of 0.92. In terms of task-specific distraction detection, its accuracy was 0.88. Further analysis on the individual difference in detection performance showed that drivers' EEG performance differed across individuals, which suggests that adaptive learning for each individual driver would be needed when developing in-vehicle distraction detection applications.

In software defined networking, the flat design of distributed control plane enables the management of multi-domain networks that are incapable of deploying a root controller. However, it is very difficult to avoid policy conflicts between independent local controllers due to the lack of centralized arbitration. Moreover, domains without trust may not be always cooperative and could even cheat to maximize their own interests. In this paper, we first consider the cooperative scenario and address the problem of traffic engineering in a flat distributed control plane. We propose a fully distributed algorithm, called DisTE, which can provide max-min fair bandwidth allocation for flows and maximize resource utilization. DisTE also preserves the local topology of each domain and achieves policy consistency by multiple rounds of synchronizations. We then consider the non-cooperative scenario, where selfish domains may discriminate bandwidth requests from other domains or overstate their owns to squeeze more bandwidths. DisTE is enhanced to defend against such behaviors. Experiments on a 717-switches 5-domain network demonstrate that DisTE could drive the link utilization ratio to more than 93%, and reduce up to 95% convergence time at cost of 3% relative error on fairness, compared to the centralized approach. The countermeasures against selfish behaviors are also demonstrated to be extremely effective.

The increased availability of quantitative historical datasets has provided new research opportunities for multiple disciplines in social science. In this article, we work closely with the constructors of a new dataset, CGED-Q (China Government Employee Database-Qing), that records the career trajectories of over 340,000 government officials in the Qing bureaucracy in China from 1760 to 1912. We use these data to study career mobility from a historical perspective and understand social mobility and inequality. However, existing statistical approaches are inadequate for analyzing career mobility in this historical dataset with its fine-grained attributes and long time span, since they are mostly hypothesis-driven and require substantial effort. We propose CareerLens, an interactive visual analytics system for assisting experts in exploring, understanding, and reasoning from historical career data. With CareerLens, experts examine mobility patterns in three levels-of-detail, namely, the macro-level providing a summary of overall mobility, the meso-level extracting latent group mobility patterns, and the micro-level revealing social relationships of individuals. We demonstrate the effectiveness and usability of CareerLens through two case studies and receive encouraging feedback from follow-up interviews with domain experts.

Enhanced recovery after surgery (ERAS) is a perioperative management protocol used to accelerate patient recovery. This study evaluated its benefits in patients with resectable esophageal cancer.This retrospective study compared patients before (January 2013 to December 2016) and after (June 2018 to December 2020) ERAS protocol implementation in our hospital. A propensity score-matched analysis was used to compare short-term surgical outcomes between ERAS and non-ERAS groups. After propensity score matching each group included 243 patients.There were significant differences in hospital length of stay after surgery (7.40 vs 11.17 days, P.001) and hospitalization cost (¥69380 vs ¥78075, P.001) between the ERAS and non-ERAS groups. The time to chest tube removal (4.91 vs 7.16 days, P.001) and first bowel movement (2.87 vs 3.97 days, P.001) was significantly shorter in the ERAS group. However there was no significant difference in total postoperative complication morbidity (20.2% vs 25.1%, P = .193). The complication of postoperative atelectasis or pneumonia was significantly lower in the ERAS group (P = .003), but there was no significant difference in occurrence of at least grade III complications between the 2 groups (12.3% vs 11.5%, P = .889).We demonstrated that ERAS could reduce hospital stay, numerical pain scores, and hospitalization costs without increasing postoperative complication and readmission. Furthermore subgroup analyses revealed that ERAS was safe for older people (70 years old).

Automatic crack detection is vital for efficient and economical road maintenance. With the explosive development of convolutional neural networks (CNNs), recent crack detection methods are mostly based on CNNs. In this article, we propose a deeply supervised convolutional neural network for crack detection via a novel multiscale convolutional feature fusion module. Within this multiscale feature fusion module, the high-level features are introduced directly into the low-level features at different convolutional stages. Besides, deep supervision provides integrated direct supervision for convolutional feature fusion, which is helpful to improve model convergency and final performance of crack detection. Multiscale convolutional features learned at different convolution stages are fused together to robustly represent cracks, whose geometric structures are complicated and hardly captured by single-scale features. To demonstrate its superiority and generalizability, we evaluate the proposed network on three public crack data sets, respectively. Sufficient experimental results demonstrate that our method outperforms other state-of-the-art crack detection, edge detection, and image segmentation methods in terms of F1-score and mean IU.

Aims Despite the recent prosperity of shrimp cultivation in China, very little is known about how different shrimp farming models influence the dynamics of Vibrio parahaemolyticus populations and the antibiotic resistance of this bacterium. Methods and Results To this end, we conducted continuous surveillance of V. parahaemolyticus on four farms over 3 years: two traditional shrimp farms with daily water exchange and two farms operated in the recirculating aquaculture systems (RAS). No antibiotics were used in these farms to exclude the potential impacts of antibiotics on the emergence of antibacterial resistance. Multilocus sequence typing was utilized to characterize the dynamics of V. parahaemolyticus populations. Whole-genome sequencing (WGS) was conducted to determine the representative sequence types (STs) at each farm. Results revealed that the population structure of V. parahaemolyticus remained stable over time in both RAS farms, with only nine and four STs observed at each. In contrast, annual replacement of V. parahaemolyticus populations was observed in traditional farms with 26 and 28 STs identified in rearing water. WGS of 50 isolates divided them into five clusters, of which ST917a isolates harboured a genomic island that disrupted the gene recA. Pair-wised genomic comparison of isolates from the same STs showed that they were genetically related but belonged to different clones associated with geographical distribution. Conclusions These results suggested that RAS presented a specific ecological niche by minimizing the water exchanges with the external environment. In contrast, traditional farming might pose a food safety issue by introducing new V. parahaemolyticus populations with antibiotic resistance genes. Significance and Impact of the Study Our results expose the potential food safety issue associated with conventional agriculture and should encourage the development of preventive strategies to reduce the emergence of resistant V. parahaemolyticus populations.

Transcatheter aortic valve implantation is a fast-growing procedure. Expanding to low-risk patients, it has surpassed surgical aortic valve implantation in frequency and has been associated with excellent outcomes. Stroke is a devastating complication after transcatheter aortic valve implantation. Silent brain infarcts identified by diffusion-weighted magnetic resonance imaging are present in most patients following transcatheter aortic valve implantation. Postoperative delirium and cognitive dysfunction are common neurologic complications. The stroke and silent brain infarcts are likely caused by particulate emboli released during the procedure. Intravascularly positioned cerebral embolic protection devices are designed to prevent debris from entering the aortic arch vessels to avoid stroke. Despite promising design, randomized clinical trials have not demonstrated a reduction in stroke in patients receiving cerebral embolic protection devices. Similarly, the association of cerebral embolic protection devices with silent brain infarcts, post-operative delirium and cognitive dysfunction is uncertain. Monitored anesthesia care or conscious sedation is as safe as general anesthesia and is associated with lower cost but different anesthetic techniques have not been shown to decrease stroke risk, postoperative delirium, or cognitive dysfunction. Anesthesiologists play important roles in providing perioperative care including management of neurologic events in patients undergoing transcatheter aortic valve implantation. Large randomized clinical trials are needed that focus on the correlation between perioperative interventions and neurologic outcomes.

Methicillin-resistant Staphylococcus aureus (MRSA), biofilms, and persisters are three major factors leading to recurrent and recalcitrant implant infections. Although antibiotics are still the primary treatment for chronic implant infections in clinical, only few drugs are effective in clearing persisters and formed biofilms. Here, felodipine, a dihydropyridine calcium channel blocker, was reported for the first time to have antibacterial effects against MRSA, biofilm, and persisters. Even after continuous exposure to sub-lethal concentrations of felodipine, bacteria are less likely to develop resistance. Besides, low doses of felodipine enhances the antibacterial activity of gentamicin by inhibiting the expression of protein associated with aminoglycoside resistance (aacA-aphD). Next, biofilm eradication test and persisters killing assay suggested felodipine has an excellent bactericidal effect against formed biofilms and persisters. Furthermore, the result of protein profiling, and quantitative metabonomics analysis indicated felodipine reduce MRSA virulence (agrABC), biofilm formation and TCA cycle. Then, molecular docking showed felodipine inhibit the growth of persisters by binding to the H pocket of ClpP protease, which could lead to substantial protein degradation. Furthermore, murine infection models suggested felodipine in combination with gentamicin alleviate bacterial burden and inflammatory response. In conclusion, low dose of felodipine might be a promising agent for biomaterial delivery to enhance aminoglycosides efficacy against implant infections caused by MRSA, biofilm, and persisters.

Protein aggregation causes alpha-synuclein (α-syn) to change from its original physiological role to a pathological state, which is a potential pathogenic mechanism in Parkinson’s disease (PD). Chi...

In the natural language processing family, learning representations is a pioneering study, especially in sequence-to-sequence tasks where outputs are generated, totally relying on the learning representations of source sequence. Generally, classic methods infer that each word occurring in the source sequence, having more or less influence on the target sequence, should all be considered when generating outputs. As the summarization task requires the output sequence to only retain the essence, classic full consideration of the source sequence may not work well on it, which calls for more suitable methods with the ability to discard the misleading noise words. Motivated by this, with both relevance retaining and redundancy removal in mind, we propose a summarization learning model by implementing an encoder with copious contextual information represented and a decoder with a selecting mechanism integrated. Specifically, we equip the encoder with an asynchronous bi directional parallel structure, in order to obtain abundant semantic representation. The decoder, different from the classic attention-based works, employs a self-aware context selecting mechanism to generate summary in a more productive way. We evaluate the proposed methods on three benchmark summarization corpora. The experimental results demonstrate the effectiveness and applicability of the proposed framework in relation to several well-practiced and state-of-the-art summarization methods.

Federated learning (FL) has been widely recognized as a promising approach by enabling individual end-devices to cooperatively train a global model without exposing their own data. One of the key challenges in FL is the non-independent and identically distributed (Non-IID) data across the clients, which decreases the efficiency of stochastic gradient descent (SGD) based training process. Moreover, clients with different data distributions may cause bias to the global model update, resulting in a degraded model accuracy. To tackle the Non-IID problem in FL, we aim to optimize the local training process and global aggregation simultaneously. For local training, we analyze the effect of hyperparameters (e.g., the batch size, the number of local updates) on the training performance of FL. Guided by the toy example and theoretical analysis, we are motivated to mitigate the negative impacts incurred by Non-IID data via selecting a subset of participants and adaptively adjust their batch size. A deep reinforcement learning based approach has been proposed to adaptively control the training of local models and the phase of global aggregation. Extensive experiments on different datasets show that our method can improve the model accuracy by up to 30\%, as compared to the state-of-the-art approaches.

Real-time inference of deep learning models on embedded and energy-efficient devices becomes increasingly desirable with the rapid growth of artificial intelligence on edge. Specifically, to achieve superb energy-efficiency and scalability, efficient parallelization of single-pass deep neural network (DNN) inference on chip multiprocessor (CMP) architectures is urgently required by many time-sensitive applications. However, as the number of processing cores scales up and the performance of cores has grown much fast, the on-chip inter-core data movement is prone to be a performance bottleneck for computation. To remedy this problem and further improve the performance of network inference, in this work, we introduce a communication-aware DNN parallelization technique called CAP, by exploiting the elasticity and noise-tolerance of deep learning algorithms on CMP. Moreover, in the hope that the conducted studies can provide new design values for real-time neural network inference on embedded chips, we also have evaluated the proposed approach on both multi-core Neural Network Accelerators (NNA) chips and general-purpose chip-multiprocessors. Our experimental results show that the proposed CAP can achieve 1.13x-1.65x speedups and 51%-90% on-chip communication energy reduction for different neural networks while maintaining the inference accuracy, compared to baseline approaches.

While existing prediction models built on popular deep architectures have shown promising results in facial depression recognition, they still lack sufficient discriminative power due to the issues of 1) limited amount of labeled depression data for deep representation learning and, 2) large variation in facial expression across different persons of the same depression score and the subtle difference in facial expression across different depression levels. In this paper, we formulate the facial depression recognition as a label distribution learning (LDL) problem, and propose a deep joint label distribution and metric learning (DJ-LDML) method to address these issues. In DJ-LDML, LDL exploits label relevance inherent in depression data to implicitly increase the amount of training data associated with each depression level without actually enlarging the dataset, while deep metric learning (DML) aims at learning a deep ordinal embedding with a specifically designed label-aware histogram loss, allowing semantics similarity between video sequences (described by ordinal labels) to be preserved for discriminative feature learning. The two learning modules in our DJ-LDML work collaboratively to enhance the representation ability and discriminative power of the deeply learned spatiotemporal feature, leading to improved depression prediction. We empirically evaluate our method on two benchmark datasets and the results demonstrate the effectiveness of our formulation.

At present, while the Internet of Things (IoT) devices bring convenience to people, security issues have led to an increasing number of threats to IoT security. Since IoT devices have a Web application system for device managers to operate, the system can view device information, control and configure device status, and its security is of great significance. Among the various authentication methods provided by IoT devices, the password information authentication mechanism is still a critical method for Web login. If the IoT device has a weak Web password, once a hacker discovers the device, it is straightforward to be attacked and implanted with malicious code to control the device and attack other devices in the network. In response to this problem, this paper designs a set of automatic detection frameworks for weak passwords for web application systems of IoT devices. Based on this framework, an automated weak password detection system was developed to detect weak Web passwords on IoT devices on the wide-area networks of Beijing, Shandong Province, and Zhejiang Province. A total of 12,179 devices with weak Web passwords were found, accounting for all discovered IoT devices of 7.58%, verifying the effectiveness of the proposed framework.

Inrush of limestone karst water through the mine floor occurs frequently in the Carboniferous–Permian coalfield in northern China. The unascertained measurement evaluation and set pair analysis methods are proposed to evaluate water inflow, by using eight indices. Firstly, ordered binary comparison quantization method and interval number fuzzy decision matrix of EA-TOPSIS ordering were used to obtain weights for these eight factors. Secondly, conflicting evidence theory is applied to couple of these weights calculated by ordered binary comparison quantization and EA-TOPSIS ordering, to ensure the effective evaluation of the relative importance with respect to each indicator for the dynamic model. Lastly, the unapprehend measurement evaluation and set pair analysis methods were utilized to evaluate the risk of limestone karst water influx from coal seam floor. Based on the type of index value, the state classification is carried out through quantitative analysis, and single index measure functions including linear and non-linear patterns, are constructed. The process is described as follows: first, the 8 index values collected from 30 water inrush points in Xinwen coalfield are substituted into the measurement function to construct the multi index comprehensive evaluation matrix. Second, the credibility identification criterion is introduced to determine the risk level of water incurred point X i Finally, the general situation of water inflow risk from coal seam floor during coal mining in Xinwen coalfield is illustrated by using set pair potential analysis method. The production practice in Xinwen coalfield has proved that the evaluation results are consistent with the actual situation.

Objective To analyze the rates of blindness with the demographics and clinical characteristics of patients with primary angle-closure disease (PACD) to provide a comprehensive epidemiologic reference in China. Methods A retrospective analysis was conducted in the Chinese Glaucoma Study Consortium database, which is a national multicenter glaucoma research alliance of 111 hospitals participating between December 21, 2015 and September 9, 2018. The diagnosis of PACD was made by qualified physicians through examination. Comparison of sex, age, family history, subtypes of PACD, and blindness were analyzed. Results A total of 5762 glaucoma patients were included, of which 4588 (79.6%) had PACD. Of PACD patients, 72.1% were female with the sex ratio (F/M) of 2.6, and the average age of patients was 63.8±9.3 years with the majority between 60 and 70 years. Additionally, 30% of these patients had low vision in one eye, 8.8% had low vision in both eyes, 1.7% had blindness in one eye, and 0.3% had blindness in both eyes. There were statistical differences with regards to age between male and female patients with PACD, with male patients being older on average. Primary angle-closure glaucoma was more commonly diagnosed in males (60%) compared to females (35.9%), whereas acute primary angle closure (APAC) was more commonly diagnosed in females (54.3%) compared to males (37.7%). The visual acuity in APAC patients was lower and the rate of low vision and blindness was higher than other subtypes. Conclusion PACD was the major type of glaucoma in Chinese hospitals. There were more female patients with PACD, mostly between 60 and 70 years old, with higher rates of APAC in women. APAC resulted in the worst visual outcomes of all PACD subtypes.

With the rapid development of Internet of Things (IoT), it becomes convenient to use mobile devices to remotely monitor the Physiological signals (e.g., Arrhythmia diseases) of patients with chronic diseases (e.g., cardiovascular diseases (CVDs)). High classification accuracy of inter-patient ECGs is extremely important for diagnosing Arrhythmia. The Supraventricular ectopic beat (S) is especially difficult to be classified. It is often misclassified as Normal (N) or Ventricular ectopic beat (V). Class imbalance is another common and important problem in electronic health (eHealth), as abnormal samples (i.e., samples of specific diseases) are usually far less than normal samples. To solve these problems, we propose a sustainable deep learning-based heartbeat classification system, called BeatClass. It contains three main components: two stacked bidirectional Long Short-Term Memory Networks (Bi-LSTMs), called Rist and Morst, and a Generative Adversarial Network (GAN), called MorphGAN. Rist first classifies the heartbeats into five common Arrhythmia classes. The heartbeats classified as S and V by Rist are further classified by Morst to improve the classification accuracy. MorphGAN is used to augment the morphological and contextual knowledge of heartbeats in infrequent classes. In the experiment, BeatClass is compared with several state-of-the-art works for inter-patient arrhythmia classification. The F1-scores of classifying N, S and V heartbeats are 0.6%, 16.0% and 1.8% higher than the best baseline method. The experiment result demonstrates that taking multiple classification models to improve classification results step-by-step may significantly improve the classification performance. We also evaluate the classification sustainability of BeatClass. Based on different physical signal datasets, a trained BeatClass can be updated to classify heartbeats with different sampling rates. At last, an engineering application indicates that BeatClass can promote the sustainable development of IoT-based eHealth.

The characteristics of patients with colorectal cancer who have benign mesenteric lymph node enlargement are not well documented.The aim of this study is to assess the clinical and prognostic significance of benign mesenteric lymph node enlargement in patients with colorectal cancer.This is a prospective cohort study.This study was conducted at multitertiary institutions.We included 601 patients with stage 0, I, and II colorectal cancer in Tianjin, Shandong, and Zhejiang from January 2010 to April 2014. Patients underwent curative surgery and were separated into 2 groups by the presence of benign mesenteric lymph node enlargement: the enlargement group (n = 275) and the control group (n = 326).Univariate log rank and multivariate Cox regression analyses were constructed to identify risk factors for recurrence and mortality.The risk of recurrence in the enlargement group after curative resection was significantly lower than in the control group, with the 1-, 3-, and 5-year disease-free survival rates being 97.1%, 91.6%, and 86.9% in the enlargement group and 95.7%, 86.2%, and 78.2% in the control group (p = 0.004). The postoperative 1-, 3-, and 5-year overall survival rates were 99.6%, 94.9%, and 90.5% in the enlargement group and 99.4%, 91.4%, and 82.1% in the control group (p = 0.001). Patients in the enlargement group had a higher percentage of patients at a younger age, family tumor history, right-sided tumors, and larger tumor size compared with the control group. For patients in the enlargement group, no significant correlation was observed between the number of enlarged lymph nodes and disease-free survival or overall survival (p = 0.113 and 0.386). Adjusted Cox regression model showed that benign mesenteric lymph node enlargement was an independent prognostic risk factor for both disease-free survival (HR, 0.587; 95% CI, 0.399-0.861; p = 0.007) and overall survival (HR, 0.506; 95% CI, 0.328-0.779; p = 0.002).No immunological results could be compared with clinicopathological findings.The study indicates that benign mesenteric lymph node enlargement can be a useful positive factor in predicting recurrence and long-term survival concerning patients with colorectal cancer. See Video Abstract at http://links.lww.com/DCR/B785.ANTECEDENTES:Las características de los pacientes portadores de cáncer colorrectal con agrandamiento benigno de los ganglios linfáticos mesentéricos no se encuentran bien documentados.OBJETIVO:El objetivo de este estudio es evaluar la importancia clínica y pronóstica del agrandamiento benigno de los ganglios linfáticos mesentéricos en pacientes con cáncer colorrectal.DISEÑO:Este es un estudio de cohorte de tipo prospectivo.AJUSTE:Este estudio se llevó a cabo en instituciones de educación superior.PACIENTES:Incluimos a 601 pacientes con cáncer colorrectal en estadio 0, I, II en Tianjin, Shandong y Zhejiang desde enero de 2010 hasta abril de 2014. Los pacientes fueron sometidos a cirugía curativa y fueron separaron en dos grupos tomando en cuenta la presencia del agrandamiento benigno de los ganglios linfáticos mesentéricos: grupo con agrandamiento (n = 275) y grupo control (n = 326).PRINCIPALES MEDIDAS DE RESULTADO:Se construyeron análisis de rango logarítmico de una variante y de regresión de Cox con variante múltiple para identificar los factores de riesgo de recurrencia y mortalidad.RESULTADOS:El riesgo de recurrencia en el grupo con agrandamiento tras la resección curativa fue significativamente menor que en el grupo de control, con tasas de periodo libre de enfermedad a los 1, 3 y 5 años de 97,1, 91,6, y 86,9% en el grupo de agrandamiento y con tasas de 95,7, 86,2, y 78,2% en el grupo control respectivamente (p = 0,004). Las tasas postoperatorias de supervivencia general a los 1, 3 y 5 años fueron 99,6, 94,9, y 90,5% en el grupo de agrandamiento y de 99,4, 91,4, y 82,1% en el grupo de control, respectivamente (p = 0,001). Los pacientes del grupo con agrandamiento tenían un porcentaje más elevado de menor edad, antecedente familiar tumoral, tumores del lado derecho y de mayor tamaño tumoral con respecto al grupo de control. Para los pacientes con agrandamiento, no se observó una correlación significativa entre el número de ganglios linfáticos agrandados y el periodo libre de enfermedad o la supervivencia general (p = 0,113 y 0,386). El modelo de regresión de Cox ajustado mostró que el agrandamiento benigno de los ganglios linfáticos mesentéricos era un factor de riesgo pronóstico independiente tanto para la supervivencia libre de enfermedad (cociente de riesgo 0,587; IC del 95%: 0,399-0,861; p = 0,007) como para la supervivencia global (cociente de riesgo 0,506; IC del 95%: 0,328- 0,779; p = 0,002).LIMITACIONES:No fue posible comparar los resultados inmunológicos con los hallazgos clínico-patológicos.CONCLUSIONES:El estudio indica que el agrandamiento benigno de los ganglios linfáticos mesentéricos puede ser un factor positivo útil para predecir la recurrencia y la supervivencia a largo plazo en pacientes con cáncer colorrectal. Consulte Video Resumen en http://links.lww.com/DCR/B785. (Traducción-Dr. Osvaldo Gauto).

Spiking Neural Networks (SNNs) use spatio-temporal spike patterns to represent and transmit information, which is not only biologically realistic but also suitable for ultra-low-power event-driven neuromorphic implementation. Motivated by the success of deep learning, the study of Deep Spiking Neural Networks (DeepSNNs) provides promising directions for artificial intelligence applications. However, training of DeepSNNs is not straightforward because the well-studied error back-propagation (BP) algorithm is not directly applicable. In this paper, we first establish an understanding as to why error back-propagation does not work well in DeepSNNs. To address this problem, we propose a simple yet efficient Rectified Linear Postsynaptic Potential function (ReL-PSP) for spiking neurons and propose a Spike-Timing-Dependent Back-Propagation (STDBP) learning algorithm for DeepSNNs. In STDBP algorithm, the timing of individual spikes is used to convey information (temporal coding), and learning (back-propagation) is performed based on spike timing in an event-driven manner. Our experimental results show that the proposed learning algorithm achieves state-of-the-art classification accuracy in single spike time based learning algorithms of DeepSNNs. Furthermore, by utilizing the trained model parameters obtained from the proposed STDBP learning algorithm, we demonstrate the ultra-low-power inference operations on a recently proposed neuromorphic inference accelerator. Experimental results show that the neuromorphic hardware consumes 0.751~mW of the total power consumption and achieves a low latency of 47.71~ms to classify an image from the MNIST dataset. Overall, this work investigates the contribution of spike timing dynamics to information encoding, synaptic plasticity and decision making, providing a new perspective to design of future DeepSNNs and neuromorphic hardware systems., This work has been submitted to the IEEE for possible publication. Copyrightmay be transferred without notice, after which this version may no longer beaccessible

As a distributed computing paradigm, edge computing has become a key technology for providing timely services to mobile devices by connecting Internet of Things (IoT), cloud centers and other facilities. By offloading compute-intensive tasks from IoT devices to edge/cloud servers, the communication and computation pressure caused by the massive data in industrial IoT can be effectively reduced. In the process of computation offloading in edge computing, it is critical to dynamically make optimal offloading decisions to minimize the delay and energy consumption spent on the devices. Although there are a large number of task offloading-decision models, how to measure and evaluate the quality of different models and configurations is crucial. In this paper, we propose a novel simulation platform named ChainFL, which can build an edge computing environment among IoT devices while being compatible with Federated Learning and Blockchain technologies to better support the embedding of security-focused offloading algorithms. ChainFL is lightweight and compatible, and it can quickly build complex network environments by connecting devices of different architectures. Moreover, due to its distributed nature, ChainFL can also be deployed as a federated learning platform across multiple devices to enable federated learning with high security due to its embedded blockchain. Finally, we validate the versatility and effectiveness of ChainFL 2 by embedding a complex offloading decision model in the platform, as well as deploying it in an industrial IoT environment with security risks.

Background: Calibrated CT fat fraction (FFCT) measurements derived from non-enhanced abdominal CT reliably reflect liver fat content, allowing largescale population-level investigations of steatosis prevalence and associations. Objective: To compare prevalence of hepatic steatosis, assessed by calibrated CT measurements, between population-based Chinese and U.S. cohorts, and to investigate in these populations the relationship of steatosis with age, sex, and body mass index (BMI). Methods: This retrospective study included 3176 adults (1985 women, 1191 men) from seven Chinese provinces and 8748 adults (4834 women, 3914 men) from a single U.S. medical center, drawn from earlier studies. All participants were at least 40 years old and underwent unenhanced abdominal CT for the earlier studies. Liver fat content measurements on CT were cross-calibrated to MRI proton density fat fraction measurements using phantoms and expressed as adjusted FFCT. Mild, moderate, and severe steatosis were defined as adjusted FFCT of 5.0%-14.9%, 15.0%-24.9%, and ≥25.0%, respectively. The two cohorts were compared. Results: Median adjusted FFCT was for women 4.7% and 4.8%, and for men 5.8% and 6.2%, in the Chinese and U.S. cohorts, respectively. Steatosis prevalence was for women 46.3% and 48.7%, and for men 58.9% and 61.9%, in the Chinese and U.S. cohorts, respectively. Severe steatosis prevalence was for women 0.9% and 1.8%, and for men, 0.2% and 2.6%, in the Chinese and U.S. cohorts, respectively. Adjusted FFCT did not vary across age decades in women or men in the Chinese cohort, though increased across age decades in women and men in the U.S. cohort. Adjusted FFCT and BMI exhibited weak correlation (r=0.312-0.431). Among participants with normal BMI, 36.8% and 38.5% of those in the Chinese and U.S. cohorts had mild steatosis, and 3.0% and 1.5% had moderate or severe steatosis, respectively. Among U.S. participants with BMI ≥40.0, 17.7% had normal liver content. Conclusion: Steatosis and severe steatosis had higher prevalence in the U.S. than Chinese cohort in both women and men. BMI did not reliably predict steatosis. Clinical Impact: The findings provide new information on the dependence of hepatic steatosis on age, sex, and BMI.

There is a strong interest in using physical waves for artificial neural computing because of their unique advantages in fast speed and intrinsic parallelism. Resonance, as a ubiquitous feature across many wave systems, is a natural candidate for analog computing in temporal signals. We demonstrate that resonance can be used to construct stable and scalable recurrent neural networks. By including resonators with different lifetimes, the computing system develops both short-term and long-term memory simultaneously.

Objective: This study aims to use loperamide by gavage in SD rats to establish the model of STC, and evaluate the degree of similarity between the model and clinical diseases. Furthermore, the effect of electro-acupuncture when stimulating BL32 acupoints in SD rats with slow transit constipation were observed to determine changes in intestinal transmission function. Methods: Thirty SD rats were selected and divided randomly into three groups. Rats in the model group were given 2 mg/kg/d of loperamide tablets by gavage for 15 days. The morphology of the feces of rats was observed, and the dry weight of the feces and the time of first black feces movement with black ink by gavage after molding were measured. After 20 days of electro-acupuncture treatment at the BL32 acupoint, the time of first black feces movement, the ratio of black ink length of the entire length of the intestine, and contraction amplitude and frequency of contraction were measured. Results: The time of black feces movement in the two model groups was significantly prolonged (P P th and 20th day of treatment, the time of black feces movement in the treatment group significantly decreased (P P P Conclusion: Through loperamide gavage, the model of slow transit constipation in SD rats was successfully established in a relatively short period of time. Electro-acupuncture treatment had a rapid onset of action, and its mechanism might have been caused through strengthening the ability of intestinal smooth muscle contraction, and are independent with increased smooth muscle contraction frequency.

Treatment delays in prostate cancer have been characterised, although not explicitly in men undergoing transperineal prostate biopsies. We aimed to determine if delays to radical prostatectomy correlate with adverse outcomes using a contemporary population-based cohort of men diagnosed by transperineal biopsies.This study analysed men with prostate cancer of the International Society for Urological Pathology grade group ≥2, diagnosed by transperineal prostate biopsies who underwent prostatectomy, using the prospectively data from 1 January 2014 to 30 June 2018 Prostate Cancer Outcomes Registry-Victoria. Data were analysed according to stratified demographic and disease characteristics. Time intervals from biopsy (28, 60, 90, 120, and 270 days) were compared using odds ratios and regression analyses for proportion of upgrading, early biochemical recurrence, pT3 disease at prostatectomy, and positive surgical margins.In total, 2008 men were analysed. There were 306 (16.7%) men with upgrading, 151 (8.4%) with biochemical recurrence, 1068 (54.1%) with pT3 disease, and 464 (23.1%) with positive surgical margins (percentages excluded patients with missing data). All adverse outcomes studied were significantly associated with higher prostate-specific antigen and grade at diagnosis. Delays of 120-270 days did not adversely alter the incidence of Gleason upgrading, pT3, or recurrence. Delays (most frequent 60-89 days, 28%) were associated with positive surgical margins but not monotonically. Regression modelling demonstrated no increased likelihood of most adverse outcomes for up to 270 days.Men with prostate cancer of grade group ≥2 diagnosed through transperineal biopsy may wait up to 270 days for a prostatectomy without a greater likelihood of upgrading, pT3 disease, positive surgical margins, or biochemical recurrence.

Rationale: The objective was to assess the cost-effectiveness of staging positron emission tomography/computed tomography (PET/CT) in early-stage follicular lymphoma from the Canadian health care system perspective. Methods: The study population was FL patients staged as early-stage using conventional CT imaging and planned for curative-intent radiation therapy (RT). A decision analytic model simulated the management after adding a staging PET/CT vs. using staging CT alone. In the no-PET/CT strategy, all patients proceeded to curative-intent RT as planned. In the PET/CT strategy, PET/CT information could result in an increased RT volume, switching to a non-curative approach, or no change in RT treatment as planned. Subsequent disease course was described using a state-transition cohort model over a 30-year time horizon. Diagnostic characteristics, probabilities, utilities and costs were derived from the literature. Baseline analysis was performed using quality-adjusted life years (QALYs), costs (2019 Canadian dollars, CAD$) and the incremental cost-effectiveness ratio. Deterministic sensitivity analyses were conducted, evaluating net monetary benefit at a willingness-to-pay threshold of $100,000/QALY. Probabilistic sensitivity analysis using 10,000 simulations was performed. Costs and QALYs were discounted at a rate of 1.5%. Results: In the reference case scenario, staging PET/CT was the dominant strategy, resulting in an average lifetime cost saving of $3,165 and a gain of 0.32 QALYs. In deterministic sensitivity analyses, the PET/CT strategy remained the preferred strategy for all scenarios supported by available data. In probabilistic sensitivity analysis, the PET/CT strategy was strongly dominant in 77% of simulations (i.e., reduced cost and increased QALYs), and was cost-effective in 89% of simulations (i.e., either cost-saving or with an incremental cost-effectiveness ratio below $100,000/QALY). Conclusion: Our analysis shows that the use of PET/CT to stage early-stage FL patients reduces cost and improves QALYs. Patients with early-stage FL should undergo PET/CT prior to curative-intent RT.

An increasing number of patients are being diagnosed with multiple ground glass opacities (GGOs), but a consensus on the treatment of these patients is still lacking. The aim of this study was to investigate the safety and feasibility of a novel technique, electromagnetic navigation bronchoscopy (ENB)-guided microwave ablation combined with uniportal video-assisted thoracoscopic surgery in patients with multiple GGOs.The clinical, radiographic, surgical, and pathologic data of patients with multiple GGOs who underwent ENB-guided microwave ablation combined with uniportal video-assisted thoracoscopic surgery from October 2018 to December 2019 were reviewed.Eleven patients with multiple GGOs underwent ENB-guided microwave ablation combined with uniportal video-assisted thoracoscopic surgery; they included 6 men and 5 women with a mean age of 61.3 ± 5.1 years (range, 53 to 68). Thirty-seven lesions were observed in the 11 patients, 21 of which were microwave ablated and 16 of which were surgically resected. Only 1 patient had postoperative pneumothorax and subcutaneous emphysema and was successfully discharged from the hospital after symptomatic treatment. The success rate and the efficiency of microwave ablation under ENB guidance were both 100%, with no other serious complications or procedure-related deaths occurring. No local metastasis or recurrence occurred in any patients during the follow-up period.Electromagnetic navigation bronchoscopy-guided microwave ablation combined with uniportal video-assisted thoracoscopic surgery is safe and feasible in patients with multiple GGOs suspected of having multiple primary lung cancers, and may represent an alternative approach for more patients, particularly patients who cannot tolerate the simultaneous resection of multiple tumors.

Purpose To quantify the damage to the corneal endothelium from the flow of balanced salt solution during phacoemulsification. Setting John A. Moran Eye Center, University of Utah, Salt Lake City, Utah, USA. Design Experimental Study. Methods Twelve New Zealand white rabbits received bilateral surgery. Six eyes served as controls with no irrigation (incision only). In 18 eyes, the Intrepid Balanced tip of the Centurion (Alcon) Ozil handpiece was inserted into the anterior chamber for continuous irrigation at 50mL/minute with either 250 mL or 500 mL of balanced salt solution (BSS). After euthanasia and enucleation, the corneas were removed, stained with trypan blue/alizarin red, and photographed in a standardized manner (X400 photographs from 5 specific areas, and 1 overview photograph from each corneal button). The ImageJ program was used to evaluate cell damage and loss in the photographs obtained from each cornea. Results Analysis of the X400 photographs showed no statistical differences between control, 250 mL, and 500 mL groups in the percentage of intact, damaged, or lost cells (P= 0.896, 0.851, and 0.972 respectively). For the overview photos, the differences in areas of intact and damaged cells among the groups were statistically significant, likely due to the peripheral areas of damage related to touches between the phaco tip and the endothelium during irrigation. Conclusion The volume of BSS flow alone does not appear to be a major contributing source of endothelial cell damage and loss, providing further insight on mechanisms of corneal endothelium damage during phacoemulsification.

The robot-held laparoscope efficiently maintains the stability of the surgical screen in comparison with the laparoscope held by the assistant. The collaboration between the robot-held laparoscope with the surgeon is achieved without the artificial manipulation by tracking surgical instruments automatically. In this work, a visual tracking space vector method is proposed to autonomously control the field-of-view (FOV) by tracking multiple surgical instruments. In the proposed method, a visual tracking space vector is introduced to track the motion of multiple instruments, and a constraint vector is defined to ensure that the laparoscope is always within the safe working space. In addition, a modified YOLOv3 algorithm is proposed to robustly detect and classify the surgical instruments. The proposed method automatically adapts to the changes in the number of surgical instruments, as compared to other visual servoing methods for tracking surgical instruments. The proposed method does not require the depth information of the tracked surgical instruments. Experimental and simulation results verify that the proposed method accurately and efficiently controls the robot-held laparoscope to track the motion of multiple surgical instruments and ensure safe tracking movement.

2D LiDAR is an efficient alternative sensor for vehicle detection, which is one of the most critical tasks in autonomous driving. Compared to the fully-developed 3D LiDAR vehicle detection, 2D LiDAR vehicle detection has much room to improve. Most of existing state-of-the-art works represent 2D point clouds as pseudo-images and then perform detection with traditional object detectors on 2D images. However, they ignore the sparse representation and geometric information of vehicle in the 2D cloud points. In this paper, we present Keypoint-Aware and Keypoint-Matching Network termed as KAM-Net, specifically focus on better detecting the vehicles by explicitly capturing and extracting the sparse information of L-shape in 2D LiDAR point clouds. The whole framework consists of two stages, namely keypoint-aware stage and keypoint-matching stage. The keypoint-aware stage utilizes the heatmap and edge extraction module to simultaneously predict the position of L-shaped keypoints and inflection offset of L-shaped endpoints. The keypoint-matching stage is followed to group the keypoints and produces the oriented bounding boxes with axis by utilizing the endpoint-matching and Lshaped-matching methods. Further, we conduct extensive experiments on a recently released public dataset to evaluate the effectiveness of our approach. The results show that our KAM-Net achieves a new state-of-the-art performance.

Diverse technologies, such as machine learning and big data, have been driving the prosperity of the Internet of Things (IoT) and the ubiquitous proliferation of IoT devices. Consequently, it is natural that IoT becomes the driving force to meet the increasing demand for frictionless transactions. To secure transactions in IoT, blockchain is widely deployed since it can remove the necessity of a trusted central authority. However, the mainstream blockchain-based IoT payment platforms, dominated by Proof-of-Work (PoW) and Proof-of-Stake (PoS) consensus algorithms, face several major security and scalability challenges that result in system failures and financial loss. Among the three leading attacks in this scenario, double-spend attacks and long-range attacks threaten the tokens of blockchain users, while eclipse attacks target denial of service. To defeat these attacks, a novel bidirectional-linked blockchain (BLB) using chameleon hash functions is proposed, where bidirectional pointers are constructed between blocks. Furthermore, a new Committee Members Auction (CMA) consensus algorithm is designed to improve the security and attack resistance of BLB while guaranteeing high scalability. In CMA, distributed blockchain nodes elect committee members through a verifiable random function. The smart contract uses Shamir’s Secret Sharing scheme to distribute the trapdoor keys to committee members. To better investigate BLB’s resistance against double-spend attacks, an improved Nakamoto’s attack analysis is presented. In addition, a modified entropy metric is devised to measure eclipse attack resistance across different consensus algorithms. Extensive evaluation results show the superior resistance against attacks and demonstrate high scalability of BLB compared with current leading paradigms based on PoS and PoW.

A dual-polarized ferrite-loaded ultra-wideband tightly coupled dipole array (TCDA) with an extremely low profile is presented. A novel ferrite grid with high relative permeability and permittivity characteristic is loaded between the antenna and ground plane, which expands the low-end of operation bandwidth and reduces the overall profile, simultaneously. An X-shaped frequency selective surface (FSS) is utilized as a special band-stop filter to improve the realized gain at higher frequencies. Meanwhile, the proposed array utilizes a microstrip (MS) to coplanar stripline (CPS) uniplanar double-Y balun to achieve unbalance-balance transformation and impedance matching. A thin meta-surface based wide-angle impedance matching (MS-WAIM) layer is introduced to improve the beam scanning ability. Simulated results show that the proposed antenna is able to operate over a 10:1 bandwidth (0.2-2GHz) with active VSWR < {2, 2.6, 3.4} while scanning up to {broadside, 45°, 60°}, respectively. Most of all, the overall antenna profile is only 62.5mm (0.042λL) at the lowest operating frequency, which is known to be the thinnest 10:1 ultra-wideband array with wide angle scanning (± 60° in E- and H-plane) performance. A prototype of 6 × 8 array is simulated, fabricated and measured. Good agreement is achieved between the simulated and measured results.

The key to hyperspectral anomaly detection is to effectively distinguish anomalies from the background, especially in the case that background is complex and anomalies are weak. Hyperspectral imagery (HSI) as an image-spectrum merging cube data can be intrinsically represented as a third-order tensor that integrates spectral information and spatial information. In this article, a prior-based tensor approximation (PTA) is proposed for hyperspectral anomaly detection, in which HSI is decomposed into a background tensor and an anomaly tensor. In the background tensor, a low-rank prior is incorporated into spectral dimension by truncated nuclear norm regularization, and a piecewise-smooth prior on spatial dimension can be embedded by a linear total variation-norm regularization. For anomaly tensor, it is unfolded along spectral dimension coupled with spatial group sparse prior that can be represented by the l2,1-norm regularization. In the designed method, all the priors are integrated into a unified convex framework, and the anomalies can be finally determined by the anomaly tensor. Experimental results validated on several real hyperspectral data sets demonstrate that the proposed algorithm outperforms some state-of-the-art anomaly detection methods.

MOV is often used to limit the overvoltage generated by the current breaking of DC breaker to protect power electronic device from breakdown. However, while suppressing overvoltage, the MOV also causes the degradation of continuous operating voltage capability. This paper proposes a novel solid-state switch scheme with high voltage utilization efficiency by using modular gapped-MOV (MG-MOV) for DC breakers. The breakable gap is first applied to the protection of power electronic devices. By using the gap to share the voltage in the static state and be broken down in the dynamic state, the overvoltage and continuous operating voltage of the solid-state switch can be decoupled. Moreover, based on the modular scheme, the instability of the gap breakdown and the voltage imbalance of the series devices can be solved well. The working mechanism, voltage-balance and stability issue of the MG-MOV are discussed in detail in this paper, and a 15kA prototype is developed to verify the feasibility of the novel solid-state switch scheme. The experiment results show that voltage utilization efficiency can be increased from 44.4% to 72.0% by using MG-MOV.