Mode Of Attack Research Articles

FGDNet: Fine-Grained Detection Network Towards Face Anti-Spoofing

With the development of facial recognition technology, face anti-spoofing as the most important security module of face recognition system becomes more and more important. As a matter of fact, face anti-spoofing is still a challenging task, especially facing various attacks simultaneously. Moreover, most of current detectors mainly focus on binary classification while usually fail to complete the task of fine-grained multiple classification, referring to as replay, print, partial mask, and full mask attacks. To fill the gap, in this context, it is proposed to design the fine-grained detection network for classifying various face spoofing attack modes. First, we propose to establish a Transformer style network structure for feature extraction, where the convolution mapping operation is adopted instead of traditional linear mapping. Specifically, we adopt the self-attention module for extracting long distance feature, and convolution mapping is used to maintain the model's ability to extract local features. Finally, the simple yet effective linear classifier is introduced for fine-grained classification. Moreover, with the help of the VGG based style-transfer network, the well-designed scheme of data augmentation module is proposed for solving the problem of insufficient training samples. In the large-scale experiments, compared with the baseline detectors, our proposed fine-grained classifier with low computation cost performs its superiority for multiple classification.

Semi-quantum bi-signature scheme based on Bell states

Currently, quantum signature protocols assume that all participants in the protocol (signers, verifiers, and arbitrators) should exhibit quantum capabilities. If quantum devices are lacking, the quantum signature protocol cannot be implemented. However, quantum devices are not widely used, and not all users can own quantum devices. Boyer et al. first proposed the concept of the semi-quantum environment. Zhao et al. proposed a semi-quantum signature protocol with double signers based on the W state. Yang et al. discovered that in the semi-quantum signature protocol with double signers developed by Zhao et al., an internal attacker could launch an impersonation attack during the verification and signature stages to capture the private key and use it to impersonate the user’s identity. In addition, the eavesdropper can obtain all the secret information of the signers by performing a man-in-the-middle attack. All three attack modes can pass the eavesdropping check. Hence, if these security problems are not solved, the signature of the semi-quantum signature protocol with double signers proposed by Zhao et al. cannot be completed safely. Therefore, based on Bell states, a new semi-quantum signature protocol of double signers was developed in this study. The proposed protocol not only resists impersonation attacks and man-in-the-middle attacks but also resists other common attack modes.

Invulnerability of the Urban Agglomeration Integrated Passenger Transport Network under Emergency Events.

Urgent natural environmental events, such as floods, power failures, and epidemics, result in disruptions to the traffic system and heavy disturbances in public requirements. In order to strengthen the ability of the transport network to handle urgent natural environmental issues, this paper simulates the disruption situation of traffic stations in the urban agglomeration by attacking nodes, and evaluates the ability of the transport network to resist disruptions (i.e., invulnerability). Firstly, the model of the urban agglomeration integrated passenger transport network is established based on complex network theory. The highway network, railway network, and coupling network are combined into a multi-layer network space structure, and the edge weight is calibrated by travel time and cost. Secondly, the invulnerability simulation process including multiple attack modes under random and deliberate attack strategies is sorted out. By improving the traditional network efficiency indicator, the network impedance efficiency indicator is proposed to measure network performance, and the network relative impedance efficiency indicator is used to evaluate network invulnerability and identify key nodes. Finally, Chengdu-Chongqing urban agglomeration is taken as a case study. The results show that the network does not collapse quickly and it shows certain invulnerability and robustness under continuous random attacks. Network performance and invulnerability are not necessarily positively correlated. The failure of individual nodes that are small in scale but act as transit hubs may significantly degrade the network performance. The identified key nodes have significance for guiding the construction, maintenance, and optimization of the urban agglomeration passenger transport network, which is conducive to promoting public safety.

“Cocktail” anti-tumor strategy through enhanced antigen exposure with photothermal/chemodynamic therapy

“Cocktail” anti-tumor strategy effectively prevent tumor escape through a variety of attack modes. To reduce the difficulty of tumor marker exposure caused by adhesion factor coverage and improve the efficiency of tumor killing, in this work, prussian@20(R)-ginsenoside@defective tungsten oxide@hyaluronic acid (PRDH) nanosystem, an in-situ “cock-tail” anti-tumor nano-generator was prepared. PRDH nanosystem attacked tumor cells not only through photothermal therapy (PTT) and chemodynamic therapy (CDT), but more importantly through removing protection of sticky factors for tumor cells and triggering precise tumor recognition by immune cells. Responsing to glutathione (GSH) in tumor microenvironment (TME) and 808 laser irradiation, PRDH nanosystem released 20(R)-ginsenoside (Rg3) and in-situ catalyzed CO2 into CO. Rg3 and CO synergistically inhibited the expression of intercellular adhesion molecule 1 (ICAM-1) which acted as a “protective umbrella” for the immune escape of the tumor cells, thereby promoting the exposure of the MUC1 antigen. Subsequently, as an “I'm here” signal, the MUC1 antigen induced the tumor cells to be targeted and eliminated by immune cells in cooperation with lipid peroxide (LPO) promoted by Fenton reaction of PRDH NPs. Therefore, anti-tumor mode of PRDH nanosystem through enhanced antigen exposure with PTT/CDT, provided a novel “cocktail” strategy for the development of tumor therapy.

Neither Theory nor Science Metaphilosophical Remarks on Philosophical Elucidations

This article examines Wittgenstein’s philosophical reflection on philosophy: its method, its scope, and its relationship with other knowledge as central elements of the philosophical proposal the author developed in Tractatus Logico-Philosophicus. Therefore, my proposal is to rehearse a metaphilosophical reading of Wittgenstein’s remarks about philosophy in TLP focusing, on the one hand, on his reflections on philosophy as an activity and not as a theory; on the other hand, on his categorical differentiation established between philosophical elucidations and scientific explanations.
 This proposal differs from the readings that interpret it is through the construction of a theory of meaning -or logical doctrine- that one can distinguish philosophical nonsense and see the world correctly. It also differs from Tractatus’s non-theoretical or quietist readings which understand philosophical elucidations as exclusively negative or critical nonsense. My aim is to highlight a positive aspect of philosophical elucidations that I will call performative. From my metaphilosophical reading, not only does the activity of clarification work as a critical sieve that separates what makes sense from what does not –with no theory; but it also makes a modification -a transformation- in the one who establishes it. These performative aspects imply emphasizing that there is a change in the way we see the world but also in the way we ‘see’ language, meaning, logic, science, philosophy, life, etc. Thus, the refusal to elaborate a theory or to offer scientific explanations does not turn the philosophical elucidations in TLP into a mode of self-destructive attack on all kinds of philosophy. On the contrary, these elucidations are part of the defense of a particular way of practicing it.

Detection and Prevention of DDoS Attacks on the IoT

The Internet of Things (IoT) system has been a hot topic in recent years. Its operation is a system that stores data in data storage and is completed by the exchange of network information about things. Therefore, the security of information between network transmissions is very important. In recent years, the most likely cause of information security problems has been a distributed denial of service (DDoS) attack. In this paper, we proposed an autonomous defense system that combines edge computing with a two-dimensional convolutional neural network (CNN) to recognize whether the data server in IoT suffers from DDoS attacks and identify the attack mode. The accuracy of trained two-dimensional CNN is up to 99.5% and 99.8% for packet traffic and packet features training, respectively. A field experiment’s results show that the data server in the proposed system can effectively distinguish the difference between the DDoS attacks and the normal transmission to reduce the impact of DDoS attacks on the IoT data storage while it is under attack.

Foraging Patterns of Non-Territorial Eastern Imperial Eagle (Aquila heliaca): A Case of Successful Adaptation

The Optimal Foraging Theory predicts that, to maximize fitness, animals adapt their foraging strategy that provides the most benefit for the lowest cost, maximizing the net energy gained. While the diet of many breeding raptor populations is well known, studies on the foraging patterns of non-territorial birds of prey (floaters) are scarce. In this study, we examined the foraging pattern of non-territorial Eastern Imperial Eagle, scrutinizing different aspects of its feeding ecology and behavior. We built a simple model of the optimal foraging strategy of floater eagles including the success of foraging as a currency as well as environmental factors such as seasons, type of prey, habitat, foraging techniques, and eagle age as a limitation affecting the foraging efficiency of birds. We found that floaters focused their diet exclusively on European Souslik, accounting for almost half (44.2%) of the eagle’s prey. Diet differences between floaters and breeders were due to higher Souslik and carrion consumption and lower Hedgehog predation by floater eagles. The diet diversity of breeding eagles (H = 3.297) was much higher than that of floaters (H = 1.748). Our model suggested that the foraging mode, habitat type, and season best explained the feeding success of non-territorial eagles (ΔAIC = 0.00, w = 0.42). Of all explanatory factors, “Kleptoparasitism” (β2 = −4.35), “Rodents” (β2 = −4.52), “Pasture” (β2 = 2.96), “Wheat” (β2 = 4.41), “In the air” (β2 = 4.16), and “Other habitats” (β2 = 4.17) had a pronounced effect. The factors “Spring–summer season” (β2 = −0.67) and “European Souslik” (β2 = −2.76) had a marginal effect in our models. Generally, the mean success rate of attack modes used by non-territorial eagles was 0.54 ± 0.50. Floaters successfully obtained food through: kleptoparasitism (43.10%), carrion feeding (24.14%), and high soar with vertical stoop (14.66%). Several important issues for the conservation of non-territorial Eastern Imperial Eagles arose from our research. The strong relation of floaters with the European Souslik calls for specific conservation measures aimed at the conservation of this type of prey and the restoration and appropriate management of its grassland habitats. The importance of the scavenging behavior of juvenile birds requires increased control of the use of poison baits and subsequent prosecution by state institutions. Protecting the most important temporary areas, improving institutional control against the use of poison baits, and intensifying awareness-raising campaigns among pigeon-fanciers and hunters are also of crucial importance for effective species conservation.

Personality predicts mode of attack in a generalist ground spider predator

AbstractPersonality traits, such as boldness and/or aggressiveness, have long been accepted to have a profound influence on many aspects of the lives of animals, including foraging. However, little is known about how personality traits shape the use of a particular attack strategy. Ground spiders use either venom or silk attack to immobilize prey. In this study, we tested the hypothesis that behavioral differences among individuals (namely boldness, measured as the time spent exploring a novel environment; and aggressiveness, measured as the number of killed but not consumed prey) drive the use of a particular attack strategy. We used a generalist ground spider, Drassodes lapidosus, and recorded the mode of attack on two types of prey, dangerous and safe. Moreover, we measured the size of the venom gland to test the relationship between the size of venom volume and the personality, as well as the mode of attack. Drassodes individuals showed consistent behavioral differences in the way they attacked prey. Venom attack was significantly related to increased aggressiveness when attacking spider (dangerous) prey and to increased boldness when attacking cricket (safe) prey. Silk attack was more frequently used by shy (for cricket prey) and docile (for spider prey). The volume of venom was not related to the attack strategy. We conclude that personality traits are important drivers of prey-capture behavior in generalist ground spiders.

HTTP Traffic Analysis based on Multiple Deep Convolution Network Model Generation Algorithms

In recent years, with the development of the Internet, social networking, online banking, e-commerce and other network applications are growing rapidly. At the same time, all kinds of malicious web pages are constantly emerging. Under the new situation, the network security threats are distributed, large-scale and complex. New network attack modes are emerging. With more and more diverse devices access to the Internet, our life is more intelligent and convenient, but also brings more loopholes and hidden dangers. Some malicious web pages through a variety of means to lure users to open URL links and conduct malicious behavior. However, if we can detect the URL of the malicious web page and identify the malicious web page, we can avoid the problems of content variability and behavior tracking. Therefore, traffic analysis based on various deep convolution network model generation algorithms arises at the historic moment, and becomes an important research issue in the field of Internet security.

Coordinated cyber-physical attack considering false overload of lines

The application of many sensor devices and complex information communication networks in a power system brings new potential threats to the electrical cyber-physical system. In this paper, a coordinated cyber-physical attack method considering false line overload is proposed. Through the combination of physical side attack and information side attack, the false data injection attack is used to cover up the physical attack and create false overload events, leading the dispatcher to make unreasonable scheduling strategies to achieve the purpose of affecting the economic operation of the system. In response to false overload events, this paper proposes a method to select the target line of false overload. At the same time, the output cost of the generator, the cost of load reduction and the number of overload lines are used as the evaluation indices to measure the damage capability of the coordinated attack, and the evaluation indices of different dimensions are normalized using fuzzy methods. Finally, the system losses caused by different attack modes are compared and analyzed through simulation calculations, and the effectiveness of the proposed model is verified.

Robustness test of multiple protection strategies for ecological networks from the perspective of complex networks: Evidence from Wuhan Metropolitan Area, China

AbstractRobustness is the key for strategies to protect ecological network (ENs) against external threats. In this study, we proposed a research framework for robustness testing of EN protection strategies. First, we constructed the ENs of Wuhan Metropolitan Area (WHM) based on the advantages of circuit theory in landscape ecology. We then extended complex network theory to the migration application of ENs. Five protection strategies of ENs were antagonistically combined with multiple attack modes to implement multi‐scenario simulations of attacks on ENs. The robustness parameters of the complex network were applied to measure the effectiveness of EN protection strategies under different attack efficiencies. The results showed that identifying the key performance of ENs makes hierarchical protection of ENs possible. Therefore, when focusing on maintaining the transmission efficiency and global connectivity, or preventing fragmentation of ENs in WHM, we recommend protecting the main trunks of ENs. When focusing on maintaining the backbone connectivity of ENs in WHM, we recommend protecting the main‐secondary trunks of ENs. We argue that specific protection strategies and directions can be determined according to the objectives of ecological construction and the scheme we put forward.

A Novel Attack Mode on Advanced Technology Nodes Exploiting Transistor Self-Heating

Self-heating (SH) is a phenomenon that can induce excessive heat inside the transistor channel. SH represents an emerging and serious concern, especially in advanced technology nodes, where excessive heat acting on elevated channel geometries will notably shift the critical transistor parameters (e.g., threshold-voltage <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{\text {th}}$ </tex-math></inline-formula> and carrier mobility <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula> ). The underlying 3-D device structures (e.g., FinFET, nanowire, or nanosheet structures), along with newly employed materials such as silicon-germanium (SiGe), which show worse thermal conductivity than traditional materials, can considerably exacerbate SH. On top of that, quantum confinement, a phenomenon that becomes dominant at sub-10nm, further increases the intensity of SH. In this article, we are the first to explore SH effects from the perspective of hardware security, rather than the performance, reliability standpoints covered in state-of-the-art (SOTA) work. As proof of concept, we devise an SH-based hardware trojan (HT) that exploits the SH-induced <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{\text {th}}$ </tex-math></inline-formula> change in 7-nm FinFET circuits. Leveraging <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{\text {th}}$ </tex-math></inline-formula> -dependent reconfigurable logic, we design a reconfigurable HT payload that maliciously changes its functional behavior once the SH-induced <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{\text {th}}$ </tex-math></inline-formula> change takes effect. Following SOTA work, we present a comprehensive modeling and analysis of SH effects at the device level and highlight its impact on transistor <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{\text {th}}$ </tex-math></inline-formula> . Next, we study how fabrication-time changes in the transistor doping and geometry can promote the SH-assisted degradation. We then describe various payload configurations for the proposed HT, quantify its overheads, and discuss its resilience against standard HT detection techniques. Finally, we demonstrate two case studies using the proposed HT, one to leak the secret key from a pipelined design of an advanced encryption standard (AES) circuit, and another to showcase denial-of-service for a Gaussian-blur filter circuit. Our work utilizes industry-standard models with parameters extracted from measurements and calibrated with experiments. Our results are obtained from meticulous study and optimization across the device-, circuit-, and system-levels.

Security in Edge-Centric Intelligent Internet of Vehicles: Issues and Remedies

Propelled by the growth of automotive industry, and the ubiquity of smart sensors, intelligent transport systems such as the Internet of Vehicles (IoV) have seen significant research interest in recent times. The emerging distributed IoV networks support real-time vehicular applications through on-device computing, communication-efficient data processing, edge computing, and cloud aggregation. While enriching the user experience by minimizing the end-to-end latency through efficient energy management, IoV deployments face the fundamental challenge of security attacks. In this article, we discuss various security attack modes in an edge-centric intelligent IoV framework, consisting of distributed smart vehicles, and remote processing units. We highlight various attack detection and mitigation mechanism for the proposed IoV framework, to address the security challenges. Finally, we shed light on several future research directions to ensure security of sensory data in edge-centric IoV systems.

To defend or to attack? Antagonistic interactions between Serratia plymuthica and fungal plant pathogens, a species-specific volatile dialogue

Volatile organic compounds (VOCs) are involved in microbial interspecies communication and in the mode of action of various antagonistic interactions. They are important for balancing host-microbe interactions and provide the basis for developing biological control strategies to control plant pathogens. We studied the interactions between the bacterial antagonist Serratia plymuthica HRO-C48 and three fungal plant pathogens Rhizoctonia solani, Leptosphaeria maculans and Verticillium longisporum. Significant differences in fungal growth inhibition by the Serratia-emitted VOCs in pairwise dual culture assays and changes in the transcriptome of the bacterium and in the volatilomes of both interacting partners were observed. Even though the rate of fungal growth inhibition by Serratia was variable, the confrontation of the bacterium with the VOCs of all three fungi changed the levels of expression of the genes involved in stress response, biofilm formation, and the production of antimicrobial VOCs. Pairwise interacting microorganisms switched between defense (downregulation of gene expression) and attack (upregulation of gene expression and metabolism followed by growth inhibition of the interacting partner) modes, subject to the combinations of microorganisms that were interacting. In the attack mode HRO-C48 significantly inhibited the growth of R. solani while simultaneously boosting its own metabolism; by contrast, its metabolism was downregulated when HRO-C48 went into a defense mode that was induced by the L. maculans and V. longisporum VOCs. L. maculans growth was slightly reduced by the one bacterial VOC methyl acetate that induced a strong downregulation of expression of genes involved in almost all metabolic functions in S. plymuthica. Similarly, the interaction between S. plymuthica and V. longisporum resulted in an insignificant growth reduction of the fungus and repressed the rate of bacterial metabolism on the transcriptional level, accompanied by an intense volatile dialogue. Overall, our results indicate that VOCs substantially contribute to the highly break species-specific interactions between pathogens and their natural antagonists and thus deserving of increased consideration for pathogen control.

Resilient control for T-S fuzzy systems with multiple transmission channels under asynchronous denial-of-Service attacks

This paper investigates the resilient control problem of the T-S fuzzy system with multiple transmission channels under asynchronous denial-of-service (DoS) attacks. A part of the transmission channels are chosen as ones corresponding to the premise variables of the T-S fuzzy system. The considered asynchronous DoS attacks occur on the sensor-to-controller transmission channels, which may DoS attacks cause the system state to be unstable. The stability of the T-S fuzzy system under random DoS attack mode is considered, including the influence of the premise variables attacked. A class of parallel distributed compensation controllers are proposed, which adopts the strategy of discarding the attacked channels information. By using a piecewise Lyapunov function method and a class of linear matrix inequalities, the system stability sufficient conditions related to the duration and frequency of DoS attacks on the transmission channels are presented. Compared with the existing method, the more general relationship between premise variables and transmission channels are considered in this paper. Finally, a numerical example is used to illustrate the effectiveness of the proposed scheme.

An insight into algicidal characteristics of Bacillus altitudinis G3 from dysfunctional photosystem and overproduction of reactive oxygen species

Cyanobacterial blooms negatively affect aquatic ecosystems and human health. Algicidal bacteria can efficiently kill bloom-causing cyanobacteria. Bacillus altitudinis G3 isolated from Dianchi Lake shows high algicidal activity against Microcystis aeruginosa. In this study, we investigated its algicidal characteristics including attack mode, photosynthesis responses, and source and the contribution of reactive oxygen species (ROS). The results showed that G3 efficiently and specifically killed M. aeruginosa mainly by releasing both thermolabile and thermostable algicidal substances, which exhibited the highest algicidal activity (99.8%, 72 h) in bacterial mid-logarithmic growth phase. The algicidal ratio under full-light conditions (99.5%, 60 h) was significantly higher than under dark conditions (<20%, P < 0.001). G3 filtrate caused photosystem dysfunction by decreasing photosynthetic efficiency, as indicated by significantly decreased Fv/Fm and PIABS (P < 0.001) values. It also inhibited photosynthetic electron transfer as indicated by significantly decreased rETR (P < 0.001), especially QA− downstream, as revealed by significantly decreased φEo and ψo, and increased Mo (P < 0.001). These results indicated that the algicidal activity of G3 filtrate is light-dependent, and the cyanobacterial photosystem is an important target. Cyanobacterial ROS and malondialdehyde contents greatly increased by 37.1% and 208% at 36 h, respectively. ROS levels decreased by 49.2% (9 h) when diuron (3-(3-4-dichlorophenyl)-1,1-dimethylurea) partially blocked photosynthetic electron transport from QA to QB. Therefore, excessive ROS were produced from disrupted photosynthesis, especially the inhibited electron transport area in QA− downstream, and caused severe lipid peroxidation with significantly increased MDA content and oxidative stress in cyanobacteria. The ROS scavenger N-acetyl-l-cysteine significantly decreased both cyanobacterial ROS levels (34%) and algicidal ratio (52%, P < 0.05) at 39 h. Thus, excessive ROS production due to G3 filtrate administration significantly contributed to its algicidal effect. G3 could be an excellent algicide to control M. aeruginosa blooms in waters under suitable light conditions.

AIBot: A Novel Botnet Capable of Performing Distributed Artificial Intelligence Computing

As an infrastructure platform for launching large-scale cyber attacks, botnets are one of the biggest threats to cyberspace security today. With the development of network technology and changes in the network environment, network attack intelligence has become a trend, and botnet designers are also committed to developing more destructive intelligent botnets. The feasibility of implementing distributed intelligent computing based on botnet node resources is analyzed with regard to the aspects of program size, communication traffic and resource occupancy. AIBot, a botnet model that can perform intelligent computation in a distributed manner, is proposed from the attacker’s perspective, which hierarchically deploys distributed neural network models in the botnet, thereby organizing nodes to collaboratively perform intelligent computation tasks. AIBot enables the distributed execution of intelligent computing tasks on a cluster of bot nodes by decomposing the computational load of a deep neural network model. A general algorithm for the distributed deployment of neural networks in AIBot is proposed, and the overall operational framework for AIBot is given. Two classical neural network models, CNN and RNN, are used as examples to illustrate specific schemes for deploying and running distributed intelligent computing in AIBot. Experimental scenarios were constructed to experimentally validate and briefly evaluate the performance of the two AIBot attack modes, and the overall efficiency of AIBot was evaluated in terms of execution time. This paper studies new forms of botnet attack techniques from a predictive perspective, aiming to increase defenders’ understanding of potential botnet threats, in order to propose effective defense strategies and improve the botnet defense system.

A Switched Newton–Raphson-Based Distributed Energy Management Algorithm for Multienergy System Under Persistent DoS Attacks

The security and economy of multienergy systems (MESs) are directly threatened by the potential cyberattacks. It is of great importance to investigate the effects of cyberattacks, e.g., denial of service (DoS) attacks, on distributed energy management algorithms. To this end, this article focuses on exploring how the frequency and the time of duration of DoS attacks influence the behavior of Newton–Raphson-based distributed energy management (NRBDEM) algorithm for MES and in which condition the optimal operations can still be obtained. First, a switched NRBDEM algorithm is presented, which is composed of the normal operation mode and the attack mode. In the attack mode, the attackers are able to change the communication structure at will and make it unconnected to destroy the convergence of the switched NRBDEM algorithm. Then, by making use of automatons to generate the hybrid time domain, the switched NRBDEM algorithm is further modeled and formulated as a hybrid dynamical system, which provides a mathematical model for the subsequent convergence analysis. Therein, the generated hybrid time domain satisfies the average dwell-time constraint and time-ratio constraint to limit the persistent attacks. Furthermore, we analyze the restrained conditions for persistent attacks, under which the optimality and convergence of the switched NRBDEM algorithm can be guaranteed still. Finally, simulation results demonstrate the effectiveness of the proposed method. Note to Practitioners—The multienergy system (MES) is viewed as a typical cyber-physical system. Since it works in a networked environment, the convergence and optimality of the corresponding distributed energy management algorithms are easily destroyed by various cyberattacks, such as the DoS attack. Few noticeable studies have been documented to investigate the distributed energy management problem for MES under persistent DoS attacks. To address this issue, this article proposes a switched NRBDEM algorithm subject to persistent DoS attacks, which is made up of normal operation mode and attack mode. Moreover, sufficient conditions are derived for the tolerable frequency of attack and the time of duration. It is helpful for the practitioners to know in which conditions the global convergence and optimality can be maintained still. Several simulations are provided to verify the correctness of the proposed theoretical analysis results. Our future work will aim at analyzing the performance of different energy management algorithms under different cyberattacks.

Compound adversarial examples in deep neural networks

Although deep learning has made great progress in many fields, they are still vulnerable to adversarial examples. Many methods for generating adversarial examples have been proposed, which either contain adversarial perturbation or patch. In this paper, we explore the method that creates compound adversarial examples including both perturbation and patch. We show that fusing two weak attack modes can produce more powerful adversarial examples, where the patch covers only 1% of the pixels at random location in the image, and the perturbation changes only by 2/255 in the original pixel value (scale to 0–1). For both targeted attack and untargeted attack, compound attack can improve the generative efficiency of adversarial examples, and can attain higher attack success rate with fewer iteration steps. The compound adversarial examples successfully attack the models with defensive mechanisms that previously can defend perturbation attack or patch attack. Furthermore, the compound adversarial examples show good transferability on normal trained classifiers and adversarial trained classifiers. Experimental results on a series of widely used classifiers and defense models show that the proposed compound adversarial examples have strong robustness, high effectiveness, and good transferability.

Analysis of the optimal shooting angle in football matches based on network data mining

As an important research field in video understanding, human behavior recognition is widely used in intelligent monitoring, intelligent nursing, interactive behavior, and robot control. ObjectiveThe purpose of this study is to explore the analysis method of interaction behavior of soccer video. MethodsIn this study, we use scene simulation, big data analysis, experimental comparison, and other methods to analyze the interaction between players in a football video. The resultsof experiment and analysis of attack mode under interactive behavior show that the accuracy of attack direction determination results in traditional video analysis method and accurate analysis method is 75.60 % and 85.2 %, respectively. In this study, taking into account the two factors of staff proportion and average distance, the accuracy is improved to a certain extent, which can reach 91.5 %. It can be concluded that the deviation rate proposed in this study can better describe the position information of the football, can realize the detection of the basic attack mode, and can enable the computer to identify the attack direction of both players in the shortest time, which shows that the method is efficient and accurate. It provides a good analysis method for the research of human interaction behavior in football video and even in all kinds of videos.