Length matters:Scalable fast encrypted internet traffic service classification based on multiple protocol data unit length sequence with composite deep learning

As an essential function of encrypted Internet traffic analysis,encrypted traffic service classification can support both coarse-grained network service traffic management and security supervision.However,the traditional plaintext-based Deep Packet Inspection(DPI)method cannot be applied to such a classification.Moreover,machine learning-based existing methods encounter two problems during feature selection:complex feature overcost processing and Transport Layer Security(TLS)version discrepancy.In this paper,we consider differences between encryption network protocol stacks and propose a composite deep learning-based method in multiprotocol environments using a sliding multiple Protocol Data Unit(multiPDU)length sequence as features by fully utilizing the Markov property in a multiPDU length sequence and maintaining suitability with a TLS-1.3 environment.Control experiments show that both Length-Sensitive(LS)composite deep learning model using a capsule neural network and LS-long short time memory achieve satisfactory effectiveness in F1-score and performance.Owing to faster feature extraction,our method is suitable for actual network environments and superior to state-of-the-art methods.

Research on the Drunk Driving Traffic Accidents Based on Logistic Regression Model

Most of the current studies on drunk driving accidents focus on law making and public education. However, especially in China, there is less statistical analysis on the severity of drunk driving accidents between driving under the influence of alcohol (DUI) and driving while intoxicated (DWI). 3368 drunk driving related crashes were collected from the blood-alcohol test report in a city of China at 2012 and 2013. After data pre-processing, Chi-square tests were used to analyze the association between different variables and the type of drunk driving. The logistic regression model is conducted to estimate the effect of the variables under DUI and DWI. The results show that Hour of the day, Driver’s age, Driver’s casualties and Accident area have significant correlation with drunk driving. There was a slightly decrease by 0.995 per year with age and a slightly increase by 1.014 with time in the possibility of DWI. DWI is more likely to cause death in traffic accidents (OR = 1.316) than DUI. Driver’s deaths (OR = 2.346) is more likely to happen than the injuries (OR = 1.910) under DWI cases. These findings show that more attention should be paid to strengthen controls on the DWI. It also can provide important basis for accident prevent, traffic law enforcement and traffic management.

Rethink Left-Behind Experience: New Categories and Its Relationship with Aggression

Left-behind experience refers to the experience of children staying behind in their hometown under the care of only one parent or their relatives while one or both of their parents leave to work in other places.College students with left-behind experience showed higher aggression levels.To further explore the relationship between left-behind experience and aggression,the current study categorized left-behind experience using latent class analysis and explored its relationship with aggression.One thousand twenty-eight Chinese college students with left-behind experience were recruited,and their aggression levels were assessed.The results showed that there were four categories of left-behind experience:“starting from preschool,frequent contact”(35.5%),“less than 10 years in duration,limited contact”(27.0%),“starting from preschool,over 10 years in duration,limited contact”(10.9%),and“starting from school age,frequent contact”(26.6%).Overall,college students who reported frequent contact with their parents during the left-behind period showed lower levels of aggression than others did.Females were less aggressive than males in the“starting from preschool,frequent contact”left-behind situation,while males were less aggressive than females in the“starting from school age,frequent contact”situation.Thesefindings indicate that frequent contact with leaving parents contributes to decreasing aggression of college students with left-behind experience.Meanwhile,gender is an important factor in this relationship.

Tomographic particle image velocimetry measurement on three-dimensional swirling flow in dual-stage counter-rotating swirler

Three-Dimensional(3D)swirling flow structures,generated by a counter-rotating dualstage swirler in a confined chamber with a confinement ratio of 1.53,were experimentally investigated at Re=2.3×10^(5)using Tomographic Particle Image Velocimetry(Tomo-PIV)and planar Particle Image Velocimetry(PIV).Based on the analysis of the 3D time-averaged swirling flow structures and 3D Proper Orthogonal Decomposition(POD)of the Tomo-PIV data,typical coherent flow structures,including the Corner Recirculation Zone(CRZ),Central Recirculation Zone(CTRZ),and Lip Recirculation Zone(LRZ),were extracted.The counter-rotating dual-stage swirler with a Venturi flare generates the independence process of vortex breakdown from the main stage and pilot stage,leading to the formation of an LRZ and a smaller CTRZ near the nozzle outlet.The confinement squeezes the CRZ to the corner and causes a reverse rotation flow to limit the shape of the CTRZ.A large-scale flow structure caused by the main stage features an explosive breakup,flapping,and Precessing Vortex Core(PVC).The explosive breakup mode dominates the swirling flow structures owing to the expansion and construction of the main jet,whereas the flapping mode is related to the wake perturbation.Confinement limits the expansion of PVC and causes it to contract after the impacting area.

A review of low-temperature plasma-assisted machining:from mechanism to application

Materials with high hardness,strength or plasticity have been widely used in the fields of aviation,aerospace,and military,among others.However,the poor machinability of these materials leads to large cutting forces,high cutting temperatures,serious tool wear,and chip adhesion,which affect machining quality.Low-temperature plasma contains a variety of active particles and can effectively adjust material properties,including hardness,strength,ductility,and wettability,significantly improving material machinability.In this paper,we first discuss the mechanisms and applications of low-temperature plasma-assisted machining.After introducing the characteristics,classifications,and action mechanisms of the low-temperature plasma,we describe the effects of the low-temperature plasma on different machining processes of various difficult-to-cut materials.The low-temperature plasma can be classified as hot plasma and cold plasma according to the different equilibrium states.Hot plasma improves material machinability via the thermal softening effect induced by the high temperature,whereas the main mechanisms of the cold plasma can be summarized as chemical reactions to reduce material hardness,the hydrophilization effect to improve surface wettability,and the Rehbinder effect to promote fracture.In addition,hybrid machining methods combining the merits of the low-temperature plasma and other energy fields like ultrasonic vibration,liquid nitrogen,and minimum quantity lubrication are also described and analyzed.Finally,the promising development trends of low-temperature plasma-assisted machining are presented,which include more precise control of the heat-affected zone in hot plasma-assisted machining,cold plasma-assisted polishing of metal materials,and further investigations on the reaction mechanisms between the cold plasma and other materials.

Effect of sulfur source on photocatalytic degradation performance of CdS/MoS2 prepared with one-step hydrothermal synthesis

Cd S/Mo S2,an extremely efficient photocatalyst,has been extensively used in hydrogen photoproduction and pollutant degradation.Cd S/Mo S2can be synthesized by a facile one-step hydrothermal process.However,the effect of the sulfur source on the synthesis of Cd S/Mo S2via one-step hydrothermal methods has seldom been investigated.We report herein a series of one-step hydrothermal preparations of Cd S/Mo S2using three different sulfur sources：thioacetamide,L-cysteine,and thiourea.The results revealed that the sulfur source strongly affected the crystallization,morphology,elemental composition and ultraviolet（UV）–visible-light-absorption ability of the Cd S/Mo S2.Among the investigated sulfur sources,thioacetamide provided the highest visible-light absorption ability for Cd S/Mo S2,with the smallest average particle size and largest surface area,resulting in the highest efficiency in Methylene Blue（MB）degradation.The photocatalytic activity of Cd S/Mo S2synthesized from the three sulfur sources can be arranged in the following order：thioacetamideL-cysteinethiourea.The reaction rate constants（k）for thioacetamide,L-cysteine,and thiourea were estimated to be 0.0197,0.0140,and 0.0084 min^（-1）,respectively.However,thioacetamide may be limited in practical application in terms of its price and toxicity,while L-cysteine is relatively economical,less toxic and exhibited good photocatalytic degradation performance toward MB.