Search

In an online social network (OSN), users share their personal media such as hobbies, photos, and videos. These information can help users find potential friends, but the exposure of their characteristics may also cause problems like their features being mined and utilized. To solve these problems, many social network providers choose to hide users’ sensitive information like their real names and exact positions. To ensure the effectiveness of the methods that can protect users’ information, there are several attacks to infer users’ hidden information. In this paper, based on the previous analysis of the relevance between two attributes, we analyze the relevance of multiple attributes and find that there is also a strong correlation between multiple attributes as well as the correlation between attributes and behaviors, which also has an impact on the attribute inference. Therefore, based on the correlation between multiple attributes and behaviors, we propose a new attribute inference scheme. We use the Apriori algorithm based on data cube to dig the correlation between multiple attributes and behaviors and analyze them with Kulczynski measure and Cosine measure. We model a social network as a Markov Random Field (MRF) and use Loopy Belief Propagation (LBP) for attribute inference attacks. We compare our method with the previous algorithm. The results show that our method has better performance than the traditional methods.

Error recognition, as the basis of error compensation, is very important for improving the forming accuracy of CNC machines. To improve the accuracy of error recognition, this paper provides an overview of the current research algorithms. Firstly, it briefly introduces the importance of geometric error recognition. Secondly, it classifies and summarizes the geometric error recognition algorithm based on five related devices such as ball bars, R-test device, on-machine measurement system, machine vision system, and others. Finally, the research trend of the geometric error recognition algorithm for CNC machine tools has prospected. It is pointed out that the application of machine vision technology on geometric error recognition will be the focus of future research.

The quality of the machined surface largely affects some physical properties such as wear resistance, corrosion resistance and fatigue resistance of parts, as well as functions such as service life and reliability. Surface quality is one of the requirements of customers for parts processing. The biggest indicator of the surface quality of machining is the size of the roughness. A neural network is used to predict the surface roughness of the machining workpiece, and stainless steel is selected as the workpiece material. Different cutting experiments were carried out at different cutting speeds, different feed rates, and different cutting depths. Aiming at the change of cutting conditions in the cutting process, the neural network model is used to predict the surface roughness during the process. The results of neural network prediction of workpiece surface roughness show that the numerical error predicted by this method is small and the accuracy is high. The prediction model has good prediction accuracy, and its maximum error does not exceed 2%. The model can predict the surface roughness after processing under different combinations of cutting speed, feed and depth of cut parameters, and has important guiding significance for the control of the surface quality of parts.

Social network is a critical component in mobile multimedia systems, where users share their videos, photos, and other media. However, the information (e.g., posts, user profiles, etc.) shared on the social network platforms usually reflects many users' personal (private) information, which could be mined and abused for malicious purposes. To address privacy concerns, many social network service providers adopted privacy-preserving mechanisms, e.g., anonymizing user identity, hiding users' profiles, etc. As a result, the attributes in user profiles are usually set up to be accessed only by friends to prevent privacy leakage. Several attacks have been proposed to infer the hidden attributes to Several the efficiency of current privacy-protecting mechanisms. Most of these solutions are based on the social links among users or their behaviors. In this paper, we systematically analyze the social features related to user privacy inference and found that there are relevances among social attributes, which has a great impact on inferring users' hidden attributes. According to our findings, we propose an efficient social attribute inference scheme based on social links and attribute relevance properties. We develop a relevance attribute inference method (ReAI) using random walks with restart. We analyze attribute relevance on inference performance and use Kulczynski measure to quantify attribute relevance as edge weights of attribute nodes in an improved social-attribute network. We evaluate our method and compare it with the traditional attribute inference method. The results show that our method performs better than the traditional method. We also use Kulczynski measure and Information Gain Ratio to evaluate the improvements. The results show that the bigger relevance between attributes contributes to higher improvements.

The unintentional electromagnetic (EM) radiation from electronic information equipment will cause the leakage of information. EM information leakage identification is an important research area of EM information security. Previous studies only focused on a particular type of information leakage, ignoring that there are multiple types of information contained in the leaked EM signal. This paper presents a multi-dimensional information leakage model for the first time, which reveals the multiple information leakage risks in an unintentional EM radiation. The EM information leakage of computer monitors is taken as an example to verify the model. Furthermore, a Convolutional Neural Network (CNN) based multi-dimensional information identification method is proposed. This novel method designed a unified neural network architecture to identify multiple information leaked by the same EM radiation from the monitor. It overcomes the limitation of available identification methods that need to definite leakage characteristics for each type of information leakage. The effectiveness of the method is validated by contrast experiments.

JavaScript applications are widely used in a range of scenarios, including Web applications, mobile applications, and server-side applications. On one hand, due to its excellent cross-platform support, Javascript has become the core technology of social network platforms. On the other hand, the flexibility of the JavaScript language makes such applications prone to attacks that inject malicious behaviors. In this paper, we propose a detection technique to identify malicious behaviors in JavaScript applications. Our method models an application’s normal behavior on function activation, which is used as a basis to detect attacks. We prototyped our solution on the popular JavaScript engine V8 and used it to detect attacks on the android system. Our evaluation shows the effectiveness of our approach in detecting injection attacks to JavaScript applications.

In order to improve the search speed of AGV path planning and shorten the search time, the author summarizes the algorithm and characteristics of AGV on path planning. Firstly, the AGV is briefly introduced, and the path planning technology is summarized. Secondly, the AGV path planning is classified and summarized. From the perspective of the AGV's environmental mastery, the AGV path planning is divided into two types: global planning and local planning. Then the related algorithms of global planning and local planning are reviewed, and the development status, advantages and disadvantages of related algorithms are summarized. Finally, the future development trend of AGV path planning technology in further research, hybrid algorithm, multi-AGV collaboration, complex environment and multi-dimensional environment is pointed out.

Based on the operation process of cold chain logistics center and the relationship between different businesses, this paper plans the existing storage center reasonably. On the basis of SLP method, the basic elements of logistics objects, logistics flow and logistics operation route are analyzed. Make clear the relationship between logistics operation units and non-logistics operation units, use the weighted sum of their comprehensive relations to generate the comprehensive proximity ranking, and determine the area and location of the operation units based on the actual situation and site conditions. Finally, the reasonable layout is obtained by comparing and analyzing with the existing scheme.

In this paper, A correction algorithm is proposed for QR-code that cannot be scanned by common tools. First, we use the concept of contour tree to filter the contours and cluster them to extract the required positioning markers. Secondly, the three methods of slope comparison, coordinate comparison and angle rotation are using to determine the correction point. Finally, the image is corrected using an affine transformation. The experimental results show that the QR-code corrected by the algorithm proposed in this paper can be scanned by common tools, which shows that the proposed algorithm has certain useful value.

For the detection of moving targets, the background complexity is high, and when the target is acquired, it is easy to be affected by the surrounding environment and illumination, so that the obtained target contour is incompletely selected, and the moving target can be accurately tracked. Then, a target detection method based on HSV fusion Gaussian mixture model is proposed. Firstly, the HSV is used for color conversion, then the ROI region is selected, and the obtained ROI region is thresholded to obtain a binarized graph; then the median filtering and kernel noise reduction method are used to remove the background noise in the video target motion feature; The morphological operation is used to process the image; then Gaussian mixture modeling is performed according to the current image, and the noise of similar colors in the background is removed to obtain a complete moving target. The moving target obtained by this method is more accurate, avoiding the influence of the surrounding environment and illumination, and is better than the traditional Camshift tracking effect.

The domain name resolution system provides support service for website visits as the basic service of the Internet. With the increase of DNS attacks, it has brought copious challenges to network security. The paper studies on the key defense technologies against DNS attacks based on the DNS principle. The multi-core customized to the DNS is adopted to analyze hardware kernel, while AI algorithms being realized for malicious flow cleaning and intelligent routing running on the cloud system established specifically for DNS. The designed DNS intelligent cloud system can provide high-efficiency domain name resolution in practice, while ensuring the network security.

This paper proposes a novel education paradigm named Wisdom Class (WisClass), comprising three main pillars: Virtual Learning Environment, Mobile Learning Community and Ubiquitous Network. The WisClass teaching mode boasts the new characteristics of independent learning; inquiry learning; and collaborative learning. The paper contains the following contributions: (1) Define WisClass: Due to the diversity of WisClass, it is necessary to comprehensively cover WisClass from many angles, including Intelligent Contents, System Implementation and Classroom Infrastructure. (2) Summarize the Innovation of WisClass: We propose to shift from the conventional Knowledge Class (KnoClass) paradigm to a new WisClass paradigm. More elaborately, a) WisClass improves teaching theory from teacher-oriented to student-oriented; b) WisClass improves teaching paradigm from static state to dynamic state; c) WisClass improves teaching strategy from teaching knowledge to cultivating wisdom; d) WisClass improves teaching method from closed class to open class. (3) Analyze the Characteristics of WisClass: Through massive researches and deep analysis, we explore the WisClass from three characteristics. a) WisClass Teaching: WisClass has a process of enlightening, situational, dynamic, and innovative teaching. b) WisClass Learning: WisClass has a process of constructivist, inquiring, blended and cooperative learning. c) WisClass Environment: WisClass has a digital, interactive, dynamic and situational environment. (4) Construct a real WisClass: we construct a real WisClass by cooperatively building hardware environment and software system as a holistic design system. a) The hardware environment includes cloud server, WIFI server, hardware equipment; b) the software system contains cloud platform, online/offline learning environment, leaning evaluation system and Trinitarian learning material system. Our works will play a promising role in the future development of several system design perspectives: (1) Intelligent Contents: In order to better cultivate wise learners, the teaching environment and objective are both focused on intelligent teaching methods. Specifically, a) it takes the educators’ wisdom teaching method as essential condition; b) it takes the digital teaching environment as material basis; c) it takes the learners’ wisdom development as teaching objective. WisClass incarnates the think of “Transforming Knowledge In to Wisdom”. (2) Classroom Infrastructure: The intelligent teaching in WisClass hinges upon perceptive, dynamic, active, open and wisdom of physical space. In order to provide a human-friendly space, it is important to have WisClass Infrastructure equipped with modern IT devices, e.g. audio-visual equipment, computer and interactive whiteboard, internet-accessible teaching materials. Especially, the last system encourage educators and learners to timely access the resources and update learning materials. In addition, an ideal infrastructure should also support various WIFI-accessible and internet-enhanced teaching tools such as Distance Learning and Learning Community. WisClass facilitates intelligent teaching environments. Moreover, it can seamlessly integrate people, devices, resource, environment, thought and methods. All this together make it feasible to realize multiple intelligent scenes teaching and man-machine interaction. (3) System Implementation: In order to provide an intelligent and informational classroom, we incorporate modern and most advanced IT. The purpose is that WisClass may improve teaching paradigm in the aspects of three-dimensional interaction, smart evaluation and information push. WisClass provides a natural platform which integrates IT and class teaching. (Abstract)

Digital electronic devices emit weak electromagnetic signals and cause information leakage during operation. In this paper, stochastic resonance is used to detect the weak electromagnetic leakage signals. stochastic resonance can enhance the signal amplitude by transfer the noise energy to the signal, thus improve the signal-to-noise ratio and realize feature extraction of weak electromagnetic signals from complex noise backgrounds.

The similarity of the user interfaces (UIs) among Android apps is an important factor to indicate abnormal behaviors in Android apps. For example, phishing apps use UIs similar to those of their target apps to lure users to input sensitive information. As another example, repackaged apps preserve the UIs of the original apps while adding malicious code. In this paper, we propose a novel solution, GeminiScope, to robustly detect similar UIs among Android apps. Our approach analyzes the UI layouts of Android apps, extracts the fundamental features of the UIs' visual appearance, and rates the similarity among the UIs of apps. We evaluated GeminiScope using a few sets of apps from various sources. It reliably detected apps with similar UIs, showing the potential of using visual similarity as a unique feature to detect malicious apps.

The thin-walled parts of the tank are widely used in aerospace filed and the accuracy is the basis requirement for ensuring the usability. Due to the low rigidity of the thin-walled parts, the milling force becomes a major factor for the machining deformation. Based on analysis the relationship between feed speed and the length of the cutting arc, the feed speed optimization method is proposed to reduce the machining deformation. To reduce the machining deformation the optimal machining path is put forward which is base the finite element analysis results on the rigidity distribution law of the parts. At the same time, an efficient compensation method based on laser measurement is proposed. Through machining the aluminium alloy 2219 thin-walled tank parts as an example, experimental results indicate that the effectiveness of the proposed control method in machining process. This research provides an effective approach to reduce the machining deformation for the tank thin-walled parts.

To facilitate apps to collaborate in finish complex jobs, Android allows isolated apps to communicate through explicit interfaces. However, the communication mechanisms often give additional privilege to apps, which can be exploited by attackers. The Android Task Structure is a widely-used mechanism to facilitate apps' collaboration. Recent research has identified attacks to the mechanism, allowing attackers to spoof UIs in Android. In this paper, we present an analysis on the security of Android task structure. In particular, we analyze the system/app conditions that can cause the task mechanism to leak privilege. Furthermore, we identify new end-to-end attacks that enable attackers to actively interfere with victim apps to steal sensitive information. Based on our findings, we also develop atask interference checking app for exploits to the Android task structure.

Practical skills are critical requirements of cyber security professionals. Adding practical skill training in university curriculums not only better prepares students for their future career, it also deepen their understanding to knowledge. To support such trainings, we designed infrastructures and curriculums to support practical experimentation in cyber security training. In this paper, we share our experiences over the past five years in two universities.

Information equipment industry is knowledgeintensive and technology-intensive industry. The ability of independent innovation is the key driver for industry development. This paper studied modeling of the information equipment industry development. The academic attributes and the technical attributes be introduced into the model, through defining the theoretical density, the technology density and the new product density. It is proposed that establishing information equipment industry model with the innovation index, equity index and productivity index, as well as these expressions. Additionally, the paper put forward new way to calculate the weight of indicators which related to specific research problems' context-relevant. Analyzed the batch data of information equipment industry from 2010 to 2014 based on the established model, and through a Python program to achieve the sorting and global optimization of decision-making, shows part of the data analysis results as visualize graphs.

Digital electronic equipment emitting unintentional electromagnetic signals in working state need to be tested to ensure data security. The ordinary testing method using traditional devices, such as spectrum analyzer and oscilloscope, cannot provide multi-functions and it is inconvenient outside the laboratory. The novel testing method using virtual instrument can provide more portability and flexibility and the leakage signal reconstruction effect is more intuitive.

Cloud-based data storage is a popular cloud services. It can alleviate data management burden of data owner. However, after data file is transferred to the cloud, data owner is not able to physical dominate these outsourced data file. To efficiently ensure these data intact, several ReD-based integrity checking schemes were presented to ensure data integrity. Due to periodical checking, many data integrity checking schemes makes that the verifier takes high computational cost or communication overhead to verify data intactness. This results in a heavy burden for some schemes. Recently, to ensure secure cloud storage, a dynamic-hash-Table based public auditing protocol was put forward to achieve data privacy protection and data dynamics. An outstanding feature of the protocol is to significantly decease computational cost and communication cost in terms of cloud server and the verifier. Very regretfully, in this work, we will point out that their protocol is not secure. Our attack enables a malicious cloud server to arbitrarily delete or alter the outsourced data without being inspected by the verifier. And We analyze the reason to produce such attack and give a advice to fix the problem. Finally, we also point out that their protocol still exist a security threat: malicious auditor attack, and analyze the corresponding reason to produce such threat.

Software-defined networking (SDN) supports flexibly routing and switching by splitting the control plane and data plane in network devices. However, the security resource utilization now in SDN is inefficient, especially in the process of routing path construction. In this paper, we present a novel model for SDN controller to construct secure routing paths based on Network Security Virtualization (NSV) and the latest network status. Our new scheme considers various security devices' functionalities and the trust of different network devices in routing path establishment. We also introduce an identity-based authentication module to ensure the legitimacy of routing policies in SDN. We implement the prototype system, NSV-GUARD, and conduct theoretical analysis and performance evaluation. The results demonstrate that our approach can optimally utilize existing security devices and mechanisms in SDN. In addition, NSV-GUARD can efficiently verify malicious and fake routing policies generated by unregistered routing applications.

Hybrid mobile applications (or apps) are based on web technologies, such as HTML5 and JavaScript, and run in a browser environment. They facilitate cross-platform development. However, the security issues of web technologies are inherited by hybrid mobile apps, where the injected code may execute with the system-level privilege. In this paper, we propose a behavior model to detect malicious behaviors in hybrid mobile apps. Our model uses function-level information to describe how an app's behaviors are activated. Furthermore, once script injection happens, the behaviors made by the injected code can be detected according to the deviation from the app's behavior model.

In a typical two-step sequential deposition of perovskites such as MAPbI 3 (MA= CH 3 NH 3 +, PbI 2 is first deposited on the substrate (mesoporous or planar scaffold) by spin-coating or vacuum evaporation, subsequently transformed into the perovskite (MAPbI 3 ) by exposing it to an anhydrous isopropanol (IPA) solution of MAI. For the two-step sequential deposition, the conversion and film morphology of the final perovskite film strongly depend on the initial PbI 2 film during the first step of the process. Conventionally, PbI 2 from dimethyl formamide (DMF) solution tends to form a layered and dense crystalline film on a flat substrate However, the complete conversion of PbI 2 to perovskite on exposure to the MAI solution usually requires several hours. However, this long reaction time in MAI solution could lead to the dissolution of perovskites. These drawbacks make it difficult to fabricate planar-structured PSCs by sequential deposition method. To our knowledge, a simple and effective method for fast conversion of PbI 2 film into perovskite on flat substrate via two-step sequential deposition process at room temperature has not been reported. Although many groups have demonstrated the PbI 2 residue in perovskite film has a positive effect on the efficiency of PSCs, the effect of PbI 2 residue on the long-term stability of PSCs is unclear. In this work, we demonstrate a new approach for forming the PbI 2 nanostructure and the use of high CH 3 NH 3 I concentration which are adopted to form high quality (large crystal size and smooth) perovskite film with better photovoltaic performances. On one hand, self-assembled porous PbI 2 is formed by incorporating small amount of rationally chosen additives into the PbI 2 precursor solutions, which significantly facilitate the conversion of perovskite without any PbI 2 residue. On the other hand, through employing a relatively high CH 3 NH 3 I concentration, a firmly crystallized and uniform CH 3 NH 3 PbI 3 film is formed. As a result, a promising power conversion efficiency (PCE) of 16.21% is achieved in planar-heterojunction PSCs. Furthermore, we experimentally demonstrate that the PbI 2 residue in perovskite film has a negative effect on the long-term stability of devices.

Cloud computing is a kind of internet-based computing, it outstanding property is on-demand computing. As an important service, cloud storage is able to provide users and enterprises with various capabilities to store and access their data in the centers. It relies on sharing of resources to achieve coherence and economies of scale. However, in most of data sharing protocols, to update the outsourced data, the data updating operation is only executed by data owner. Obviously, it is far from practical due to the tremendous computational cost on data owner. To support multi-user modification, we propose a novel public auditing for data sharing with multi-user modification by using ring signature in this paper. Our scheme does not only support public checking and efficient user revocation, but provide backward security. Efficiency analysis shows that our scheme is very efficient.

Indexing metric spaces have long been a heated subject in the area of data processing. This paper proposes the method to accelerate similarity images retrieval in airport video monitor system via distributed indexing schema D-MVP (Distributed MVP-Tree), which divides the metric space into several pieces. Experiments on large high-dimensional metric spaces data show that D-MVP indexing model has more significant performance improvement compared with the traditional indexing structure and is highly scalable with the explosive growth of data.

In this paper, we present a new solution, BaitAlarm, to detect phishing attack using features that are hard to evade. The intuition of our approach is that phishing pages need to preserve the visual appearance the target pages. We present an algorithm to quantify the suspicious ratings of web pages based on similarity of visual appearance between the web pages. Since CSS is the standard technique to specify page layout, our solution uses the CSS as the basis for detecting visual similarities among web pages. We prototyped our approach as a Google Chrome extension and used it to rate the suspiciousness of web pages. The prototype shows the correctness and accuracy of our approach with a relatively low performance overhead.

In distribution of data encrypted to multiple users, there is a problem that the system will become low efficient if a single key center has to generate keys for a large number of users. Besides, it is a risk that some users could deliberately disclose their keys for some benefits. We in this paper give a key leakage discovering scheme where users are partitioned into groups and groups are hierarchically organized. In our scheme, users in upper-level groups can delegate keys for users in lower-level groups, which alleviates the key generation burden of the trusted third party. As an interesting feature, our scheme provide a key leakage discovering measure that if some users in groups leaked their decryption keys then at least one of them can be found out. This enables the data owners to accuse the illegal users when they infringed the copyright. At last, we analyze the performance of our system.

Behavior models of applications are widely used for diagnosing security incidents in complex web-based systems. However, Ajax techniques that enable better web experiences also make it fairly challenging to model Ajax application behaviors in the complex browser environment. In Ajax applications, server-side states are no longer synchronous with the views to end users at the client side. Therefore, to model the behaviors of Ajax applications, it is indispensable to incorporate client-side application states into the behavior models, as being explored by prior work. Unfortunately, how to leverage behavior models to perform security diagnosis in Ajax applications has yet been thoroughly examined. Existing models extracted from Ajax application behaviors are insufficient in a security context. In this paper, we propose a new behavior model for diagnosing attacks in Ajax applications, which abstracts both client-side state transitions as well as their communications to external servers. Our model articulates different states with the browser events or user actions that trigger state transitions. With a prototype implementation, we demonstrate that the proposed model is effective in attack diagnosis for real-world Ajax applications.

Recent research revealed that wind farms may produce obvious interference to primary surveillance radars (PSR). The echoes simulation of wind farm for the PSR is important for research about the evaluation and mitigation to wind farm interference. A novel wind farm echoes signal simulation method for PSR considering multipath effects was presented in the paper. First of all, the solution was implemented by modeling the wind turbine blades as a set of multiple slender cylinders and dividing the cylinder into a series of thin slices along the axial direction. Secondly, the blades echoes were equivalent to the coherent sum of a series of thin slices echoes. Thirdly, two basic models of reflection path echoes were derived by modeling distance functions of reflected path based on the geometry of reflector, wind turbine and radar. The experimental results indicated that, the simulated data from wind turbine is consistent with common radar cross-sections (RCS) variability and Doppler features of the measured data and the simulation of multipath echoes are in good agreement with theoretical analysis.

Distributed storage is a scheme to store data in networked storage services, which is the basis of popular cloud storage solutions. Although this scheme has huge benefits in reducing maintenance and operation cost, it has several security concerns. Among them, malicious file deletion by the storage providers is a top concern. In this paper, we develop a novel error-tolerant solution, ET-DME, to effectively detect malicious file deletion behaviors in distributed storage services. Our approach prevents malicious servers from forging evidence to bypass data auditing test. In addition, our approach does not limit the number of challenges made by the client. Our approach also has low computation and communication overhead.

This paper presents the first version of the NIST Cloud Computing Reference Architecture (RA). This is a vendor neutral conceptual model that concentrates on the role and interactions of the identified actors in the cloud computing sphere. Five primary actors were identified - Cloud Service Consumer, Cloud Service Provider, Cloud Broker, Cloud Auditor and Cloud Carrier. Their roles and activities are discussed in this report. A primary goal for generating this model was to give the United States Government (USG) a method for understanding and communicating the components of a cloud computing system for Federal IT executives, Program Managers and IT procurement officials.

The method was established for determination of PCBs in sediments by accelerated solvent extraction, Florisil column purification and selected ion GC/MS. It was good linear relationship between PCBs concentrations and instrument responses, high sensitivity, and the recovery from 84. 3% to 103.4%, RSD from 9. 5% to 12. 2%. The test result of standard reference soil sample was correct.

Supply chain management is a strategy that recognizes the cost and service improvements to be gained by viewing the supply chain as a whole process from initial supplier to final customer. A supply chain system based on dynamic alliance enterprise is proposed in this paper. Firstly, the basic activities, unit model, solid model and system model are analyzed. Then, the system model is introduced and an expansive organization structure model is developed. Stock control is also studied for minimization the cost. Finally, the integration supply chain management implementation model is presented. The proposed method provides a supply chain system model based on dynamic alliance enterprise so as to rapidly respond the requirements of the market.