Earticle

With the rapid and bursting development of computer science and related technology, the design and optimization of network topology has been a most important issue for thee researchers to solve. However, the increasing needs of the current control schemes have already been shown. In this paper, we theoretically analysis the features of neural network structure and convex optimization related method. There are a lot of shortcomings but the cable network should be restricted the wiring in some situations, for example, wiring, reroute workload big, easy to damage, in the network, each node cannot move and so on. With the mathematical optimization, game theory, mathematical economics and optimal control theory to the development of disciplines, convex analysis are increasingly brought to the attention of the people. Our proposed methodology combines the neural network and optimization theory to obtain better performance, the experimental result shows the performance of our approach compared with other state-of-the-art algorithms.

In the setting of (m,n) rational secret sharing, n rational players wish to share a secret s , arbitrarily m players can reconstruct it, they always choose the strategies which can bring them more utilities. The security requirement includes privacy, correctness and fairness. Fairness is a central objective of the rational secret sharing scheme, complete fair means either all players get the secret s or none of them get it, but most existing schemes do not meet this nature. In this work, a rational secret sharing protocol is proposed, which neither need particular communication channel, nor assume the existence of honest minority, in addition, the scheme can resist the collusion attack with k(k  m) players, and can achieve sequential equilibrium. Theoretical analysis shows that the proposed protocol is complete fair.

Attribute-Based Encryption provides an encrypted access control mechanism. A user applies and obtains a private key that is associated with various attributes of the applicant. A secret message is encrypted with an access control policy tree representing a logical combination of different attributes. Only the users whose attributes satisfy the policy tree can decrypt the message. There are many scenarios that require dynamic attributes. Whenever a dynamic attribute updates, a new key is regenerated and issued even if there are hundreds of attributes involved in the key. This is not efficient since the cost of generating new private key is proportional to the number of attributes associated with that private key. To resolve the efficiency problem, this paper introduces the concept of Dynamic Attribute Based Encryption, a key revocation mechanism, which is necessary to prevent a user from keeping a private key with expired attributes. We show that the new concept improves the efficiency of the encryption scheme and helps us develop a new trust management method.

Network Intrusion Detection System (NIDS) is an important network security system. There always appears high packet loss rate in NIDS, especially when the network traffic is high. We raised Multimedia Packets Processing Method to reduce the packet loss rate and received good results. On this basis, Dynamic Programming is applied to optimize the decision-making steps of this method. This improvement can help the system to find an optimum solution to select the highest risk of multimedia data packet sequence in each time slice, moreover, the system load capacity can also be considered. In this way, the limited processing power of NIDS can be focused on the more dangerous multimedia data packets. A series of experimental results indicate this optimization can help the system to improve the detection rate of the high risk of multimedia information.

Recently, as Internet is widely used due to the increased spread of the internet network, the software with malicious intent is distributed via the internet and its infection path gets various too. In particular, attacks by Bot mainly work at C&C (command-and-control) server but it can be secured just by blocking IP because C&C server runs in form of IP. However, this attacker too gets gradually intelligent as they try to connect periphrastically in order to avoid server blocking. Once these malicious codes infiltrate user’s system, it is not easy to detect it through general detection method while it is running. In this paper, we propose malicious process detection system based on security agent in order to prevent damage caused by malicious code infection from spreading.

In order to improve the WSN, a novel WSN protocol based on the event-driven named PB-TDMA is proposed. One of the problems in the WSN protocol is that each node’s probability of data delivery is usually diverse at different time and environment. This paper puts forward a method to resolve such a problem. It makes two nodes with different probability share a time slot which could use to compress the frame length and enhance the utilization ratio of channel. In addition, through running statistics on the probability of each node’s data sending, it proposes a new allocation method of time slot to avoid data collision. On the basis of the above strategy, we can constantly update the delivery rate to achieve the goal of adaptive. Simulation results show that PB-TDMA protocol performs better for wireless sensor network with high-density deployment and moderate load.

The driver takes leading part in the complex model of "human-vehicle-road-environment". The driving behavior has the characteristics of randomness, impulsivity and autonomy etc. which make it difficult to study the driver's driving behavior directly. Monitoring vehicle attitude is an important element of proactive safety management of public transport vehicles. This paper constructs a data acquisition system by using an InvenSense's 6-Axis inertial measurement unit (IMU) as the center of this system, which can sense the vehicle attitude information to make up for the lack of driver senses, and then builds fuzzy synthetical evaluation model which is combined with ISO 2631-1:1997/Amd 1:2010 standard. Finally, prediction of the driver's driving behavior in transient term has been established with Elman neural network method.

In addition to addressing the scarcity of IP address space, Internet Protocol version 6 (IPv6) also addressed some of the shortcomings of Internet Protocol version 4(IPv4). These include neighbor discovery, address auto-configuration, and others. Many of this message exchange are done via the Internet Control Message Protocol (ICMP) and the use of this protocol in the IPv6 paradigm, i.e. ICMPv6 plays a bigger role compared to ICMPv4. One of the key process that is carried during neighbor discovery process is to check if the address generated already exists. This process is called the Duplicate Address Detection (DAD). Nevertheless, the design of this process has led to a severe security vulnerability allowing attackers to easily carry out Denial-of-Service (DoS) attack by causing every address generated to be a duplicate leading to new hosts unable to join the network. Various techniques and mechanisms have been introduced to address this vulnerability such as NDPMon, SeND, and SAVA. Nevertheless, these techniques are either not robust or have performance implications vis-à-vis with the DAD DoS detection and mitigation. In this paper, we put forward a novel framework that is able to detect, mitigate DoS attacks while being light-weight at the same time.

In recent years, with the rapid development of internet technology, enterprise business model has entered a new century. The emergence of network economy has forced retail enterprises to change the original marketing mode, global information resources can be shared instantly, which greatly improve the operational efficiency and reduce the cost. In this paper, we analyze the influence factors that impact on retail enterprises based on consumers online shopping behavior, survey data from small and medium enterprises. The result shows that macro environment factors, market competition, enterprise self factors will affect the enterprise's e-commerce performance, and significantly promote the enterprise's market strategy innovation. Therefore, it is an important way to improve the competitiveness of enterprises by carrying out the E-business market strategy.

Botnet have turned into the most serious security dangers on the present Internet framework. A botnet is most extensive and regularly happens in today's cyber-attacks, bringing about the serious risk of our system resources and association's properties. Botnets are accumulations of compromised computers (Bots) which are remotely regulated by its creator (BotMaster) under a typical Command-and-Control (C&C) framework. Botnets cannot just be implemented utilizing existing well-known applications and additionally developed by unknown or inventive applications. This makes the botnet detection a challenging issue. In this paper proposed an anomaly detection model based on genetic neural network system, which joined the significant global searching capability of genetic algorithm with the precise local searching element of back propagation feed forward neural networks to improve the initial weights of neural network.

The world has actually been attacked from other countries by means of cyberattacks against national infrastructure and cyber-terror weapons. In December, 2014, a hacker stole the nuclear power data including key nuclear power technology to infringe people’s assets and further increased damages and people’s uneasiness about cyber-terror attacks against national infrastructure. The National Cybersecurity Act in response to cybercrime and cyber-terrorism and cyberwarfare citizens, public agencies, and experts in the military to establish a comprehensive response system for participation. National Cybersecurity Council should respond to cyberattacks, defense, media psychological warfare. National Cybersecurity Council is mutually share relevant information and real-time analysis, information sharing and coordination support. National Cybersecurity Act aim to build a general national action system by private, public and military organizations, detect cyber-terror attacks in advance to block potential cyber-terror attacks and warfare early and integrate national capacity for fast action if they occur.

In order to solve the frame overhead to the influence of frame transmission performance, improve the throughput, lower hybrid scenario user acces delay, improve user (802.11 n) especially the user experience, this paper aimed at different rate of terminal equipment, designs the RAS algorithm, this algorithm through optimizing scheduling, in does not affect user delay at low speed, reduce the high-speed users send time delay.The algorithm is reasonable configuration of channel resources, make full use of the high speed channel rate, rate of various users (especially in high speed rate of users) to improve the user experience .In the Chariot on Atheros9341 chip using TCP service test 802.11 n AP topology network end-to-end delay and throughput, thus the proposed method was verified. The measured results showed that the proposed method can improve the AP client in a high-density environment effective machine capacity, and increase the throughput, reduce the delay.

Many key management schemes have been proposed in systematically deployed Wireless Sensor Networks (WSNs).The sensor nodes are equipped with inadequate battery power, low memory, limited computation and communication range. Energy efficient secure routing is major issue in wireless sensor networks. In this paper, the hybrid key management scheme is proposed by combining public key cryptography scheme with symmetric scheme. The symmetric keys are generated by using genetic algorithm. Initial input for the genetic algorithm is the seed key generated by Hyperelliptic Curve Cryptography (HECC).The proposed scheme addresses the energy efficiency, resilience against node capture attack, key refreshment between the cluster head and member nodes in the cluster. The simulation results show that the hybrid scheme is more robustness, efficient energy with reduced key size.

Process service is an important segment of cloud manufacturing ecosystem, in which virtualized process resources are furnished as consumable services. In this paper, a service-oriented approach cloud process service is proposed to achieve a collaborative processing environment for distributed participants. In order to implement the cloud process service, it’s important to identify the existing process resources and capabilities, then virtualize them as processing services. In the first place, basic features of process service are summarized, and a standardized business model and process knowledge base are built to describe concept, cloud service and relevant features. Besides, a seven layer structure of cloud platform is proposed with a virtualization methodology to describe process service, source provider and process service requests. In the end, a case is presented to demonstrate the ability of integrating cloud process services.

NFR determination of the software products is an important link in the requirement development. Considering the correlation and inter-dependence between FR and NFR of software, this paper puts forward a NFR system identification method of the software products based on the fuzzy-QFD model through constructing the FR-NFR correlation matrix of software. The model express correlation evaluation with the 7-level natural language, build the software FR-NFR correlation matrix, fuzzy the matrix vector by the trapezoidal fuzzy number, then calculate the relative importance and absolute importance of NFR, at last get the importance NFR by the NFR screening algorithm.

Cloud computing has emerged as a powerful and viable architecture to meet large-scale and complex computational needs of the organizations. It extends the IT capability by providing on-demand access to computer resources for dedicated use. Data security and privacy are the major concerns over the cloud from user perspective. Consequently, the organizations which deal with cloud computing should amicably address the key data security risks. In this paper, a generic scheme of user authentication and data confidentiality has been introduced. We introduce a tool that encrypts/decrypts the user data at application layer using public key infrastructure. The information security experts of both the sides i.e. cloud users and service providers must comprehend and address the risk and security issues in detail before actually benefiting high-end computing power offered by the cloud computing paradigm.

Data center network (DCN) as the core of the cloud platform receives a widespread concern attention. Recently, the main study of these new architectures is limited to load balancing or improvement of architectures while the security study of DCN is still in its preliminary, especially for the vulnerability study. In our previous researches, we mainly concern about the invulnerability study from the protection of important nodes and vita edges. Differ from the previous one, in this paper; we engaged in the invulnerability studies from the perspective of network attack. Taking typical instances of DCN for example, based on invulnerability measurement of the average shortest path distance, we analysis the invulnerability of DCN by choosing deliberate attack and random attack. Experimental results show that for most architectures of DCN, the deliberate attacks of degree get the same result with the deliberate attacks of betweenness, especially, the deliberate attacks of degree are often harmful than betweenness attacks when attack a few nodes. However, from the perspective of random attacks, random attacks of betweenness are more harmful than degree attacks. The main contribution can be summarized as follows. We firstly engaged in the invulnerability research of DCN from the perspective of network attack and then conclude the network invulnerability of DCN by amount of experiments. The result in this paper can be widely used for the protection of DCN.

This paper presents a comparative analysis of the performance of three estimation algorithms: Expectation Maximization (EM), Greedy EM Algorithm (GEM) and Figueiredo-Jain Algorithm (FJ) - based on the Gaussian mixture models (GMMs) for a Dynamic Biometrics Face Authentication Systems. An automated biometric systems for human identification measure a “signature” of the human body, compare the resulting characteristic to a database, and render an application dependent decision. A Dynamic Face From eNTERFACE 2005 Database is used and Simulation shows that finite mixture modal (GMM) is quite effective in modelling the genuine and impostor score densities. Hence, the still face information scheme based on dynamic biometrics face is robust and could be explored for identity authentication.

There are many key parts in the field of aerospace with aluminum alloy thin-walled parts, the problem of breaking the low efficiency in production that’s also followed. There will be a lot of deformation in processing because of its poor rigidity, poor process, which will lead to low processing efficiency. It is of great significance to improve the machining efficiency of the aluminum alloy thin-walled parts, and a series of analysis of the surface quality of aluminum alloy thin-walled parts there are carried out clamping, cutting tools, take the knife route so on, and the problems of low machining efficiency and poor surface quality are solved.

Research on word sense disambiguation (WSD) is of great importance in natural language processing. In this paper, a new method based on beam search algorithm for Chinese WSD is proposed. By mining potential knowledge between phrase and semantic category in a sentence, this approach can construct its semantic network. It searches an optimal semantic category sequence from a Chinese sentence's semantic network with beam search algorithm, so that correct meanings of ambiguous words can be found from the optimal sequence. Experiments show that a better WSD performance is gotten.

The lack of data security in public mobile telecommunication system increases the need for a strong data protection and security mechanisms. Cryptography is considered as one of the key elements that provides security for mobile systems. Signcryption algorithm, which is based on public key cryptography, provides both confidentiality and authenticity in one step. Moreover, hardware implementations of cryptographic algorithms provide scalable solutions to enhance the level of security of the existing hardware. In this paper, we present a case study on using a DSP board to construct a secure communication channel. We use the constructed DSP board as a hardware cryptosystem to increase the security of transmitted data using any mobile communication system. We evaluate the performance by computing the consumed time by encryption/decryption process, while implementing the Schnorr Signcryption scheme on a DSP. Furtheremore, we propose an enhanced model of applying multiple DSP using pipelines and parallelization technique to decrease the consumed time in the whole process.

The traditional PageRank algorithm can’t efficiently dispose large data Webpage scheduling problem. This paper proposes an accelerated algorithm named topK-Rank .It is based on PageRank on the MapReduce platform. Owing to this algorithm ,Top k nodes can be found efficiently for a given graph without sacrificing accuracy. It can iteratively estimate lower/upper bounds of PageRank scores, and construct subgraphs in each iteration by pruning unnecessary nodes and edges. Theoretical analysis shows that this method guarantees result exactness. Experiments show that it can find top k nodes much faster than the existing approaches.

As network techniques have become one of the most significant aspects of our daily lives, network security has been a major concern. One common application is network intrusion detection. From the perspective of data oriented consideration, intrusion detection can be formulated as a clustering task, which aims to differentiate normal and insecurity behaviors and categorize into several groups. In this paper, we employ ensemble clustering method to improve the generalization and robustness of basic clustering. Specifically, we employ fuzzy kernel C-means (FKCM) as basic clustering, which improves the fuzzy C-means (FCM) clustering by introducing kernels from the support vector machines (SVM) to optimize the features of sample data by mapping the sample pattern into a higher dimensional feature space. Then, we formulate the ensemble problem as the optimization of the mutual information among all clusterings and introduce Ant Colony Optimization (ACO) as the solution. Experiments prove the efficiency of our method.

Without the science of metrology, we lost the way of any research field. The science of metrology affects the science of network security in many sides. You have to measure in network security as you have to do in the other fields. You have to test and analyze and detect as in the other research areas. You must work hard to reach the standards level in network security as you make your best efforts to reach them in other science branches. From the other side, network security has a great effect on developing the science of metrology. In the present paper, many trends are introduced including the two viewpoints. The necessary requirements to research and development in network security and metrology are proposed. Finally, we declared how different metrology labs around the world are interested in this field.

Focusing on the file pollution problem exists in unstructured P2P file sharing systems, we propose a strategy named TrustIs by combining the trust mechanism and the immune strategy of complex network to resist the propagation of polluted files. With the help of peer reputation, trusted immune peers and backup peers can be screened out. Also, the files and indices can be identified based on their object reputation and index reputation so as to provide reference for immune pees to filter out the polluted files or indices. Meanwhile, an immune token ring network is set up for monitoring peers’ behavior and implementing the refreshment of the immune resources. The results of the simulations show that TrustIs can effectively resist the propagation of polluted files.

In this paper, a novel image encryption method which is based on the coupled map lattice (CML) and two-stage diffusion is proposed. The author employs the two-stage diffusion to process images. The plain image is expanded into two composed of selected four bit-planes and diffuse them at bit-level as first stage diffusion, then reconstruct them as the input of block diffusion, which is served as second stage diffusion. The chaotic coupled map lattice employed in this method generates pseudo-random sequences in block diffusion phase. The experiment results and analysis have proved the novel image encryption method is practical and effective for encryption applications.

This paper considers a scenario in which a sender who holds a k -bit to k -bit trapdoor permutation f wants to transmit a message x to a receiver who holds the inverse permutation f-1; with the condition that encryption should require just one computation of f , decryption should require just one computation of f , the length of the enciphered text should be precisely k and the length n of the text x that can be encrypted is close to k . Our scheme takes the encryption of x to be f (rx) , which is a simple probabilistic encoding of x . Assuming an ideal hash function and an arbitrary trapdoor permutation, we describe and prove secure a simple invertible enmesh scheme that is bit-optimal in that the length of the string x that can be encrypted by f (rx) is almost k . Our scheme achieves semantic security, which implies chosen-cipher text security and non-malleability. Keywords: Asymmetric encryption

In order to reduce the node position error of DV-Hop algorithm in wireless sensor network, the artificial bee colony algorithm is introduced to design the DV-Hop algorithm. A new ABCDV-Hop (Artificial Bee Colony DV-Hop) algorithm is proposed in this paper. Based on the traditional DV-Hop algorithm, by using the minimum hops of nodes and position information of anchor nodes, the average distance per hop is solved by artificial bee colony algorithm to make it more close to the actual value. The simulation results show that compared with the traditional DV-Hop algorithm, the improved algorithm can effectively reduce the positioning error without increasing the node hardware overhead.

A scheme of grouping key distribution based on polynomial protocol is proposed and experimentally demonstrated. The grouping idea is ingeniously used in the design of key distribution. The performance is experimental demonstrated by comparing with kinds of random key pre-distribution schemes. The results show that it can not only guarantee the key connected rate but also strengthen the security performance of the network, and this scheme is very suitable for static underwater acoustic communications network.

Today, Cloud computing is becoming a household technology. In cloud, a shared pool of computing resources can be accessed on demand through internet and web applications. Since outsourced data is in third party premises i.e. cloud, accountability of user data is paramount. To preserve privacy and security of user data in cloud, many cryptographic techniques have been proposed by many scientists. One among them is searchable encryption. Searchable encryption allows secure search over encrypted data. In our system, a noble approach has been made using the Elliptic Curve Cryptography (ECC), a cryptographic techniques to reduce the overall computation overhead. Dynamic B+ tree data structure is used to perform multi-keyword search over the encrypted data. To retrieve appropriate data files, ranking will be done based on relevance score. Finally, we compare the effectiveness and efficiency of our proposed scheme with our previous work on CRSA/B+ tree through extensive experimental evaluation using Microsoft azure platform.