- EI核心期刊
Disaster backup and recovery is an important approach for system security and reliability, which can be used as an important support for information assurance in the development policy of China informationization. The measurements used for assessing the performance of disaster-tolerant systems are introduced firstly. The key technologies and realization methods of data-level disaster tolerance, system-level disaster tolerance, and application-level disaster tolerance are then summarized. Three future directions of disaster backup and recovery techniques are presented, which is the service-model shift from the data-storage-center to application-service-center, the architecture shift from a centralized structure to a shared mode based on the combination of distribution and virtualization, and the application mode shift from an isolate and specified system to a comprehensive service system, respectively.
Based on a planar parallel mechanism, a forward kinematics model is built and solved by geometric algebra （conformal geometric algebra）. Six position and pose result are eventually obtained through the establishment the Maple program. Finally, the effectiveness and accuracy of this method are verified by using Wu method. And this does good for finding a unified mathematical language in mechanism analysis, and for reducing the nonlinearity and enhancing the geometric intuition.
An efficient identity-based threshold signcryption scheme was presented. With bilinear pairings technology, a proved secure threshold signcryption scheme was designed in the standard model, based on Waters and Paterson’s signature scheme. We prove its semantic security on the hardness of decisional bilinear Diffie-Hellman problem and its unforgeability on the hardness of computational Diffie-Hellman problem. Compared with other schemes in the random oracle model, this scheme is more meaningful.
localization scheme using mobile bacon nodes is proposed. One or several mobile bacon nodes traverse the whole physical space of the sensor area by straight line or random walking mode. Mobile beacon nodes periodically broadcast their physical location information which can be received by unknown nodes. The locations of unknown nodes can be estimated from bound lines which are derived from the broadcast information. Extensive simulations show that the proposed scheme achieves significantly better performance than conventional typical algorithms in terms of computation cost, localization accuracy and convergence.
Based on the statistical learning theory (SLT), the margin scale reflects the generalization capability to a great extent. Inspired by oneclass support vector machine (SVM), doublemargin SVM is put forward to classify two classes by two margins separately. Instances can be classified correctly as well as margin maximization, and its superiority is theoretical proved by both generalization performance and imbalanced class distribution. Experiment on benchmark data sets shows that classification margin obtained by doublemargin SVM is larger than SVM, improving the generalization apparently, and analysis on imbalanced data shows that it has a higher recognition ratio. Finally real intrusion detection data shows that the detection precision is increased by 2% against boundary samples selection method.
The ecological relationships between creative enterprises including competition, mutualism, predatorprey and so on are explored based on niche theory from the perspective of the ecological organization. According to the characteristics of creative enterprises, a classical Logistic model is extended, a creation factor is introduced in the model. Although the model only considers the ecological relationships between two enterprises, it can be extended to the situation of a number of enterprises. In addition, the model also can be used to research the ecological relationships between creative enterprise clusters.
Considering that the channel state information is often imperfect due to the noise and interference, a robust joint linear transceiver design scheme is proposed based on minimum mean square error criterion for a nonregenerative multipleinput multipleoutput（MIMO） relay system. Assuming that the channel estimation errors are independent complex Gaussian with zero mean and known secondorder statistics, the processing matrices at the source, relay and destination nodes could be derived through an iterative algorithm under average power constraint at both the source and the relay nodes. Simulations show that the joint design scheme is more robust and it can achieve a better bit error rate performance compared with the conventional zero forcing and minimum mean square error transceiver design schemes.
A dynamic multidimensional evaluation system was developed in order to give a better solution on the management of telecom service providers （SP） for telecom operators. Depending on the multidimensional dynamic credit evaluation and confidence interval significant factor identification, this model introduced credit as an important dimension, summarized several factors and distinguished significant factors in the evaluation of SP. This model will help telecom operators make better decisions to manage and choose SP.
One type of orthogonal frequency division multiplexing (OFDM)-N minimumshift keying (NMSK)/N quadrature MSK (NQMSK) system based on discrete cosine transform (DCT) is proposed. By combining the NMSK modulation scheme with the OFDM technology based on DCT, although the frequency efficiency is reduced to some extend, it can improve the outofband attenuation greatly and the performance deterioration caused by carrier frequency offset (CFO) will be mitigated. In the presence of CFO, analysis and simulation results show the excellent performance of the OFDMNMSK/NQMSK system based on DCT over both additive white Gaussian noise (AWGN) and Rayleigh fading channels.
For formally analyzing untraceability of tags in radio frequency identification protocols, the strand space model is extended so that it can be applied to analyze untraceability. The theorems about untraceability are also given. Based on the extended strand space model, Feldhofer protocol and O′FRAP protocol are analyzed. It shows that Feldhofer protocol is untraceable. On the contrary, O′FRAP protocol is traceability for active attackers.
Performance of orthogonal spacetime block codes (OSTBC) in multipleinput multipleoutput (MIMO) relay networks under fading channels is analyzed. Closed form symbol error probability and outage probability expression for OSTBC in MIMO relay network by deriving and applying some new closedform expressions to the statistics of the harmonic mean of two independent, nonidentical Chisquare distributed random variables are presented and validated by simulations. Furthermore, using the Lagrangian method, the optimal power allocation aiming at minimizing the outage probability is derived. Analysis shows that by adopting OSTBC, MIMO relay network can achieve full diversity gain. And the proposed power allocation scheme can further improve the system performance.
An analysis method of gene expression data based on immnue clustering algorithm is presented. A modified consine coefficient is put forward to measure comparability of genes in accordance with the characteristic of gene expression data matrix. Inspired by the biology immune system，a new clustering algorithm based on immunodominance cloning(ICCA) is designed. In comparison with Kmeans algorithm and genetic Kmeans algorithm, the proposed ICCA given can achieve good class compactness and separability.
Blind source separation is one of the hot spots of signal processing domain. Considering that mixed images cant be separated with sparse component analysis (SCA) model for image doesnt always satisfy sparse conditions, a method based on wavelet transformation and SCA (WTSCA) is proposed to extract source images. WT is used to transform mixed images to frequency domain. SCA is used to estimate mixing matrix, and to reconstruct source images. The experiments manifest that WTSCA can accurately and effectively extract sources from mixed images. The visual result and correlation coefficient analysis verify that, compared to classical FASTICA, separating precision of WTSCA is higher, and separating effect is better.
In mobile Ad hoc networks (MANET), a multipath load balancing mechanism was proposed in order to distribute the traffic evenly into the network and alleviate partial congestion. Load balancing method was realized at the source node after detecting the current flows in each node by using Bloom filter. The estimated link lifetime, the endtoend delay and the existing traffic along the forwarding paths were used together as the routing metric of multipath load balancing mechanism. The proposed multipath load balancing mechanism was evaluated and analyzed through simulations, which showed that the performance was better than the traditional multipath routing mechanism, such as the packet delivery ratio, the endtoend delay and the control traffic; the results also showed that multipath load balancing mechanism suited for the scenario with small node density.
A power allocation algorithm is proposed for multicell orthogonal frequency division multiple access (OFDMA) systems. The optimization objective is to minimize the system transmission power under the constraint of the users requirements on data rate. The power allocation on cofrequency subchannel of multiple cells is solved by the combination of the genetic algorithm and the tabu search algorithm. A new fitness function is defined and the process of this algorithm is given. Simulation shows that, compared with other algorithms, the proposed algorithm can reduce the system transmission power while guaranteeing the data rate requirement of all users, which improves the energy efficiency.
To analyze the performance of cooperative automatic repeat request (CARQ) network, which adopts equal gain combining over Nakagamim channels, an approximation strategy is utilized to derive frame error probability at high signaltonoise ratio. It is applied for the networks with arbitrary relay number. Simulations verify the reliability of the theoretical analysis and the superiority of CARQ when compared with traditional automatic repeat request （ARQ）.
For odd n, the crosscorrelation function between a decimated DillonDobbertin (DD) sequence and WelchGong (WG) sequence is investigated, and the crosscorrelation function between a pair of Hyperoval sequences is studied. It is shown that the calculation of the crosscorrelation functions between a decimated DD sequence and WG sequence, and a pair of Hyperoval sequences could be reduced to the calculation of the crosscorrelation between msequences and their decimations. The crosscrrelation function between a decimated DD sequence and WG sequence could be 3valued under a 2k+1decimation.
Aggregation reconfiguration assisted by the cooperation of terminals is a new paradigm for heterogeneous networks convergence. The marketbased incentive pay mechanism is investigated, and dynamic pricing is used to promote global optimization on resource allocation autonomously. With the consideration that the pricing decision will confront incomplete information, an adaptivecontrolbased pricing solution based on artificial immune system is proposed. Experiment shows that the artificial immune system model based automatic pricing (AISMAP) algorithm can improve resource utilization of whole system and autonomously guide the traffic resources allocation toward global balance.
An Ethernet passive optical network (EPON) architecture and peer to peer （P2P） redirecting mechanism for optimization P2P traffic transmitting are put forward. Its scheme is analyzed and verified. Simulations indicate that the scheme can effectively reduce the transferring distance of P2P protocol packet and alleviate the flow burden of core network.
To obtain peer to peer (P2P) flow information accurately in a realtime and lowcost way, NetFlow based P2P flows analysis system called netFlowbased P2P flows analysis system(NPFAS) is proposed, it loosely integrates systems for flow aggregation, flow identification and flow analysis respectively in a distributed manner. By utilizing the key information of P2P flows provided by P2P identification systems, NPFAS can extract P2P flow statistical information from NetFlow records generated by NetFlowenabled devices and analyze it by some existing analysis systems. NPFAS provides a feasible solution for analyzing P2P flows based on the LinuxIntel framework. Experiments of the prototype system have verified feasibility and availability of NPFAS.
In light of the congestion control system with timevarying parameters and nonlinear property, a neuron control algorithm with fuzzy selftuning gain (FNAQM) is proposed for active queue management. Both queue length and traffic rate are employed as congestion indicators which detect both current and incipient congestion states. Combining the advantages of neuron control and fuzzy control strategies, the endtoend mark probability is calculated by the neuron controller, in which the weights are adjusted online by supervisory Hebb learning rule. Additionally, fuzzy logic control is used to tune the gain of the neuron dynamically for improved network performance. The proposed scheme exhibits good adaptability and selflearning ability, being simple in form and easy to implement. Simulation in network simulator2(NS2) demonstrates that FNAQM can quickly stabilize the queue length to the target with small jitter, and shows strong robustness against dynamic traffics and nonresponsive flows.
The security and query effectiveness must be considered in the extensible markup language (XML) encryption. To address these problems, an approach based on curve interpolation is presented. Firstly, more flexible encryption granularity can be obtained through a XML nodehide mechanism; secondly, more efficiency to resist various kinds of database attack due to the changing of ciphertext distribution and data size. A XMLQuery (XQuery) is established. The proposed model offers greater security, and the ciphertext query performance can amount to 65% of the plaintext query based on the result of experimental simulation.
Based on heterogeneous Ad hoc network, study of unidirectional link, low overhead routing based on unidirectional link (LORUL) is designed, LORUL can improve the network connectivity by the detection and use of unidirectional link, it also decrease control overheard by special method. LORUL is evaluated in simulation, it is shown that the performance is better than the traditional routing mechanism on the control overheard, the decrease on the endtoend delay and the throughput is quite small. It is also shown that LORUL suits for the scenario with low node density.
A lightweight router using optimized methods is proposed to meet the requirements of realtime and reliable transmission in logging networks. In the router, the pointtopoint protocol is simplified for three states and five events in state machine, and the number of escape characters will be reduced to improve transmission efficiency. The memory of link frame is preallocated with the maximum header of protocol data unit to decrease movement of buffered data. A lowcomplexity method based on linear static routing table is designed to fast lookup. The forwarding engine driven by direct memory access controller in hardware timer can fast process Internet protocol packet. Experiment shows that the forwarding delay and jitter is low and the router is stable and efficient for logging networks.
A mdeia access control (MAC) protocol, utilizing the optimal idle licensed channel, is proposed for cognitive radio wireless networks. The protocol enables the secondary (unlicensed) users to select the optimal idle licensed channel with spectrum sensing of multiple licensed channels, and solves the problems of hidden terminals and exposed terminals. An analytic model is proposed to analyze the throughput of our proposed MAC protocol under the saturated network case. Simulations show that the model is very accurate in evaluating the saturation throughput of our proposed MAC protocol.
A general model of central resource distribution is proposed by studying the designs of blockingresistant anonymity system and considering the problem of access point information distribution. Based on this model, analysis of its specialized model with randomchoice strategy is given, which could be used for evaluating the antiattack ability of such systems and how much it costs to counter for the attacker. Simulation shows that the practical blockingresistant systems could select the appropriate amount of access points for each user and have a large number of active access points to improve the ability, and the censor could make appropriate investment according to the amount of access points to counter.
Network resource coordinated scheduling strategy supporting multichannel transmission is presented to implement highperformance optical network with high bandwidth and flexibility for distributed services under the researching of optical grid architecture. The divisible load theory is introduced into the strategy to divide data or bigbulk file. The parallel multiple wavelength reservation is combined with the applicationdriven mode. An optical grid testbed is designed to verify the feasibility of the coordinated scheduling strategy. The performance of multichannel transmission is analyzed through an example of the distributed computing about electric power flow. Experiment shows that the strategy can provide flexible resource allocation with optimizing the transmitting performance of distributed service and using efficiency of optical network resource.