- EI核心期刊
To decrease the incomplete information between agents in MAS(multi-agent system)based e-commerce, accelerate the process of negotiation, and enhance the efficiency of negotiation, a classified learning mechanism is provided in bilateral multi-issue negotiation to learn some incomplete information especially the incomplete information about agent opponent’s negotiation issue weight. The opponent’s issue weight is learned by assigning different concession issue to different classified weight. In the experiment, two kinds of negotiation agents are used to compare, one is the agent with no learning, and the other is the agent with classified learning mechanism. The experimental results show that the agent with learning mechanism can negotiate more efficiently than that of no learning agent. The agent with classified learning mechanism can enhance the efficiency of negotiation in agent based ecommerce.
Traditional fuzzy comprehensive evaluation model must concentrate all the resources for calculation. A secure multi-party fuzzy comprehensive evaluation model (SMFCE) with basic properties was proposed for multiple users carrying out comprehensive evaluation in distributed environment. The proposed model allows users sharing data at different places to implement the evaluation, without leaking secret data of their own. A scheme under semi-honest model was proposed based on the theories of secure multi-party computation, which focuses on a common scenario, that is to say, many participants share fuzzy evaluation matrix by rows. Finally, the correctness and security of the scheme are proved.
An analytical model for power allocation based on multiple cooperative relays and multiple antennas at the destination in cooperative communication is presented. An optimal power allocation scheme (OPA), through singular value decomposition (SVD) of the channel matrix, is developed among the relay nodes under a certain power constraint to achieve the maximum channel capacity. The expression of the optimal power allocation coefficient is derived using Lagrange multiplier. Compared with the equal power allocation (EPA) scheme, the OPA can obtain higher capacity gain and outage probability gain.
A novel method to construct regular or near-regular Low-density parity-check codes from prime fields is proposed. The low-density parity-check codes constructed by the proposed method have a good structure. Compared the simulation results with those of the widely researched array codes as LDPC codes, it is shown that these Low-density parity-check codes perform well with iterative decoding.
Considering a class of networked control systems (NCS) with generalized uncertainty and nonlinearities, a control strategy is presented, which is based on robust adaptive control and remote feedback control. Firstly, a robust function η1 is introduced to compensate the generalized uncertainties and nonlinearities. The adaptive tuning rule of parameter is designed by Lyapunov stability theory and the uniformly bounded stability and exponential convergence of the closed-loop error system can be obtained. Then the state feedback control is utilized to the NCS with time-delay, and the stability of the closed-loop NCS is effectively guaranteed by Lyapunov stability theory. Finally, The simulation results show that this method is very effective.
In recent years, the coding-aware routing mechanism attracts the interest of many people to further magnify the gains achieved by network coding in wireless networks. In general, there are two steps that should be realized in coding-aware routing communication: one is finding coding opportunity in the routing discovery phase and the other is determining the optimum path from the coding-available paths and the coding- unavailable ones. The first step has been well developed while in the case of the second one, current simplification like the way of always using the coding-available path, or so called the “free ride” assumption is rather than reasonable operation. In this paper, we propose a routing metrics ECTX (Expected Coding-aware Transmission Count) which fairly quantify the network coding gain while considering the wireless transmission conditions. We implement the ECTX routing in wireless mesh system in C++ and conduct extensive evaluation, and the simulation results illustrate that compared with the ones which oblivious of coding gains using ,transmission efficiency can be sharply increased in the whole network and achieves 10% throughput gains.
A user rejection criterion was proposed for the ZFBF based greedy multi-user selection algorithm in MIMO downlink system. As compared to the original algorithm, the improved one which using the user rejection criterion has nearly no performance loss and reduces the number of candidate user set in every selection step significantly. Simulation results show that the user rejection criterion can lower down the computation complexity more than 50% when the signal to noise ratio is relatively small.
In computational geometry, determination of circle-circle and line-circle relationship are important problems. If the information of circle or line is supplied by different users and none of them would like to disclose his or her data to others, it is called the secure two-party circular computation problem. In this paper, a secure square of two numbers sum protocol, a secure two numbers relation determination protocol, a secure points distance computational protocol and a secure point-line distance computational protocol are proposed to solve the he circularity-circularity relation determination problem and line-circularity relation determination problem; the correctness, security and efficiency of these protocols are also analyzed
The concept of interval-set is introduced firstly, and then the interval compu瑃ations on Boolean and reference types are given. Finally interval computations are applied in program static analysis, and an algorithm of Range Analysis Based on Interval Computation (RABIC) is presented. The advantage of this method was verified in experiments. On the one hand, it can detect infeasible paths in program, On the other hand, it can improve the testing efficiency of defects testing tool based on this method.
A new Chain Verification signature scheme was proposed based on Elliptic Curve Cryptosystem, Which enables verifier Pn to verify the validity of the signature only after being authorized by members P1、P2、...、Pn-1 of the chain verification group. The security and effectiveness of the proposed scheme was analyzed. The proposed scheme not only possesses the properties of Chain Verification Signature, but also contains the following properties: chain authorized verifiers can be easily added or deleted and the sub-secrets can be protected in time.
The inverse kinematics for trinal-branch space robotic manipulator is researched based on the screw theory POE equation and space Jacobian matrix, a general iterative algorithm on inverse kinematics is proposed and its unified formulation is established. If each handling branch features a spherical group of joints at the wrist, a simplified inverse kinematics of the manipulator can be obtained with the position and orientation decomposition. The proposed algorithm is of the higher computation accuracy than the velocity control method, and it is not relied on the configuation of manipulator which is applicable to modular robot. Simulation of carrying a single object is analyzed, therefore, the validity of the proposed algorithm is certificated.
To solve the problem of long acquisition time of routine PN code acquisition algorithm, a PN code iterative acquisition algorithm based on Tanner graph is proposed. The Tanner graph show of PN code and the sum-product algorithm process on the graph are presented. The performance of iterative acquisition algorithm is analyzed by simulation and the sum-product algorithm is modified. The simulation results show that when the signal-noise ratio is not very low, the bit-error rate of the APP estimation will convergence to zero.
A new method to avoid the singularity is presented for the 6-Dof robot with the structure of three consecutive parallel axes. By analyzing the robot Jacobian matrix structure and rebuilding the Jacobian matrix, the six-dimensional Jacobian matrix is divided into two three-dimensional Jacobian matrixes. On this basis, the conditions of robot singularity occurrence are determined and the robot avoiding the singularity is achieved by using quadratic programming. Finally, the algorithm effectiveness is validated by the experiment. The comparison between the quadratic programming method and the existing Damped least-squares method is manifested and the former is superior.
A Fuzzy Comprehensive Evaluation based loss differentiation algorithm for wired/wireless hybrid network is brought out, which takes both average PLR（Packet Loss Rate）and the probe packet ROD（Relative One-way Delay）in loss pair as loss differentiation parameters ，arranges them dynamic weight values in different network environments, and then distinguishes the wireless error loss from the congestion loss according to the fuzzy comprehensive evaluation principle. Unlike most of the current loss discrimination algorithms, the proposed algorithm does not take one loss differentiation parameter and simply set its threshold of one loss pattern, but takes different parameters and arrange them dynamic weight values in different network conditions, and then uses fuzzy comprehensive evaluation model for distinction. Simulation demonstrated that the algorithm has better network adaptability.
Conventional equalizers could hardly work effectively in the severe fading channel with zeros on the unit circle. A novel combing technique is presented in this paper aiming at the equalization of such nonminimum phase channels. The received signal is equalized separately by the IIR equalizer and time-reversal (TR) IIR equalizer with decision feedback (DF) architecture, and then the output signals from the two equalizers are combined together before demodulation. Since IIR and TR-IIR equalizers deal with the same symbol under independent noise background, the signal-to-noise ratio (SNR) can be increased evidently after combining, and the error propagation effect in the DF-IIR equalizer can be effectively depressed. It is evaluated through computer simulation to verify the advantages of the new scheme which can surpass the perfect performance of filter equalization methods.
The anonymity and dynamics of the P2P (peer to peer) network bring many security problem, The traditional trust manage model can not satisfy the P2P environment commendably. This paper proposes a recommendation-based trust model in P2P network, the model distinguishes the ability of peer between providing service and offering recommend, uses reputation value to express the ability of offering recommend, the peer’s trust value is assessed according to direct trust value, other peer’s recommendation trust value and the corresponding penalty value, then the peer decide which to choose. The simulation experiment results show that the recommendation-based trust model is able to assess the peer’s trust value more effectively, surpass the traditional trust management model in the transaction success ratio.
Chinese Wall model can express the dynamic security policy which keeps changing at runtime, but it can only applied in a limited area. To make the model can express the dynamic policy and applied to a extensive area as well, based on an open synthetical security model(OSSM), an implement method of Chinese Wall model is proposed. In this method, a dynamic accumulative role(DAR) is built to record the necessary historical access, and then, based on both the security policy and historical access, the response to the current access request is made to meet the dynamic of the policy. The result shows that this method improves the flexibility of security model and keeps its integration as well.
The power control method for Cognitive Radio Networks (CRN) is investigated in this paper, which is based on the Nash bargaining cooperative game theory. A utility function based on (Signal-to-Interference plus Noise) SINR is proposed for this model. Based on the Nash Bargaining Solution (NBS) from the cooperative game theory, the power control method not only provides the power level of users that are Pareto optimal from the point of view of the whole system, but also are consistent with the fairness axioms of game theory. Meanwhile, we prove the NBS is essentially a general form of the proportional fairness. Under the constraints from the interference temperature of the primary user and the maximum transmit power of the cognitive users, and according to NBS Nash theorem, we convert the power control problem into the optimization problem with multiple constraints. Through introducing the Lagrange operators, the optimal power levels are obtained, and achieving the SINR threshold requirements are met. The simulation results show that our algorithm has a faster convergence rate, and compared to reference traditional method, our algorithm can effectively improve the understanding between users, the fairness of the cognitive users and the overall performance of the whole system.
To investigate the performance of network coding for distributed storage, a model of that in wireless sensor network (WSN) is brought up. An expression among the data nodes, storage nodes and query nodes is set up and proved via bipartite graph theory, which is directive in wireless sensor network with limited energy efficiency.
In 3G evolution systems, it’s an important inter-cell interference mitigation scheme for inter-cell interference coordination, which includes various frequency reuse methods. At the same time, the concept of available frequency sets is proposed for frequency reuse, which not only enables to mitigate inter-cell interference, but also enables to improve frequency utilization. On this basis, how to allocate the sub-carriers in available frequency sets is a critical problem, which directly relates with the strength of interference. But on the other hand, the existing methods are usually on the state of current channel, which easily causes extra interference due to time delay. In this situation, an adaptive subcarriers allocation method is proposed, which is based on the forecast by Kalman filter. In this method, the gain interference ratio (GIR) among users and available subcarriers are got by measurement. Then the GIR in the next time is forecasted by Kalman filter, and the subcarriers are allocated according to such forecasted GIR. The simulation results show that compared with fixed subcarriers allocation and subcarriers allocation based on current GIR, the proposed method can effectively reduce inter-cell interference under different frequency reuse methods, and improve the performance. Specially, the throughput is raised, and the blocking rate and the bit loss rate in cell-edge are also reduced too.
Based on the framework of “Multi-bank matched filter + FFT”, a new acquisition algorithm is proposed and implemented, which is based on time and frequency domain parallel search and applicable to P code direct acquisition in GPS (Global Positioning System) high dynamic situation. Its novelty is multi-bank matched filter’s implementation in the structure of “folded filters + re-order RAM” and its FPGA implementation in parallel. The pipelined implementation shows that the new algorithm occupies about 1/16 of the old one. Simulation results and field tests show the algorithm improves P code direct acquisition performance in GPS high dynamic situation.
In order to increase computing speed, MOM formulations including closed-form spatial domain Green’s functions are derived for geometry of microstrip. Based on loss substrate, radiation emission and interference suppression for signal lines are investigated. The results show that radiated power from a signal line with a shielding line is much less than that from a signal line without a shielding line. Therefore, in the EMC design of digital circuits, adding a shielding line to a signal line can effectively decrease spurious radiation.
A new uplink scheduling algorithm is proposed for voice over IP (VoIP) services, which is based on the adaptive multi-rate speech codec in IEEE 802.16 systems. This algorithm sends the special Code-Division Multiple Access (CDMA) code in the channel quality indication (CQI) channels to request bandwidth during the silence period of users. Then the performance of the proposed strategy in terms of system capacity, throughput, delay and signaling overhead of system is analyzed. And simulation experiments are implemented. Through the theoretical analysis and simulation results, the new algorithm shows that it not only increases the system capacity and throughput but also reduces the average access delay compared with the conventional algorithms.
A single sign-on scheme is proposed supporting C/S applications and B/S applications. Authentication and authorization based on the PKI and PMI, the service is implemented by means of middleware to achieve the functions of authentication, authorization and auditing, and SAML to support the exchange of the authentication and authorization information; secure Cookies, shared memory and the technique of tickets are used in the client to achieve the multi-mode and cross-domain SSO solution. The scheme has a higher security, more comprehensive solution to the multi-mode SSO and therefore has a broad prospect.
This paper introduces our way of finding semantic related Chinese word pairs from the open encyclopedia Wikipedia and analyzing the degree of semantic relations. Almost 50,000 structured documents are collected from Wikipedia pages. Then considering of hyperlinks and text overlaps etc., about 400,000 semantic related pairs are employed. We roughly measured the semantic relatedness using the position and frequency information in the documents. With comparing experiment on data sets with different degrees of semantic relations using some other classic algorithms, we analyze the reliability of our measures and other properties. Key words: Wikipedia; semantic relation; semantic relatedness.
Classic rough set approach acquires knowledge by means of indiscernibility relation, but it is powerless to resolve the sequencing classification problems which contain qualitative and quantitative attributes as well as criterias. In this case, a decision analysis method based on extension of rough set theory is proposed .This method replaces indiscernibility relation in original rough set theory with the “indiscernibility-similarity-dominance” relation and obtains rough approximation of knowledge, it will not only be able to resolve the aboved-mentioned problems but also deal with inconsistence in decision table. Finally, the effectiveness and superiority of the method are demonstrated by a real example.
A dynamic adaptive route flap suppression mechanism based on OSPF network was proposed. This algorithm tuned the hello timer locally according the changes of network state and eliminated route flap which was caused by link flap. SSFNET establishes an analytical model to investigate that our approach not only achieves better performance and consumes less resources than network-wide methods, but also improves network stability and reliability.
Blind signatures play a central role in applications such as e-cash and e-voting systems, while the notion of partially blind signature is a more applicable variant. This paper presents a model of simple partially blind signature with low computation, from which many simple partially blind signature schemes can be constructed, and gives a detail formal security proof in the random oracle model. Compared with the scheme due to Masayuki Abe and Tatsuaki Okamoto, our scheme is more efficient. The low-computation property makes our scheme very attractive for mobile client and smart-card implementation in many e-commerce applications.
Self-Electroopic Effect Device’s (SEED’s) basic structure and operation principles are proposed, then equivalent Resistance-Capacitance (RC) model are applied for analysis. Theoretical analysis and mathematic computation are utilized to obtain SEED’s Voltage-Time (V-T) characteristic expressions. Theoretical expressions are verified by simulations, and the simulations compare SEED’s switch characteristics with different radiuses of input control beams, and demonstrate that SEED’s switch characteristic can be effectively improved with small input radius. Moreover, the simulations compare SEED’s Voltage-radius (V-r) diffusion characteristics with a fixed time and different input radiuses, then the results reveal that SEED can only keep its excellent switch performance near the input light beam’s center.