20 April 2012, Volume 38 Issue 08
-
Select all|
-
Computer Engineering. 2012, 38(08): 1-3. https://doi.org/10.3969/j.issn.1000-3428.2012.08.001On the basis of constructing the fuzzy measurement cognitive bias, a novel independent quota fusion algorithm for the multi-sensor data is proposed. And the given sensor’s reliability can be obtained by using mean value and covariance value of the adjacent information, which can ensure the objectivity and sensitivity for the fusion weight assignment. Experimental result shows that compared with the mean and support degree based data fusion techniques, the more reasonable fusion weight assignment can improve the fusion precision further by the proposed method.
-
Computer Engineering. 2012, 38(08): 4-5. https://doi.org/10.3969/j.issn.1000-3428.2012.08.002For studying how shooting angle, shooting speed and shooting height affect humanoid basketball robot basket shooting hit rate, this paper puts forward a normal cloudy model to simulate the manufacture accuracy error of the basketball robot. Simulation experiments are done in Matlab, and the influence rules about humanoid basketball robot basket shooting hit rate are achieved. Results show that the influence factors about robot basket shooting hit rate from large to small are shooting angle, shooting height and shooting speed. If the entropy and super-entropy of these parameters become smaller, the shooting hit rate rises, conversely it decreases.
-
Computer Engineering. 2012, 38(08): 6-9. https://doi.org/10.3969/j.issn.1000-3428.2012.08.003In traditional numerical analysis methods, the 3D simulation of multitudinous cross structural planes’s occurrence and distribution is difficult, so a intelligent preprocessing method based on 3D combination element is proposed. The elements crossed by structural planes are pretreated as combination elements without dividing special elements, that simplifies the pretreatment work. The intelligent preprocessor is developed, which can auto-recognize the combination elements and structural plane influence zone elements, and obtain the detail information of each component. An underground engineering case show that it can improve the efficiency of Finite Element Method(FEM) simulation.
-
Computer Engineering. 2012, 38(08): 10-12. https://doi.org/10.3969/j.issn.1000-3428.2012.08.004This paper presents an improved Gaussian mixture background model to deal with dynamic surveillance scenes. Different threshold values are utilized in the updating and detection process to reduce the misclassification rate for single mode background. In the detection process, it fuses statistical difference method and time domain finite difference method to decrease the misclassification rate for multimodal background. Experimental results show that improved model can effectively solve the mistakes of complex dynamic background, and it has good detection performance.
-
Computer Engineering. 2012, 38(08): 13-15. https://doi.org/10.3969/j.issn.1000-3428.2012.08.005Aiming at optimizing walking parameters for quick and stable walking of humanoid robot based on trajectory planning method, this paper presents a walking parameters training algorithm based on reinforcement learning. By decreasing the number of walking parameters, the reinforcement learning is applied to optimize these parameters, and the reward and punishment mechanism is given. Experimental results show that the algorithm is feasible in the RoboCup3D simulation platform.
-
Computer Engineering. 2012, 38(08): 16-18. https://doi.org/10.3969/j.issn.1000-3428.2012.08.006Aiming at the problem of hand posture recognition from complex backgrounds, this paper proposes a hand posture recognition method based on depth image technoloy. The hand posture region is extracted from complex background via depth image. Appearance features are integrated to build the decision tree for hand posture recognition. Nine common postures with complex background are tested. Experimental results demonstrate that recognition rate is 98.4% and speed rate achieves 25 frames per second.
-
Computer Engineering. 2012, 38(08): 19-21. https://doi.org/10.3969/j.issn.1000-3428.2012.08.007Currently market trading mechanism in cloud environment is inflexible, and the price is not reasonable enough in some situation. Aiming at at this problem, this paper proposes a new dynamic resource pricing model based on combinatorial double auction, and gives the algorithm of cloud resource allocation and pricing. The users determine their price according to the amount of respond time, and the resource providers ask price considering the workload of the system. Simulation experimental results show that the proposed algorithm is better than the fixed pricing algorithm. The profit of user is increased by 18% and the profit of resource provider is increased by 9%.
-
Computer Engineering. 2012, 38(08): 22-24. https://doi.org/10.3969/j.issn.1000-3428.2012.08.008Aiming at the personalization problem of Web services composition, this paper introduces a Web services composition method based on Context-aware Process Network(CAPN). It supports high-level modeling of context-aware composition systems and low-level system imple- mentation. The formal semantics of context-aware composition model is described using Calculus of Communicating Systems(CCS) process algebra and Labeled Transition System(LTS). Implementation framework of context-aware system models is presented. Analysis results prove the feasibility of the composition method.
-
Computer Engineering. 2012, 38(08): 25-27. https://doi.org/10.3969/j.issn.1000-3428.2012.08.009In order to increase the efficiency of Extensible Markup Language(XML) document tree encoding and support the XML data dynamic update, this paper proposes an XML tree encoding scheme based on extended byte carry, named Extended Byte Carry Labeling(EBCL). Combining with prefix-based encoding scheme and additive operation, the EBCL uses segment and section to forming labels. The algorithm of dynamic prefix-based encoding and static prefix-based encoding are given. Experimental results on encoding size and dynamic/static performance show that EBCL is efficient, and can support dynamic update to XML document efficiently.
-
Computer Engineering. 2012, 38(08): 28-31. https://doi.org/10.3969/j.issn.1000-3428.2012.08.0010This paper analyzes the task features of ensemble prediction, based on the Web Services Resource Framework(WSRF), it designs a dynamically adaptive ensemble prediction task schedule system. The adaptive schedule algorithm is proposed, according to the dynamic changes of grid nodes, it assigns the optimal number of CPU for model prediction task dealing with initial samples. The schedule system can save computational costs, while shorten the time of prediction. The practice operation results demonstrate the feasibility of the system.
-
Computer Engineering. 2012, 38(08): 32-33. https://doi.org/10.3969/j.issn.1000-3428.2012.08.011This paper proposes a secondary font library restruction method of Liu style based on inscription of Xuan Mi Tower. Based on the inscription of Xuan Mi Tower which is written by Liu Gong-quan, it divides original characters which come from the inscription into some components. It defines the structures and district information coding of Chinese characters. With using Visual C++ language, it designs the interface program of Chinese characters inquiry, compose, entering database, then achieves the found of inscription of secondary font library. This method is achieved by software and that proves its feasibility.
-
Computer Engineering. 2012, 38(08): 34-36. https://doi.org/10.3969/j.issn.1000-3428.2012.08.012The paper analyzes the data conflict of military simulation scenario, and proposes a resolution of data conflict. It builds the scenario knowledge library which is a collection of facts and rules of military action presented by the production rule representation, then detects the data conflict of scenario by reasoning logically through the scenario knowledge library founded. Analysis results prove that this method can detect the data conflict efficiently and can be the reference of the simulation training and evaluation.
-
Computer Engineering. 2012, 38(08): 37-40. https://doi.org/10.3969/j.issn.1000-3428.2012.08.013The classical Term Frequency and Inverse Documentation Frequency(TFIDF) algorithm neglects the proportion of distribution of terms in categories and between categories of the text collection. Aiming at this problem, this paper introduces the information entropy, and the TFIDF algorithm based on information gain(TFIDFIG) is improved. It proposes a TFIDF algorithm based on information gain and information entropy (TFIDFIGE). Experimental results show that the TFIDFIGE algorithm is more effective than the traditional algorithm, namely TFIDF, TFIDFIG, in terms of precision and recall.
-
Computer Engineering. 2012, 38(08): 41-43. https://doi.org/10.3969/j.issn.1000-3428.2012.08.014In order to obtain user’s query intent more accurately, this paper proposes a method to infer target node. It considers frequency of Extensible Markup Language(XML) node in the node type, and the effect when the same word appears at different positions, gives different weights to them. It keeps the level factor of XML document tree. Experimental results verify that the method is more efficiently to infer user’s intention and improves the accuracy of the search.
-
Computer Engineering. 2012, 38(08): 44-46. https://doi.org/10.3969/j.issn.1000-3428.2012.08.015Traditionally energy saving can be attained by switch of components’ energy states, but the switch itself costs a lot of energy. This paper proposes an energy efficient data access technique based on readahead strategy. It extends the readahead technique to the energy saving field, reduces the power consumption by integrating data that need to be read to reduce the amount of switches among different energy states of computer components. Experimental results show that in the case of accessing multimedia data, this technique can reduce energy consumption of disk by over 70%.
-
Computer Engineering. 2012, 38(08): 47-49. https://doi.org/10.3969/j.issn.1000-3428.2012.08.016The characteristics of L-shape tile and the distribution of a kind of dual optimal directed double-loop network which has both the least diameter and the least average distance in the group of is studied, based on Cartesian coordinates. The number of dual optimal double-loop networks DG(N;1, s) for any given N is caculated. The n-N dual optimal distribution char for N varied from 4 to 1 000 is simulated. Result of simulation indicates that the characteristic of n-N distribution is reposeful undulation.
-
Computer Engineering. 2012, 38(08): 50-53. https://doi.org/10.3969/j.issn.1000-3428.2012.08.017This paper presents an IPv6 lookup algorithm based on Hash and Content Addressable Memory(CAM) by analyzing the prefix length distribution of routing table and the growth trend of routing table entries. The prefixes whose length can be divided by 8 are stored in 8 Hash tables, and the remaining prefixes are stored into expanded Random Access Memory(RAM). Analysis result shows that the algorithm has high efficient storage utilization, searching rate and updating rate. It is easy to be scalded in hardware.
-
Computer Engineering. 2012, 38(08): 54-56. https://doi.org/10.3969/j.issn.1000-3428.2012.08.018To improve the Bit Error Rate(BER) performance of Multi Input Multi Output(MIMO) X channel, this paper proposes an Interference Alignment(IA) scheme. It transmits data symbols according to Alamouti scheme at senders, and changes the operation mode of reconfigurable antennas at receivers. The final equivalent channel matrix is orthogonal, thus interference can be eliminated through matrix operation, and low complexity Partial Interference Cancellation(PIC) group decoding can be employed. Simulation results proves the validity of the scheme.
-
Computer Engineering. 2012, 38(08): 57-60. https://doi.org/10.3969/j.issn.1000-3428.2012.08.019Aiming at the disadvantages of existing coverage optimization algorithms, such as early maturity, poor convergence performance and easy to fall into local extreme, this paper combines the application environment of the Three Gorges Reservoir water quality monitoring, presents a new coverage optimization algorithm for Wireless Sensor Network(WSN). The Particle Swarm Optimization(PSO) model within constriction factors is adopted. It uses chaotic sequence generated by chaotic Tent mapping instead of the original random parameter, introduces the conception of aggregation degree for judging, and realizes parameters adaptive adjustments. Experimental results show that the algorithm can improve network coverage.
-
Computer Engineering. 2012, 38(08): 61-63. https://doi.org/10.3969/j.issn.1000-3428.2012.08.020This paper proposes a hybrid Medium Access Control(MAC) protocol based on Wireless Sensor Network(WSN) by combining with the work pattern of Time Division Multiple Access(TDMA) and Carrier Sense Multiple Access(CSMA). It Introduces Fast Collision Resolution(FCR) into MAC protocol. Sensor-MAC(SMAC) protocol is improved in the mode of CSMA. Under such condition, the contention window changes according to the changes of network traffic. The hybrid MAC protocol can switch the working way self-adaptively and smoothly between the mode of CSMA and TDMA. Simulation results on OMNet++ show that the improved protocol can enhance the utilization efficiency of channel and prolong lifecycle of network.
-
Computer Engineering. 2012, 38(08): 64-66. https://doi.org/10.3969/j.issn.1000-3428.2012.08.021In order to reduce the energy consumption for communication come from network management and prolong the lifetime of the network, this paper designs a Reduced Encoding Rule(RER) based on Wireless Sensor Network(WSN) management primitives. In RER, one byte or more bytes can be used to indicate the length of the Protocol Data Unit(PDU) according to the length of the management message. Theoretical analysis and experimental results show that the encode rule can effectively reduce the datagram length and the communication energy, increase the encoding efficiency.
-
Computer Engineering. 2012, 38(08): 67-69. https://doi.org/10.3969/j.issn.1000-3428.2012.08.022This paper studies data transmission between different communication ways in heterogeneous network environment, and presents an improved Ad hoc On-demand Distance Vector Routing(AODV) protocol. In order to provide universal network services, realize data transmission and interoperability between heterogeneous network, it makes use of the complementary characteristics of different networks. Experimental results show that the protocol can reduce network congestion and average end to end packet delay, improve packet delivery ratio of data when communication load increases.
-
Computer Engineering. 2012, 38(08): 70-72. https://doi.org/10.3969/j.issn.1000-3428.2012.08.023Data aggregation can obviously cut energy consumption of Wireless Sensor Network(WSN), but there is no reasonable clustering solution which may cause more energy cost form the beginning. Aiming at the problem, this paper proposes a cluster-tree routing construction method based on energy consumption model in WSN. It proves the exist of optimal-clustering based on this model, gives the Optimal Energy Aware Cluster Tree(OEACT) algorithm, and realizes the construction of cluster-tree routing. Simulation results demonstrate that the method can improve the whole energy consumption of network, and increase the lifetime of the network.
-
Computer Engineering. 2012, 38(08): 73-75. https://doi.org/10.3969/j.issn.1000-3428.2012.08.024For the campus network and other small-scale local area network, the abnormal traffic can not be detected rightly by calculating self-similarity value directly. Based on analyzing the characteristics of the campus network traffic, it decomposes traffic into trend and random components and other items, uses Empirical Mode Decomposition(EMD) to eliminate traffic trend items, and makes the self-similarity value direct response to the state of random items. Experimental results show that this method can improve the accuracy of abnormal traffic detection.
-
Computer Engineering. 2012, 38(08): 76-78. https://doi.org/10.3969/j.issn.1000-3428.2012.08.025Sensor networks with limited nodes are deployed randomly, and thus there exist a number of monitoring holes which lead to incomplete sensor data sets. Aiming at the problem, this paper prposes a Kriging interpolation algorithm for sensor data based on irregular grid. An irregular grid meshing algorithm is designed to adapt to the random nodes deployment. A fast search algorithm is proposed to sift the neighboring nodes of the interpolation points. Kriging matrix is solved and fast Kriging data interpolation is obtained. Experimental results based on Intel-Berkeley data set show that this algorithm has higher accuracy.
-
Computer Engineering. 2012, 38(08): 79-81. https://doi.org/10.3969/j.issn.1000-3428.2012.08.026This paper proposes a fault-tolerant data center network structure——DCUBE. It uses modular definition way to enhance extensibility. It makes the server nodes failure treatment to improve fault tolerance, and adopts the parallel path between to improve the bandwidth throughput between the nodes. Experimental results show that DCUBE can better satisfy the network architecture requirements of data center, fault-tolerant, and it can provide high throughput.
-
Computer Engineering. 2012, 38(08): 82-84. https://doi.org/10.3969/j.issn.1000-3428.2012.08.027In limited data length, an improved algorithm based on data division and overlap reservation is proposed to realize detection of Direct Sequence Spread Spectrum(DS-SS) signal in low Signal Noise Ratio(SNR). The signal detection is realized by one-dimensional search in non-zero cyclic frequencies without prior knowledge. Experimental results show that the algorithm can restrain noise effectively, and it has a good detecting ability of DS-SS signal in low SNR.
-
Computer Engineering. 2012, 38(08): 85-88. https://doi.org/10.3969/j.issn.1000-3428.2012.08.028Aiming at the problem that many existing schemes have poor resistance to node capture attack. This paper presents a new deployment knowledge-based key pre-distribution algorithm, the deployment area is divided into several non-overlapping triangular grids and multi-key space scheme is applied to each pair of adjacent grid. Theoretical analysis and simulation show that compared with the existing algorithms, network has good resistance to node capture attack while maintaining a higher key connectivity through using deployment knowledge.
-
Computer Engineering. 2012, 38(08): 89-91. https://doi.org/10.3969/j.issn.1000-3428.2012.08.029This paper proposes a host intrusion prediction model named Fuzzy Wavelet Neural Network Intrusion Prediction(FWNN-IP) by using fuzzy methodology, neural network and wavelet technology. System calls are classified according to their dangerous degrees and higher dangerous system calls are assigned greater number. Programs(processes) traces are analyzed by applying fuzzed short sequences of system calls, and the aim of intrusion prediction can be achieved. Experimental results show that FWNN-IP can predict abnormal behaviors of programs(processes) more quickly, and takes more active action to protect host.
-
Computer Engineering. 2012, 38(08): 92-94. https://doi.org/10.3969/j.issn.1000-3428.2012.08.030Botnet based on Kademlia can easily be hidden in the legitimate traffic, but Botnet which uses Kademlia as its protocol may be detected because of firewall’s blocking. This paper designs a new semi-distributed Botnet based on Kademlia. By changing hybrid Botnet’s backbone from unstructured network to Kademlia network, it can circumvent firewall and has small flow. Simulation comparison experiments show that the new Botnet has better flow characteristics and robustness than traditional Botnets. And three kinds of defensive mechanisms against the designed Botnet are proposed.
-
Computer Engineering. 2012, 38(08): 95-97. https://doi.org/10.3969/j.issn.1000-3428.2012.08.031Aiming at the problems of network security, connectivity and large memory and communication costs in classic key pre-distribution scheme, this paper presents a Wireless Sensor Network(WSN) chaos key management scheme based on district-divided. It updates key by group partition method to reduce the network storage, and the ergodicity and initial value sensitivity of chaos are used to transmit safety information. Simulation results show that both the security and connectivity of WSN are improved, and memory cost and communication cost are reduced.
-
Computer Engineering. 2012, 38(08): 98-100. https://doi.org/10.3969/j.issn.1000-3428.2012.08.032This paper presents a broadcast encryption scheme based on lattice and hierarchical identity, whose encryption and decryption algorithms are based on Learning with Error(LWE) hard problem. Each user’s identity is associated with a lattice matrix, and the relationship between the lattice and its sub-lattice can easily represent the hierarchical identity construction. Short basis of a lattice is treated as a private key of a user, and based on the lattice delegation algorithm, the private key for each level’s user can be derived. Theory analysis proves that the scheme is Indistinguishable-Chosen Ciphertext Attack 2(IND-CCA2) secure against the attacks by outsider users and Forward Safe-Chosen Ciphertext Attack (FS-CCA) secure against the attacks by authorized receivers. Compared with the broadcast encryption scheme based on the bilinear mapping, it is more efficient in the encryption and decryption.
-
Computer Engineering. 2012, 38(08): 101-104. https://doi.org/10.3969/j.issn.1000-3428.2012.08.033Microsoft Office2007 and later version adopt ECMA-376 document encryption format. In this format, the file security is mainly protected by user authentication and files encryption, and Password Based Key Derivation Function(PBKDF) is the core of the Office security mechanism. In order to analyze the security of the PBKDF in ECMA-376 format, this paper proves the upper bound of the adversary’s advantage between the Key Derivation Function(KDF) and ideal random function through the Game-Playing technology, and discusses the safety of the Office encrypted files based on that theoretical result and the actual attack experiment. Based on the analysis results, it obtains that Office is with a certain degree of security while the user password is longer than 6 characters.
-
Computer Engineering. 2012, 38(08): 105-106. https://doi.org/10.3969/j.issn.1000-3428.2012.08.034By using discrete logarithm method, this paper proposes a RSA-based threshold proxy signature scheme based on the security problems of previous RSA-based threshold proxy signature scheme. Analysis results show that the new scheme can resist the original signer’s insider attack and the proxy signers’ conspiracy attack. Besides, the new scheme satisfies non-repudiation, verifiability, recognizability and other security requirements. And it has the advantages of simplicity of signing process, and low communications.
-
Computer Engineering. 2012, 38(08): 107-110. https://doi.org/10.3969/j.issn.1000-3428.2012.08.035The database zero-watermark has ambiguous, which may lead up to database copyright dispute. Aiming at this problem, this paper presents a novel database copyright image zero-watermark algorithm. The algorithm uses zero width and invisible character as the watermark character, embeds the copyright image into the host database, and establishes a database zero-watermark. At the same time, the Unicode value in character fields is used to extract the other database zero-watermarks. Experimental results show that the dual database zero-watermark can improve the database copyright protection ability.
-
Computer Engineering. 2012, 38(08): 111-113. https://doi.org/10.3969/j.issn.1000-3428.2012.08.036In order to optimize test efficiency of Intrusion Detection System(IDS) based on Support Vector Machine(SVM), a new intrusion detection method based on Graphics Processing Unit(GPU) and feature selection is proposed. During the process of intrusion detection, GPU-based parallel computing model is adopted and features of samples are reasonable selected. Experimental results demonstrate that the proposed method can reduce time consumption in the training procedure of IDS and the performance for intrusion detection is kept as usual.
-
Computer Engineering. 2012, 38(08): 114-116. https://doi.org/10.3969/j.issn.1000-3428.2012.08.037When using Covert Flow Tree(CFT) method to search covert channel, only few operation sequences really constitute covert channel. It increases the subsequent manual analysis work. In order to solve this problem, this paper proposes an improved Covert Flow Tree(CFT) method. It uses the information flow graph to constitute trees, designs the statement information flow rule, and gives the tree traversal algorithm for automated analysis. It uses the pedagogical file system as analyzing example, and result shows the accuracy of the improved method.
-
Computer Engineering. 2012, 38(08): 117-119. https://doi.org/10.3969/j.issn.1000-3428.2012.08.038There are some problems in Bell-La Padula(BLP) model, including the loss of integrity protection, the indetermination of trusted subject and the neglect of the environment factors of the platform, so this paper proposes an improved BLP model based on trusted level, which is named TL-BLP. In TL-BLP, security property, sensitive label of subject and object, and state transfer rules are improved by introducing the trusted level of subject, object and platform. It realizes the dynamic measurement of trusted degree and ensures access platform security, and protects the information integrity by the restriction to “read down and write up”. Analysis shows that the model not only can ensure the confidentiality, but also can enhance the integrity and availability of the system effectively, and it implements the access control based on trusted degree.
-
Computer Engineering. 2012, 38(08): 120-122. https://doi.org/10.3969/j.issn.1000-3428.2012.08.039In order to solve security problem of downstream data transmission of Ethernet Passive Optical Network(EPON), this paper presents a new encryption scheme based on timestamp function. It combines Advanced Encryption Standard(AES) algorithm with timestamp to form keys for encrypting downstream data. By adding an encryption/decryption module in both Optical Line Terminal(OLT) and Optical Network Unit(ONU), unsafe keys transmission through channel can be avoided. The extraction and synchronization of timestamp, along with the dynamic update and synchronization of keys, are released. Experimental results show that the method has the advantages of low delay, high throughput and high security compared with the traditional encryption algorithm.
-
Computer Engineering. 2012, 38(08): 123-124. https://doi.org/10.3969/j.issn.1000-3428.2012.08.040Existing VoIP security technology can not effectively detect Spam over Internet Telephony(SPIT) attack. To resolve the problem, this paper uses social network method to analyze the history of communication behavior data of the user node. Average call length, call frequency, call does not exist the user number, hang up the number/passive hang up the number and one-way outgoing/two-way communication nodes are extracted as characteristic properties to establish a Bayesian model. Experimental results verify that the method is effective.
-
Computer Engineering. 2012, 38(08): 125-127. https://doi.org/10.3969/j.issn.1000-3428.2012.08.041In iris recognition, eyelash occlusions can reduce the accuracy of iris recognition, an algorithm of eyelash effective region extraction based on endpoint identification is proposed. Based on analyzing the structural characteristics of eyelashes, this algorithm identifies the endpoints of eyelashes projection in the outer circle of iris, and uses a sector region to represent the region of eyelash occlusions. Removing the sector region and the pupil, the rest of the outer circularity is the effective region of iris. Experimental results show that the region of eyelash occlusions detected by the method can contain all eyelash pixels in the outer circularity and the effective region of iris extracted is “pure”.
-
Computer Engineering. 2012, 38(08): 128-130. https://doi.org/10.3969/j.issn.1000-3428.2012.08.042The traditional tf.idf algorithm can not take full advantage of the characteristics of data set for classification, which can not reflect the relationship of the term among the classes. On the basis of analyzing the tf.rf which is a supervised text feature weighting method, this paper proposes an improved supervised algorithm of text feature weighting which is called tf.ridf. The algorithm combines the advantages of two ideas, considers the relationship of the term in the various categories and in the overall documents, implements text feature weighting. Experimental result shows that the classification accuracy of tf.ridf increases significantly higher than tf.rf.
-
Computer Engineering. 2012, 38(08): 131-133. https://doi.org/10.3969/j.issn.1000-3428.2012.08.043This paper proposes a new Cascade Support Vector Machine(CSVM) classification algorithm, using AdaBoost algorithm framework and Support Vector Machine(SVM) combination to deal with the problem of multiple classifier. For the problem of consuming time in the multi-classification problems with SVM, introduces the Minimum Enclosing Ball(MEB) algorithm to extract the original sample data to shorten the training time for SVM. CSVM is applied in the Chinese thematic classification. Experimental results show that CSVM algorithm has similar accuracy with AdaBoost algorithm, but the computation time is only 35% of the AdaBoost-SVM algorithm.
-
Computer Engineering. 2012, 38(08): 134-136. https://doi.org/10.3969/j.issn.1000-3428.2012.08.044Particle Swarm Optimization Particle Filtering(PSOPF) algorithm is easy to fall into local optimum, so the particles can not move to the global optimal location, and reduce algorithm precision. According to this problem, the paper proposes a Particle Swarm Optimization Particle Filtering based on Chaotic(CPSOPF) algorithm. Through the chaotic search algorithm, this algorithm makes particles find the global optimal location, dispels particle swarm at local optimum location and makes them move to global optimal location. So the number of effective particles increases, which can effectively restrain particles degradation and exhaustion. Simulation results show that the CPSOPF algorithm can remarkably improve the estimation accuracy compared with of the conventional Particle Filtering(PF) and the traditional PSOPF algorithm.
-
Computer Engineering. 2012, 38(08): 137-140. https://doi.org/10.3969/j.issn.1000-3428.2012.08.045This paper mainly studies the problem of serpentine locomotion gait optimization and control for the snake-like robot. It analyzes the motion gait model of the snake-like robot with the friction model, and uses PID controller to control and track the gait according to the gait control system structure based on closed-loop feedback. Applying the Non-dominated Sorting in Genetic Algorithm(NSGA-Ⅱ) into the gait optimization on the above basis, it implements the parameter optimization for the motion gait track and control system based on closed-loop feedback. Simulation results show that NSGA-Ⅱ algorithm can achieve the objective to optimize variables efficiently, find the optimal value between the power and speed, and is feasible and effective to solve the problem of the multi-objectives optimization for the motion gait of the snake-like robot.
-
Computer Engineering. 2012, 38(08): 141-143. https://doi.org/10.3969/j.issn.1000-3428.2012.08.046A fast localization algorithm is presented for the human eye with single camera. The face is located adopting the improved AdaBoost constructed strong cascade classifier. The eye region can be detected according to the obvious gray changes in the area of the eye before eye-blink and after. The method of edge extraction and local Hough transform for detecting the center of circle is adopted to detect the center of pupil. Experimental results indicate as follows: simple equipments used in the algorithm, precise localization algorithm, fast and accurate positioning the eye even in the complex background, as well as good robustness in the case of illumination and head rotation.
-
Computer Engineering. 2012, 38(08): 144-146. https://doi.org/10.3969/j.issn.1000-3428.2012.08.047In view of the distinct levels of the association of a term with different classes, an improved difference-based CHI method is proposed to extract discriminative affective words form lyrics text. Support Vector Machine(SVM) classifier is constructed to apply the selected features, obviously increasing lyric sentiment classification performance. The lyric features selected by the improved method are combined with audio features. Experimental results verify the efficiency of this fusion.
-
Computer Engineering. 2012, 38(08): 147-149. https://doi.org/10.3969/j.issn.1000-3428.2012.08.048In order to make use of different information of sentiment text. This paper proposes sentiment classification method based on status and action description. It aims to classify the sentimental text into two views called status views and action views and employs them in supervised sentiment classification. In order to get the two views, it manually annotates corpus and trains a binary classifier with the corpus to detect the two views. Each view is trained to be a single classifier and an ensemble of the individual classifiers is employed to improve the effect of the classification. Experimental results show the classification effectiveness of the proposed method has improvement on three domains of Chinese sentiment classification tasks.
-
Computer Engineering. 2012, 38(08): 150-152. https://doi.org/10.3969/j.issn.1000-3428.2012.08.049The concept of granular entropy is introduced in dominance information systems, the granular entropy based on dominance relation is defined, its related properties and theorems are gained, the significance of attribute is defined through granular entropy, with the significance of attribute as heuristic information. On this basis, a heuristic algorithm based on granular entropy for reduction of attribute is proposed, and the time complexity of the algorithm is analyzed. It illustrates that this algorithm is effective through an example, which puts forward new theoretical basis for further researching of rule extraction in target information systems based on dominance relation.
-
Computer Engineering. 2012, 38(08): 153-155. https://doi.org/10.3969/j.issn.1000-3428.2012.08.050Aiming at the problem of attribute reduction, this paper proposes a new multi-objective particle swarm optimization algorithm which aims at the least reduction of attributes sets and maximum of the dependency of the attributes. It searches best particle based on non-dominated sorting and weighting method. Combining the new motion equation with ε-neighborhood perturbation, the algorithm has strong global and local searching ability. Experimental results with UCI data sets show that the proposed algorithm is more effective than the compared algorithms.
-
Computer Engineering. 2012, 38(08): 156-158. https://doi.org/10.3969/j.issn.1000-3428.2012.08.051On the basis of Pyramid Histogram of Oriented Gradient(PHOG), a new layered vote method is presented. It is used in the area of face recognition, object recognition, and object detection. By using the layered vote, it can concentrate the attention on the levels which have high recognition rate, and the weights of vote are set to these chosen levels. The final result is got by vote for the object category. This method is tested on ORL and UIUC Car database, and the results show that it is better than traditional classification method.
-
Computer Engineering. 2012, 38(08): 159-160. https://doi.org/10.3969/j.issn.1000-3428.2012.08.052Most of the rectangle recognition methods due to the low speed of detection, low recognition rate and lack of precision, they can not meet the needs of modern industrial development. To solve the above problems, a gray projecting integral extreme value method to detect rectangular is presented. It gets the image gray value projected in all directions to form a Gray Projection Integral(GPI) matrix, obtains the GPI matrix extreme points saved as a new matrix GPIEV, and searches for parallel lines in the GPIEV matrix to reconstruction rectangle. Experimental results show that the method has better anti-jamming ability, quick operation speed, and it can meet the requirement of real-time processing.
-
Computer Engineering. 2012, 38(08): 161-163. https://doi.org/10.3969/j.issn.1000-3428.2012.08.053This paper presents a novel classifier called Total Margin v-Fuzzy Support Vector Machine(TM-v-FSVM) based on the idea of FSVM and Total Margin(TM). Although it can be seen as a modified class of the classic FSVM, TM-v-FSVM has better theoretical classification performance than FSVM. The proposed method solves not only the overfitting problem resulted from outliers with the approaches of fuzzification of the penalty and total margin algorithm, but also the imbalanced datasets by using different cost algorithm, thus obtaining a lower generalization error bound. Exprimental results obtained with UCI real datasets show that the algorithm is more stable and superior than other related diagrams.
-
Computer Engineering. 2012, 38(08): 164-166. https://doi.org/10.3969/j.issn.1000-3428.2012.08.054The traditional Online Sequential Extreme Learning Machine(OS-ELM) has variations in different trials of simulations and the over-learning problem. Using wavelet substitutes network’s traditional activation function, structural risk minimization is used to modify the problem, so a novel algorithm called regularized wavelet extreme learning machine is proposed. Motivated by online learning method, Online Sequential Regularized Wavelet Extreme Learning Machine(OS-RWELM) is designed. Experimental results show that this algorithm avoids the local minimum and over-learning problem, has a fast online learning speed and a good generalization and robustness.
-
Computer Engineering. 2012, 38(08): 167-169. https://doi.org/10.3969/j.issn.1000-3428.2012.08.055In order to improve the performance of Multiple Classifier System(MCS), a new classifier selection ensemble algorithm called MCC-SCEN is proposed. This algorithm first finds classifier subsets which have the biggest single accuracy and diversity to generate the unexpand classifier set, then selects classifiers that have largest MCS to construct the ensemble system and uses weighted vote as ensemble method. Experimental results show this method is better than AdaBoost and Bagging method.
-
Computer Engineering. 2012, 38(08): 170-172. https://doi.org/10.3969/j.issn.1000-3428.2012.08.056Aiming at the problem of lack of information in ground targets detection of unmanned airship, this paper proposes a novel method for both moving and nonmoving ground objects detection. The nonmoving object is detected successively in the image sequence by tracking the feature points selected in the initial static object region with the Lucas-Kanade technique. For the moving object, the overall image motion between the adjacent frames is estimated by matching the feature points, and the overall frame motion is compensated. Appling the frame difference approach to the compensated frame and the current frame, the moving object is detected. The proposed method is evaluated and experiments are carried out with the real data provided by the unmanned airship, experimental results show that the effectiveness of the proposed methods.
-
Computer Engineering. 2012, 38(08): 173-176. https://doi.org/10.3969/j.issn.1000-3428.2012.08.057In transfer learning process, noise makes the result unreasonable when you classify slow changing dataset. Here is an algorithm called Extended Support Vector Machine(ESVM) proposed to solve this problem. Because it makes full use of probability distribution of original data and uses the learning experience of the previous dataset to classify the latter dataset, ESVM can correctly classify the changing dataset with inheriting the characteristics from the previous dataset. Experimental result shows the antinoise performance of the algorithm.
-
Computer Engineering. 2012, 38(08): 177-179. https://doi.org/10.3969/j.issn.1000-3428.2012.08.058It can not satisfy the accurate need of the estimation of the multiple dimension state vector by using a single observational value. So more kinematic observational information must be used for the calculation. On the basis of this, this algorithm based on dynamic weighted multiple observational values for particle filtering is presented. This algorithm uses multiple observational values and can enhance the precision of target tracking by combining multiple estimated state vectors. Simulation experiment verifies the effectiveness of the algorithm.
-
Computer Engineering. 2012, 38(08): 180-182. https://doi.org/10.3969/j.issn.1000-3428.2012.08.059This paper aims at the parallelism of Quantum Evolutionary Algorithm(QEA), makes full use of MapReduce’s the highly abstract, the preparation of serial program automatically running in parallel, simple programming interface and easy parallel programming, realizes the parallelization of QEA in MapReduce, puts forward parallel QEA based on MapReduce model and runs the algorithm on Hadoop platform. Using 0-1 knapsack problem for test, experimental results prove the feasibility of MRQEA, and it has good speed-up ratio and parallel efficiency in dealing with large data set.
-
Computer Engineering. 2012, 38(08): 183-185. https://doi.org/10.3969/j.issn.1000-3428.2012.08.060In order to achieve serve tracking of cigarette package PV belt and shear the PV belt more accurately and efficiently, a visual servo tracking algorithm based on motion estimation is proposed. The centroid of the moving object is obtained by using image moments, the speed and accelerator of the moving object can be obtained, and the location of the moving object can be predicted in the next time through Kalman filter. The operation of the servo motor depends on the speed, accelerator and location. Experimental results show that the speed error is less than 2 pixel/s, and the position error is less than 5 pixel. So the algorithm has the virtue of accuracy and real-time property through analyzing the speed and the location.
-
Computer Engineering. 2012, 38(08): 186-188. https://doi.org/10.3969/j.issn.1000-3428.2012.08.061The X-ray films for nondestructive flaw detection in industry have the character of low Signal to Noise Ratio(SNR), bad definition and contrast. In order to improve the quality of the X-ray images, an improved Multi-scale Retinex(MSR) algorithm is proposed in this paper. Wavelet transforms the original image and MSR is filtered in the low frequency band. It offers a new threshold function and threshold filtering in the high frequency band to reconstruct the image. Experimental result shows that this MSR algorithm can efficiently enhance the image enhancement while reduce its noise.
-
Computer Engineering. 2012, 38(08): 189-191. https://doi.org/10.3969/j.issn.1000-3428.2012.08.062In order to improve efficiency of image retrieval, a retrieval method based on visual saliency map is proposed in this paper. The background information unrelated to retrieval task in original image is removed according to the visual saliency map, and remains the region of interest. The BDIP-BVLC moments in the wavelet domain is used to extract the image features. The secondary inquires measure strategy is adopted as the distance function. Experimental result shows that the proposed method outperforms the method based on saliency weighted. It has higher retrieval accuracy.
-
Computer Engineering. 2012, 38(08): 192-194. https://doi.org/10.3969/j.issn.1000-3428.2012.08.063This paper researches the shortest path algorithm and the weighted histogram image segmentation method. Combined with rapid fuzzy clustering thought, it presents an improved rapidly Fuzzy C-means(FCM) image segmentation algorithm. Through the shortest path algorithm, it initializes the image to make it accurately reflect the pixel neighborhood. It changes a weighted histogram to influence degree of gray strong regional pixel in the image segmentation and identifies a clustering center by adaptive weighted histogram. Experimental result shows that this algorithm can rapidly and accurately segments images and has strong antinoise.
-
Computer Engineering. 2012, 38(08): 195-197. https://doi.org/10.3969/j.issn.1000-3428.2012.08.064In process of concrete CT image segmentations, the compute complexity is higher when belief messages are transmitted. Aiming at this problem, this paper proposes a fast segmentation algorithm based on multiscale belief propagation. According to the feature of belief propagation algorithm, it structures grid figure with pyramid shape for message transmission, so as to speed up the convergence speed and reduce iteration times. It uses ambigram technology to reduce faith the quantity of information. Experimental results show that the proposed method can improve efficiency and precision of the image segmentation.
-
Computer Engineering. 2012, 38(08): 198-200. https://doi.org/10.3969/j.issn.1000-3428.2012.08.065This paper researches classical HIS transformation, wavelet transformation and fusion method, and proposes a new method for HIS wavelet remote sensing image fusion method with dynamic weighting of regional multi-feature. It combines the advantages of the IHS transform and wavelet transform to achieve a better fusion result, which comes from new full-color components after matching and brightness component of multispectral image after IHS transformation. Experimental results indicate that the proposed method has apparent advantage in reservation of spectral information and spatial details enhancement than other methods.
-
Computer Engineering. 2012, 38(08): 201-203. https://doi.org/10.3969/j.issn.1000-3428.2012.08.066For traditional curve compression algorithms can not solve curve smoothing problem and avoiding topological error, an algorithm of compression and multi-scale representation for vector data based on spline is proposed. The algorithm selects smooth factor as threshold to compress curves and it can well keep the curves smooth features. In addition, the algorithm only needs a little control points to describe the whole curves, so it can improve the velocity and quality for processing vector curves. Experimental result shows that the algorithm only needs nine control points to well approximate an given original curve. It proves that the algorithm is effective.
-
Computer Engineering. 2012, 38(08): 204-206. https://doi.org/10.3969/j.issn.1000-3428.2012.08.067In order to overcome the problems such as slow convergence speed and fuzzy solution, this paper proposes an improved Iterative Blind Deconvolution(IBD) algorithm. Moment of momentum is utilized to obtain the limited support domain of image, and alternating iterative calculation is achieved in the frequency and space domain. Blind image restoration is completed. Compared with traditional iterative blind deconvolution and IBD based on wavelet, simulation results show that the algorithm speeds up the convergence rate, and obtains a better restoration result.
-
Computer Engineering. 2012, 38(08): 207-209. https://doi.org/10.3969/j.issn.1000-3428.2012.08.068At present, the measurable temperature range of sensors which is standard configured by Wireless Sensor Network(WSN) nodes is limited. In order to fulfill the demand of coalfield fire area monitoring, a high temperature sensor node of WSN based on JN5139 node that is widely used is designed. Type-K thermocouple and MAX6675 amplification and digital converter are used in the high temperature gathering part. MAX6675 and JN5139 are connected through SPI interface and the shifting and data transmission processing are controlled by JN5139 through software. Results show that the temperatures are confirmed correct in different actual environment and it also has expansibility and portability.
-
Computer Engineering. 2012, 38(08): 210-212. https://doi.org/10.3969/j.issn.1000-3428.2012.08.069In order to improve hardware resources utilization and system maintenance efficiency of the distributed management information system, and provide application system service on demand for business users, this paper puts forward a solution based on virtualization technology ARP Private Cloud Services(APCS). The ARP Cloud Services Infrastructure Platform(ACSIP) realizes the integration and utilization of hardware resources, and improves the system resources utilization, guarantees the continuity of business system, provides dynamic configuration of the application services. The ARP Cloud Services Management Platform(ACSMP) realizes the dynamic management of distribution system resources, centralized maintenance management, monitoring management and safety management.
-
Computer Engineering. 2012, 38(08): 213-216. https://doi.org/10.3969/j.issn.1000-3428.2012.08.070Aiming at containers on the container-train being unloaded and transferred high efficiently, two separate problems, inter trucks schedule and container slots allocation are integrated. On the base, a two-objects optimal model is set up which thinks of the factors on the operation efficiency, which is traveling time between the rail gantry cranes and assistant containers area and waiting time at those work places. The model solved by hybric genetic algorithm is effective and reliable by comparing the simulation results with the actual data and analyzing the mean time and its reliance by 20 times calculation.
-
Computer Engineering. 2012, 38(08): 217-220. https://doi.org/10.3969/j.issn.1000-3428.2012.08.071Through the requirement analysis for the Ground Control Station(GCS) of Small-scale Unmanned Helicopter(SUH), not only the structure and function of GCS are introduced for SUH, but also the solution to the key technologies is analyzed. The GCS is based on VC++ development environment and embedded Matlab technique. The GCS for SUH is applied in a small-scale unmanned helicopter, and the flight experimental results show that it has perfect real-time property, good stability, friendly man-machine interface, strong expandable ability, etc.
-
Computer Engineering. 2012, 38(08): 221-223. https://doi.org/10.3969/j.issn.1000-3428.2012.08.072Location and navigation technologies are the key of autonomous ship in lakes. This paper establishes an autonomous ship navigation system, designs the global static coordinate and the local static coordinate, the ship dynamic coordinate and kinematics model. According to human driving technologies, a cruise navigation mode is designed and a region navigation mode is given for saving energy, and navigation strategies using if-then rules. Reasons of practical measurement location data noises of GPS and electronic compass are analyzed. In order to improve the precision of location, FIR digital filtering is designed using Hamming window function. Monitoring area calculating equation is derived. Navigation algorithms and interfaces are programmed based on VC++. And navigation experiments are accomplished on self-developed autonomous ship. Results show that effect of FIR filtering to electronic compass is better than to GPS signals, and the cruise mode can achieve a wide range of scanning and monitoring and the given region mode can achieve the monitoring of key areas using lesser energy consumption.
-
Computer Engineering. 2012, 38(08): 224-227. https://doi.org/10.3969/j.issn.1000-3428.2012.08.073For security issue of personal terminal operating system, this paper proposes a security enhancement model based on the system virtualization technology on operating system, and in-depth study implementation of graphics card pass-through technology to improve virtual machine performance, based on Kernel-based Virtual Machine(KVM) virtual machine. Experimental results show that graphics card pass-through technology can breakthrough display performance bottlenecks of the virtual machine client operating system. This technology can make client operating system meet the graphics and processing applications as a real operating system, and can expand more space for development of terminal security field.
-
Computer Engineering. 2012, 38(08): 228-231. https://doi.org/10.3969/j.issn.1000-3428.2012.08.074Based on the model of single producer multiple consumers, the drawback of the writing concurrency and reading processes in traditional ring buffer is analysed. An algorithm with tagged buffer unit status is presented to solve the synchronization problem of writing and reading processes in the ring buffer. And the condition resulted in the bottleneck of the ring buffer performance is analysed quantificationally. The performance of the ring buffer is promoted greatly if the condition does not be met. Contrastive experiment and mathematical analysis verify that the performance of data packet processing in the ring buffer is better.
-
Computer Engineering. 2012, 38(08): 232-234. https://doi.org/10.3969/j.issn.1000-3428.2012.08.075This paper proposes a research on time and space consistency problem in HLA bridge multiple federations system. It divides the data forwarded by bridge federate into three data types, defines the corresponding structure of entity class and interaction class, and puts forward three processing methods, the delay compensation method(used to solve delay of entity attributes update), the impact point parallel remove method(used to solve delay of Precision Guidance Weapons(PGW) hit Interactions) and the rollback method(used to solve delay of entity anti-attributes update). Experimental results show the methods can efficiently contents the demand of time and space consistency by practical running of the bridge multiple federations system.
-
Computer Engineering. 2012, 38(08): 235-238. https://doi.org/10.3969/j.issn.1000-3428.2012.08.076Aiming at the dominant aging mechanism induced by Negative Bias Temperature Instability(NBTI) and the soft error, e.g. Single Event Upset(SEU), affect the normal operation of the device, this paper presents a sensor called Soft Error and Aging Online Sensor(SEAOS) that can on-line detect soft error and aging when the device is under its normal operation, to against SEU and the timing violations induced by aging. Due to the reuse of Concurrent Built-in Logic Block Observer(CBILBO), the hardware overhead below 33.46% is well tolerated. Experimental results demonstrate that, compared with several classical structures by employing 0.18 μm technology process, SEAOS has a good capacity of detection and a lower hardware overhead.
-
Computer Engineering. 2012, 38(08): 239-241. https://doi.org/10.3969/j.issn.1000-3428.2012.08.077The traditional dual-frequency three-dimensional profilometry measurement cannot be directly used due to its low resolution and long time of data processing. Therefore, this paper presents a 3D vision system based on the mean method of spatial domain. This method processes the dual-frequency structured light images which include depth information with mean image of spatial domain, obtains real depth information of the surface. It not only improves the measurement efficiency, but also improves the measurement accuracy. As a result, through measuring of the natural objects, linear relationship of the input-output of the system which is posed by this method is proved, and this method can efficiently work in three-dimensional profile measurement with high resolution.
-
Computer Engineering. 2012, 38(08): 242-245. https://doi.org/10.3969/j.issn.1000-3428.2012.08.078A triangular mesh simplification algorithm is proposed for the mesh simplification in Computer-aided Engineering(CAE) computation. In this algorithm, mesh is simplified by edge collapses and triangle collapses, and different simplifying schemes are adopted to gain adaptive Level-of-Detail(LOD) mesh according to geometric features of mesh recognized by the node discrete degree. During the simplification, in order to optimize mesh and preserve features the goal points in collapse are predicted by optimizing the triangular elements’ aspect ratio and lastly extracted by fitting surfaces around. It is proved that this algorithm can simplify mesh efficiently, and reduce the computation time up to 75%.
-
Computer Engineering. 2012, 38(08): 246-248. https://doi.org/10.3969/j.issn.1000-3428.2012.08.079Aiming at the effective active period and its influencing factors of the normal video and adverse video in the Internet, this paper uses the method of survival analysis to discuss the network video propagation. In order to find the risk factors which influence the effective time, it takes advantage of the COX’s proportional hazard regression model. Research results indicate that the effective time of adverse videos is longer than the normal. It is also found that the video content has great impact on the active time of network video.
-
Computer Engineering. 2012, 38(08): 249-251. https://doi.org/10.3969/j.issn.1000-3428.2012.08.080This paper proposes a new Radio Frequency Identification(RFID) anti-collision algorithm to resolve the collision problem in the process of tag identification. The proposed algorithm makes the use of binary collision algorithm to estimate the number of the partial tags which are to be identified and use the multi-branch tree algorithms to identify those tags. Experimental results show that the algorithm reduces the probability of the conflict and the identification tag efficiency is 42.5%~42.8%.
-
Computer Engineering. 2012, 38(08): 252-255. https://doi.org/10.3969/j.issn.1000-3428.2012.08.081In the paper, a programming task planning for the problem of mould programming task planning is presented, a multi-objective optimization model for duration, quality and programmer load balancing is built. Horizontal search ability of virus evolution genetic algorithm and elite reservation strategy of Non-dominated Sorting Genetic Algorithm-Ⅱ(NSGA-Ⅱ) are adopted for better solutions with diversity. Multi-group strategy is introduced for solving multi-objective problem. Simulation result shows that approximate solution of the algorithm has good distribution and diversity is better, and compared with the NSGA-Ⅱ, the solution distribution for the algorithm is more close to the front of the optimal solution.
-
Computer Engineering. 2012, 38(08): 256-259. https://doi.org/10.3969/j.issn.1000-3428.2012.08.082Because satellite imaging reconnaissance system can just partly meet the user’s requirement, this paper puts forward a new evaluation method for satisfied degree. The evaluation is implemented from three aspects such as image quality, reconnaissance time and reconnaissance region. For continuous monitoring requirements, it puts forward a respectively quantitative method for reconnaissance time satisfaction. According to regional reconnaissance contains important point target, it suggests calculation method for reconnaissance regional satisfaction. The evaluation method is validated to be effective by experiment, and finds the reasons which reduce the satisfied degree.
-
Computer Engineering. 2012, 38(08): 260-263. https://doi.org/10.3969/j.issn.1000-3428.2012.08.083An efficient adaptive algorithm is proposed as heuristic algorithm based on quasi-rectangle packing and the Minimun Local Distance Position(MLDP) strategy, through the introduction of compactness and fitness. Combining heuristic algorithm with genetic algorithm based on natural number coding, a hybrid algorithm for the 2D unequal circle packing problem is presented. The hybrid algorithm gives full play to the global search capability of genetic algorithm, and takes full advantage of local search capability of adaptive heuristic algorithm. Experimental results of standard test instances from literatures indicate that more satisfactory results can be achieved with the new approach than related classic algorithms in a relatively shorter time.
-
Computer Engineering. 2012, 38(08): 264-267. https://doi.org/10.3969/j.issn.1000-3428.2012.08.084This paper presents a verification method of a kind of Ethernet for Plant Automation(EPA) network chip. It Introduces the function and structure of the EPA network chip, designes a virtual equipment based on the EPA network environment. From the angle of network communication interface and scheduling verify correctness, this paper puts the EPA network chip into EPA network to test. Test results shows that the simulation system of synchronous precision is less than 1 s, can satisfy the using requirement, and verifies the stability of the Field Programmable Gate Array(FPGA) as subordinate clock program.
-
Computer Engineering. 2012, 38(08): 268-270. https://doi.org/10.3969/j.issn.1000-3428.2012.08.085Most of the extraction methods mainly focus on the extraction of the subject information block, and pay no attention on the individual information piece. A video metadata extraction system based on DOM tree is proposed to solve this problem. Combining with the node type of Web DOM tree, it extracts the metadata of Web pages thorough individual subject information block by improving the links filter functions of Heritrix and queue management strategy of URL. Experimental results show that the average precision ratio of Web page and the average extraction ratio of the system are 95.7% and 98.4%, greatly higher than the similar systems.
-
Computer Engineering. 2012, 38(08): 271-273. https://doi.org/10.3969/j.issn.1000-3428.2012.08.086This paper proposes a Qext-algorithm with the capture effect according to the Q-algorithm introduced in EPC-Gen2 protocol. In process of computing the value of Q, it is proposed that parameter C should be taken into account independently for idle cases and collided cases. Capture effect is considered when computing parameter C. Experimental results show that this algorithm can get a more precise Q value, and an optimal frame size for ALOHA protocol can be acquired efficiently. It improves the identification performance of Radio Frequency Identification(RFID) system further.
-
Computer Engineering. 2012, 38(08): 274-276. https://doi.org/10.3969/j.issn.1000-3428.2012.08.087According to the numerical control system of real time property, embedded characteristics and for the special needs of specific manufacturing technology, this paper puts forward the a construction method of numerical control system for aspect-oriented orthogonal structure. The system structure designed includes basic component, aspects component and connector as the core elements constitute hardware control center and time control center. Hardware control function from the basic components and connectors, time control function is composed by the time aspect component and connectors. The components correlation in time between different threads is realized by aspect-oriented component crosscutting. Analysis results show that this method is beneficial to the system performance analysis and assembly application, and easy to the reconstruction of the numerical control users.
-
Computer Engineering. 2012, 38(08): 277-279. https://doi.org/10.3969/j.issn.1000-3428.2012.08.089This paper researches one-way motion estimation algorithm, and presents a motion compensated Frame Rate Up-conversion(FRUC) method. It merges compensating frames of forward and backward motion estimation and fills up holes of the merging frame. The matching criterion of motion estimation includes correlation factor of motion vector and edge matching error to get more accurate motion vectors. It adopts an easy method to estimate covered/uncovered area using two frames to solve the occlusion problem. A re-estimating method is adopted to fill up the holes existing in merging frame. Experimental results confirm that the proposed method is easy to compute and has a better frame interpolation performance.
-
Computer Engineering. 2012, 38(08): 280-282. https://doi.org/10.3969/j.issn.1000-3428.2012.08.090
According to lithium battery pack equalization control algorithm, this paper proposes a novel control scheme for lithium battery pack that can realize energy equalization. The design of fuzzy-PI is represented in detail, including the controller’s inputs and outputs, fuzzy logic rules base and the set of PI parameters. PI control is introduced into fuzzy control to form fuzzy-PI control, which integrates the advantage including fast response of fuzzy-PI, having no steady-state static error of PI and so on, the equalization efficiency and precision are improved. Experimental results show that this scheme can improve the problem that the control precision of fuzzy control application is poor in battery charging and discharging energy equilibrium process, and realize more efficient equalization.
-
Computer Engineering. 2012, 38(08): 283-286. https://doi.org/10.3969/j.issn.1000-3428.2012.08.091
Aiming at the coupling relationship among electronic equipment’s feature parameters, a hybrid Support Vector Regression(SVR) based fault prediction model is proposed in the paper, both the time relativity and space relativity of the feature parameter are taken into account, and it designes the algorithm process of hybrid SVR which improves the prediction accuracy by fusing the vertical historical state data and the horizontal correlation parameter together, applying D-S evidence theory. Experimental results show that, compared with vertical SVR or horizontal SVR, the proposed method is more accurate, and is capable of performing fault prognosis on the complicated electronic equipment effectively.
-
Computer Engineering. 2012, 38(08): 287-290. https://doi.org/10.3969/j.issn.1000-3428.2012.08.092
The point cloud normal vector calculation is sensitive to distribution density, and the calculation error is big in sharp border presently. In order to solve this problem a method of normal vector calculation based on Self Organization Map(SOM) is presented. The geometrical and topological information on scattered point cloud are employed to estimate the normal vector. A sphere SOM is trained to approximate the sampled surface with triangular meshes. Point cloud is clustered on the nodes of SOM, after that plane fitted by the k-neighbor points gives an estimation of the point normal. And the estimated point cloud normal vectors are aligned by adjusting patch normal. Experimental results show that the relative error is less than 0.08 and the standard deviation is 0.009. The method has high calculation precision.
-
Computer Engineering. 2012, 38(08): 291-292. https://doi.org/10.3969/j.issn.1000-3428.2012.08.093
For the universal adapter problem between Enterprise Digital Rights Management(E-DRM) system and reader software which is caused by wide ranges of them, this paper proposes an E-DRM universal adapter model. This model is high level hierarchical abstraction of E-DRM system and reader software, and completes unified management of rights by right control level. The model implements lasting security by monitor of reader and priority management of levels. Experimental result shows this model has good universal adaptation ability and lasting security.
2024
Volume 50 No 3
Volume 50 No 3
(Started in 1975, Monthly)
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428