05 August 2010, Volume 36 Issue 15
-
Select all|
-
Computer Engineering. 2010, 36(15): 1-3. https://doi.org/10.3969/j.issn.1000-3428.2010.15.001Three basic problems: the minimum points set, converging mapping, dynamic link table of Randomized Hough Transform(RHT) of zero baseline sinusoidal curve are studied. The improved method for the fitting of a sine curve with three sample points is proposed. A detection algorithm of a sort of zero baseline sinusoidal curve based on RHT is given. The performance of computation and the storage is analyzed. Simulation experiment shows that the method proposed is effective.
-
Computer Engineering. 2010, 36(15): 4-6. https://doi.org/10.3969/j.issn.1000-3428.2010.15.002A combined index method, integrated with the inverted index and the function compatibility index, is designed to improve the efficiency of Web service discovery. The combined index is composed of the Weighted Compact Function Compatibility-Graph(WCFC-Graph) for Web service and the inverted index for the nodes in WCFC-Graph based on ontology-annotated output. Using the combined index, a service discovery algorithm is proposed. Comparison to discovery algorithm based on inverted index and function compatibility index, shows that the proposed algorithm is able to significantly reduce the times of function-compatibility check and the average length of service per output, and has best performance.
-
Computer Engineering. 2010, 36(15): 7-8,11. https://doi.org/10.3969/j.issn.1000-3428.2010.15.003Current traffic sampling methods are based on mathematic theory without considering about real network traffic. So according to traffic prediction to dynamically adjust sampling rate, this paper proposes a new sampling method based on Fractal Auto Regression Integrated Moving Average(FARIMA) traffic prediction. The method not only reduces CPU load, but storage space as well. Compared with sample mean and Hurst parameter of traffic data produced from currently used sampling method, the experiment results can generate more accurate traffic statistics of sampling traffic.
-
Computer Engineering. 2010, 36(15): 9-11. https://doi.org/10.3969/j.issn.1000-3428.2010.15.004Aiming at the problem of the lack consideration of Peer node heterogeneity and the dynamics of networks in data scheduling of P2P-VoD based on network coding, a data scheduling algorithm of P2P-VoD based on adaptive random network coding is proposed. Network coding window is adaptively adjusted by the Peer node according to its cache, the link available bandwidth and the deadline of playable block. The simulation is conducted on the general P2P-VoD simulation platform based on NS2 design. Experimental results show that the node heterogeneity and the dynamics of networks are adaptive, the playback continuity index in the peer node is enhanced.
-
Computer Engineering. 2010, 36(15): 12-14. https://doi.org/10.3969/j.issn.1000-3428.2010.15.005This paper researches the utility maximization problem of network coding scheme-based packet network (ie. coded packet network). Depending on the correspondence between network coding and network flows and multicast tree decomposition method, the coded packet Network Utility Maximization(NUM) model is established. A distributed projection subgradient algorithm for the problem is derived, and sufficient condition that can make the algorithm converge to the globally optimal solutions is found through finding an effective Lipschiz constant. Numerical example is provided to validate the correctness of the algorithm.
-
Computer Engineering. 2010, 36(15): 15-16,19. https://doi.org/10.3969/j.issn.1000-3428.2010.15.006This paper concerns the study of reliable transport protocols employed in cislunar telecommunication environments from the two perspectives of transport layer and application layer. Based on OPNET software, CCSDS process-level modules are developed. Four kinds of reliable transport mechanism including TCP, SCPS-TP, CFDP and MDP are configured, and protocol emulation and performance analysis are conducted. Experimental results show that, with the long propagation delay, high error rate and a high degree of asymmetry in front/back to the channel rate of the links, SCPS-TP losts the advantage of TCP performance enhancement, MDP is in general, while the CFDP reliable transmission protocol model can well adapt to this environment.
-
Computer Engineering. 2010, 36(15): 17-19. https://doi.org/10.3969/j.issn.1000-3428.2010.15.007This paper proposes a simple and fast failure detection mechanism——the loopback detection mechanism for Multi-Protocol Label Switching(MPLS) network failure. The “OAM loopback module” is added into the MPLS protocol module, and the implementation and deployment are very simple. The detection packets are transferred below the MPLS level, and the detection operations are executed fast. The detection methods with fixed period and dynamic period are both provided. The mechanism is implemented on the Linux platform and a prototype named LSP Fast Detecting(LSPFD) is developed. Experiments show that the mechanism works effectively.
-
Computer Engineering. 2010, 36(15): 20-22. https://doi.org/10.3969/j.issn.1000-3428.2010.15.008This paper uses machine learning techniques combined with the rules to solve sentiment classification of news text, and researches affection of feature selection and feature weights based on Support Vector Machine(SVM) classifier. Experiments show that combined with the rules, experimental result is improved. In order to test universality of the combined method, more experiments based on KNN and Bayes classifier are done. Results show that combined method does better than not combined ones.
-
Computer Engineering. 2010, 36(15): 23-24,28. https://doi.org/10.3969/j.issn.1000-3428.2010.15.009TCP Friendly Rate Control(TFRC) protocol over wireless networks has poor performance because it can not judge the reason of packet loss exactly. To solve this problem, this paper proposes a step marking method based on Explicit Congestion Notification(ECN) and designs a new wireless stream media transformation protocol. Simulations show that, with appropriate loss differentiation and congestion notification, the new protocol achieves high throughput while keeps reasonable friendliness to TCP flows.
-
Computer Engineering. 2010, 36(15): 25-28. https://doi.org/10.3969/j.issn.1000-3428.2010.15.010Aiming at these features of Web service environment such as open, deceptive and uncertain, a Web service trust evaluation model——WSTrustM based on the fuzzy theory is proposed. The model uses the fuzzy theory method to measure trust and combination of weight distribution of trust evaluation of factors, and uses membership degree of fuzzy sets theory to divide grade of entities trust relation. Simulation experimental result shows that the model on the evaluation of Web service trust is more accurate and more reliable, reduces the interaction risk and ensures the success rate of interaction.
-
Computer Engineering. 2010, 36(15): 29-31. https://doi.org/10.3969/j.issn.1000-3428.2010.15.011A latency hiding algorithm, called HASH-Latency Hiding(HASH-LH), is promoted. By leveraging the streaming process characteristics of the Hash algorithms used for P2P file integrity verification and the asynchronous receiving buffer of TCP, it is reasonable for HASH-LH to divide a big file block into multiple smaller sub-blocks, which makes the latency much shorter for calculating Hash values on each sub-block. Furthermore, the Hash calculation starts just after the first sub-block is received. As the latency of each sub-block Hash calculation becomes as short as it can be hidden by the asynchronous receiving buffer of TCP, it makes calculation and receiving processed nearly at the same time. As a result, all the latency from file verification is hidden and performance boosts up.
-
Computer Engineering. 2010, 36(15): 32-35. https://doi.org/10.3969/j.issn.1000-3428.2010.15.012In the deployment of real time Domain Name System(DNS) monitoring and analysis system, the large volume CN top level domain DNS log heavily burdens its transmission from the distributed sites to the data processing center. The redundancy of DNS query type and the repetitiveness of DNS query time, IP and domain name are utilized for the DNS log compression to facilitate its transmission. For IP and domain name compression, the algorithm relies on their concentration property as well as their time and space locality.
-
Computer Engineering. 2010, 36(15): 36-39. https://doi.org/10.3969/j.issn.1000-3428.2010.15.013After introducing some current research about configurable index and ontology-based configuration, this paper presents a configurable information index method based on semantic mapping to solve the problems of building indexes in multi-source environment and storing the huge index data in the dynamics distributed structure. The rules of the index method are generated by protégé, and result shows the configurable and expansibility of this method.
-
Computer Engineering. 2010, 36(15): 40-42,48. https://doi.org/10.3969/j.issn.1000-3428.2010.15.014To effectively deal with the document clustering problem, a novel document clustering algorithm based on marginal manifold learning is proposed. The high dimensional document space is reduced into the lower dimensional feature space with marginal fisher analysis. The support vector clustering algorithm is applied to cluster documents herein. Experimental results on the benchmark document sets show the algorithm achieves much better clustering performance than tradition document clustering algorithms.
-
Computer Engineering. 2010, 36(15): 43-45,48. https://doi.org/10.3969/j.issn.1000-3428.2010.15.015To resolve the question of insufficient consideration for users’ preference Quality of Service(QoS) in the research of service grid resource mapping mechanism, this paper studies a mapping algorithm that ensures the multi-dimension QoS requests of users. After analyzing the multiple QoS properties of the grid resources, a method to screen out the correct grid resources which hits to the spot of QoS restricts of sub-tasks is presented. The corresponding mapping result and a QoS-Sufferage algorithm are put forward. This QoS-Sufferage algorithm is proved to be effective by the simulations taken on the GridSim platform.
-
Computer Engineering. 2010, 36(15): 46-48. https://doi.org/10.3969/j.issn.1000-3428.2010.15.016Aiming at the analyzing process of software binary patch, this paper proposes an instruction normalization algorithm to distinguish whether the corresponding basic blocks are reordered. Compared with the normally small prime algorithm, this algorithm is not limited to instruction numbers and does not fail to report the difference information. It has high efficiency and good effect. Initial experiments show that this algorithm can reduce the manual work to analyze difference information and help reverse engineering person quickly locate software difference.
-
Computer Engineering. 2010, 36(15): 49-50,54. https://doi.org/10.3969/j.issn.1000-3428.2010.15.017In order to support various queries for XML data, a double index method is proposed. Using the reverse order, an absolute index model and a relative one are built. A query algorithm is presented. The absolute index model reduces the number of comparison by shortening the path expressions. The relative one completes the path expressions by setting up parent-child index table and replaces original queries with small index structure. Experimental results to DBLP dataset show the method works well.
-
Computer Engineering. 2010, 36(15): 51-54. https://doi.org/10.3969/j.issn.1000-3428.2010.15.018Embedded software of telecommunication area must deal with increasing communicating businesses under restricted hardware environment, and only depending on the task mechanism of commercial embedded operating system can’t supply enough parallel processing ability and throughput any more. To solve the problem, this paper proposes a new solution scheme, in which the smaller executing units called process are designed and scheduled based on tasks. Compared with tasks, processes occupy less memory and cost less time to be switched, so system can support large number of running processes. It provides effective ways on monitoring system resources and diagnosing system failures, which brings robustness.
-
Computer Engineering. 2010, 36(15): 55-56. https://doi.org/10.3969/j.issn.1000-3428.2010.15.019An automatic program flow analysis method is presented, which uses static single assignment to simplify data dependence relation in program slicing, and uses a simple and fast program slicing algorithm to delete statements and control predicate which has no effect on loop control, and uses abstract interpretation to get precise program flow information automatically. Experimental results show this method is more fast than the normal method up to 25%, while keeping precision. Moreover, there is no any assume on program form, so the flow analysis method is suitable for any program form.
-
Computer Engineering. 2010, 36(15): 57-59. https://doi.org/10.3969/j.issn.1000-3428.2010.15.020Aiming at the function alternative model of the periodic transaction, Conflict-free Concurrency Control Protocol for Supporting Periodic transaction(SP-CCCP) which is based on CCCP, adds the compatible test between the new transaction and its future active periodic transaction with higher priority, and improves the performance of compatible test between the new transaction and the current active transaction. SP-CCCP selects the compatible alternative which has no conflict with other transactions, and improves success ratio of the system transaction and utilization ratio of the CPU. Simulation results show that in most situations, the performance of SP-CCCP is better than CCCP.
-
Computer Engineering. 2010, 36(15): 60-62,65. https://doi.org/10.3969/j.issn.1000-3428.2010.15.021This paper improves the hierarchical and k-means clustering, builds the two-stage clustering method. It gets the outliers and high-quality clustering results by the two-stage clustering method. It uses Clementine to model the process of realization of clustering by clients’ real transaction records in securities companies and manual data, identifies outliers and calculates the suspicious degree of the records, and provides high-quality survey data for the financial intelligence departments.
-
Computer Engineering. 2010, 36(15): 63-65. https://doi.org/10.3969/j.issn.1000-3428.2010.15.022The duration of cache writing is always long in the segment-based cache. To holding the cache proxy server performance, this paper limits the writing bandwidth by judge method of writing bandwidth and introduces several judge methods. Testing results show that these methods have the capability of keeping cache effect and reducing writing workload.
-
Computer Engineering. 2010, 36(15): 66-67,71. https://doi.org/10.3969/j.issn.1000-3428.2010.15.023To find out the important materials automatically in supply chain Material Requirement Planning(MRP), this paper proposes a material requirement planning method based on data field and cloud model. The data field clustering method is based on the distribution of equipotential line (surface) of the natural nested data structures in the field and the aggregation properties of self-organization among data objects to achieve the level division. By means of data field, the materials can be clustered automatically and find out different kinds of classifications in material requirement planning. Cloud model can reflect the uncertainty of clustering knowledge, especially the randomness and fuzziness. An actual case study and its results analysis show that this approach is feasible in the area of supply chain material requirement planning.
-
Computer Engineering. 2010, 36(15): 68-71. https://doi.org/10.3969/j.issn.1000-3428.2010.15.024According to workflow constraint problem existing in the process of sending mail and requesting service among different business system, this paper presents a Workflow and Service oriented Multithread E-mail System model(WSMES). Introducing mail work flow unit into Service Oriented Architecture(SOA), the model uses semaphore and thread pool to realize ordered and granulated service composition between mail service and workflow.
-
Computer Engineering. 2010, 36(15): 72-73,76. https://doi.org/10.3969/j.issn.1000-3428.2010.15.025Java language standard does not provide any message, which makes message mechanism in object-oriented model hard to implement. This paper uses interface, inheritance and polymorphism of Java to implement a message mechanism for object model, which is similar to that of framework in Microsoft Foundation Class(MFC). The mechanism is applied in a collaborative intelligent multi-robot system based on LEGO robot, and the result proves it is reasonable, effective and easy to implement.
-
Computer Engineering. 2010, 36(15): 74-76. https://doi.org/10.3969/j.issn.1000-3428.2010.15.026Aiming at the security problems of Windows terminal, a schema of security Windows terminal which is called main and accessorial dual-processors and double operating systems is proposed. The overall construct and hardware architecture of the schema are described, software structure are designed. The design and implementation of key technologies such as device management, system administration and security control are analyzed. The key points include the main and accessorial kernel co-operating management of devices, the design of multi internal channel and translation of network data.
-
Computer Engineering. 2010, 36(15): 77-79. https://doi.org/10.3969/j.issn.1000-3428.2010.15.027The traditional rough set theory is based on the set theory. It can be found that upper and lower approximation operators have some shortcomings in the description of function. This paper puts forward the definition of upper and lower rough function in view of scale, which can confirm strict partition of a monotone real function defined in the integer, then form rough function model of real domain. It establishes the corresponding Galois concept lattice in light of the analysis of rough function model. It simplifies the property of lattice concept by discernibility matrix.
-
Computer Engineering. 2010, 36(15): 80-82,85. https://doi.org/10.3969/j.issn.1000-3428.2010.15.028In the distributed environment, the high frequency of remote data access leads to a decline in system performance. Aiming at this problem, this paper improves the SBM model, and designs a Distributed Cache system(D-Cache). A new cache replacement algorithm, OCV is given. Digital simulation results demonstrate that D-Cache system can shorten the response time of system and increase the throughput of system more effectively, which improves performance in the distributed environment, compared with the distributed cache system DartCache.
-
Computer Engineering. 2010, 36(15): 83-85. https://doi.org/10.3969/j.issn.1000-3428.2010.15.029Aiming at the problem of multi-label text classification, this paper proposes a multi-label text classification algorithm based on frequent item sets. It uses FP-growth algorithm for mining frequent item sets between labels, calculates prototype vector and similarity threshold for each class, if the similarity between prototype vector and text are greater than the corresponding threshold, then classifies the text into corresponding category. After classifying, the association rules between the class are utilized to verify the result of classification. Experimental results show that the algorithm has a higher ability of classification performance.
-
Computer Engineering. 2010, 36(15): 86-88. https://doi.org/10.3969/j.issn.1000-3428.2010.15.030This paper proposes a generation method of path coverage test based on Gene Expression Programming(GEP) algorithm, compiler technology and Virtual Machine(VM). The static control flow paths of the program are generated by the compiler, the test cases covering every specific path are generated by GEP algorithm, and the results are calculated by VM, of which the effective one is the case needed. Experimental result proves that the test cases covering the specific expected control flow paths can be calculated efficiently and accurately with the method.
-
Computer Engineering. 2010, 36(15): 89-91. https://doi.org/10.3969/j.issn.1000-3428.2010.15.031This paper applies scenario patterns to design and generate embedded software test cases. By analyzing and modeling the requirements of the system under test, the scenario models constructed are classified into different patterns. The state diagrams of test scenarios are created according to scenario patterns. Test executing paths are derived by traversing the state diagrams, and test cases are designed and generated from test executing paths and test data. Experimental result proves that the method can design and generate test cases rapidly, enhance the reusability of test cases, and realize efficient test for embedded software.
-
Computer Engineering. 2010, 36(15): 92-93,96. https://doi.org/10.3969/j.issn.1000-3428.2010.15.032According to the relationship between the interface parameters of the tested system, this paper studies the relation between interface set and bipartite graph, proposes the definition, classification and properties related to interface set, and proves these properties. It presents a graph-based test suite optimization algorithm for test suite reduction and optimization according to the properties. Through the analysis of a case, the approach can reduce the size of test suite without reducing the coverage rate of test suite and enhance the efficiency of test.
-
Computer Engineering. 2010, 36(15): 94-96. https://doi.org/10.3969/j.issn.1000-3428.2010.15.033In some applications with intensive writing, Flash becomes invalid earlier than expected if SQLite writes database file too frequently. In order to solve the problem, by analyzing the transaction processing procedure and auto-commit mode of SQLite, this paper uses the method of delay commit to reduce the frequency of erasing and writing Flash. It divides data into the key or un-reproducible data and reproducible data, and treats them with appropriate methods, which ensures the security of data.
-
Computer Engineering. 2010, 36(15): 97-99,102. https://doi.org/10.3969/j.issn.1000-3428.2010.15.034According to the demand of the validity period of data in application system, this paper proposes a real-time schedule model based on producer-consumer model by using model-based software design method. The model relaxes the deadline restriction for the task period in Earliest Deadline First(EDF) algorithm to resolve the problem of precedence constrain in the model. Real-time schedule simulation program designed can truly reflect the schedule situation of the tasks.
-
Computer Engineering. 2010, 36(15): 100-102. https://doi.org/10.3969/j.issn.1000-3428.2010.15.035The paper introduces a routing algorithm, which supports publish/subscribe system in Ad Hoc social networks—SocialCast, and proposes some improvements which aims at some shortcoming of SocialCast. Improved algorithm can take full advantage of the characteristics of social networks to improve the routing performance for the calculation of utilities, reduce the redundant transmission, improve transmission efficiency for the message delivery, and reduce the burden on publishers to facilitate the transmission of information for the message release.
-
Computer Engineering. 2010, 36(15): 103-104,107. https://doi.org/10.3969/j.issn.1000-3428.2010.15.036Aiming at the problem that single node network card throughput can not satisfy the bandwidth of the tested link, based on TCP protocol, this paper outputs high bandwidth load for tested system by using multiple nodes to parallel generate network flow, and goes on bandwidth test for link. Experiment validates the feasibility of this parallel measurement method, and gets the main factors which are TCP window size and buffer size in TCP protocol by analyzing.
-
Computer Engineering. 2010, 36(15): 105-107. https://doi.org/10.3969/j.issn.1000-3428.2010.15.037This paper presents a lower layer full interconnection rate to evaluate network performance based on the principle of computing local property. It uses this criterion to go on performance evaluation about computing speed and physical layout of Direct Interconnection Network(DIN) of Triplet Based computer Architecture(TriBA). Result shows that lower layer full interconnection structure of TriBA can reduce network complexity and enhance communication performance, diagonal VLSI layout of TriBA saves about 11% power compared to 2D Mesh structure, this advantage of power consumption is in large scale Multi Processor System on Chip(MPSoC).
-
Computer Engineering. 2010, 36(15): 108-110. https://doi.org/10.3969/j.issn.1000-3428.2010.15.038Aiming at the network heterogeneity and node group mobility in tactical communication, based on building group movement model, this paper uses cluster-head stable clustering algorithm to dynamically divide and manage hierarchical structure in heterogeneous network. It presents hierarchical routing strategy based on link state to integrate heterogeneous wireless resource. The strategy consists of intra-group routing strategy and inter-group routing strategy, adopts different routing strategies for hierarchical network structure. OPNET simulation results show that this strategy has better performance of group success delivery ratio and average end-to-end delay over hierarchical AODV.
-
Computer Engineering. 2010, 36(15): 111-113. https://doi.org/10.3969/j.issn.1000-3428.2010.15.039For the characteristics of short path life and network disconnection in Vehicular Ad-hoc NETwork(VANET), this paper designs a low latency routing protocol. This protocol sets up a delay model based on vehicle density, divides the routing into two parts of road routing and intersection routing. The protocol chooses minimum delay path to forward in intersection routing. Experimental results show that this protocol can reach high group delivery ratio and low transmission delay in city disconnected VANET.
-
Computer Engineering. 2010, 36(15): 114-116. https://doi.org/10.3969/j.issn.1000-3428.2010.15.040Aiming at the structural property of layered clustering network, this paper proposes a WSN clustering protocol based on connectivity coverage. It makes use of maximum entropy principle to select temporary cluster head by pre-clustering for whole network, and selects Connected Dominating Set(CDS) of active nodes in each cluster under the premise of coverage connectivity, and selects real and substitute cluster head by optimizing and updating temporary node to realize communication between intra-cluster and inter-cluster. Simulation result shows that this protocol can improve the connectivity coverage performance, balance energy in clustering, and achieve strong robustness compared with HEED and SPAN protocol.
-
Computer Engineering. 2010, 36(15): 117-119. https://doi.org/10.3969/j.issn.1000-3428.2010.15.041In the field of wireless sensor network self-localization, positioning accuracy of sensor node is directly related to the collection of data availability. Concerning the weak point of node average localization error, based on DV-Hop localization algorithm of the least square method principle, method of algebra reconstruction techniques is introduced. Core idea of the algorithm is used as successive iterative algorithm from the problem of image reconstruction, algebra reconstruction. Simulation result proves that the improved algorithm a obviously reduces the average localization error of the node in the Wireless Sensor Network(WSN).
-
Computer Engineering. 2010, 36(15): 120-122. https://doi.org/10.3969/j.issn.1000-3428.2010.15.042An energy efficient clustering routing algorithm(EERA) for Wireless Sensor Networks(WSN) is proposed for the “hot spot” problem. The entire region is divided into unequal radial level with the base station at the center. Cluster head is chosen based on the residual energy and the relative position. Multi-hop data routing is formed in inter-cluster, routing to build not only considers the energy consumption of nodes send data, but also takes care of forward nodes to receive energy consumption. The distance of the data sentting directly is limited to within the threshold at the same time minimizing the number of data transfer. And at the stage of stable transmission, each cluster head uses dynamic routing to avoid the relay node rapid death due to excessive energy consumption. Simulations results show that EERA can effectively reduce network energy consumption, balance the energy consumption of network nodes, and extend the network life period.
-
Computer Engineering. 2010, 36(15): 123-125,128. https://doi.org/10.3969/j.issn.1000-3428.2010.15.043This paper proposes a Link-Lifetime based Multipath Source Routing(LTMSR) protocol for multimedia traffic over Mobile Ad Hoc Networks(MANETs). Two different classes of routing discovery processes are deployed for multimedia traffic and non-multimedia traffic respectively. The former is assigned with a higher possibility to discover link-disjoint multipath than the latter. The routing protocol chooses route for packets in consideration of path lifetime to prevent from transmitting over already broken-routes. The link-broken detection mechanism distinguishes the cause of a lost frame properly, and coordinates with the routing protocol to avoid unnecessary route updating. Simulation results demonstrate that the LTMSR improves the quality of multimedia playing significantly, in terms of decodable frame ratio and end-to-end frame delay. In addition, route overhead is reduced.
-
Computer Engineering. 2010, 36(15): 126-128. https://doi.org/10.3969/j.issn.1000-3428.2010.15.044In view of node heterogeneity and large transmission delay problems of current application layer multicast, an application layer multicast algorithm is proposed based on topology architecture and peer performance. Network will be layered and clustered according to the topology-aware approach. It can fully consider node heterogeneity when constructing the multicast tree. The algorithm makes the high performance peer close to the media source and the low performance one close to the leaf, which improves the efficiency of the multicast tree and decreases the delay of peer-to-peer, link stress, link stretch and control overhead. Experimental results show that algorithm can efficiently improve the multicast tree performance on large scale peer conditions.
-
Computer Engineering. 2010, 36(15): 129-130,133. https://doi.org/10.3969/j.issn.1000-3428.2010.15.045A new active/passive sensor management algorithm based on KL distance is proposed, in which only one sensor works each frame. For getting the better tracking accuracy, the optimal sensor is selected based on the KL distance. Particle filtering method is used to calculate the KL distance and identification of motive model is added to improve accuracy. Simulation result shows the proposed method can achieve the desired tracking accuracy compared with the traditional management method based on covariance control.
-
Computer Engineering. 2010, 36(15): 131-133. https://doi.org/10.3969/j.issn.1000-3428.2010.15.046Abstraction and description of attribute, attribute predicate and attribute-value pair are studied, and a policy model is proposed for Attribute-Based Access Control(ABAC) with formalization of policy and policy evaluation. Access control decision is described under policy combining algorithm and default authorization, an improved policy framework is devised and tested by simulation. Experimental result shows the framework has high scalability, which provides a basis for implementation of ABAC.
-
Computer Engineering. 2010, 36(15): 134-136,139. https://doi.org/10.3969/j.issn.1000-3428.2010.15.047To research and analyze the theory of framework for modeling trojans and trojans model about cooperative concealment, this paper gives BIOS Rootkit’s formal model about cooperative and presents a BIOS Rootkit detection technology based on the character of cooperation in Windows environment. Current detection technology of BIOS Rootkit can detect but lacks the function of resuming, so the method uses characteristic of codes searching and dynamic resuming to solving it. Experimental results show that this is a detection method of higher reliability, efficiency and integrity.
-
Computer Engineering. 2010, 36(15): 137-139. https://doi.org/10.3969/j.issn.1000-3428.2010.15.048In order to strengthen the confidentiality of data, this paper proposes a kind of Kernel-level Cryptographic File System(KCFS) combining the optimized data read/write flow. The kernel-level cryptographic file system can overcome the inconvenience of encryption application and the low efficiency of user-level encryption file system, so as to provide the encryption/decryption function in the kernel-level file system, protecting the data stored in the storage system. The result of comparative experiment shows that KCFS can reach better read/write performance compared to user-level encryption file, such as CFS.
-
Computer Engineering. 2010, 36(15): 140-141,144. https://doi.org/10.3969/j.issn.1000-3428.2010.15.049Using too many or too too few characters(variable) in Intrusion Detection System(IDS) leads to reduce recognizing correctness of IDS. To resolve the contradiction and to improve the whole performance of IDS, an approach of choosing optimal characters used to IDS is presented. With the approach, new characters made of original characters, “contributions” of new characters for recognizing intruders are computed, and the characters with larger “contributions” value are chosen as the characters of IDS. Number of the characters used to IDS is reduced, and the information belonging to original characters are kept largely to improve recognizing correctness. The characters with larger “contributions” are optimal characters. Tests show that the approach is useful.
-
Computer Engineering. 2010, 36(15): 142-144. https://doi.org/10.3969/j.issn.1000-3428.2010.15.050The security of RSA is designed based on the difficulty of large integer decomposition. In the RSA cryptanalysis, according to RSA public key encryption characteristics that the public key n as the product of two large prime numbers, contrary to the form n=pq(in which p, q as large prime numbers) of the large integer n decomposition. A determine algorithm of large integer decomposition on RSA cryptanalysis is given. The decomposition algorithm is proved by mathematical method, the corresponding algorithm design is given, the complexity of the algorithm is made below O(plogn), n prime factor characteristics and the relationship between the effectiveness of the decomposition, as well as RSA safety impacts are analyzed.
-
Computer Engineering. 2010, 36(15): 145-147. https://doi.org/10.3969/j.issn.1000-3428.2010.15.051Based on the digital watermark and information theory, a rudimental research for the Gaussian watermark capacity is given. Considering that Gaussian distribution has the maximum uncertainty with the maximum entropy among all given second-order distributions, this paper focuses on the watermark communication process with Gaussian channel, and uses the average mutual information to analyze the communication capacity. Also the additive noise channel is introduced to expand the non-Gaussian case. The formula of maximum transmission rate for Gaussian watermark channels is given and simplified, and the 2D and 3D curves for the relationship between the distortion and channel capacity by Matlab are illustrated. It analyses the results at different distortions in two different usages of watermark.
-
Computer Engineering. 2010, 36(15): 148-150,155. https://doi.org/10.3969/j.issn.1000-3428.2010.15.052This paper analyses the security problems existed in low-cost RFID system, and presents a security protocol based on PUF. The protocol uses Physically Unclonable Functions(PUF) to identify the tag, and uses the Linear Feedback Shift Register(LFSR) to generate random series, encryptions the communication between the reader and tag, so it can resist replay attack, tracking attack, physical attack, eavesdropping attack and other attacks. Using the FPGA, PUF and LFSR are implied on the Altera DE2 board. The FPGA is programmed through Quartus II 8.0. Experimental result shows that the presented protocol meets the requirement on processing time and the number of gates on low-cost tags.
-
Computer Engineering. 2010, 36(15): 151-152,155. https://doi.org/10.3969/j.issn.1000-3428.2010.15.053According to the technology for realizing chaos communication, a scheme for the implementation of voice chaotic secrecy communication is proposed by means of the first type piecewise-linear Lorenz system and drive-response method, including utilizing digital processing technology, continuous time system discretization, and the system design principle of the scheme by using 5509 series digital signal processors. Theoretical analysis and hardware implementation results demonstrate the effectiveness of the presented scheme.
-
Computer Engineering. 2010, 36(15): 153-155. https://doi.org/10.3969/j.issn.1000-3428.2010.15.054To solve the problem of data distribution and management in the distributed environment, this paper proposes a secure communication model based on smart card and password authentication. The model is also analyzed from the point of view of adversary attacks and oneself security. Results show that the model not only guarantees the integrality, non-repudiation, and authentication, but also realizes the confidentiality during the data transmitting.
-
Computer Engineering. 2010, 36(15): 156-158. https://doi.org/10.3969/j.issn.1000-3428.2010.15.055Aiming at the problem of traditional honeypot have shortcomings of low camouflage and low interaction, a high camouflage and highly interactive honeypot system is designed, including data processing layer, camouflage layer, data acquisition layer, data redirection layer and cluster servers. By the technology of ARP agent technology, virtual routing technology, protocol processing technology of libcap realizing, data direction technology, the system receives routing camouflage, IP address camouflage, operating system camouflage, highly interactive services camouflage, and hole spoof.
-
Computer Engineering. 2010, 36(15): 159-161. https://doi.org/10.3969/j.issn.1000-3428.2010.15.056MDAV(Maximum Distance to Average Vector) algorithm is an efficient microaggregation algorithm. However, it does not capture diversity of sensitive values in each equivalence class, so the anonymity table generated by the algorithm cannot resist homogeneity attack and background knowledge attack. To solve the problem, the paper proposes a (k,e)-MDAV algorithm. The algorithm groups at least the k nearest tuples to cluster center into one cluster, and further requires the range of the distinct values in one cluster to be no less than e. Experimental results show the algorithm can generate anonymity table satisfying (k,e) anonymity model efficiently.
-
Computer Engineering. 2010, 36(15): 162-163,167. https://doi.org/10.3969/j.issn.1000-3428.2010.15.057Aiming at the lack of malicious nodes order in unfair recommendation filtering algorithm, through the judgment which is important factor in final trust value, this paper finds malicious nodes, the greater the standard deviation of malicious nodes, the greater of impact final trust value, gives an order of filtering malicious nodes according to the standard deviation. The algorithm is compared with algorithm of Mui et al, when there are a large number of malicious recommendations, the improved algorithm can better filter out unfair recommendation trust.
-
Computer Engineering. 2010, 36(15): 164-167. https://doi.org/10.3969/j.issn.1000-3428.2010.15.058Aiming at the shortness of single node moving mode, poor flexibility and real-time in common stretching tree, this paper proposes a model of intellectualized stretching tree based on biological memory principles. The model establishes corresponding math model through the short memory cell’s characteristics of decreasing and updating and the relationship of short memory, long memory and comprehensive memory. The node accessing range reflects the step number of node forward moving. Experimental result shows that intellectualized stretching tree can track the latest development in real time, dynamic character and has application value to some degree.
-
Computer Engineering. 2010, 36(15): 168-170. https://doi.org/10.3969/j.issn.1000-3428.2010.15.059Relevance Vector Machine(RVM) classification method uses the probabilistic output to overcome Support Vector Machine(SVM) shortage as well as has more sparsity. Whereas RVM has overloaded computation complexity and memory storage when it is applied for the text-independent speaker identification because of the mass training samples. For solving the problem, a hybrid GMM and RVM approach is proposed which effectively abstracts the speaker feature vector as well as solve the mass storage problem. Further more, this hybrid approach combines the robustness of generative model and the powerful classification of discriminative model to improve the performance and robustness of identification. Experimental results prove that the method has better error identification rate than the GMM system and more sparsity than state-of- the-art GMM/SVM system.
-
Computer Engineering. 2010, 36(15): 171-173. https://doi.org/10.3969/j.issn.1000-3428.2010.15.060Through the study on the glass-block cutting problem, a new hybrid algorithm of Quantum-behaved Particle Swarm Optimization and Ant Colony Optimization(QPSO-ACO algorithm) is proposed. The algorithm modifies the model of QPSO and ACO to solve Traveling Salesman Problem(TSP) in glass-block cutting. It makes full use of the positive feedback mechanism and high solution efficiency of ACO, as well as the fast convergence of QPSO. Experimental results show that QPSO-ACO algorithm has stronger optimization ability in solving the glass-block cutting problem.
-
Computer Engineering. 2010, 36(15): 174-175. https://doi.org/10.3969/j.issn.1000-3428.2010.15.061In order to improve the performance of detector generation algorithm, an improved algorithm is proposed. Coverage estimates are used as the control parameters of the algorithm, and a candidate detector generation mechanism of combination random with gene lib is presented. It ensures diversity of detector while increasing the efficiency of the detector maturity. Experiments of synthetic and real data sets results show that the algorithm has low false alarm rate and reduces the number of required detectors while the detection rate is the same.
-
Computer Engineering. 2010, 36(15): 176-178. https://doi.org/10.3969/j.issn.1000-3428.2010.15.062According to the problem of effective extraction of facial expression information in space-time domain, one kind facial recognition method based on multi-frequency Local Binary Patterns from Three Orthogonal Panels(LBP-TOP) features and Maximal-margin Spherical- structured Support Vector Machine(MSSVM) is proposed. It adapts wavelet decomposition to enhance information of each frame in image sequence. It extracts improved LBP-TOP features of sub-images on the same frequency. MSSVM is applied for sample training and recognition. Experimental result indicates that, this method can extract movement expression feature more effectively, as well as recognition rate is better, and it meets the requirement of real-time.
-
Computer Engineering. 2010, 36(15): 179-181. https://doi.org/10.3969/j.issn.1000-3428.2010.15.063For the problem whether Graphic Processing Unit(GPU), the stream processor with high performance of floating-point computing is applicable to neural networks, this paper proposes the parallel recognition algorithm of Convolutional Neural Networks(CNNs). It adopts Compute Unified Device Architecture(CUDA) technology, definites the parallel data structures, and describes the mapping mechanism for computing tasks on CUDA. It compares the parallel recognition algorithm achieved on GPU of GTX200 hardware architecture with the serial algorithm on CPU. It improves speed by nearly 60 times. Result shows that GPU based the stream processor architecture are more applicable to some related applications about neural networks than CPU.
-
Computer Engineering. 2010, 36(15): 182-184. https://doi.org/10.3969/j.issn.1000-3428.2010.15.064This paper presents an efficient simulated annealing Genetic Algorithm(GA) for the constrained Airport Gate Assignment Problem (AGAP). The algorithm encodes gate-allocations as vectors whose element of it represents the gate allocated to the flight. Crossover operator and mutation operator are designed, and offspring generated by crossover and mutation is feasible. Simulated annealing algorithm and mutation operator are parallelly executed to each individual in subpopulation generated by crossover to increase abilities of local search of the algorithm. The convergence of the proposed algorithm is studied. Experimental results show the performance of the proposed algorithm is better than that of the other search and the genetic algorithm.
-
Computer Engineering. 2010, 36(15): 185-187,190. https://doi.org/10.3969/j.issn.1000-3428.2010.15.065The ?-insensive Smooth Support Vector Regression(?-SSVR) is solved by a fast iterative algorithm, which improves the performance and efficiency of regression, but the problem of convergence remains still. In the present paper, ?-SSVR’s global convergence for an arbitrary given penalty parameter is proved by the method of set theory, and the upper bound of the convergence is worked out, which provides the smooth support vector machine with a basic theoretical support.
-
Computer Engineering. 2010, 36(15): 188-190. https://doi.org/10.3969/j.issn.1000-3428.2010.15.066To solve the problem of high time complexity and high learning complexity of traditional AdaBoost algorithms, this paper puts forward an improved algorithm named AdaBoostFISP. It uses fixed increment single sample perceptron as weak learners for AdaBoost, and applies fixed increment instead of variable increment in weight updata of perceptron, so that the complexity of time and learning is decreased. Experimental results demonstrate that the algorithm achieves better performance in prediction accuracy, learning complexity and time complexity compared with other AdaBoost algorithms.
-
Computer Engineering. 2010, 36(15): 191-193. https://doi.org/10.3969/j.issn.1000-3428.2010.15.067Aiming at the problem of eye tracking under the head movement varied in the process of drowsy driving detection, this paper proposes an improved Mean-shift algorithm. By using pixel gray value distribution feature and the density distribution feature of target gradient direction, it performs an alternate iterated operation for the two characteristics iteration formula, and carries the rotation and translation movement tracking better for the moving target, so the method can improve the eye tracking under the head rotation in depth. Experimental results show that the method can effectively track and position the human eye under multi-pose head movement.
-
Computer Engineering. 2010, 36(15): 194-195. https://doi.org/10.3969/j.issn.1000-3428.2010.15.068In order to detect edges from noisy images, a method for edge detection based on fuzzy reasoning is proposed. In the method, a fuzzy reasoning system is constructed, basic gradients and the correlated gradients are considered as the inputs of the fuzzy reasoning system to determine noisy membership degrees for candidates for edges. Experimental results show more edge information is observed and more noise is suppressed by the method compared with recent edge detection methods.
-
Computer Engineering. 2010, 36(15): 196-198. https://doi.org/10.3969/j.issn.1000-3428.2010.15.069Synthetic Aperture Radar(SAR) image mosaic effect evaluation can be used to evaluate mosaic effect and mosaic algorithms. Aiming at the registration effect and mosaic-sew elimination effect of mosaic image, this paper proposes a new evaluation criteria based on structural similarity. The evaluation experiments of different kinds of registration algorithms and fusion algorithms are made to demonstrate that the evaluation criteria are effective.
-
Computer Engineering. 2010, 36(15): 199-201. https://doi.org/10.3969/j.issn.1000-3428.2010.15.070Aiming at the problem of large number of terrain data and low speed of terrain rendering, this paper presents a terrain rendering method based on data block and enclosing ball error metric. It organizes data block partition, indexes data blocking based on the order of row and line, implements the local loading of terrain data. It implements the real-time loading of terrain data by visibility estimation. For the data blocking in view field, an error metric based on enclosing ball is designed. The metric constructs Level Of Detail(LOD) model by using triangle bin-tree. The terrain rendering method implements large scale terrain real-time rendering. Experimental results show that the method can achieve a high frame speed and a good visual effect.
-
Computer Engineering. 2010, 36(15): 202-203. https://doi.org/10.3969/j.issn.1000-3428.2010.15.071According to low-frequency domain average method, this paper proposes a multifocus wavelet image fusion method based on low-frequency edge characteristics and energy. It adopts wavelet multi-scale decomposition in the image needs fusion. High-frequency details of the components of image are fused according to region energy maximization principle, while the low-frequency approximation component of image edge characteristics are combined with the energy of the fusion method. Experimental results show that compared with existing methods, this method can embody low frequency domain edge characteristics and regional characteristics better.
-
Computer Engineering. 2010, 36(15): 204-205,210. https://doi.org/10.3969/j.issn.1000-3428.2010.15.072A new Lucy-Richardson(LR) image restoration algorithm is proposed to avoid the noise amplification problem of the original LR algorithm.The problem is solved via regular method for separating the residual noise in the original LR iterative process and Nonsubsampled Contourlet Transform(NSCT) domain local neighborhood correlation denoising method. Comparative experimental results show the algorithm can obtain satisfactory restoration visual effects.
-
Computer Engineering. 2010, 36(15): 206-207,210. https://doi.org/10.3969/j.issn.1000-3428.2010.15.073This paper presents techniques and results on semantic annotation of sport video with active learning. It takes tennis video as examples and defines 6 scenes containing semantic concept. The mid-level features such as court line, court color percentage, domain motion, camera move pattern etc are extracted through analyzing the visual difference of semantic scenes. The Support Vector Machine(SVM) is used to classify the shots. The proposed method employs a clustering based active learning scheme. It performs most representative samples selection as well as to avoid repeatedly labeling samples in the same cluster. Experimental results show encouraging result compared with the traditional methods.
-
Computer Engineering. 2010, 36(15): 208-210. https://doi.org/10.3969/j.issn.1000-3428.2010.15.074In order to obtain accurate model parameters for rate control algorithm in H.264, rate point selection and model parameters estimation are improved for rate control algorithm in H.264. This paper utilizes a strategy based on Manhattan and 2D sliding window mechanism to implement efficient rate control algorithm. It applies a weighted linear regression technology to actualize effective model parameter prediction. Experimental results show that the algorithm can improve the quality of video encoding and increase the control precision of the output rate.
-
Computer Engineering. 2010, 36(15): 211-213. https://doi.org/10.3969/j.issn.1000-3428.2010.15.075Video segmentation and video index favour video browsing and retrieval. This paper proposes a video segmentation and index method based on integration of the film script and subtitles. The method obtains the initial video clips’ boundary of different scenes of the video through matching the subtitles and the dialogue text of the scene in the film script. It determines the final boundary of different scenes of the video by analyzing the color distribution difference between video frames. Experimental result shows that the method enhances the video segmentation precision and speed compared with the methods completely depending on content-based analyzing measure.
-
Computer Engineering. 2010, 36(15): 214-215. https://doi.org/10.3969/j.issn.1000-3428.2010.15.076In this paper, a new motion estimation algorithm based on multi-template switching according to prediction direction is presented. It uses vectors’ sum of optimal point direction and movement direction to forecast the next movement direction and switches templates real-timely. During the realization process, in order to reduce search time, looking-up table method is used to achieve template switching and the threshold is set to exit. Text results indicate that this algorithm is effective.
-
Computer Engineering. 2010, 36(15): 216-218,221. https://doi.org/10.3969/j.issn.1000-3428.2010.15.077A virtual assembly system is constructed according to frame theory. The process frame encapsulates knowledge about typical assembly processes that is a kind of data structure and saved in the system. A method of assembly scene matching is presented to perceive user’s intention, which enhances the perception ability of the virtual assembly system to a large extent. The system makes use of the matching process frame to guide the following assembly process. The part is located precisely by implementing actions at the time that are described in the matching process frame. The problem is resolved effectively which is difficult to control the motion of the assembling parts precisely with current virtual reality I/O devices during virtual assembly.
-
Computer Engineering. 2010, 36(15): 219-221. https://doi.org/10.3969/j.issn.1000-3428.2010.15.078A Global Positioning System(GPS) compass data acquisition and processing system based on multi-thread technology is designed. This paper describes the principle of GPS attitude determination and related communication protocol. With the example of the SDI-COMPASS, a solution based on the API communication function of Win32 is realized, which achieves the multi-thread and overlapped I/O communication between GPS compass and PC via serial port. The static and dynamic experiments show that the system is capable of gathering and processing GPS compass data quickly and correctly, and performs with high stability and flexibility.
-
Computer Engineering. 2010, 36(15): 222-224. https://doi.org/10.3969/j.issn.1000-3428.2010.15.079Referring to the possible problem of gyro drift effect on the line of sight stabilization performance, a method for gyro correction based on visual attitude measurement is proposed. A monocular vision system is established by four coplanar reference points on the ground and a double-focus CCD camera in the carrier. The quaternion is used to represent the transform relation between geodetic coordinate system and carrier coordinate system, and on this basis, the model of gyro correction in the DLOSS is derived by the Levenberg-Marquardt algorithm and the rigid body kinematics equation. Simulation results show that the algorithm is effective.
-
Computer Engineering. 2010, 36(15): 225-227. https://doi.org/10.3969/j.issn.1000-3428.2010.15.080Aiming at the opening up problem in the field of industrial control, by analyzing the principles of conformance and interoperability test for field bus, this paper designs and implements an EPA conformance and interoperability test system, and expresses its design ideas in the function, procedures framework and test process. Input specification documents before test are designed. Consistency and interoperability test on the products based on EPA standard with the system verifies the feasibility of the design scheme.
-
Computer Engineering. 2010, 36(15): 228-229,232. https://doi.org/10.3969/j.issn.1000-3428.2010.15.081For the problems that Structural Health Monitoring(SHM) system has the field nodes with fixed functions, which makes it difficult for flexible configuration and post-maintenance, this paper proposes a reconfigurable architecture for SHM system based on CAN bus. Some reconfigurable techniques are designed including node function reconfiguration, network structural reconfiguration and resource allocation reconfiguration. The functional reconfiguration of nodes is realized by collaboration of the hardware which is functional adaptation interface and the software which is embedded operation system. The self-organizing feature map net is used for the priority clustering of network structural reconfiguration. The resource allocation in accordance with the requirements is realized by the monitoring software based on the component object model technique. The system designed with this method has the abilities such as flexible, efficient and autonomy.
-
Computer Engineering. 2010, 36(15): 230-232. https://doi.org/10.3969/j.issn.1000-3428.2010.15.082In order to accomplish unit maintenance training quickly, effectively and economically, this paper builds a virtual dismantling system based on Pro/E by analyzing the demands of the system, which includes I/O devices, virtual dismantling environmental management, database and performance test module. The key technologies such as communication with I/O devices and virtual dismantling environmental management are given. Feasibility and validity of the system are testified through Pro/E secondary development.
-
Computer Engineering. 2010, 36(15): 233-235,239. https://doi.org/10.3969/j.issn.1000-3428.2010.15.083Aiming at blind equalization problem of QAM signal, this paper proposes a hybrid blind equalization algorithm named GMMA-DD by combining Generalized Multi-Modulus Algorithm(GMMA) and Decision-Directed(DD) algorithm. It combines the benefits of GMMA and DD algorithm, and utilizes the mean square error to control the proportion of two kind of error signals in total error signals to improve the performance of blind equalization. An improved algorithm with lower computational complexity and less FPGA resources is presented and implemented to simplify FPGA-based implementation by using serial and pipelining structures. Experimental results show that the algorithm has good performance for high-order QAM signals.
-
Computer Engineering. 2010, 36(15): 236-239. https://doi.org/10.3969/j.issn.1000-3428.2010.15.084Aiming at the lack of current emergency rescue mechanisms for oilfield accidents, this paper proposes a visualization decision-making support system for oilfield accident emergency rescue based on WebGIS. It combines all levels of emergency rescue command organizations in different locations by network, thereby cooperating to emergency rescue. And it provides all-side, visualization and high-efficiency information services, along with decision-making support so as to improve emergency rescue in terms of scientific and real-time characters. Its architecture and key techniques of the system are presented, including asynchronous communication between the client and the server, load and display of map, and model of optimal rescue path algorithm.
-
Computer Engineering. 2010, 36(15): 240-241,244. https://doi.org/10.3969/j.issn.1000-3428.2010.15.085This paper presents a novel degree distribution function that is entirely distinct from Robust Soliton Degree(RSD) distribution function in the aspect of mathematical expression and the effect on the construction of LT codes. By using the function, original data can be reconstructed with high probability, and decoding overhead is negligible. Theory analysis and simulation results prove that encoding and decoding complexity of LT codes with the function is on the same order of the corresponding complexity of LT codes with RSD function.
-
Computer Engineering. 2010, 36(15): 242-244. https://doi.org/10.3969/j.issn.1000-3428.2010.15.086Based on Minimum Mean Squared Error(MMSE) method, this paper presents Signal to Noise Ratio(SNR) estimation algorithm by using periodic PN sequence as pilot under Double Fading Channel(DFC). It inserts guard space at each edge of pilot to reduce the influence of Inter-Carrier Interference(ICI), and avoids noise interference by going on correlation operation in time domain to get more precise channel parameter. Simulation result shows that this algorithm can more accurately estimate SNR, and it can satisfy performance requirement under different channel environment by changing periodic PN sequence.
-
Computer Engineering. 2010, 36(15): 245-247. https://doi.org/10.3969/j.issn.1000-3428.2010.15.087Aiming at the problem of much data redundancy, slow data loading of traditional data loading algorithm, according to status of map showing technology and user need, this paper presents a dynamic on-demand loading algorithm. It blocks graphic element and establishes sub-block index to achieve rapid querying, loading and showing. Test and analysis results show that the algorithm reduces amount of loading data and data redundancy to a large extent, and improves the data loading rate.
-
Computer Engineering. 2010, 36(15): 248-249,252. https://doi.org/10.3969/j.issn.1000-3428.2010.15.088Brain-Computer Interface(BCI) based on neural firing rates of cerebral motor cortex can control the paralysed body or the motion of the prosthetic device by recording neuronal firing rates of cerebral motor cortex. The core of the software and hardware are the neurons decoding method and the detection technique of the potential of neuronal. The neuronal decoding methods are divided into inferential algorithm and classifiers method, the detection technique is recording the neuronal activity of single or population neuronal using the chronic microelectrodes planted in the cerebral motor cortex. Analysis shows that BCI technology should apply continuously chronic working implanted electrodes on more areas of the brain to control equipment better; different decoding method should unite and add feedback signal to improve the performance of the decoding.
-
Computer Engineering. 2010, 36(15): 250-252. https://doi.org/10.3969/j.issn.1000-3428.2010.15.089An investigation of the signal detection scheme with Grover algorithm in MIMO-OFDM system is developed, and Grover algorithm is applied to find the minimum value in order to decide the sending sequence. In order to test the efficiency and reliability of this algorithm, the analysis and comparison between Grover algorithm, improved Grover algorithm and other traditional algorithms are made by Matlab simulation. Experimental results show that it can reduce the complexity, while achieving the same performance of the traditional optimum detection algorithms.
-
Computer Engineering. 2010, 36(15): 253-255,258. https://doi.org/10.3969/j.issn.1000-3428.2010.15.090In order to solve the problem of CDMA signal pseudocode estimation under the condition of low Signal to Noise Ratio(SNR), this paper presents a blind estimation method based on blind beamforming and Singular Value Decomposition(SVD). It can estimate spread spectrum for each user, accomplish multi-user detection, and give complete mathematical expression of blind multi-user detection under Gaussian noise without the information of channel synchronizing and signal Direction Of Arrive(DOA). Theory analysis and simulation results show that this method can work well in lower SNR surrounding by utilizing spreading gain and the antenna array gain, and the performance of this method is better than single array detection.
-
Computer Engineering. 2010, 36(15): 256-258. https://doi.org/10.3969/j.issn.1000-3428.2010.15.091Aiming at the problem of overwhelmed information, weak individualized service and the lack of information retrieval ability of current learning system, this paper presents a personalized recommendation learning system based on multi-agent. The system uses Java Agent Development Framework(JADE) to design Learner Agent(LA) and Recommendation Agent(RA), adopts Lucene to build personalized search engine to support recommendation, and combines three kinds of recommended methods to take advantage of cooperation and negotiation among Agents. Experimental results show that this system has better recommendation effect and efficiency compared with single recommendation algorithm.
-
Computer Engineering. 2010, 36(15): 259-261. https://doi.org/10.3969/j.issn.1000-3428.2010.15.092
This paper analyses the structural characteristics of Xilinx Platform Flash PROM XCF32P, gives a configuration scheme for multi-FPGA system based on the design revisions of XCF32P. The scheme adopts slave SelectMAP model, uses XCF32P and a CPLD XC9572 to configure four Virtex XCV200 FPGAs, due to adopt a single vendor solution program, the system software and hardware design to be simplified, and it can receive a higher configuration rate.
-
Computer Engineering. 2010, 36(15): 262-264. https://doi.org/10.3969/j.issn.1000-3428.2010.15.093In order to improve the analysis accuracy of electromagnetic waves propagation characteristics in the stratified stratum, Kriging interpolation is adopted to estimate the electromagnetic parameters distribution in stratum and set up the stratum model, and the Finite Difference Time Domain(FDTD) method is used to simulate the process of electromagnetic waves transmission in the real world stratum. Simulation results show that the method can explain the electromagnetic transmission phenomenon in the through-the-earth communication.
-
Computer Engineering. 2010, 36(15): 265-267. https://doi.org/10.3969/j.issn.1000-3428.2010.15.094Under partially shaded conditions, the P-V curve of PV arrays has the characteristics of multi-summit, which makes the Maximum Power Point Tracking(MPPT) failed. Aiming at above problem, this paper proposes a control algorithm based on Particle Swarm Optimization(PSO) algorithm for solving maximum power point tracking problem. Experimental results show that output voltage of PV system is maintained near maximum power point, and the algorithm is effective.
-
Computer Engineering. 2010, 36(15): 268-270. https://doi.org/10.3969/j.issn.1000-3428.2010.15.095Disadvantage of exiting distributed Modeling and Simulation(M&S) is analyzed. Model Driven Architecture(MDA) is applied to the domain of distributed M&S and a Distributed M&S framework based on MDA is designed. The framework consists of four layers (model repository layer, PIM modeling layer, Platform Specific Model(PSM) modeling layer and simulation runtime layer). Development procedure and key technologies of framework are researched, plug and play weapon system distributed simulation platform is designed. Simulation results show that the framework makes distributed simulation system more extensible, improves simulation module’s interoperability and reusability.
-
Computer Engineering. 2010, 36(15): 271-273. https://doi.org/10.3969/j.issn.1000-3428.2010.15.096To solve the evaluation problem of sites-setting for short wave air-to-ground communications, two evaluating indicators, coverage degree and coverage efficiency are given. Grid scanning algorithm and Monte Carlo method are used respectively to calculate the two evaluating indicators. According to the simulation, two special sites-setting structures (square and hexagon) are evaluated by the two indicators, and finds out that the hexagon structure is more reasonable comparing with each other. The feasibility and availability of the two evaluating indicators and their calculating methods are validated.
-
Computer Engineering. 2010, 36(15): 274-276,279. https://doi.org/10.3969/j.issn.1000-3428.2010.15.097
For problems of standardization in the field of science and technology plan project management system, this paper researches and designes a flexible, scalable, loosely coupled lightweight Web framework. Based on analysis and abstraction of the areas, it proposes management model with a whole life-cycle process, builds a J2EE stratified architecture that takes Spring framework as the core, and integrates Struts2, JPA and JBPM3. In the basis of two points, a domain framework is constructed, which has good scalability, maintainability and portability. The practical applications of the framework show that it has a significant effect on normalization of technology management processes and improving efficiency of system development.
-
Computer Engineering. 2010, 36(15): 277-279. https://doi.org/10.3969/j.issn.1000-3428.2010.15.098Aiming at the disadvantages of the current shelf-life prediction device, a new intelligent prediction device is proposed in this paper. The shelf-life prediction algorithm of penaeus vannamei is based on the theory of Time-Temperature Indicator(TTI) and K-means algorithm. Different prediction algorithm is adopted to predict shelf-life according to the corresponding temperature and stage of penaeus vannamei. Compared with the previous, this device can not only accurately predict the shelf-life, but also has low power consumption and small size.
-
Computer Engineering. 2010, 36(15): 280-282,285. https://doi.org/10.3969/j.issn.1000-3428.2010.15.099This paper puts forward a Verification Purpose Operating System(VPOS) based on the verification for Memory Management Unit(MMU), which is capable of static memory management, static case scheduling and pseudo interrupt handling. Designed for hardware verification only, VPOS provides an executable and controllable platform for MMU test-program development. Verification practice shows that VPOS based simulation can cover 94% of the MMU bugs very early in the project, which releases the debugging pressure on later FPGA verification by porting General Purpose Operating System(GPOS). Therefore, it guarantees the one time success of transplanting general operating system.
-
Computer Engineering. 2010, 36(15): 283-285. https://doi.org/10.3969/j.issn.1000-3428.2010.15.100In this paper, an adjacent substitution rate estimation method based on weighted Euclid distance is proposed in order to improve the disadvantage of estimating weight by experience. It is based on Support Vector Machine(SVM). The proposed method is applied to the demand forecast of facial cleaning category. Experimental results identify that the predict precision is improved and the objectivity and robustness are both well.
-
Computer Engineering. 2010, 36(15): 286-287,290. https://doi.org/10.3969/j.issn.1000-3428.2010.15.101An automatic modulation recognition algorithm based on decision-theoretic approach for analog and digital communication signal is presented. The signal processing architecture based on in phase the quadrature channels is introduced. The instantaneous amplitude, instantaneous phase and instantaneous frequency are calculated by Coordinated Rotation Digital Computer(CORDIC) algorithm with an expanded convergence range. It is combined with software defined radio technology and some recognition parameters are modified. The recognition flowchat based on decision tree is put forward. Simulation indicates when the Signal-Noise Ratio(SNR) is no less than 6 dB, the average correct recognition ratio is above 94%.
-
Computer Engineering. 2010, 36(15): 288-290. https://doi.org/10.3969/j.issn.1000-3428.2010.15.102The distribution of solar radiation is an important factor to the grouth of plant. In order to reduce the complexity and time-consuming in modeling the distribution of radiation in canopy, this paper aims to improve this model by changing the projection plane and making certain of the density of solar radiation. It uses a software based on advance to model the a boot stage paddy rice to evaluate the improvement, and results show that the advance model can provide the accuracy and reduce the complexity.
-
Computer Engineering. 2010, 36(15): 291-封三. https://doi.org/10.3969/j.issn.1000-3428.2010.15.103Limited by the computing speed and the memory space of embedded system, a new ECG data processing mechanism is proposed. It is based on the features of real-time ECG telemonitoring. It includes ECG sampling, ECG storage, ECG transmit, ECG encryption, ECG compression and ECG analysis. The processing mechanism is applied in the remote wireless multiple physiological parameter real-time monitoring and analyzing network platform. It can send the complete, accurate, real-time ECG data to the monitoring server through the GPRS/CDMA/3G wireless network and get the analysis result on the monitor.
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