Abstract:Analysises on factors which impact the performance of the multi level algorithms have been made, and on the basis of which an immune algorithm with selfadaptive reduction has been proposed for the TSP problems. By using an evolutionary reduction set, the proposed algorithm refines the reduction edges which gradually increase in the number and enhance in the forecasting accuracy. As a result, the probability that the refined algorithm finds the global optimal solution can be improved. Experimental results show that the proposed algorithm can achieve better solutions than other approaches.

Abstract:In order to optimize data locality, communication and synchronization overhead, this paper proposes a multi-layers symmetric Gauss-Seidel method. Then the serial execution model of this iterative method is given, which introduces the sequence of iterative space tile as the sequence of execution, and divides iteration space by time skewing. In this model, nodes of the tile can be updated many times to improve data locality. The parallel GS execution model based on iteration space tiling is presented, which uses an improved iteration space partition algorithm and reorders the tiles of iteration space to reduce cache misses, communication and synchronization cost. Finally the numerical results are presented to confirm the effectiveness of Gauss-Seidel parallelized with alternate tiling method, specifically compared with owner-computing and red-black Gauss-Seidel methods, and show that the new parallel iterative method has better parallel efficiency as well as scalability.

Abstract:A cluster is a collection of data objects that are similar to one another within the same cluster and are dissimilar to the objects in other clusters. Moreover, there will exist more or less similarities among these large amounts of initial cluster results in real life data set. Accordingly, analyzer may have difficulty to implement further analysis if they know nothing about these similarities. Therefore, it is very valuable to analyze these similarities and construct the hierarchy structures of the initial clusters. The traditional cluster methods are unfit for this cluster post-processing problem for their favor of finding the convex cluster result, impractical hypothesis and multiple scans of the data set. Based on multifractal theory, this paper proposes the FCHO (fractal-based cluster hierarchy optimization) algorithm, which integrates the cluster similarity with cluster shape and cluster distribution to construct the cluster hierarchy tree from the disjoint initial clusters. The elementary time-space complexity of the FCHO algorithm is presented. Several comparative experiments using synthetic and real life data set show the performance and the effectivity of FCHO.

Abstract:This paper introduces a new technique for assessing the diversity of approximation of exact set. The diversity is assessed by the "exposure degree" of the exact set against the approximation set in the new technique where the projection of the approximation set on the exact set, not the approximation set, is used to cover the exact set. Therefore, the new technique still works well for the problems with 3 and more objectives, significantly improving the existing techniques. Computational results show that it assesses the diversity well. Rigorous analysis is made for the new method.

Abstract:Point-Based algorithms are a class of approximation methods for partially observable Markov decision processes (POMDP). They do backup operators on a belief set only, so linear programming is avoided and fewer intermediate variables are needed, and the bottleneck turns from selecting vectors to generating vectors. But when generate vectors, there will be a great deal of repeated and meaningless computing. This paper will propose a preprocessing method for point-based algorithms (PPBA). This method preprocesses each sampled belief point, and before generating α-vectors it estimates which action and α-vectors to be selected first, in so doing repeated computing is eliminated. Base-vector is also defined in this paper, which cancels meaningless computing with sparseness of problem. Experiments on Perseus show that, PPBA accelerates the performance greatly.

Abstract:With the popularity of UML (unified modeling language) and MDA (model driven architecture), models are becoming the core artifacts of software development and maintenance. As a result, modeling languages and meta-models which are used to define modeling languages, become more and more important. Software development may cover quite a few domains, and different domains may require different modeling languages and their supporting modeling tools. But it is very expensive to develop modeling tools manually for every domain. Metamodeling is one of the technologies to facilitate the design of domain modeling languages and the development of modeling tools. In the approach of metamodeling, people design domain modeling languages according to domain request by metamodeling. And then, metamodeling tools automatically generate modeling tools, which support the designed domain modeling languages. As shown by experimental results, metamodeling, combined with MDA, can increase productivity of software development. This paper makes a survey of the current research on metamodeling, compare metamodeling tools, and discuss further directions of metamodeling and their supporting tools.

Abstract:Software architecture evaluation is an important technology used to assure the quality of software products early in the software lifecycle. This paper classifies three types of software architecture evaluation methods: scenario-based, metric and prediction based, and ADL-based. Software architecture evaluation method characteristics (such as method goal, quality attribute, key technique) are then combined with these classifications to produce a comparison framework. This paper utilizes this framework to analyze various existing software architecture evaluation methods and point out problems which need to be resolved. Finally, potential research directions of software architecture evaluation methods are discussed.

Abstract:In this paper, a dynamically customizable framework for supporting the autonomy of components is introduced. To implement autonomous components, the framework adopts an approach to repacking existing components. By combining Agent technology into components, the framework can customize behavior rules and plans for components to enable components to adjust their behaviors according to the states of the environment. By integrating a rule-engine that can interpret and execute declarative rules, the framework supports the autonomous behaviors of components without recoding and re-deploying of components.

Abstract:Based on the idea of trust evaluation, this paper proposes an Internetware software architecture oriented trust-driven mechanism for selecting services to solve the problem from the aspect of service selection: Firstly, a general, machine-readable description mechanism is proposed to express user requirements and trust evolution policies; secondly, a feed-back trust formation and decision mechanism is introduced and a trust-driven algorithm for selecting services is given; finally, an Internetware software architecture oriented trust-driven service selection framework is proposed for the development of trustable Internetwares. Some development examples indicate that this mechanism can support the development of the trustable Internetwares effectively.

Abstract:According to the principle of "Separation of Concerns", by investigating the similarity between multi-view software process modeling and object Petri nets, this paper proposes the MOPN-SP-net model which is a multi-view software process model based on object Petri nets and enhances the reusability of software process model. During process modeling, MOPN-SP-net is a multi-dimensional Petri net, which is difficult to analyze directly. So, this paper provides a translation rule from an object Petri net to an equivalent traditional flat Petri net. The translation preserves the soundness property. According to the translation rule, the soundness property of the MOPN-SP-net can be indirectly analyzed by its translated flat net.

Abstract:This paper studies statistical test's assurance ability on different programs and brings forward to incorporating effectiveness information into software reliability assessment. This can improve the accuracy of reliability assessment. This paper empirically investigates the reasonableness of this method. The conclusion provides a new way for high reliable software assurance.

Abstract:This paper gives a survey on current Skyline queries techniques. It first introduces the background in which Skyline queries appear. Then it presents in-memory algorithms in Skyline query problem. Facing to the situation of large data sets, it further presents the techniques about Skyline query processing by two cases, with or without indices respectively. Evaluations of Skyline query methods are discussed after that. This paper also introduces the novel query model-SKYCUBE which is applied to process multi-Skyline queries in various subspaces and related research based on it. Additionally, it introduces the efficient algorithm to solve Skyline queries in various applicant environment and the extension of the Skyline query processing. Finally, this paper proposes several directions for further research on the topic of Skyline query processing.

Abstract:In order to improve the query efficiency, K-means cluster approach is often used to estimate the data distribution in the context of high dimensional metric space index. But in previous work, the parameters of clustering are usually selected according to some heuristic manner. This paper presents a new high dimensional index approach—cluster splitting based high dimensional B⁺-tree. Through cluster splitting, the data space is partitioned more finely to reduce the cost of data access. The relationship between cluster and the query cost is discussed, and based on the query cost model, this paper give formulas to compute the "optimal" parameters of the cluster which can minimize the query cost in theory. Experiment results show that the efficiency of the methods is better than iDistance, M-Tree and sequence scan, and the parameters computed by the formulas are very close to the real optimal one.

Abstract:A new predict model was contrived, which involves local stream energy prediction, the energy distribution pattern mining, the predictive series reconstruction and measurement method of stream energy. A new method was designed to forecast stream by energy regression and wavelets decomposing based on frequent pattern, and extended to multi-streams with strong coincidence. The concept of the nearest maximum frequent pattern was proposed to decompose local stream energy. The validity of new algorithm was demonstrated by extensive experiments on real data.

Abstract:A novel method of discovering relation information among entities buried in different nest structures of XML documents is proposed. The method is able to identify relations among different types of entities given by users, and extract relation instances and their occurrence patterns in XML documents. The solution is as follows: identify and collect XML fragments that contain all types of entity given by users at first, then calculate similarity between fragments based on semantics of their tags and their structures, and cluster fragments with a adaptively selected similarity threshold so that the fragments containing the same relation are clustered together, finally extract relation instances and patterns of their occurrences from each cluster. The experimental results show that the method can identify and extract relation information among given types of entities correctly from all kinds of XML documents with meaningful tags.

Abstract:In this paper the multi-task interaction scenario is analyzed and an activity-centered method to manage personal information is presented. An activity model is discussed from the aspects of the static structure and dynamic involvement of an activity and relation between activities. A method of computing the relativity between activities from the aspects of the interaction characteristic in multi-task scenario and the content of the object involving an activity is proposed. Moreover, a personal information management centered-activity tool ACPIM (activity-centered personal information management) is designed. An evaluation of this ACPIM system showed that the tool could reduce cognition overload for users and be useful in their own work.

Abstract:This paper gives an introduction to the transmission control problem and a survey to the recent novel protocols and their taxonomy. Firstly, the research background of the WSN transmission control problem is presented. Then a comprehensive survey with analysis is made to different methods and representative protocols in two aspects: Congestion control and reliability guarantee, respectively. In the end, the protocols are compared and future research directions are suggested.

Abstract:This paper presents a survey on the study of the scalable router and proposes a hierarchical model based on the study of the architecture and models of the scalable router. The model splits the scalable router into six layers from bottom to top, which are interconnection network and data switching layer, routing lookup layer, standard interface layer, distributed operating system layer, distributed routing behavior layer and single image management layer. The paper provides surveys on the research of each layer, and finally concludes and analyses the difficulties of current development of the scalable router.

Abstract:As a policy-based routing protocol, BGP (border gateway protocol) allows each AS to choose local routing policy independently. Possible policies conflict may result in BGP route persistent oscillation. This paper proposes an adaptive mechanism to guarantee BGP convergence with policies conflict, which neither impairs the flexibility of choosing routing policy, nor inserts additional information in BGP messages. With the mechanism, route stability is taken into BGP decision process so that unstable route is degraded to cause more stable route to be chosen to stop policies dispute. The new mechanism also can adapt to topology change and converge to new stable route.

Abstract:This paper presents a search algorithm called probabilistic search team (PST). In PST, all nodes advertise their resource sharing information, maintain and broadcast the information based on DDBF (distributed discarding bloom filter), which discards some information when transmitted to their neighbors. During the search process, PST extends the concept of walker in RW to search team. PST realizes collaborative and parallel search of multiple search teams by aggregating the resource information obtained in search process. Experimental results show that PST achieves a good tradeoff between performance and overhead.

Abstract:There are many research projects based on grid in many research fields like high energy physics (HEP). Grid applications including massive data transfer have very high requirement for network on QoS parameters such as bandwidth and delay. The traditional IP routing network can not satisfy these requirements. However, the connection-oriented optical network has this capability, but it still faces many challenges. The authors firstly analyzed the features of grid applications and their special requirements for optical network control and manage- ment, and emphatically analyze several existing methods of optical network control and management. Subsequen- tly, current issues which are open are summarized. At last, some novel research issues such as Lambda Grid and OVPN are put forward.

Abstract:Mobile node (MN) is reachable based on home agent (HA) in mobile IPv6 (MIPv6). This paper proposes the active detection and transfer HA fault-tolerant method (ADTM) for MIPv6 networks. It defines the conception of ring detection & backup chain (D&B chain) among HAs in the same link. An MN only backups its registration information on the next HA in the D&B chain. For any HA, its previous HA and its next HA in the D&B chain detect the HA's validation based on neighbor discovery mechanism. If an HA is detected failure, its next HA will take over the failed HA temporarily, then inform the failure information to MNs served by the failed HA. An MN received the failure information will select a new valid and perform home registration again. For implementing ADTM, this paper defines the data structures, designs the fault-tolerant algorithm and describes the procedure in detail. It is proved that this method has less fault-tolerant time and less signal cost than the method in RFC 3775, especially in the scenario of MN moving more frequently.

Abstract:This paper first presents an analysis model named flow-competing congestion model (FCCM) for this type of congestion. Based on FCCM, the paper derives the distribution of competing flows at the congested link, and analyzes the conditions under which the flow-competing congestion would happen. This paper also explores how much buffers a congested link requires to keep full link utilization when the flow-competing congestion occurs. This paper proves that when sizing buffers for a congested internet link with the aim of keeping full link utilization, the buffer requirement of the flow-competing congestion is not bigger than the minimum buffer requirement of the famous BSCL (buffer sizing for congested internet links) scheme.

Abstract:This paper provides a complete packet-level topology aware worm simulation model based on worm targets discovery, worm code transmission and worm activation, and designs a directed Small World topology generation algorithm. Then, through selective abstraction, this paper implements a complement packet-level topology aware worm simulation system. Finally, with simulation experiments, the paper analyzes the impacts of topology structure and worm activation on worm propagation. The results of simulation experiments show that this simulation system can provide great support for topology aware worm research.

Abstract:The paper presents a code-injection attack automatic analyses and response system. By means of analyzing abnormal scene and syntax of data payload, it generates vulnerability-oriented signatures which are used to filter off variant form of code injection attack based on same attack scheme of same vulnerability. By combining with protocol state and process state during signatures generating and attack responding process, very low false positive rate and lagging without elevating false negative rate can be gained. Some technical details of the protocol type system on Linux and Windows 2000 and experimental results are also presented.

Abstract:To improve performance of TCP (transmission control protocol) Reno, a widely used TCP version, an improved TCP Reno named DAA-TCP is proposed. DAA-TCP (dual AMID-based active TCP) combines dual AIMD (additive-increase multiplicative-decrease) algorithms and adds an active congestion control manner, i.e., the congestion window may be decreased when some given conditions are satisfied before the congestion really occurs. DAA-TCP performances are investigated through simulations. Comparing with TCP Reno, the throughput may be improved; the retransmitted packets and the retransmit probability may be decreased. DAA-TCP can coexist with TCP Reno very friendly. In addition, DAA-TCP is obtained through a little modification from TCP Reno at the sender side, so DAA-TCP can be easily developed and the additional overheads are very low.

Abstract:The approach is to include forwarding state scalability as one of the optimal objective when constructing new multicast trees. This multi-objective optimization approach can be applied to many existing multicast state reduction methods. In this paper, the approach is illustrated by applying it to aggregated multicast (AM) and dynamic tunnel multicast (DTM). Both AM and DTM routing problems are formulated as multi- objective optimization problems, and both heuristic and genetic algorithms are proposed for solving them. Based on the experimental results, the approach can further improve the forwarding state scalability of both approaches by reducing the number of aggregated trees required by the AM method, and by increasing the number of non-branching nodes for the DTM method.

Abstract:It is a worthwhile challenge to deal with the key-exposure problem in identity-based signatures. To deal with this problem, this paper adopts Dodis, et al.'s key-insulation mechanism to identity-based signature scenarios, and proposes an identity-based key-insulated signature scheme. The proposed scheme enjoys two attractive features: (i) it is strong key-insulated; (ii) it is provably secure without random oracles.