Abstract:Two ways to sort externally are Multi-Line Merging Sort and Bucket Sort, both with two passes. The Bucket Sort burdens the CPU less and is more efficient, while its usage is restricted heavily by the High-Bit scheme that distributes records into subfiles: the keys have to be integers; the sizes of subfiles may vary too much; the number of subfiles cannot be chosen freely. Based on statistical theory, this paper presents a sample-seperators scheme to broaden the ussage of bucket sort algorithm. A brief discussion on the convergance of sample-seperator estimation is given and the probability to avoid memory overflow is calculated. This scheme enables the bucket sort algorithm to be applied in the SheenkSort system to win the 2003 PennySort (the Indy category) competition.

Abstract:In order to solve the problems with computing resource and high-speed network traffic, it is necessary to deal with the network traffic by some measuring technologies, such as sampling measurement and load balance, etc, while the hash algorithm is one of the key measuring technologies. In this paper, firstly, a random metric is provided to evaluate the performance of the hash algorithms. Secondly, the randomicity of XOR and shift operations are analyzed, and it is proved that the two operations can improve the bit randomicity. Thirdly, this paper analyzes the four fields of IP packet, such as source IP, destination IP, source port, and destination port, and a hash algorithm named XOR_SHIFT is provided based on the analysis. Finally, using the CERNET backbone traffic and PMA traffic, this paper analyzes the character of the XOR_SHIFT hash algorithm and compares with the performance among XOR_SHIFT, IPSX and CRC32 hash algorithms. This study shows that the XOR_SHIFT hash function provided in this paper has two advantages: algorithm performance and hash randomicity, and it can be applied to measure the high-speed network traffic.

Abstract:Constraints involving Boolean and numerical variables are used widely, but it is difficult to solve especially when they contain nonlinear numerical expressions. Many existing methods for solving such constraints are incomplete. A new method is presented in this paper to solve Boolean combinations of nonlinear numerical constraints completely. This method transforms the nonlinear constraints into a special-formed optimization problem to solve them. A prototype tool is implemented, and some experiments are made. The experimental results show that the method is effective.

Abstract:The existing automatic mesh simplification algorithms at present always ignore some important shape features of the original model, such as the corners and high-curvature regions, in the low-level model, and this will lead to the degeneration in the sense of sight. On the base of Garland’s simplification algorithm, a method of changing the order of edge collapses in the simplification is presented by introducing the concept of sharp degree into the error metrics. The results can not only preserve the important features of the model but also distribute meshes reasonably. Finally a better simplified model is obtained which has dense meshes in the high-curvature regions and sparse meshes in the flat regions.

Abstract:Existing string matching algorithms typically set the sliding window size as the pattern length. This paper presents a Linear DAWG Matching (LDM) algorithm, which divides the text into [n/m] overlapping windows of length 2m-1. In the windows, the algorithm attempts at m positions in batches. It firstly searches pattern prefixes from middle to left with a reversed suffix automaton, shifts to next window directly when it fails, otherwise, scans the corresponding suffixes forward with a finite automaton. Theoretical analysis shows that LDM has optimal time complexities in the worst (O(m)), best (O(n/m)) and average cases (O(n(1og_σm)/m)). Experimental comparison of LDM with the existing algorithms validates this theoretical claims of average case for searching long patterns. It further reveals that LDM is also efficient for searching short patterns on large alphabets. Thus, LDM algorithm not only suits for off-line pattern matching, but also fits in high-speed online pattern matching.

Abstract:The Fork-Join structure is one of the basic modeling structures for parallel processing. Although some algorithms are able to find an optimal schedule under certain conditions, they ignore to economize processors and minimize the total completion time. This paper presents a Task Duplication based Balance Scheduling(TDBS)algorithm which can generate an optimal schedule for fork-join task graph with a complexity of O(vq+vlogv), where v and q are the number of tasks and processors respectively. By considering workload and idle time slots of the used processors, TDBS algorithm tries to assign tasks to scheduled processors and maximize their utilization. Simulation results show that TDBS algorithm has better speedup and efficiency than other compared algorithms. Therefore,TDBS algorithm is a viable option for practical high performance applications.

Abstract:In this paper, an example-based color scheme generating and authoring method is presented, and a color scheme system target to office applications is introduced. This system can automatically extract main colors from the input image or picture (say, a commercial logo or brand), and then automatically produce a group of harmonious colors. By employing intelligent optimization, the system can suggest a set of reasonable color themes. The experimental results show that this system can generate customized results to match the user’s personalized application.

Abstract:Anaphora Resolution is playing more and more important role in Natural Language Processing. There is an increasing need for the development of effective and robust strategies of anaphora resolution to meet the demands of practical applications. However, traditional approaches to anaphora resolution rely heavily on multilevel linguistic knowledge, such as syntactic, semantic, contextual and domain knowledge. It is undoubtedly difficult to acquire such knowledge at present. This paper presents a two-step approach with limited knowledge to resolve pronominal anaphora within Chinese text, which only uses number features, gender features and the features of grammatical roles. In this approach, a filter is firstly used to eliminate those expressions whose features are inconsistent with the pronoun, and thus form a set of potential antecedent candidates; then, a scoring algorithm is employed to calculate score of the candidates, and the candidate with the highest score is selected as the resultant antecedent. The algorithm does not examine each candidate in the set, but automatically determine whether to end the calculation or not by dynamically testing a termination condition, therefore the computational complexity is low. In addition, the approach does not need a deep analysis of the text, and can easily be implemented. Experiment shows the result is satisfactory.

Abstract:An algorithm based on disparity points to solve the occlusion problem in the process of building high-quality stereo disparity map is presented in this paper. It is firstly proved that the disparity curve corresponding to a pair of epipolar-line images may be approximated by a group of piece-wise straight lines, and then the definition of disparity point is introduced. In the parameterization of a disparity point, two parameters are used to describe left and right occlusions so that the occlusion problem can be successfully solved in the approach. By analyzing intensity property of a disparity point and its neighbor points, an approach which combines stepwise hypothesis-verification strategy and Marquardt-Levenberg (M-L) algorithm is devised to extract the candidate disparity points from the epipolar images, and then aperiodic dynamic programming is employed to search the epipolar-optimal disparity function. The proposed method is tested by using the international standard image data and compared with other methods, and the experimental results show that its performance is the best among epipolar-optimal methods and worse than some excellent global-optimal approaches, but its complexity is much lower than the global-optimal approaches.

Abstract:In this paper, a novel example-based automatic face alignment strategy has been proposed for facial features alignment, i.e. facial shape extracting. The method is motivated by an intuitive and experimental observation that there exists an approximate linearity relationship between the image intensity difference and the shape difference, that is, similar face image intensity distribution implies similar face shape. Therefore, given a learning set of face images with their corresponding face landmarks labeled, the shape of any other face image can be learned by estimating its similarities to the training images in the learning set and applying these similarities to the shape reconstruction of the unknown face image. Concretely, if the unknown face image is expressed by an optimal linear combination of the training images, the same linear combination coefficients can be directly applied to the linear combination of the corresponding training shapes to construct the optimal shape for the novel face image. Experiments have shown that, compared with traditional methods, the proposed method can achieve a comparable alignment accuracy in less time. Furthermore, the same strategy has been extended to extract the shape of face images with varying poses.

Abstract:This paper introduces an effective algorithm for color edge features extraction and proposes a novel edge orientation encoding, biaxial symmetry orientation encoding. The average performance of human face detection system, which is based on the support vector classifier using the histograms of color and color edge features, is evaluated with ROC in multi_fold cross validation. Experimental results show that color edge features outperform gray edge features evidently; the classification accuracy of the novel edge orientation coding outperforms the traditional edge orientation coding when they are linearly combined with color histogram respectively; the face detection accuracy can be significantly improved when both color and color edge histograms are used, non-deep rotated human face can be correctly detected in color image under different illuminations, with different expressions and partial occlusions.

Abstract:In factored Markov decision process (FMDP) such as Robocup system, the effect to value evaluation of various states is different from each other within state attributes. There are some important state attributes that can determine the whole state value either uniquely, or at least, approximately. Instead of using the relevance among states to reduce the state space, this paper addresses the problem of curse of dimensionality in large FMDP by approximating state value function through feature vector extraction. A key contribution of this paper is that it reduces the computation complexity by constraints reduction in linear programming, speeds up the production of joint strategy by transplanting the value function to the more complex game in reinforcement learning. Experimental results are provided on Robocup free kick, demonstrating a promising indication of the efficiency of the approach and its’ ability of transplanting the learning result. Comparing this algorithm to an existing state-of-the-art approach indicates that it can not only improve the learning speed, but also can transplant state value function to the Robocup with more players instead of learning again.

Abstract:A new method for fast building the rough data model (RDM) by means of supervised fuzzy clustering in the product space of input and output variables is proposed. The approach incorporates the RDM’s classification quality performance index with Gustafson-Kessel (GK) clustering algorithm and is of many good properties. The way to convert the fuzzy cluster models to rough data models by introducing the concept of putative membership degree of a data point to a fuzzy cluster is suggested. Hence, an efficient algorithm that can obtain RDMs by just iteratively computing two necessary condition equations is worked out. It minimizes the objective function and turns the multi-dimensional search process of the Kowalczyk’s method to one dimensional search strategy (in terms of the number of clusters). This technique reduces the searching time greatly. Compared with the traditional rough set theory and the Kowalczyk’s method, the approach has more powerful ability to handle data contaminated by noise and better generalization ability. Finally, different examples of data sets illustrate the effectiveness of the approach.

Abstract:The need to solve complicated problems in the real world is the direct driving force for the quest of flexible logic system, which is the future of current development of logics. The theory of universal logic is a new flexible logic system, developed by He Hua-Can during his exploration of the flexibility of logic. This paper discusses the flexibility of probability logic based on the ideas and methods in the implementation of flexible logic. Theoretically, probability logic is a special example of the universal logics. So it is possible to build flexible probability logic in the framework of universal logics.

Abstract:Recently various evolutionary approaches have been developed for multi-objective optimization. Most of them take Pareto dominance as their selection strategy and do not require any preference information. However these algorithms cannot perform well on problems involving many objectives. By introducing preferences among different criteria, a multi-objective concordance genetic algorithm (MOCGA) is proposed to deal with the problems in the paper. As the number of objectives to be simultaneously optimized increases, the weak dominance is used to compare among the individuals with decision-maker's information. It is proven that the algorithm can guarantee the convergence towards the global optimum. Experimental results of the multi-objective optimization benchmark problems demonstrate the validity of the new algorithm.

Abstract:The study of “Implicit Desire” is essential to agent theory. This paper first analyzes insufficiency and inadequacy of the existing theories of implicit desire. Then logical semantics is employed to define a new kind of implicit desire―pm-desire consequence, to investigate its main properties, to compare it with the relevant work, and thereby to argue its rationality such as its accordance with intuitions and ability to improve the autonomy intelligent-agents, etc.

Abstract:This paper discusses the application and performance of multidimensional pattern classification problems using Michigan approach based on fuzzy genetics-based machine learning mechanism, and proposes a new approach. In the approach, each fuzzy if-then rule is handled as an individual, and a fitness value is assigned to it. The approach not only retrieves fuzzy if-then rules, but also tunes the membership functions of each dimension, meanwhile the selection mechanism based on the similarity of individuals is involved to reduce the high selective pressure, keep the diversity of population, and avoid the premature convergence problem consequently. Finally the experiments prove that the approach has a better correct classification rate and a better adaptability on multidimensional pattern classification problems.

Abstract:Currently, OCR (optical character recognition) classifiers are generally designed for one character set (or language). On the other hand, multilingual document increases drastically due to the globalization. Therefore, designing a document processing system with multilingual capability is very important. A general scheme is presented in this paper: two OCR techniques, a system, and a language classification. For embodying the scheme, a Chinese/English mixed document processing system is implemented. Three key problems are considered: the control of the system flow, the classification of Chinese/English regions, and the segmentation of English characters. Compared with old systems presented in other papers, the module of the classification of Chinese/English regions is added in the system, and a novel approach based on the equidistance is applied to the module. To verify the effectiveness of the system, another system is implemented according to the methods presented in other papers. Experiment shows, the new system is more effective than the old system. The recognition rate increases from 98.48% to 99.13% on magazine samples and from 98.68% to 99.25% on book samples, respectively.

Abstract:Incorporating instance-level constraints into K-means algorithm can improve the accuracy of clustering. As the partition generated is represented by K centers and a cluster is represented by only one center, the representation model prevents further improvement of the accuracy. Based upon the instance-level constraints, two types of constraints between instance and class are presented, three types of constraints between classes are presented too, and the constrained partition model is presented and analyzed. In this model, based upon the constraints between sub-clusters, more centers are utilized to represent one cluster, which makes the representation of partition flexible and precise. An algorithm CKS (constrained K-means with subsets) is presented to generate the constrained partition. The experiments on three UCI datasets: Glass, Iris and Sonar, suggest that CKS is remarkably superior to COP-K-means in accuracy and robustness, and is better than CCL too. The time for running CKS is neither significantly influenced by the number of constraints compared with COP-K-means, nor remarkably increased when the number of instances is increased compared with CCL.

Abstract:Most of the XML query processing strategies are based on some numbering schemes. Nodes on the XML tree will be assigned a unique code by the numbering scheme, and ancestor-descendant relationship could be directly told through the codes. The most famous numbering scheme is Region Based Numbering Scheme. However, XML data will be updated. Once the data is updated, the region code should be adjusted to keep the indexing and query processing techniques working. Unfortunately, few studies have been reported on the issue of the numbering scheme. This paper focuses on this issue, proposing a series of space preserving and updating algorithms. Extensive experiments are conducted to test the effectiveness of the algorithms.

Abstract:WebGIS servers send digital maps to users. For each request-response round, the servers access map data in batches. The access has a feature called multiscale, that is, the map scales selected by users determine the map detail levels. The access method based on R-tree is not adaptive to the multiscale and batch feature. It has two problems: (1) the data records of features of the same level are not clustered in disks; (2) the granularity of data I/O is too small. So accessing map data for display is unefficient. This paper presents a novel spatial index called Multilevel R-tree, which can solve the two problems. The statistics from experiments show that for range queries, the access method based on multilevel R-tree is much more efficient than the one based on R-tree and can improve the response speed of WebGIS Servers.

Abstract:XML query languages take complex path expressions as their core. To facilitate path expression processing, the processing strategy based on path decomposition and structural join operation needs to be investigated more deeply. In this paper, a target node aimed at path expression processing framework for XML data is proposed. This approach makes use of the extended basic operations to reduce the number of join operations. In the procedure of path decomposition and query plan selection, target node in the query tree is utilized to avoid the transfer of the intermediate results. In addition to decomposition rules and strategies, a set of extended basic operations and implementation algorithms are proposed. Preliminary experiments indicate this approach has good performance. It provides path query processing with more choices.

Abstract:This paper discusses the semantics on temporal variable “Now”, that is, “Now” may express the meaning about “current”, “past” and “future” time in a valid time, and based upon this semantic analysis, the paper studies the corresponding issues for determining the temporal value of “Now” and gives a formal algebra system on the calculus of temporal relations.

Abstract:A detailed understanding of the structural properties of Internet topology will benefit the further design and development of the Internet. It seems infeasible to study the whole Internet at router level due to its extremely large size and the difficulty in obtaining a whole topology at this level. Studying each national or continental Internet service provider (ISP) topology individually becomes an alternative method for this goal. In this paper, the measured China Education and Research Network topology, a nationwide ISP topology, is basically taken as an example. The results of mapping the topology from multiple vantage points are briefly presented. The properties of the degree distribution, large eigenvalues, and the spectral density of the measured topology graphs are analyzed. The characteristics of the signless Laplacian spectra (SLS), the normalized Laplacian spectra (NLS), and the clustering coefficients of the measured graphs are also presented. The results suggest that some power laws indeed hold in some large-scale ISP topologies; in contrast to the case of autonomous system level topologies, the power law fit is not the best choice for some ISP topologies in terms of the complementary cumulative distribution function of the degree; some real ISP topologies are a kind of scale-free graphs which are not consistent with the Barabási-Albert (BA) growth model; router level topologies are distinguishable in terms of the SLS or the NLS; router level Internet topology may have developed over time following a different set of growth processes from those of the BA model.

Abstract:Wireless Sensor Networks, a novel technology about acquiring and processing information, have been proposed for a multitude of diverse applications. The problem of self-localization, that is, determining where a given node is physically or relatively located in the networks, is a challenging one, and yet extremely crucial for many applications. In this paper, the evaluation criterion of the performance and the taxonomy for wireless sensor networks self-localization systems and algorithms are described, the principles and characteristics of recent representative localization approaches are discussed and presented, and the directions of research in this area are introduced.

Abstract:Because XML (extensible markup language) is self-described, there is much redundant structural information in XML data stream. How to compress XML data so as to reduce the network transfer cost and support XPath evaluation on the compressed data is a new area of research. The existing methods on XML compression require the multi-pass scan on data or can not support real time query processing on compressed data. In this paper, a novel compression method XSC (XML stream compression) is proposed to compress and decompress XML stream in real time. XSC constructs XML element event sequence dictionary and outputs the related index dynamically. When DTD is available, XSC can generate the XML element event sequence graph for producing more reasonable encoding before XML data stream is processed. The compressed XML data stream can be decomposed directly for XPath evaluation. Experimental results show that XSC outperforms other methods in compression ratio and compression efficiency, and the cost of XPath evaluation on compressed data stream is acceptable.

Abstract:Traditional Transmission Control Protocol (TCP) works well in wired network but suffers from performance degradation in mobile ad-hoc networks (MANET) due to the fact that it cannot distinguish packet losses due to congestion from packet losses due to link breakage, channel error and route changes. In this paper, an enhanced TCP, named TCP Freeze-Probing, is proposed to improve the TCP performance in mobile ad-hoc networks. TCP Freeze-Probing is an end-to-end approach that does not need the cooperation of the intermediate nodes in the network. Besides, a throughput model for TCP Freeze-Probing is given, which is validated through simulation. It is shown by analysis and simulation that the proposed approach can greatly improve the TCP performance in MANET.

Abstract:Engineering a large IP backbone network without a view of internal link state is challenging. Previous algorithms assume that probes experience fixed routes in networks. As there are load-balancing equipments in networks, probes are delivered across random routes. This results in invalidation of the prevous algorithms. New algorithm proposed in this paper uses CGF (cumulate generating function) to infer delay characteristics of the internal link under stochastic routes. Simulation results prove that the algorithm could resolve the delay inference in the networks with load-balancing route. Based on the delay characteristics of the internal link, the bottleneck link can be located.

Abstract:According to the requirement of network massive storage applications, this paper puts forward a P2P based objects distribution and location model, supporting the logic network dynamically composed by a large number of voluntary nodes. The model is discussed in detail as follows: global mapping relation, routing table, object locating and routing algorithm, object indices distribution scheme, and maintenance algorithm when nodes join and leave the network. In particular, a novel scheme for distributing objects indices is provided to improve the average success rate of objects access, and each part of the model is improved. Through analysis, the model fulfills the five objectives given in the introduction. Finally, a simulation program built on this model verifies the expected abilities for balancing the load distribution and improving the objects access efficiency.

Abstract:Active queue management (AQM) is a very active research area in networking. Compared with drop-tail, AQM can provide smaller average queue delay and higher bandwidth utilization. Although the performance of proportional integral (PI) controller is superior to that of random early detection (RED), its convergence speed is slow. This paper proposes an adaptive proportional integral (API) algorithm based on the original PI. API obtains load information by measuring the current packet-dropping rate, then sets PI parameters accordingly. Verified by using NS-2 simulations, API can achieve faster convergence speed and smaller queue oscillation than PI and PIP (proportional integral based series compensation and position feedback compensation) which is an improved algorithm of PI.

Abstract:Proxy signature allows an original signer to delegate his/her signing capability to a proxy signer such that the proxy signer can sign messages on behalf of the original signer. Blind signature allows a user to have a given message signed by the signer without revealing any information about the message. By using Schnorr blind signature, Tan et al. recently proposed a digital proxy blind signature scheme. They claimed that it satisfies the security properties of both blind signatures and proxy signatures. However, it is not the fact. This paper shows that the proposed scheme is vulnerable to universal forgery as well as linkability attacks. It also explains why their proofs of the security are incorrect.

Abstract:Building a virtual network topology named P2P overlay network on top of Internet’s physical topology layer based on P2P computing mode can lead to the effective building of a full-decentralized internet based self-organizing network routing model—hierarchical aggregation self-organizing network (HASN). The target and architecture of HASN are described in this paper, as well as a detailed description of the P2P decentralized naming, route discovering and updating algorithm-HASN_Scale. Simulation results testify the performance of HASN.

Abstract:This paper presents the design of an end-to-end adaptive smoothing and TCP-friendly transmission for stored MPEG-4 fine-grained scalable (FGS) videos over the best-effort Internet. The goal is to minimize the playback quality variation when the network conditions are constantly varying. A novel framework for FGS video delivery over a TCP-friendly connection is first presented. In the context of this scheme, and under the assumption of complete knowledge of bandwidth evolution, an offline quality adaptive smoothing algorithm is derived, and an online adaptive smoothing algorithm is also developed based on the predicted available bandwidth to stream FGS video over the TCP-friendly rate control (TFRC) Protocol with the enhanced ARAR model. Through simulation experiments, it has been shown that the online adaptive algorithm performs almost as well as the offline version for a wide-range of the bandwidth scenarios, and a smooth and TCP-friendly video transfer can be accomplished by the proposed scheme.

Abstract:Linear complexity is an important design index for assessing the cryptographic strength of a sequence.Pseudorandom sequences with large linear complexity and large k-error linear complexity is a hot topic in cryptography and communications. Niederreiter found many such periodic sequences over F_q firstly. In this paper,the authors construct some periodic sequences over F_q with very large 1-error linear complexity by the GDFT of a periodic sequence. The result is much better than the known ones.

Abstract:Optimal BGP route selection on traffic demand is one of the problems in interdomain traffic engineering. Determining bottleneck area will give important heuristic information to the problem. As the problem of determining bottleneck area is NP-hard, a bottleneck area predicting algorithm on traffic demand in polynomial time is proposed, which deals with interdomain peering links and intradomian links simultaneously. Moreover, this paper also analysis the relationships between the traffic and the bottleneck area, as well as the relationships between the topology and the bottleneck area. Simulation results show the accuracy of the algorithm is more than 90%. In addition, a conclusion is drawn from the simulation that the topology is a very important factor in determining the bottleneck area.

Abstract:The characteristic that nodes can enlist into the network topology freely and independently without any fixed infrastructure makes mobile Ad Hoc networks (MANET) widely used in various environments such as disaster rescue, battlefield and so on. Traditional Ad Hoc network routing protocols are mainly based on the condition “Shortest Path”, and possibly form many local “hotspots” to influence the performance of the network. To this point, a new metric “Leisure Degree” is presented for denoting the transmission state of the node. Based on this metric, a cross-layer design method that involves the MAC (medium access control) layer and the network layer is used to present a new routing protocol Leisure Degree Adaptive Routing (LDAR), which uses a heuristic routing select mechanism to efficiently control the congestion and share the load. This method can improve the performance of the Ad Hoc networks. Simulation results show that LDAR performs much better than the DSR routing protocol under the circumstances of both static networks and mobile networks.

Abstract:Delegation is an important security policy that should be supported by RBAC model. The basic idea of delegation is that some active entity in a system delegates authority to another active entity to carry out some functions on behalf of the former. The grantor of the delegated roles should be responsible for the usage of them, so constraints on the usage of the delegated roles are critical components of the whole delegation model. Currently, there’re some models that extend RBAC model to support role delegation. However, their supports for constraints on the usage of delegated roles are very limited. This paper presents the requirements of role-based delegation, including temporary constraints, regular role dependency constraints, partial delegation constraints and propagation constraints. The former two kinds of constraints are modeled with a formal model – CRDM, which provides the foundation for applications in need of the constrained delegation.

Abstract:A kind of interference, called direct access interference, is found in the calculus of Mobile Resources(MR), which will cause more damage than the grave interference one finds in the calculus of Mobile Ambients,because in MR malicious environments or contexts can freely access the sensitive resources inside a process. This kind of interference should be regarded as a program error. To control the direct access interference, we devise a variant of MR, the calculus of Safe Mobile Resources (SR). The authors use a type system to avoid the occurrence of all direct access interferences. Due to the study, the grave interference is a special form of the direct access interference, which is also controlled in SR. At the end of the paper, several examples are provided to illustrate how to use the new calculus and how robust it is.

Abstract:This paper presents a novel peer-to-peer structure overlay network SmartBoa. Compared to previous protocols, SmartBoa nodes have routing tables with different sizes, which are determined by the local nodes individually. It is ensured that the bandwidth cost of a node is proportional to its routing table size. Therefore, from the view of the whole system, all the allowable bandwidth are fully utilized to improve the routing efficiency. SmartBoa does not increase the capacity requirement for nodes when the system expands, so it can achieve higher scalability than the one-hop protocol. Furthermore, SmartBoa nodes can adjust its level at runtime, and thereby can warm up gradually when starting. This avoids the long-time initiation which is an important problem in one-hop overlay. In a word, SmartBoa is a general structure overlay network that can be deployed in any environments, not matter what the system size is, how dynamic the nodes are, and what the node-capacity distribution is like.

Abstract:Joint access to shared resources (e.g., objects, applications, and services) among autonomous domains that form a coalition has recently become important in both military and commercial areas. The brass tacks in coalition are that these domains have different self-interests although they focus on achieving a common goal. In this paper, to enable effective protection of jointly-owned resources, the notion of trust into coalition access control is built, and a fine-grained access control policy based on quantifying permission idea is proposed. The usage format of permission in this policy is meta-permission that is a share of access permission to coalition resources and is owned by multiple domain users. When accessing jointly owned resources, the sum of participants'meta-permission value must attain a predefined permission quantity called "permission-threshold" and an assigned participant member number. In addition, permissible time span of the meta-permission is also taken into account to achieve the above goals and access requesting time must fall into their common permissible time span. Doing this enables the coalition to retain control over the access to coalition resources in distributed environments. Lastly, the preserving security property of the fine-grained access control policy with respect to state transition is proven.

Abstract:In Web information retrieval (IR), input queries are too short and fuzzy to describe user request, which leads to the mismatch problem between user query and the documents full of redundancy and noise. This paper first studies the feature of web documents information and proposes the concepts of primary feature word, primary feature field and primary feature space (PFS). Then a new PFS query term weighting scheme is proposed, which takes document frequency (DF) into account instead of the traditional IDF factor. Finally, a combination strategy of term weighting is given. Using this PFS model, three groups of experiments have been performed on 10G and 19G large scale Web collections with TREC9, TREC10 and TREC11 standard tests of Web tracks. Comparative studies indicate that the new DF-related PFS term weighting improves the system performance consistently and effectively in terms of recall, top n precision and mean average precision. At most 18.6% improvement has been made.

Abstract:Relay routing system consists of a set of relay routers, providing relay routing for routing domains that cannot exchange routing information. The key point of the system is to configure appropriate relay router for routing domains. In this paper, all the relay routers are taken as a virtual node by assigning them an anycast address which is accessed along the shortest path by anycast routing. In addition, source routing is used to route the data packets to relay router. The anycast-based relay routing system enables auto-configuration of relay routing,improves the performance and the reliability of relay routing, and it is compatible with existing network infrastructure with quite low overhead.

Abstract:Disk Array is adopted widely because of its high performance I/O. To adapt the need of applications’ changeable I/O performance, I/O storage subsystem should be highly scalable. So DDA (dynamic disk array), which can scale adaptively, is an ideal system. The key technology of DDA is its data placement algorithm and online data reorganization algorithm. The main contribution of the paper is: first, a detailed study on DDA data placement is conducted and a new placement method, D/H, is presented. In D/H placement, the space in DDA is balanced after scale, and the reorganization cost is minimized; then, an online data reorganization algorithm based on control feedback theory is provided. With this strategy, the reorganization in DDA does little impression to the system QoS, and under this condition, data reorganization can be accomplished as quickly as possible; finally, simulation results show that Online Data Reorganization based on Control Theory is useful.