Abstract:This paper presents an approach named Multiple Goals Oriented Data Prefetching (MGODP) to satisfy the data prefetching requirements from different users. MGODP does not only take the user preferences into account to prefetch appropriate amounts of data, but also adopts global coordination for Client/Server data access model to greatly improve the quality of service from the server’s perspective. Moreover, MGODP balances the workload between the mobile client and the backend server to achieve proper allocation of resources and to guarantee the system performance. Experimental results demonstrate that MGODP could satisfy diverse user requirements, and reduce the consumption of battery energy and network bandwidth through global coordination and workload balance.

Abstract:This paper suggests an ant-like agent service discovery mechanism. There are two types of agents cooperating to search target services: Search Agent and Guide Agent. Search Agent simulates the behavior of an ant that searches for services on the network. Guide Agent is responsible to manage a service route table that consists of pheromone and hop count, instructing Search Agent’s routing. Volatile pheromones make Search Agent sense the change of topology and service resource, and hop count makes them know the distance. Semantic similarity is also introduced in routing selection as a heuristic factor, which improves the recall. The life-span control policy makes query traffic controllable. With system size increasing, the query traffic would increase slightly and has an upper bound. The result of simulation shows that the suggested mechanism is scalable and adaptable enough to be suitable for large-scale dynamic computing environments.

Abstract:To address state space explosion and the inability to automatically generate the BPEL (business process execution language) codes of the existing methods of composing services based on mediators, this paper presents an approach to verify the Petri net models of service composition by modular reachability graphs. In this approach, the Petri net models of service composition are divided into sub-models in a modular way, and verify the feasibility of composition by analyzing the state spaces of individual sub-models, without unfolding to the ordinary state space. Using this modular technique can avoid the state space explosion. After verification of the feasibility, the paper proposes a method of automatically generating the BPEL codes of the whole composite service from the Petri net models of composition. The main idea is to generate the BEPL codes from the fused transitions between the sub-models based on ECA rules. Finally, an application of the methods is illustrated though a case study in an e-business enterprise.

Abstract:This paper proposes and implements a demand prefetching algorithm, which uses relaxed sequentiality criteria as well as page and page cache states as reliable sources of information. It can discover and prefetch sequential streams mixed in random accesses. It can also support the interleaved access patterns created by concurrent sequential reads on one file descriptor. Experimental results show that it performs much better than legacy Linux readahead: sequential reading intermixed with random ones are faster by 29%; I/O throughput of interleaved streams could be 4~27 times better, with application visible I/O latencies improved by up to 35 times. This demand prefetching algorithm has been adopted by Linux kernel 2.6.24.

Abstract:This paper surveys the state of the art of sentiment analysis. First, three important tasks of sentiment analysis are summarized and analyzed in detail, including sentiment extraction, sentiment classification, sentiment retrieval and summarization. Then, the evaluation and corpus for sentiment analysis are introduced. Finally, the applications of sentiment analysis are concluded. This paper aims to take a deep insight into the mainstream methods and recent progress in this field, making detailed comparison and analysis.

Abstract:This paper proposes to combine the global and local facial features in both serial and parallel manner. Firstly, global features are used for coarse classification. Then, global and local features are integrated for fine classification. In the proposed method, global and local features are extracted by Discrete Fourier Transform (DFT) and Gabor Wavelets Transform (GWT) respectively. Experiments on two large scale face databases (FERET and FRGC v2.0) validate that the proposed method can not only greatly increase the system accuracy but also improve the system speed.

Abstract:This paper presents an approach to check the satisfiability of acyclic SHOIN(D)-concepts—CDNF (complete disjunctive normal form) algorithm. This calculus can make a direct judgment on the satisfiability of acyclic SHOIN(D)-concept by restructuring it into a hierarchical disjunctive normal form group on concept descriptions which is satisfiability self-telling, and reusing description clauses to block unnecessary extensions. CDNF algorithm eliminates description overlaps to the largest extent for it works on concept description directly. Therefore, CDNF algorithm has much better performance than Tableau as it saves a lot of spatial and temporal costs.

Abstract:To learn a good distance metric without any class label information, an algorithm named HDDI-FCM (double indices fuzzy C-means with hybrid distance) is proposed in this paper. In detail, the unknown distance metric is firstly represented as the linear combination of several known distance metrics. Then the algorithm is executed to perform the clustering task as well as learn the most suitable metric simultaneously. To guarantee the convergence of the algorithm, the Steffensen iteration is introduced into the process of updating cluster centers. The selection of parameter for the algorithm is also discussed. The experimental results on a collection of UCI (University of California, Irvine) datasets demonstrate the effectiveness of the proposed algorithm.

Abstract:According as the main factor deciding the performance of ensemble learning is the diversity of component learners, clustering technology is used to speed up AdaBoost in this paper. The performance of the new algorithm is very close to the AdaBoost in learning deferent noise levels data sets. The new algorithm can be used to detect and eliminate noisy data quickly, and could achieve rapid learning on data sets after eliminating noise. It overcomes the noise-sensitive shortcoming of AdaBoost. The general performance and efficiency of the new algorithm are much better than AdaBoost in processing data sets containing noise.

Abstract:Feature extraction of data lying on disconnected manifold is an open problem in the field of manifold learning, and decomposition-composition (D-C) algorithm is the most effective method so far to deal with this problem. However, the biggest limitation of D-C method is edge problem, that is when the nearest data points of different clusters are located in the inner part instead of the edge part of the corresponding cluster, D-C method always behaves poorly. To tackle this key issue, a method, called transition curve method, is presented in this paper. The main idea of the method is to make all clusters on the underlying manifold connect more effectively by constructing smooth transition curves which connect the nearest edge points of different clusters, and in this way the global shape of the data can be preserved better in the low-dimensional space. Experimental results on a series of synthetic and image data sets verify that the transition curve method performs evidently better than D-C method. Particularlly, the edge problem is alleviated. In this way, the application scope of D-C method is expanded remarkably.

Abstract:This paper proposes a network coding-aware multi-path routing (CAMP) protocol, which forwards packets over multiple paths dynamically based on path reliability and coding opportunity. CAMP employs a route discovery mechanism which returns to the source multiple paths along with ETX (expected transmission count) of all links on each path. With its unique forwarding mechanism, CAMP splits the traffic among multiple paths and actively creates, while not merely waiting for, the coding opportunity by switching its path to maximize the switching gain and thus improves the network throughput. The experimental results demonstrate that CAMP can achieve much higher throughput than peer schemes for delivering packets in wireless networks.

Abstract:This paper proposes a query in wireless sensor networks: Peak Region Query (PRQ). Given the shape and size of the query region, i.e., a disk region with radius R, peak region query finds out a region with this shape in the network field, in which the aggregation value of the data of the sensors can be maximized. This paper first gives the definition of PRQ, and then proposes a centralized algorithm for the problem. Because the sensors have limited energy, a distributed approach EXQ (an algorithm for extreme value query processing) is proposed, which not only reduces the energy cost but also balances the workload of the sensors, so as to prolong the lifetime of the network. The basic idea is to divide the network field into overlapped sub-regions, compute a local result for each sub-region and aggregate these results to obtain the query answer. The paper compares the energy efficiency and load balance between EXQ and the centralized approach analytically and experimentally.

Abstract:This paper proposes an approximate aggregation algorithm based on Bernoulli sampling to satisfy the requirement of arbitrary precision in wireless sensor networks (WSN). Besides, two sample data adaptive algorithms are also provided. One is to adapt the sample to the varying precision requirement. The other is to adapt the sample to the varying sensed data in networks. Theoretical analysis and experimental results show that the proposed algorithms have good performance in terms of accuracy and energy cost.

Abstract:This paper proposes CET, a community based event transmitting protocol in publish/subscribe system for Delay Tolerant Sensor Networks (DTSN). The core idea of CET is that event transmitting is based on communities which are formed by sensors in the network according to their connectivity. CET consists of two key components data transmission and queue management. In data transmitting, not only events are transmitted to mobile subscribers as best as possible, some events in subscribers are also retransmitted to sensors in community, for enhancing the data delivery ratio. The queue management employs both the event survival time and the successful delivery time to decide whether the event should be transmitted or dropped for minimizing the transmission overhead. Simulation results have shown that the proposed CET achieves a higher event delivery ratio with the lower transmission overhead and event delivery delay than DG (direct gathering protocol).

Abstract:In this paper, an adaptive optimization scheme for DCF (distributed coordination function) is proposed to enhance the network performance. The scheme is based on the channel sensing result for network state information and thus it is named CSB (channel sensing backoff). The key idea to approach optimal performance dynamically in the proposed scheme is that the transmission attempt from the DCF is filtered by an adjustable probability P_T, which is dynamically adapted to reflect the current channel competing level among the network stations. Unlike other proposals for the DCF optimization, CSB does not need to perform complex on-line estimation of the number of active stations in the network, and can make adaptive tuning always toward a certain optimization object under various network states. Detailed performance evaluation of the scheme is carried out through NS-2 simulation. Simulation results show that the scheme can effectively adapt to both network size and packet length changes in the network, and simultaneously achieve performance improvements in several aspects including system throughput, collision probability, delay, delay jitter, fairness and so on.

Abstract:This paper presents a time-variant coverage mechanism, proactive coverage. In proactive coverage, all sensor nodes work in lower power surveillance manner which can detect target intrusion with high probability to save energy when the target flow doesn’t arrive. As soon as the target flow arrives, sensor nodes are awakened to build a local high quality coverage networks to sense intervening targets. When target flow leaves the target field, sensor nodes converge to a quiet surveillance state. Proactive coverage is more energy efficient than static area coverage, higher sensing quality than event-driven coverage. This paper analyzes preliminary problems in proactive coverage and finds theoretic results on initial detecting delay, awaking nodes strategy and active sensing duration. Numeric results from simulation reveal that initial detecting delay in proactive coverage is trivial. Compared with static area coverage, this coverage mechanism for an adequate scale target flow (above 30 targets) can prolong the network’s lifetime to near 4～7 times.

Abstract:In this paper, an admission control algorithm called CDPF (call dropping proportional fairness) is proposed to realize the proportional fairness of call loss probability. It dynamically computes the admission threshold for each class call to control the accepted call number according to the average call arrival rate, the average call duration time and the average cell dwell duration time of each class, and thus guarantees the call loss probability proportion of each class. In CDPF, the call loss probability of each class changes with the network load, while their proportion keeps unchanged. Simulation results show that with the change of network load, CDPF guarantees the proportional fairness of call loss probability. Compared with the existing algorithms, CDPF has lower time complexity.

Abstract:This paper proposes a topology-aware clustering model (called TCM). Furthermore, it proposes a TCM-based application layer multicast scheme (called TCMM). In TCMM, many nearby nodes are clustered, which localizes the transport of some nodes and alleviates the negative impact caused by different join sequences. Analysis and experiments show that TCMM can effectively group nearby nodes and build multicast trees with similar gross performance in different join orders. In addition, TCMM can improve some of other multicast performance in some degree.

Abstract:This paper gives an introduction to the energy-efficient broadcast/multicast routing problem and makes a survey on recent existed algorithms. Firstly, an introduction is made to the broadcast/multicast routing problem. Secondly, the communication and energy models, the definition of energy-efficiency and the optimization objectives are clearly presented. Finally, comparison is made for existing representative algorithms in two aspects: minimum energy and maximum lifetime. Open research issues and research trends are also discussed.

Abstract:Radio resource allocation and call admission control are studied in heterogeneous wireless network. A theoretical model of allocating bandwidth and connections is proposed on basis of non-cooperation game theory. Combining with the utility function of network connections, the existence and uniqueness of Nash equilibrium in the processes of non-cooperation game allocating is proved. Furthermore, a call admission algorithm is presented to ensure communication reliability based on the analysis of relationship of traffic intensity and block probability. Simulation results reveal that the allocated mechanism resolves the issues of allocated bandwidth and connections. And it ensures the reasonableness and fairness in general. In addition, the results also show that the call admission control algorithm could guarantee the communication reliability by dynamic adjusting allocated number of connections.

Abstract:The characteristics of cascading dynamics of Internet are analyzed. Different from betweenness centrality, a congestion function to represent the congested extent of node is proposed to assign a dynamic weight to every node. By introducing the concept of “delay time”, the intergradation between permanent removal and nonremoval is built in order to improve the flexibility of the model. A new evaluation function of network efficiency, based on congestion effects, is given in order to measure the damage caused by cascading failures. Finally, based on Statnet and Webgraph topologies the effects of network structure and size, delay time, processing ability and traffic generation speed on congestion propagation are investigated. The congestion propagation process composed of three phases and some factors affecting transition phenomenon are also uncovered.

Abstract:Due to the absence of actual obligation description and implementation abilities in existing DRM (digital rights management) mechanism, this paper presents an obligation authorization model and its implementation framework that can be applied in DRM. The model is based on distributed temporal logic and Active-U-Datalog rules, which empowers the model to express event-driven, time-driven, obligation compensation and other semantics of obligation descriptions, and also give the model a favorable feasibility of implementation. The semantics and syntax of the model are analyzed and explained. And the implementation mechanism of the model is discussed. Finally, the implementation, application and expressiveness of the model are showed and illustrated. The model improves the flexibility and capability of usage control of data in DRM system.

Abstract:This paper proposes a Direct Anonymous Attestation (DAA) scheme from the bilinear maps based on the decisional Diffie-Hellman (DDH) assumption and q-SDH assumption. Compared to other schemes, the scheme’s signature length is much shorter. Meanwhile, the scheme reduces the computational cost of the Trusted Platform Module (TPM) in the signing process. It gives a practical solution to ECC-based TPM in protecting the privacy of the TPM. This paper gives a detailed security proof of the proposed scheme in ideal-system/real-system security model which shows that the scheme meets the security requirements of unforgeability, variable anonymity and unlinkability.

Abstract:The paper presents a simple proxy re-signature scheme and its two equivalent security model. One is based on the universal composability framework, another is game-based security model. The proposed scheme is bidirectional, multi-use, transitive and key optimal. It is very attractive for its simplicity. Its security can be reduced to the Computational Diffie-Hellman assumption in the Random Oracle Model. It is also secure under the universal composability framework.