Abstract:To obtain a service composition approach that satisfies the multiple requirements of users and can be easily realized, a novel method based on events is proposed in this paper. Firstly, a simple event language for services is introduced, which can be regarded as a language based on ECA (event-condition-action) rule. Secondly, a composite scheme that describes the composite services is constructed with the modularizing method based on the rule language. The scheme can not only solve the representation problem of services composition domain when adopting AI planning (artificial intelligent planning), but also the insufficiency problem in description capability when the technologies such as UML (unified modeling language) are adopted. Thirdly, to effectively present the composite scheme, the paper accomplishes the scheme’s semantics definition and codes the scheme with answer set programs. Finally, the ASP (answer set programming) technology is utilized to present the composite trajectory.

Abstract:Nowadays many researches have focused on structural ET based on statement and branch coverage and there are few researches on path-oriented ET. To solve this problem, this paper provokes an approach to construct the fitness function for test case generation in path-oriented ET based on the similarity evaluation techniques. First, a basic model for fitness function design is provided. The core of the model is to evaluate the similarity between the execution track and the target path. Accordingly three different algorithms for the similarity evaluation are provided. This model can automatically generate fitness function for each target path. The empirical studies present the superiority of the approach over several other path-oriented testing techniques, especially for the complex paths. Besides, the limitation and the applicable scope of the approach are pointed out.

Abstract:Personal Software Process (PSP) was introduced by Watts Humphrey in CMU/SEI. It is a measured software process aiming at individual software engineers. With the increasing industrial demand for software process improvement, PSP has become a hot topic for software organizations to achieve the goal of total (from macro to micro) quantitative process management. Since higher process capability is recognized as a determinant of better project performance, it is a critical step to assess the personal software process. However, the assessment of PSP capability exhibits Variable Return to Scale (VRS), Multi-Input-Multi-Output (MIMO) and Decision-Making preference problems, which makes existing traditional assessment methods ineffective. In this paper, a novel Personal Software Process Assessment method by synthesizing Data Envelopment Analysis (DEA) and Analytical Hierarchy Process (AHP)—PSPADA is proposed. PSPADA’s hybrid model and fundamental assessment algorithms (incorporating decision-making preferences and estimating return to scale) are introduced. Experimental results show that the proposed PSPADA model would be particularly helpful in assessing the capability of personal software processes under the MIMO and VRS constraint, by incorporating Decision-Making preferences.

Abstract:A QoS guaranteed resource management system architecture is proposed. The theoretical analysis demonstrates that this system architecture can well adapt to dynamic behavior of resources; Secondly, service resource co-allocation problem is formulated as MMKP (multiple-choice multiple-dimension knapsack problem) and an exact algorithm RA_BBLP and a heuristic algorithm RA_MHEU are both presented. Experimental results show that RA_BBLP which ensures the optimal solution can be used as reference of other algorithms. However, RA_MHEU can obtain near-optimal solution at high speed, which is an ideal service resource co-allocation algorithm.

Abstract:In this paper, a ring-based multi-resolution storage scheme for sensor networks is proposed. Combined with the hierarchical scheme for data storage and query, it leverages characters of the ring-based structure to support multi-resolution data storage and query operations, achieving a good performance in terms of energy consumption. It adopts optimal parameters for ring-based storage structures, which can well handle the hierarchical data storage and query scheme with the minimized overall communication cost. Furthermore, the authors do theoretical analysis on the performance of the ring-based multi-resolution storage scheme, including energy efficiency, load balance. The experimental results indicate that significant benefits can be achieved with this ring-based multi-resolution storage scheme.

Abstract:The choice of physical carrier sensing threshold affects the performance of the MAC (medium access control) layer protocol in multi-hop wireless networks. Previous research ignored the impact of the collisions correlated by ack frame and cumulative interference in determining the optimal carrier sensing range, and the collisions caused by hidden terminal were exaggerated to some extent. This paper presents a new analytical model to get the optimal carrier sensing threshold and transmit power while maximizing the network aggregate throughput. Through analyzing the cumulative interference in the network based on the carrier sensing threshold and transmit power, the channel rate and the number of the concurrent transmissions can be concluded. Additionally, four types of collision happened inside of the carrier sensing range and two types of collision contributed by the accumulative interference are proposed in this paper. Then, the channel activities can be modeled as a Markov chain that reflect the collision referred to above and the channel utilization by this Markov chain model is figured out. This paper investigates the impacts of the carrier sensing threshold and transmit power on the network aggregate throughput that determined by the channel capacity, channel utilization and the concurrent transmissions. Different from previous research, this research identifies the optimal carrier sense range without considering the collision from acknowledgement packet, accumulative interference and hidden terminals. The analysis proves that the aggregate throughput can suffer a loss, if the collision from acknowledgement packet and accumulative interference is not taken into account in determining the optimal carrier sensing range.

Abstract:A new index structure for network constrained moving objects, network-constrained moving objects dynamic trajectory R-Tree (or NDTR-Tree for short), is proposed in this paper. NDTR-Tree can index the whole trajectories of moving objects, including their historical, current, and near future positions. In order to compare the performance of the NDTR-Tree with other index structures for network constrained moving objects, such as MON-Tree and FNR-Tree, a series of experiments have been conducted, and the experimental results show that the NDTR-Tree outperforms the previously proposed network constrained moving objects index methods in dealing with dynamically maintained trajectories of moving objects.

Abstract:The mobile ad hoc network (MANET) is a temporarily organized computer multi-hop communication network which is composed of mobile hosts equipped with wireless transmitters and receivers. These hosts can move at will, which results in a highly dynamic topology and imposes significant effect on the route stability and overall performance of the network. Based on the observation of the velocity and direction of these mobile hosts, this paper proposes a routing algorithm which can predict the link stability with fuzzy logic. The simulation results have proven that this protocol works well in extending the average lifetime of data connections, in increasing the data rate as well as the end-to-end throughput, and in decreasing the end-to-end latency and overhead substantially.

Abstract:After an analysis and comparison of three most commonly used link-measuring standards in wireless sensor networks at present, it is concluded that there exists the lowest threshold of received signal strength (RSS) to satisfy a certain packet reception rate (PRR). Before taking RSS as the condition to construct topology, all nodes need to use the uniform transmitting power. Thus, the path-loss is determined as the condition of topology construction, and a path-loss based distributed topology control algorithm (PLBD) is designed. This algorithm does not only guarantee PRR, but also maintain the least loss links for correspondence between nodes. The simulation results have indicates that the topology constructed by PLBD algorithm keeps the network connective, and it has more advantages like low correspondence time-delay, good robustness and relatively balanced energy consumption.

Abstract:The paper proposes a secure inter-domain routing protocol which adopts identity-based cryptographic system—id²r (identity-based inter-domain routing). id²r consists of a key management mechanism, an origin AS verification mechanism LAP (the longest assignment path), and an AS_PATH authenticity verification mechanism IDAPV (Identity-based Aggregate Path Verification). The key management mechanism adopts a distributed and hierarchical key issuing protocol DHKI (distributed and hierarchical key issuing) to solve the inherent key escrow problem in the identity-based cryptographic system. The basic idea of LAP is that all ASes must provide the assignment path and attestations of their announced prefixes, and for a prefix, the AS which provides the longest valid assignment path is its legitimate origin AS. With identity-based aggregate signature scheme, IDAPV generates a route aggregate attestation to guarantee the authenticity of AS_PATH. Performance evaluation results indicate that based on RouteViews data on December 7, 2007, an id2r router only consumes 1.71Mbytes additional memory, which is 38% of S-BGP router; id2r has shorter UPDATE message than S-BGP; convergence time of id2r with hardware implementation of cryptographic algorithm approximately equals BGP.

Abstract:Classical saliency-based visual attention models are adapted for embedding real-time systems with less time and space costs based on approximate Gaussian pyramids of the input image. Firstly, the circular window and discrete Gaussian convolution are approximated by rectangular window and rectangular average operator respectively. Then, rectangular average operator is implemented through “row accumulation followed by column accumulation”. And conspicuity maps of each channel are calculated and sampled at desired intervals directly with linear computational complexity on the number of the input pixels. At last, a fast algorithm for inhibiting the saliency of extracted regions in the saliency map is proposed. Experimental results in the images from Berkeley segmentation dataset validate that the proposed methods have much less computational costs with acceptable outputting errors. The two approximate methods in this paper can also be applied in other image processing problems in embedding real-time systems.

Abstract:This paper presents geometric texture Wang Tiles to generate geometric textures on different object surfaces at run-time. A set of Wang Tiles is pre-computed from a specific geometric texture and used to generate geometric textures on different object surfaces. Although tile-based approaches are used for image textures, constructing geometric texture Wang Tiles presents additional challenges due to the differences between geometric textures and image textures. The geometric texture Wang Tiles is built automatically by using constrained synthesis to ensure geometric continuity across all possible combinations of tile arrangements. Once a tile set is built, it can be reused for texturing different output objects. This texturing algorithm consumes much less storage and computation than the existing methods.