Abstract Heterogeneous high-performance computing (HPC) systems have been proposed as a power efficient alternative to traditional homogeneous systems. In heterogeneous HPC system, fast CPUs which have complex pipelines, high clock frequencies as well as high power consumption are combined with slow ones which have simple pipelines, low clock frequencies as well as low power consumption. Different types of applications should be... Heterogeneous high-performance computing (HPC) systems have been proposed as a power efficient alternative to traditional homogeneous systems. In heterogeneous HPC system, fast CPUs which have complex pipelines, high clock frequencies as well as high power consumption are combined with slow ones which have simple pipelines, low clock frequencies as well as low power consumption. Different types of applications should be distributed to run on different type of CPUs, in order to achieve a trade-off between energy and performance. In this paper, we use experimental research to investigate the applications classifications which categorize each application as the CPU-intensive, memory-intensive, or phase-change application. We also conduct some experiments to measure the current, power, and energy of different types of applications. Afterwards, a scheduling method for applications on heterogeneous HPC systems is proposed. The experiment shows that the scheduling method can trade off the latency and energy consumption based on the applications classification and measurement. Read More Read Less

Abstract The Ubiquitous Computing envisions a world of communicating mobile devices and sensors. Advances in wireless communications raise users’ expectations about the accessibility of information in Ubiquitous Computing environments. Getting meaningful information from disparate sensor nodes and/or mobile devices spread in Ubiquitous Computing environment is an urgent task. On the one hand, the heterogeneity of the sensor nodes makes share and... The Ubiquitous Computing envisions a world of communicating mobile devices and sensors. Advances in wireless communications raise users’ expectations about the accessibility of information in Ubiquitous Computing environments. Getting meaningful information from disparate sensor nodes and/or mobile devices spread in Ubiquitous Computing environment is an urgent task. On the one hand, the heterogeneity of the sensor nodes makes share and finding data at the mobile devices and sink nodes much harder. On the other hand, nodes in Ubiquitous Computing environment are highly dynamic which means query planning in Ubiquitous Computing environment should be able to dynamically re-make their plan when the network conditions change due to intermittent connectivity. To address these two issues, this paper proposes a new service-oriented framework(called UCDS, standing for Ubiquitous Computing Data Services ) at application-level, which supports for semantic integration and querying of heterogeneous data sets distributed over the Ubiquitous Computing environment. We propose a RDF-View-based approach for mobile data mediation, and describe the service-composition-based query planning algorithm implemented in UCDS. We explore an ontology based approach to, mapping mobile data to data services (DS for short), publishing DS with the shared domain ontology. We also explore dynamically evolving service interface enabled by richly service description capability using the semantic web languages, and answering queries through DS composition. Experimental results show that the approach adopted by UCDS provides a better performance compared with the conventional approach. Read More Read Less

Abstract In large-scale grid environment, the authorization plays a vital role in access control to resources. For the demand of higher dynamicity, complexity and granularity on grid environment, a novel model named dynamic role-based access control (DRBAC) based on RBAC model is presented. DRBAC introduces conceptions of several objects such as rolegraph initial structure, atom role and middle role, and it... In large-scale grid environment, the authorization plays a vital role in access control to resources. For the demand of higher dynamicity, complexity and granularity on grid environment, a novel model named dynamic role-based access control (DRBAC) based on RBAC model is presented. DRBAC introduces conceptions of several objects such as rolegraph initial structure, atom role and middle role, and it can automatically adjust the role-graph’s structure through dynamically adding new roles or deleting original roles. The authorization mechanism which combines DRBAC model and CAS servers provides an effective method to solve loading and security problems. Theoretical analyses and example demonstrate that it is of high safety and has good time and space complexity when authorizing. Read More Read Less

Abstract The Ubiquitous Computing envisions a world of communicating mobile devices. As mobile devices usually have limited resources, it brings up the challenge to search and invoke services at any time and any place in an accurate, efficient and adaptive way. In this paper, we propose to filter services by matchmaking the precondition of service invocation and the context information of... The Ubiquitous Computing envisions a world of communicating mobile devices. As mobile devices usually have limited resources, it brings up the challenge to search and invoke services at any time and any place in an accurate, efficient and adaptive way. In this paper, we propose to filter services by matchmaking the precondition of service invocation and the context information of service consumer, in order to increase the accuracy and adaptivity of service discovery and recommendation in mobile environment. We construct context ontology to describe the environment when a service consumer submits service request first. And then, we propose to formalize the precondition of a successful service invocation based on a lightweight DL theory. Finally, reasoning on context satisfiability can be used for service recommendation. A case study shows our approach works well Read More Read Less

Abstract At present, the online network reprogramming faces two challenges: 1) How to save the energy consumed during the code dissemination; and 2) How to enhance the security of online network reprogramming. In this paper, we modify the work flow of online reprogramming system and separates an alone reprogramming component from the whole program image which need not be transmitted in... At present, the online network reprogramming faces two challenges: 1) How to save the energy consumed during the code dissemination; and 2) How to enhance the security of online network reprogramming. In this paper, we modify the work flow of online reprogramming system and separates an alone reprogramming component from the whole program image which need not be transmitted in order to reduce the energy consumption. According to the security vulnerabilities in the image management aspects of code dissemination, a cipher puzzle-based image management algorithm is proposed. According to the based-request attacks, several counter measures are introduced in our online reprogramming system, such as secure broadcast authentication based on cluster keys and the request count technique. By integrating several key techniques, our online reprogramming system can securely and energy-efficiently disseminate the new program image into the whole WSNs. Read More Read Less

Abstract This paper constructs a teaching resources sharing platform based on campus network, which has integrated the existing digital teaching resources, and collected the scattered resources together and improved the whole unreasonable resources layout, repetitive construction and so on. Since the network is convenient and real-time, it is suited to be applied to the traditional classroom teaching. The teaching resources share... This paper constructs a teaching resources sharing platform based on campus network, which has integrated the existing digital teaching resources, and collected the scattered resources together and improved the whole unreasonable resources layout, repetitive construction and so on. Since the network is convenient and real-time, it is suited to be applied to the traditional classroom teaching. The teaching resources share platform can not only help students to make full use of information technology tools for the realization of personalized learning, but also can strengthen the learning interests and enthusiasms of the students. Read More Read Less

Abstract In order to control reasonably the acceleration and dynamic tension in full-load starting process of conveyor system, accurate design techniques should be adopted. This paper selects the viscoelasticity model suiting to the belt by comparing dynamic performance of some kinds of models, and develops simulation model of driving device with feedback control, which indicating that motor rotation speed and its... In order to control reasonably the acceleration and dynamic tension in full-load starting process of conveyor system, accurate design techniques should be adopted. This paper selects the viscoelasticity model suiting to the belt by comparing dynamic performance of some kinds of models, and develops simulation model of driving device with feedback control, which indicating that motor rotation speed and its output torque vary with load and time. For belt conveyors, it has high nonlinear trait, PID controller with fixed parameter can not achieve good performance index. So the PID controller based on neural network is proposed. This design regards belt conveyors as a control object, by analyzing danamic model of belt conveyors, the design of related neural network PID control system based on BP neural network is provided. The BP-ANN is designed with three layers which are input-layer, hidden-layer and output layer in consideration of the control system real time requirement. The results show that the controller based on the neural networks can improve the robustness of the system and has better adaptabilities to the model and environments, compared with the classical PID control, therefore the simulation results are more close to practice. Read More Read Less

Abstract Reference simplifies the ODP concept, establish a user hierarchical structure can be controlled autonomously interest preference expression of user interests model. Meanwhile, from the user’s browsing behavior can reflect the user’s interest in starting, this paper analyzes the relationship between the user’s browsing behavior and user interest. Considering both the times of page visiting and the browsing speed for the... Reference simplifies the ODP concept, establish a user hierarchical structure can be controlled autonomously interest preference expression of user interests model. Meanwhile, from the user’s browsing behavior can reflect the user’s interest in starting, this paper analyzes the relationship between the user’s browsing behavior and user interest. Considering both the times of page visiting and the browsing speed for the degree of use interest, as well as the similarity of the page keywords and a single user interest point, the index can be obtained for evaluating whether a page meets the user interest preference. Utilizing it as search results of evaluation indexes, search results sorting. Further analysis under different integrity of user interests of description and different results processing under the way of sorting effect, through individual apiration is decided according to user’s personal interest in server-side describe file integrity of, will be willing to disclosure of personal information stored in the search engine server in order to improve the quality of the search results, sensitive personal information retained in local client, reach users privacy protection effect. Read More Read Less

Abstract Color consistency is one of the key techniques for color marketing in color reproduction in network transaction system. A new color management method is presented based on analyzing the color rendering principles of scanned input images. Firstly, the method substitutes color blocks in color shade district in standard color target for complete color space to solve the difficulties of color... Color consistency is one of the key techniques for color marketing in color reproduction in network transaction system. A new color management method is presented based on analyzing the color rendering principles of scanned input images. Firstly, the method substitutes color blocks in color shade district in standard color target for complete color space to solve the difficulties of color sample selection. Secondly, contraction-expansion curve fitting algorithm is improved to improve its search efficiency and decrease its calculation. Thirdly, parameters of Yule-Nielson equation are reinterpreted, making them applied for non-printing dot image which only can be used for printed dot image originally and then the color conversion model for scanned images is deduced gradually though single color, double color and tricolor conversion correction. Finally the experimental results show that the model can improve scanner color conversion accuracy and can guarantee color consistency in color marketing in electronic business system in practice. Read More Read Less

Abstract DDC (Digital Down Converter), as a key part of digital receivers, aims to get low-speed digital baseband signals by extracting and filtering the digital signals transformed by high-speed ADC (Analog Digital Conversion). This paper employs the multi-rate process technique to design the DDC, i.e. firstly conducting data CIC (Cascade Integrator Comb) extraction, then HBF (Half- Band Filtering) and at last... DDC (Digital Down Converter), as a key part of digital receivers, aims to get low-speed digital baseband signals by extracting and filtering the digital signals transformed by high-speed ADC (Analog Digital Conversion). This paper employs the multi-rate process technique to design the DDC, i.e. firstly conducting data CIC (Cascade Integrator Comb) extraction, then HBF (Half- Band Filtering) and at last low pass filtering. CIC, which doesn’t need multiplication, can realize high speed filtering and the HBF has high operation efficiency and good realtime performance. FPGA realization is taken into full consideration during the project design of the top. In the end, this paper adopts the Simulink instrument in MATLAB and combines Altera’s signal processing toolbox- DSPBuilder to implement the design scheme by the FPGA. It focuses on the description of FPGA implementation of the digitally controlled oscillator, CIC filter, half-band extracting filter and the low pass filter. Read More Read Less

Abstract Because of its data centralized management characteristic, which conveniently realize resources sharing and the data fast clone, the service deployment technology based on the centralized storage can achieve fast deployment for the system service through the network boot methed or the method of downloading disk image from the remote storage server. However, for the data centralism the I/O load is... Because of its data centralized management characteristic, which conveniently realize resources sharing and the data fast clone, the service deployment technology based on the centralized storage can achieve fast deployment for the system service through the network boot methed or the method of downloading disk image from the remote storage server. However, for the data centralism the I/O load is also congested on the storage server, which creates the drop of the overall system performance and the reducing of the service quality. So we need to test and analyze its performance factors, which provide the clue for the solution of the performance bottleneck. In this paper, we use the SonD deployment system, which is based on the centralized storage, as the analysis prototype to carry on the analysis to the client boot process, evaluate its performance influence factors such as the memory size of storage server, data distribution on the storage server through the real or simulated method, and give suggestion of the performance optimized. Read More Read Less