A timely and accurate prediction of the vehicle's trajectory can render a good service, help predict traffic flow, and in the more interesting case, detect potentially dangerous situations as early as possible. It...
详细信息
ISBN:
(纸本)9781538637906
A timely and accurate prediction of the vehicle's trajectory can render a good service, help predict traffic flow, and in the more interesting case, detect potentially dangerous situations as early as possible. It is widespread to use historical data to predict trajectories. However, due to the sparsity of trajectory, sometimes it is difficult to predicate trajectory. A novel approach to trajectory prediction is proposed which has the capability to predict the trajectory with sparse data. To achieve this, we denote city route and trajectory as a series of fixed points. Under normal conditions, only the most similarly trajectory is used to predict trajectories, which may mislead the result. To remedy this problem, we use several most similarity trajectories to predict the next point, and regard the most points as our result. The experimental results show that the approach improves the prediction accuracy 8%, coverage 2%, and greatly outperform the baseline algorithm (30 ms) compared with Spatial Iterative Grid Partition (SIGP).
The proceedings contain 84 papers. The topics discussed include: impact of NUMA effects on high-speed networking with multi-opteron machines;an ensemble method using hybrid real-coded genetic algorithm with pruning (H...
The proceedings contain 84 papers. The topics discussed include: impact of NUMA effects on high-speed networking with multi-opteron machines;an ensemble method using hybrid real-coded genetic algorithm with pruning (HRGA/P R);P2P video broadcast based on per-peer transcoding and its evaluation on planetlab;dynamic modification of wireless multihop transmission route in message-by-message manner for shorter transmission delay;an efficient partitioning and scheduling algorithm for streaming applications on FPGA with resource constraint;associativity-based adaptive weighted clustering for large-scale mobile ad hoc networks;directional node-disjoint multipath routing in wireless ad hoc network;toward more parallel frequent itemset mining algorithms;schedulability of aperiodic tasks in hybrid process model;GRID-enabled ensemble subsurface modeling;and parallel double divide and conquer and its evaluation on a super computer.
parallel recombinative simulated annealing (PRSA) is applied to the static lightwave establishment network operation problem for wavelength division multiplexing (WDM) networks. It is assumed that the cross-connects d...
详细信息
ISBN:
(纸本)1932415068
parallel recombinative simulated annealing (PRSA) is applied to the static lightwave establishment network operation problem for wavelength division multiplexing (WDM) networks. It is assumed that the cross-connects do not perform wavelength conversion. An asynchronous version of PRSA is implemented on a multiprocessor computer with multiple interacting populations. The cooling coefficient, population size, and the number of migrants is varied for PRSA to determine the affect on the convergence rate and the quality of the final solution. It behaves like a simulated annealing (SA) algorithm, but has the advantage of being easy to implement on multiple processors.
Digital signal processing as a key and difficult point of network technology research and development, currently commonly used content such as LabVIEW. But from a practical point of view, while these techniques can be...
详细信息
Unlike the conventional networks, the unique characteristics of mobile ad hoc networks (MANETs) pose a number of nontrivial challenges for pervasive service provision. Particularly, mobility of users/devices causes un...
详细信息
Unlike the conventional networks, the unique characteristics of mobile ad hoc networks (MANETs) pose a number of nontrivial challenges for pervasive service provision. Particularly, mobility of users/devices causes un-predefined and unpredictable changes in physical location and in available resources and services, event at runtime and during the same service session, thus forcing us to consider very dynamic aspects of evaluation when designing a security access control model. Alternatively, there is generally no a priori trust relationship among entities interacting in ad hoc networks which makes it essential to establish trust from scratch. This task becomes extremely challenging when it is simultaneously necessary to protect the privacy of the users involved. In this study, we show how trust evaluation process of a system can be based on previous interactions and peer recommendations. Regarding the combination of these two factors, our trust-aware access control model can establish appropriate trust values for different situations, providing a confident supervision mechanism for ad hoc users.
The proceedings contain 152 papers. The topics discussed include: swarm intelligence based data selection mechanism for reputation generation in social cloud;trust aware multi-objective metaheuristics for workflow sch...
ISBN:
(纸本)9781665496025
The proceedings contain 152 papers. The topics discussed include: swarm intelligence based data selection mechanism for reputation generation in social cloud;trust aware multi-objective metaheuristics for workflow scheduling in cloud computing;deploying mobile application on cloud;threshold based dynamic resource balancing (TDRB) algorithm in cloud computing;user knowledge modelling through azure machine learning studio;challenges and issues in energy efficient load balancing in the cloud computing environment;leaf disease detection using convolutional neural network;wind power deviation charge reduction using time series models;proposed methodology for early detection of lung cancer with low-dose CT scan using machine learning;comparison of different machine learning algorithms based on intrusion detection system;potato plant disease classification through deep learning;and face-mask recognition and detection using deep learning.
This track started in 2009 with opening remarks from the Chair observing that the computing cloud evolution depends on research efforts from the infrastructure providers creating next generation hardware that is servi...
详细信息
ISBN:
(纸本)9781538617595
This track started in 2009 with opening remarks from the Chair observing that the computing cloud evolution depends on research efforts from the infrastructure providers creating next generation hardware that is service friendly, service developers that embed business service intelligence in the computing infrastructure to create distributed business workflow execution services and service providers who assure service delivery on a massive scale with global interoperability. The state of the art architecture and evolution of the cloud at that time was already increasing datacenter complexity by piling up new layers of management over the many layers that already exist. Since then, the scale of distributed applications and their management have taken a new dimension demanding tolerance to wild fluctuations both in workloads and available computing resource pools. There are many calls to go cloud native and architect applications using the many services provided by the cloud providers such as Amazon Web Services. On the other hand, there are also calls for avoiding vendor lock-in by going multi-cloud and becoming cloud agnostic. In this conference there is a new paper that proposes cloud agnostic approach with globally interoperable cloud network using private or public network while reducing the complexity of Virtual machine image motion across clouds. In addition, there are 7 papers describing advances in current distributed and cloud computing practices dealing with quality of service, adaptive algorithms and software defined network architectures.
A wireless Mesh network is a multi-radio network with the uncertainty of node heterogeneity. This paper proposed a multi-channel allocation algorithm based on Time-Domain Greedy heuristic approximation algorithm. The ...
详细信息
ISBN:
(纸本)9781538622094
A wireless Mesh network is a multi-radio network with the uncertainty of node heterogeneity. This paper proposed a multi-channel allocation algorithm based on Time-Domain Greedy heuristic approximation algorithm. The main idea of this algorithm is to make use of the characteristics of different channels to work at the same time, withoutinterference withanother channel. It considers the different interfaces in the geographically distributed channels. The channel assignmentallows time slot to be allocated in parallel. This not only can improve the overall throughput of the whole network, but also can reduce the other overhead that caused by the time of channel reallocation and channel changes. The simulation results show that the algorithm is improved on the basis of the original, and achieves the desired effect.
parallel Discrete Event Simulation (PDES) allows to speed up the execution of a simulation by distributing the simulation's workload between multiple processors. The mapping of the simulation entities over the pro...
详细信息
ISBN:
(纸本)9789549641523
parallel Discrete Event Simulation (PDES) allows to speed up the execution of a simulation by distributing the simulation's workload between multiple processors. The mapping of the simulation entities over the processors is very important with regard to the efficiency of the overall process of simulation. This paper presents an approach for automation of the process of mapping over distributed environment (network of workstations) for optimistic simulation based on the synchronization protocol Time Warp.
Social Viewpoint Finder (SVF) is a visual analytics tool for social networks and complex networks. SVF lays out the input network data on a HIGH-dimensional (500-3,000 dimensional) Euclidean space, offers a simple, bu...
详细信息
ISBN:
(纸本)9781450355711
Social Viewpoint Finder (SVF) is a visual analytics tool for social networks and complex networks. SVF lays out the input network data on a HIGH-dimensional (500-3,000 dimensional) Euclidean space, offers a simple, but unique dragging-based UI to trigger computation-intensive, high-dimensional rotation of the presented network, and let the user investigate the clustering structure of the network. This demonstration presents the effectiveness of SVF as well as the employed implementation techniques that enabled the fluid, complex user interaction. To achieve fluidness, SVF heavily relies on modern OpenGL technologies. It achieves massively parallelcomputing through the use of the compute shader, and graphic pipelines. A large volume of the network data and its layout information is stored in a shader storage buffer. A fragment shader-based, efficient object identification method is devised.
暂无评论