Search results for " computing"
showing 10 items of 2075 documents
Towards adaptive localization in wireless sensor networks
2012
It is often useful or even indispensable to know the locations of the nodes in a wireless sensor network (WSN), from both network functionality and applications point of view. A potential yet one of the most challenging method is to utilize received signal strength indicator (RSSI) for distance estimation. Albeit easily adopted, using of RSSI is notorious for imprecise distance estimates which can lead to inaccuracies in location estimates. In this paper, we study RSSI-based localization in WSNs through simulation experiments. We evaluate and compare a hybrid algorithm and its modifications with well-known range-based and range-free localization algorithms. Simulation results indicate that …
Rapid prototyping platform for stream-oriented reconfigurable computing applications
2010
In this paper we present a methodology and tool for rapid prototyping of real time image processing applications. We describe our design flow of multiprocessor system on chip (MPSoC) architectures based on hardware/software components. This methodology provides automated methods to specify, generate the hardware, software, and the architectural interfaces between them. Our methodology starts from system level specification of the application with parallel processes described in C-code. The processes communicate through an abstract channel called streams. We describe also the solution that we proposed to synthesize a custom bus architecture for the reconfigurable computing applications, whic…
Approximation algorithm for constrained coupled-tasks scheduling problem
2014
International audience; We tackle the makespan minimization coupled-tasks problem in presence of compatibility constraints. In particular, we focus on stretched coupled-tasks, i.e. coupled-tasks having the same sub-tasks execution time and idle time duration. In such context, we propose some complexity results according to several parameters and we design an efficient polynomial-time approximation algorithm.
Two Job Cyclic Scheduling with Incompatibility Constraints
2001
The present paper deals with the problem of scheduling several repeated occurrences of two jobs over a finite or infinite time horizon in order to maximize the yielded profit. The constraints of the problem are the incompatibilities between some pairs of tasks which require a same resource.
Time and work generalised precedence relationships in project scheduling with pre-emption: An application to the management of Service Centres
2012
Abstract In this paper we present an application of project scheduling concepts and solution procedures for the solution of a complex problem that comes up in the daily management of many company Service Centres. The real problem has been modelled as a multi-mode resource-constrained project scheduling problem with pre-emption, time and work generalised precedence relationships with minimal and maximal time lags between the tasks and due dates. We present a complete study of work GPRs which includes proper definitions, a new notation and all possible conversions amongst them. Computational results that show the efficiency of the proposed hybrid genetic algorithm and the advantages of allowi…
Exact Response Time Analysis of Hierarchical Fixed-Priority Scheduling
2009
Hierarchical scheduling has recently been used to provide temporal isolation to embedded virtualised systems. Response time analysis is a common way to derive a schedulability test for these systems. This paper points out that response time analysis for hierarchical fixed-priority scheduling found in the literature is only exact for tasks of the highest priority domain. For the rest of the tasks is an upper bound. In our work, we provide the exact analysis and we compare it with previously published works.
A Real time locating system for local fleet management
2006
Abstract Local locating systems are necessary where a big number of units are handled. In this case, we propose a management system for parking spaces for vehicles that are distributed among eight different-sized lots on a 1.8 million square meter premises. In order to quickly locate each particular vehicle in this area, a new real time location system (RTLS) has been successfully installed. The vehicle location system Moby-R automatically monitors every vehicle movement and tracks its current parked position. All the vehicles are given a data carrier which is attached to the inside rear-view mirror with a special hanger, precisely locating the vehicle within approximately five meters
Accelerating H.264 inter prediction in a GPU by using CUDA
2010
H.264/AVC defines a very efficient algorithm for the inter prediction but it takes too much time. With the emergence of General Purpose Graphics Processing Units (GPGPU), a new door has been opened to support this video algorithm into these small processing units. In this paper, a forward step is developed towards an implementation of the H.264/AVC inter prediction algorithm into a GPU using Compute Unified Device Architecture (CUDA). The results show a negligible rate distortion drop with a time reduction on average up to 93.6%.
Optimizing H.264/AVC interprediction on a GPU-based framework
2011
H.264/MPEG-4 part 10 is the latest standard for video compression and promises a significant advance in terms of quality and distortion compared with the commercial standards currently most in use such as MPEG-2 or MPEG-4. To achieve this better performance, H.264 adopts a large number of new/improved compression techniques compared with previous standards, albeit at the expense of higher computational complexity. In addition, in recent years new hardware accelerators have emerged, such as graphics processing units (GPUs), which provide a new opportunity to reduce complexity for a large variety of algorithms. However, current GPUs suffer from higher power consumption requirements because of…
Fitting linear models and generalized linear models with large data sets in R
2009
We present an estimating algorithm to fit linear and generalized linear models not involving the QR decomposition. Some new R functions are presented and discussed. For large data sets, comparisons with respect to the well-known lm() and glm(), as well as to biglm() and bigglm() from the package biglm, show that the proposed functions speed up computation while preserving numerical stability and accuracy