Mobile Patient Monitoring System (MPMS) is positioned to provide high quality healthcare services in the near future. the gap between its application demands and resource supplies, however, still remains and may hinde...
详细信息
ISBN:
(纸本)9781424449316
Mobile Patient Monitoring System (MPMS) is positioned to provide high quality healthcare services in the near future. the gap between its application demands and resource supplies, however, still remains and may hinder this process. Dynamic context-aware adaptation mechanisms are required in order to meet the stringent requirements on such mission critical applications. the fundamental model underlying an MPMS includes a set of biosignal data processing tasks distributed across a set of networked devices. In our earlier work, we designed and validated a task distribution framework to support dynamic system reconfiguration of MPMS by means of task redistribution. this paper focuses on its decision-making component that can calculate the optimal task assignment by taking into account the reconfiguration costs. this paper has three major contributions. Firstly, we study a context-aware scenario and derive the design requirements for a task assignment algorithm in MPMS. Secondly, using a graph-based system model, we proposed an A*-based task assignment algorithm that minimizes the system end-to-end delay while guaranteeing required system battery lifetime and availability. We introduce a set of node expansion rules and a pre-processing procedure to calculate the heuristic function (h(n)). thirdly, we evaluate the algorithm performance with experiments and compare this A*-based algorithm with other heuristic approaches, e.g. greedy and bounded A*.
Laplacianfaces is a recent addition to appearance based face recognition algorithms with promising future potential. Unlike Eigenfaces algorithm, Laplacianfaces algorithm finds an embedding that preserves the locality...
详细信息
Laplacianfaces is a recent addition to appearance based face recognition algorithms with promising future potential. Unlike Eigenfaces algorithm, Laplacianfaces algorithm finds an embedding that preserves the locality information of the subjects in feature space. In this study we have comprehensively evaluated the performance of Laplacianfaces against PCA on FERET face-image database using csuFaceIdEval as the testing platform. the effect of internal parameters, including size of locality to be preserved, the choice of distance measure to determine locality and the number of leading eigenvalues to be used for matching has been thoroughly studied for the first time. the impact of illumination, face expression and age variations on the relative performance of Laplacianfaces and Eigenfaces has been shown to be very significant and best parameter settings for enhanced performance have been proposed.
Multi-threaded real-timeapplications run on platforms that offer services such as concurrency, timing and synchronization. In a model driven context, separation of concerns between functional specification and design...
详细信息
Multi-threaded real-timeapplications run on platforms that offer services such as concurrency, timing and synchronization. In a model driven context, separation of concerns between functional specification and design choices is one of the key challenges. For that purpose, annotations defined by the MARTE sub-profile for software resource modeling (SRM) provide high level abstractions of target software platforms helping to deploy the same application model on various targets. However, deployment of an application on a platform is not only characterized by the API of the targeted OS, but it is also influenced by possible design restrictions set by the programming languages and their related libraries. We examine key characteristics that exist in two widely used programming languages (C++ and Java) and their libraries to analyze how this can be handled with an MDE approach based on the use of SRM.
Modem real-timeembeddedsystems are highly heterogeneous and distributed. As a result, compositional methods play an important role in the design and analysis of such complex systems. One such compositional analysis ...
详细信息
ISBN:
(纸本)9780769533490
Modem real-timeembeddedsystems are highly heterogeneous and distributed. As a result, compositional methods play an important role in the design and analysis of such complex systems. One such compositional analysis method is based oil real-time Calculus [4, 14]. In this paper we present an analysis of fixed priority non-preemptive scheduling withthe real-time Calculus. Although fixed priority non-preemptive scheduling was modeled withthe real-time Calculus previously [7], we show that the model gives overly pessimistic results. We also compare our analysis withthe existing holistic scheduling analysis [10, 3] through an example of a system using a Controller Area Network (CAN) bus [5]. the proposed method can be automated by incorporating it in the RTC Toolbox [13].
Designing safe and predictable software for embeddedapplications is a considerable challenge, which has been addressed by the COMDES-II framework. this is a component-based software framework employing a timed multit...
详细信息
ISBN:
(纸本)9780769533490
Designing safe and predictable software for embeddedapplications is a considerable challenge, which has been addressed by the COMDES-II framework. this is a component-based software framework employing a timed multitasking model of computation supported by a run-time environment. this paper presents a real-time kernel implementing the run-time environment - the HARTEX mu kernel that has been specifically developed for COMDES-II systems. the design and implementation of the kernel take into account the limitations of embeddedsystems, e.g. a small memory footprint and reduced kernel overhead It also features Boolean vector processing that eliminates the need for linked-list queues, resulting in constant execution time of kernel functions independent of the number of tasks involved.
real-timesystems are often designed using preemptive scheduling to guarantee the execution of high priority tasks. For multiple reasons there is a great interest in exploring non-preemptive scheduling in the case of ...
详细信息
ISBN:
(纸本)9780769533490
real-timesystems are often designed using preemptive scheduling to guarantee the execution of high priority tasks. For multiple reasons there is a great interest in exploring non-preemptive scheduling in the case of hard real-timesystems where missing deadline leads to catastrophic situations. this paper presents a necessary and sufficient schedulability condition for determining whether a task will satisfy its period and precedences constraints when some tasks have already been scheduled. Tasks we are dealing with are non-preemptive and the periods considered here are strict.
the distribution of real-time tasks in a networked environment in principle has several advantages, above all a high degree of flexibility easing system extension and replacement of devices and components. Component-o...
详细信息
ISBN:
(纸本)9781424415052
the distribution of real-time tasks in a networked environment in principle has several advantages, above all a high degree of flexibility easing system extension and replacement of devices and components. Component-oriented system design concepts and suitable networks such as Ethernet additionally facilitate distributed automation applications. Yet, really distributed systems are problematic to apply in practice as debugging capabilities are limited. One essential prerequisite is that network traffic can be monitored in real-time with respect to its timing within the entire system, so that events can be ordered and their interplay analysed. this paper proposes a novel distributed measurement and debugging concept for embeddedsystems where tracing facilities are directly integrated into the network nodes so as to enable and ease failure analysis and detection on all protocol layers.
the paper presents a distributed computing environment for embedded control systems withtime-triggered and event-triggered distributed processing. We have already presented a time-triggered distributed object model a...
详细信息
ISBN:
(纸本)9780769533490
the paper presents a distributed computing environment for embedded control systems withtime-triggered and event-triggered distributed processing. We have already presented a time-triggered distributed object model and a time-triggered distributed computing environment for embedded control systems. However there are many embedded control systems withtime-triggered and event-triggered processing. In this paper we present two kinds of event-triggered distributed object models, a pure event-triggered distributed object model and a data-triggered distributed object model, in addition to the time-triggered distributed object model. We also present a distributed object computing environment based on a time-division scheduling for the mixed architecture withtime-triggered and event-triggered distributed processing. the time division scheduling divides an execution cycle into a time-triggered processing segment and a non-time-triggered processing segment. the time-triggered distributed processing is executed in the former segment, and the event-triggered distributed processing is executed in the latter segment. the distributed object computing environment consists of a real-time operating system withthe time division scheduling and distributed computing middleware to support the three kinds of distributed object models. We provide a development environment that generates stubs and configuration data to build distributed control systems.
Control systems are typically designed assuming all ideal behavior of the computing infrastructure where controllers execute. In practice, however in highly loaded computingsystems consisting of multiple concurrent c...
详细信息
ISBN:
(纸本)9780769533490
Control systems are typically designed assuming all ideal behavior of the computing infrastructure where controllers execute. In practice, however in highly loaded computingsystems consisting of multiple concurrent controllers, resource constraints may introduce delays and jitter in control loops that may degrade control performance significantly. Hence, taking resource constraints into account since the beginning of the design cycle is crucial for optimizing the performance of a control system. In this paper we propose a general framework for evaluating the performance of a control system as a function of multiple timing attributes (e.g., sampling frequencies, delays and jitter) and for selecting the proper control task parameters (e.g., periods and deadlines) taking resource constraints into account. the proposed framework is illustrated using a real control plant.
暂无评论