信息物理系统中面向安全关键应用的多优先级数据缓冲与调度机制研究

项目名称： 信息物理系统中面向安全关键应用的多优先级数据缓冲与调度机制研究

项目编号： No.61471084

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 无线电电子学、电信技术

项目作者： 丁男

作者单位： 大连理工大学

项目金额： 80万元

中文摘要： 面向安全关键应用领域需要引入信息物理系统以应对其日益复杂的应用需求。而庞杂的监控场景和海量化数据，往往导致信息物理系统中数据排队或拥塞，使其在面向安全关键应用中应急能力下降。 本项目结合信息物理系统的闭环、动态及资源自治等特性，研究适用于其终端设备的多优先级数据缓冲与调度机制，包括：（1）针对在闭环特性下任务调度问题，研究面向高优先级数据处理的多任务协同优先调度模型与策略；（2）针对在动态特性下数据缓冲排队问题，研究面向高优先级数据优先的动态缓冲队列模型与调整策略；（3）针对在资源自治特性下缓冲资源管理问题，研究面向高优先级数据缓冲需求的资源动态分配与全局优化策略；（4）依托嵌入式系统，在智能道路车辆安全行驶应用场景中对理论成果进行仿真与验证。 本项目用以解决在信息物理系统中高优先级数据的优先响应及其服务质量问题，提升在面向安全关键应用领域中应急能力，具有重要的研究意义与示范作用。

中文关键词： 信息物理系统；面向安全关键应用；多优先级数据管理；关联性任务调度；多优先级的数据缓冲

英文摘要： It is necessary to apply the Cyber-Physical System into the safety-critical applications to solve the complex sensing and controlling requirements. However, the quality of service for dealing with the emergency data in safety-critical applications will drop when they faceing the huge amounts of data in the complex applications.And it is urgent to resolve. According the characters of closed-loop, dynamics and resource self-adjustment, the mechanism of multi-priority data caching and scheduling is considered in this project.It includes: (1) analyzing the task-relationship for dealing with data by the closed-loop character, the task-cluster model and the schedule tactics are proposed;(2) the data queue model and adjustment tactics for multi-priority data are proposed to solve the problem of data caching by the character of dynamics; (3) the model of predicting cache requirement and the caching capacity management tactics are prosed to meet the huge data caching requirement by the character of resource self-adjustment; (4) the models are simulated and tested in the scene of safety driving on the intelligent road. In this project, the quality of high-priority data for dealing is mainly considered, and it is beneficial for improving the ability of answering the emergency for the Cyber-Physical System. This project has important significance and demonstration effect.

英文关键词： Cyber-Physical System;Safety-Critical Application;Multi-priority Data Management;Task Scheduling for Data Dependency;Multi-priority Data Caching