Real-time Scheduling for Cooperative Tasks in Sensor Networks
传感器网络中协作任务的实时调度

Wireless sensor networks are being developed for a variety of applications.With the continuing advances in network and application design,appropriate middleware is needed to provide the capability to support efficient real-time QoS for coordinative concurrent applications on sensor networks.In traditional simple real-time system,resource isola- tion mechanism is utilized for ensuring the real-time response of the tasks,it is futile to the problems where streaming data are processed collaboratively by concurrent fusion tasks and timely responses are important.The paper studies a schedule approach for providing efficient real-time services to concurrent coordinative tasks in sensor networks.We first propose a coordinative execution scheme for data stream processing.And then,by evolving server scheduling tech- niques and combining timestamp of data stream with the task execution time to tailor the design of schedule software, we present a three-layer scheduling architecture to provide real-time services while guaranteeing coordination consisten- cy.Finally,we design a three-phrase scheduling algorithms based on the model.Our performance evaluations show that by applying the three-phrase scheduling algorithms,the outer transformation ratio of collection data can be signifi- cantly enhanced and the amount of lost data can be reduced correspondingly.