The Research On Real-time Sensing Task Scheduling On WISP

In recent years,research on the Wireless Identification and Sensing Platform(WISP)is popular.WISP is a passive RFID tag that integrates sensing,computing and communication.With its small size,good computing function,wireless identification,and the ability to integrate multiple peripherals,WISP is increasingly used in the Internet of Things industry widely.Real-time systems are now widely used in industrial automation,military,power,and new energy industries.In a real-time system,there may be multiple tasks.Each task needs to be executed according to time constraints.To ensure the normal execution of each task,it is necessary to make a reasonable sequence for the execution of tasks,that is,real-time task scheduling.The goal of real-time task scheduling is to meet the time constraint of each task and respond to external requests in a timely manner.The multi-sensing task in WISP is a kind of real-time task.Because WISP is a passive tag,its strong computational function results in higher energy requirements than traditional passive RFID tags.Therefore,when the number of tasks increases,it is necessary to rationally schedule the task execution order to ensure the normal operation of the system.This paper studies the multi-sensing task of WISP,and proposes a real-time task scheduling method.It analyzes the non-overload and overload conditions of the system.Under the non-overload condition of the system,a dynamic wake-up task scheduling algorithm named S-EDF was proposed.Based on the EDF scheduling algorithm,the idle time and idle energy of the system are calculated to adjust the wake-up state of the processor to increase the idle time of the system,which is to obtain more charging time for the WISP.In addition,an acceleration factor is also proposed.When the system detects an abnormal sensory index,an abnormal alarm is issued and the task execution cycle is shortened to intensively monitor the abnormal index.Through the results of several groups of comparative experiments,it can be seen that the S-EDF algorithm can effectively improve the system schedulable task ratio and improve the time and energy utilization.When the system is overloaded,a dynamic priority task scheduling algorithm VTE is proposed.Comprehensively considering the task execution time,execution value and energy consumption,the priority of the task is redefined.The first scheduling subset for the first scheduling is selected through Shell sort to ensure that the system utilization does not exceed one,and meet as many tasks as possible Deadline.When a new task arrives,update the task schedule list.Experiments show that the VTE scheduling algorithm can reduce the deadline miss rate of the overloaded system,improve the overall execution revenue of the system,and improve the system scheduling efficiency.