Wireless Sensor Networks, Low-power Task Scheduling Mechanism

Wireless Sensor Network is one hot study at present.WSN's development is very quick because of its advantage of being very small and low-price,and communicating through ad-hoc net.People have tried to apply WSN into environment monitor,medical treatment,family/office automatization,industry control,and military affaires,etc. WSNOS's development provide a convenient and hardware-transparent platform for quicken this new network's broader application.But Once the WSN system is deployed and launched into work,it's hard to change the nodes' batteries.In the condition that people now have not yet made greater breakthrough in battery technology,reducing the working power of the nodes is an important approach to extend the life of WSN nodes.Researches about energy saving of WSN include every layer of the network.For example,the low-power sensor node chip,IEEE802.15.4 protocol,layered route protocol,controlling of the node's sleep,and so on.But few works through optimizing the task-scheduling mechanism to save energy is done.This paper studies the low-power task scheduling mechanism of TaraxOS on the base of our project team's new achievement,Tarax system.We break through traditional method,start with the battery's charging theory,and deduce several useful deductions according the battery's high-level analyzing model.These deductions become the paper's important basic point.The innovations of the paper are:1) Building the task model,and defining the super-period and frame.2) Proposing the periodic tasks' Frame_MaxCF scheduling algorithm and aperiodic tasks' MaxCF scheduling algorithm by making use of battery's recovery effect and nonlinear characteristics.3) Based on the CC2420 radio chip's multilevel send mode,we bring forword the improved task-scheduling algorithm,Frame_VC_MaxCF_Idle and VC_MaxCF,which send task's current is adjustable.4) Aiming at satisfying the need of quick response to high priority tasks,we develop the priority_based Frame_MaxCF_Idle.5) Designing and implementing the task power emulator,Task PowerTOSSIM, which provides the TaraxNode's energy model and battery model.Tarax applications can be run directly in the platform.Task_PowerTOSSIM can evaluate the real performance of saving energy of the different task scheduling arithmetic.Through two group experiments,we validate that the proposed task scheduling algorithm can extend the battery's lifetime up to about 4.5%～11%.We still find that the larger the average current is,the more remarkable the scheduling algorithm's effect is.In the end,we discuss the several problems about task scheduling and emulator. Then we conclude the whole reach and propose our prospect.