Abstract: In precision agriculture application，sensor nodes are required to keep working for months，involving in more severe challenges for the power management of the energy-limited wireless sensor networks.Wireless sensor networks’ power management is a complex research area referring to hardware，OS and application software.It needs to coordinate all the network layers’ operation.This paper focuses on the node level power management，introduces the relative research on sensor node power management and the dynamic power management strategy in TinyOS，designs a CPU duty cycle analysis mechanism for solar greenhouse application named CPUSA，providing basis for further power management optimization.Oscilloscope is used to acquire the CPU status wave to verify the veracity of the analysis method.The relative analysis on the result of the two methods shows that the method works perfectly.

Key words: precision agriculture, TinyOS, wireless sensor network, power management, CPU duty cycle analysis

摘要： 在精准农业应用中，传感器节点需持续工作数月，这对能量受限的传感器网络的能量管理提出了更为严峻的挑战。传感器网络的能量管理是一个涉及硬件、操作系统、应用软件的复杂研究领域，需要协调网络各个层次的操作。主要研究节点层的能量管理机制，介绍了节点能量管理的相关研究和TinyOS下的动态能量管理策略，设计了日光温室应用场景中CPU占空比分析机制CPUSA，为进一步的能量优化提供依据。利用示波器获取CPU状态波形并与CPUSA的结果进行对比分析，验证了CPUSA的准确性。

关键词: 精准农业, TinyOS, 无线传感器网络, 能量管理, CPU占空比分析