在無線技術快速發展的時代,藉由為數眾多的無線感測器收集並分析資訊的需求亦隨之快速成長,如何能在無線感測器間建立一套有效率的通訊方式以及電源管理的方法,已成為相當重要的課題。本論文提出一種適用於工地結構監控與生醫無線監控的網路通訊協定整合電源管理的方法。此通訊協定採用階層式網路的架構,在區域感測器網路中採用星形網路拓樸以降低無線感測器的通訊複雜度,區域感測器網路之間採用點對點網路拓樸架構,對區域感測器網路間的運作時間進行排程以避免鄰近區域感測器網路的干擾。 本研究致力於區域感測器網路內部的通訊協定設計,以提供無線感測器之非事件驅動且極低工作週期的運作模式,並透過離散事件模擬方式評估此通訊協定下的網路特性與無線感測器的壽命。在感測器網路的實測上,已經能透過區域控制中心來控制附近的無線感測器,令其依照設計的通訊協定形成區域感測器網路,並將此區域感測器網路應用於結構監控與生醫感測器訊號擷取之中。此外在無線感測器的電源供應上加入電源汲取系統,汲取環境振動所產生的能量,用以彌補無線感測器處於閒置模式的耗電量。可進一步提昇無線感測器的電池使用壽命,降低維護成本。 簡言之,本論文將對無線感測器的硬體設計,通訊協定的設計與模擬,電源設計與管理,及最終的整合與感測器網路實測進行探討。希望能建立最具能源使用效率的感測器網路架構。
In the age of rapid development on wireless technology, the need of information collection and analysis with numerous wireless sensor nodes grows correspondingly. How to build a set of efficient communication protocol and power management method becomes an essential task in this era. In this thesis, an appropriate network communication protocol that is suitable for power management of site monitoring and for healthcare monitoring was proposed. The communication protocol uses hierarchical architecture, which applies star topology in the local area wireless sensor network in order to decrease the communication complexity of wireless sensor network. Ad-hoc topology is employed among local area wireless sensor networks and the local control centers negotiate the operating schedule to avoid interference between ambient local area wireless sensor networks. The research attempts to design the interior local area wireless sensor network protocol in order to provide wireless sensor nodes non-event-driven and ultra-low duty cycle power efficient nodes. Network characteristics and battery life-time estimation of wireless sensor node under the communication protocol is simulated by discrete event simulation method. In the wireless sensor network experiment, the wireless sensor nodes around the local control center can be controlled under the implemented communication protocol to form a local area wireless sensor network. Applying the local area sensor network in structure monitoring and healthcare monitoring are demonstrated successfully. Besides power efficient protocol, power harvesting system is implemented into power supply for the wireless sensor node, which scavenges environment vibration energy for power consuming usage in the idle mode is also detailed to further increase battery lifetime. The wireless sensor node hardware design, communication protocol design and simulation, power efficient protocol and management, final integration and wireless sensor network experiments are all detailed in this thesis.