| Abstract: | 人類對自然環境的大量開發與利用，導致地球氣候變遷，並帶來了極端的天氣現象。隨著2.4GHz無線通訊與感測技術的進步，讓我們可應用無線感測網路技術來監測環境與氣候，以避免或減少災害對我們所造成的傷害與損失。無線感測網路應用系統除了需要大量的嵌入式技術，諸如處理器的選用、各類型感測器的驅動、以及無線感測網路通訊協定的軟體實作，此外，還需要整合應用領域的技能(know-how)，例如微氣候的監測應用需要地理或農業方面的背景知識。因此，開發和佈署一個無線感測網路總是耗費大量的人力、時間和成本。本研究以微氣候監測為目標，設計了一個無線感測網路嵌入式開發和佈署平台(Embedded Development and Deployment Platform, ED2P)，其包括內建GRAFCET虛擬機器的感測節點、閘道器、無線感測網路中介軟體、通訊軟體與嵌入式Web伺服器。在此平台上，開發者先以GRAFCET圖形語言描述無線感測網路應用系統的功能與行為，再轉譯成參數表，然後透過通訊軟體將它傳送至閘道器，再由閘道器傳送給感測節點，安裝於感測節點上的GRAFCET虛擬機器即會依據參數表執行應用程式，完成無線感測網路的遠端開發與即時佈署。論文最後，我們以溫室環境微氣候監測與坡地滑動監測的開發實例來驗證本論文所提出的微氣候無線感測網路開發平台，並展示其功能。A large amount of development and utilization of natural environment by the mankind, cause the Earth's climate change and bring extreme weather phenomena. With the 2.4GHz wireless communication and sensor technology advances, we can apply WSN technology to monitor environment and climate, avoid or minimize damage and loss caused by disaster for us. Applications of WSN need a lot of embedded technology, such as the choice of processor, the drivers of various types of sensors, and software implementation of WSN protocol. Besides, it also needs to integrate domain-dependent applications of skills (know-how). For example, a microclimate monitoring WSN requires the complementary background knowledge on the geography or and agriculture. Therefore, the development and deployment of WSN always spends a lot of manpower, time and cost. This study aims at micro climate monitoring, designs a WSN embedded development and deployment platform (Embedded Development and Deployment Platform, ED2P). It includes sensor nodes that are installed GRAFCET virtual machine, gateways, WSN middleware, communications software and an embedded Web server. Based on this platform, developers use GRAFCET graphical language to describe the discrete-event, behavior and function of wireless sensor network applications first. Then the coded GRAFCET model is transformed it into a coding table and send it to gateway by communications software, gateway resends it to the destination sensor nodes. The GRAFCET virtual machine is implemented on the sensor nodes which decode GRAFCET coding table and run the application. This process realizes the remote development and real-time deployment of the WSN application. Finally, we develop two WSN applications: the microclimate monitoring of greenhouse and the slope sliding supervision, to demonstrate the functionalities and performance of proposed platform. |
