物聯網無線電波功率延遲模型之建立與驗證-應用於冷鏈物流恆溫箱之連網

Abstract

物聯網是指將所有的物品結合射頻辨識與感測設備來讀取物品資訊，並連結至 網路將資訊上傳至管理平台，形成智能化的管理。在冷鏈物流中，車載的冷凍櫃是一個大金屬箱，金屬箱內有許多用以存放冷凍食品的恆溫箱，若將無線感測網路結合恆溫箱，並將恆溫箱的溫濕度資訊透過雲端傳送到監控平台即形成一個物連網。 但是當恆溫箱的溫濕度資訊要透過電波在這個物連網的無線電通道環境傳播時， 其金屬環境會造成強烈的多重路徑反射，並且恆溫箱本身會導致無線電波被散射或吸收的特性。然而，此一多路徑效應可能會對傳送訊號造成干擾，因此為了確保有良好的通訊品質，有必要對此一通道環境建立功率延遲模型並加以驗證，然而，就我們所知，目前全世界似乎未有此一通道研究。 因此本論文將針對脈衝響應之功率延遲(Power Delay Profile)建立模型，分別提出兩種方法，一種是可決定的方法(Deterministic Approach)，主要透過鏡像法則(Image Method)與閘門合併(Gating)來建立模型；另一種則是透過統計方法(Statistical Approach)，主要應用卜瓦松模型(Poisson Model)或Δ-K模型(Δ-K Model)來對功率延遲建立模型，並且我們發現以Δ-K模型對功率延遲建模更為適合，並且其參數K值介於0.66至0.77，其路徑之間相關性低，有獨立抵達的趨勢。

The Internet of things(IOT) is sending the information of objects to a platform by combining radio identification technology and wireless sensor network. The freezer is a metal cabinet in cold chain logistics, and there are many more containers inside the cabinet. Applying wireless sensor network to containers and sending information of containers are exactly the internet of things. But when the information of containers sent by radio wave which propagating in a metal environment, there is strongly multipath effect. And the radio wave may scattered or absorbed by the container. Hence, multipath effect may interfere the transmitted signal. For this reason, we need to model and verify for the power delay profile. However, it seems no one had studied this issue before. we use two approach to model PDP. One is deterministic approach, which combines image method and gating to model PDP. The other is statistical approach which comprises Poisson model and Δ-K model to model PDP. Finally, Δ-K model is suitable for modeling, and its parameter K is between 0.66 to 0.77. That means there is low correlation between multipath.