適用於農業之無線行動感測網路技術之研究---子計畫三：掃描統計在戶外無線感測網路上的應用

Abstract

因無需固定的基礎設施，無線感測網路可以低成本的方式靈活的提供在各種環 境條件下多種應用的解決方案，這其中包括了緊急的急難救助、戰場上的決策支援輔 助系統、環境監測及製造業的廠房生產機具監控等。在本計畫中，我們將針對能在戶 外進行大量隨機佈建的無線感測網路提出系統性的基礎研究，其中包括了網路拓樸特性 的基礎研究、適當的感測及傳輸半徑的設定及簡單且可靠的幾何式路徑找尋演算法， 以供農業上的應用。 在許多潛在的戶外應用中，感測元件是由不可充電及更換的電池驅動，能源的有效 用成為影響感測元件及系統使用壽命的重要因子，因此本計畫的主要課題之一即是以 最小的傳輸半徑或感測半徑提供可靠的服務品質。此外，在戶外的應用中，透過定位 系統（如: 全球定位系統等）的輔助，無線感測元件可以取得有用的地理資訊，因此 本計畫的另一個主要課題即是利用可取得的地理資訊以提供簡單可靠的路徑找尋協 定。 從目前正在執行中的國科會新人研究計畫「大型無線隨意網路中用於拓撲控制之 臨界傳輸半徑（NSC 94-2218-E-009-030-）」的部分成果中，我們發現掃瞄統計可以 有效的說明並預測大量隨機佈建無線感測網路的系統特性。因此在本計畫中，根據目 前在掃瞄統計上的研究成果，我們期望能達成下列三個研究成果：(1)針對大型隨機佈 建無線感測網路提出系統性的研究分析；(2)在無線感測網路中於節省能源的前提下，以 最小的感測半徑及傳輸半徑提供適足的網路拓樸特性，如：網路連通性及感測覆蓋率； (3)在無線隨意網路中，提出簡單且可靠的使用地理資訊的路徑找尋協定。

No need of fixed infrastructures, wireless sensor networks can be flexibly deployed in low cost for various applications, e.g., emergency disaster relief, decision making in battle fields, machine control and monitoring in manufactories, and environmental monitoring. In this proposal, we will focus on the fundamental research of random-deployed large-scale wireless sensor networks. For outdoor agriculture applications, our research works include fundamental analysis of network topology, the setup of proper transmission radius and sensing radius, and design of simple and reliable geographic routing algorithms. In many possible door applications of wireless sensor networks, sensors are powered by batteries that can not be replaced and recharged. Therefore, energcy efficiency is an important factor that would affect the lifetime of sensors and systems. One of the major objectives of this proposed research is to find the smallest transmission radius and sensing radius that can offer networking and sensing services in good qualities. In addition, in the ourdoor environment, geographic information is available to sensors through position systems, e.g., the global position system (GPS). Hence, another major objective of this proposal is how to utilize the geographic information to design simple and reliable routing algorithms. According to some reults of the undergoing NSC project for new researchers, 「Critical Transmission Radius for Topology Control in Large-Scale Wireless Ad Hoc Networks」 (NSC 94-2218-E-009-030-), we found that scan statistics can successfully explain and predict some netwok topology behavors of randomly-deployed large-scale wireless sensor networks. In this proposal, based on the scan statistics, we would like to achieve the following three objectives: (1) Develop a systematic methodology for anslysis of randomly-deployed large-scale wireless sensor networks; (2) Under the consideration of energy saving, we would like to find the smallest transmission radius and sensing radius to offer good qualities of networking and sensing services, i.e., network connectiviy and sensing coverage; (3) Develop simple and reliable geographic routing algorithms.