Wireless Sensor networks已被廣泛地應用於軍事戰略、目標追蹤及環境監測等領域。然而，在Sensor network 所佈建的環境中，常由於地形地物(如河流、峽谷) 、佈點不均、感測點毀損及外力訊號干擾等因素，使WSN中形成喪失感測能力甚至阻礙通訊的障礙區。在WSN中傳送的封包將因誤闖障礙區而造成路徑增長、耗費轉送封包之sensor 電量及增長傳送的Delay Time等問題。在本論文中，我們以主動式克服障礙物為目的，研發一Route Guiding Protocol (S-RGP)，主動的以分散式協定將障礙物資訊相對於Sensor Network予以透明化，並依障礙區的各種不同型狀設定網路的禁入區以避免封包（packet）因誤闖障礙區而造成電量消耗、頻寬浪費、delay time增加及傳輸成功率下降等問題，進而提升sensor network 中資料傳輸的效率。此外，我們亦考慮WSN中存在多個障礙物，主動導引封包避開障礙區並依最短路徑傳送至Sink，並針對障礙物可能的變化提出障礙物維護協定。實驗數據顯示，我們所提出的障礙物處理協定能有效在WSN中提供障礙物資訊並提升資料傳遞的效率。 In wireless sensor networks, a geographic region without functionality of sensing and communication can be generally treated as an obstacle, which significantly impacts the performance of existing location-based routing. The existence of an obstacle is due to unbalanced deployment, failure or power exhaustion of sensors, animus interference, or physical obstacles such as mountains or lakes. This thesis proposes a novel algorithm, namely S-RGP and M-RGP, to make existing location-based routing protocols resist obstacles. Applying the proposed S-RGP, border nodes that surround the obstacles will actively establish a forbidden region for concave obstacles and make the obstacle information transparently. Then packets will be guided to overcome the obstacle and to be moved through the shortest path from the encountered border node to the sink node. By integrating and maintaining information of multiple obstacles, the proposed M-RGP also resists multi-obstacles and enables the packets to be moved along the shortest path to the sink node. Simulation results show that the proposed protocol creates low overhead but significantly reduces the average route length and therefore improves the energy consumption and end-to-end delay for a wireless sensor network.
