應用於都會監控服務之以計數器為基礎的車用臨機傳送網路

Abstract

近年來由於車用網路豐富多變的應用，吸引了許多學術界與企業界投入研究。和傳統的移動式隨意網路（mobile ad hoc network）相比，車輛隨意網路（vehicular ad hoc network）減少了電源的限制，具備更規律的移動方向；但移動速度大幅度增加、移動模型更加複雜、網路中也更容易出現極端稀疏與極端密集的情形，這些新的限制都造成研究上的困難。 在本篇論文當中，我們設計了一個簡單、有效的臨機傳送演算法，透過都會區的巴士以及警用巡邏車，在非常低硬體限制及少量資訊交換的情況下，利用交通工具本身的特性配合臨機傳送，以達成都會環境監控的服務。最後我們使用NS2驗證，結果顯示我們提出的方法在一般的都會巴士密度中皆能達到良好的效能，另延伸探討在巴士密集與極稀疏環境下的表現。

In recent years, various kinds of applications in vehicular networks have attracted great attentions to academia and industry. Compared with MANET (mobile ad hoc network), the advantage of VANET (vehicular ad hoc network) include no significant power constraints and regular movement due to the street boundary; however the velocity is highly increasing and mobility model is more complicated. VANET also suffers from greatly high node density and extremely low density. All the constraints caused difficulties on VANET study. In this thesis, we propose a simple and effective opportunistic forwarding algorithm for disconnected vehicular ad hoc networks with low hardware constraint and only need little information in mobile nodes to provide surveillance services through buses and patrol cars in an urban area. We also evaluate the performance of our protocol via the ns-2 simulator to demonstrate that it works very well in normal bus density in an urban area. We also observe the effects upon excessive low bus density and high bus density.