頻譜認知臨機路由於感知無線電網路

Abstract

感知無線電成為一項增進頻譜使用效率的重要技術，創造出臨機傳輸於通訊鏈結上。為了傳送封包於多無線電系統組成的一般感知無線電網路，支援路由於眾多臨機鏈結上是必要的。然而，在現今文獻中缺乏對於此高度動態可使用鏈結的完整了解，且沒有一個可靠的端對端傳輸機制於感知無線電網路。為了滿足此需求，在這篇論文中，我們提出適合於衰弱通道下的感知無線電網路之頻譜認知臨機路由。擁有著由區域性量測資訊所建立出的頻譜地圖和針對臨機鏈結所推導出的路由參數，即臨機鏈結參數、臨機路徑參數和感知無線電點參數，前瞻的頻譜認知臨機路由採用合作式機制啟動多路徑傳輸，且對於實際的應用能維持雙階服務品質保證吞吐量（即臨機服務品質保證和統計服務品質保證）。效能評估的結果驗證了頻譜認知臨機路由，不僅在感知無線電網路中，具有較低的延遲時間和較多的服務品質保證下之吞吐量，在一般的無線網路中也是如此。

Cognitive radio (CR) emerges as a key technology to enhance spectrum efficiency and thus creates opportunistic transmissions over communication links. Supporting the routing function on top of numerous opportunistic links is a must to route packets in a general cognitive radio network (CRN) consisting of multi-radio systems. However, there lacks complete understanding of these highly dynamic available links and a reliable end-to-end transportation mechanism over CRN. Aspiring to meet this need, in this thesis, we propose novel spectrum aware opportunistic routing (SAOR) algorithm suited for the CRN under wireless fading channels. With innovative establishment of the spectrum map from local sensing information and the derivation of the routing metric for opportunistic links known as opportunistic link transmission (OLT), the opportunistic path metrics, and the CR node metrics, the promising SAOR employs a cooperative scheme to enable multi-path transmissions and maintains the dual-scale QoS guaranteed throughput (i.e. opportunistic QoS guaranteed throughput and statistical QoS guaranteed throughput) for practical applications. Results in performance evaluation confirm that SAOR enjoys less delay with guaranteed throughput, not only in CRN, but also in general wireless network.