多重區域無線感測網路下之最佳節點數分割

Abstract

現今大多數關於無線感測網路(WSNs)的研究，主要著重在處理單區域網路上，這在某些特殊的WSNs中是不夠有效率的，當在此網路某些區域中，事件發生的非常頻繁，而其他區域卻不是如此。當感測器的數目是有限的狀況下，將此單區域網路劃分成多重區域以得到最佳的佈放是必要的，因此多重區域無線感測網路的概念因此形成。根據這個概念，我們提出了Maximum Information Rate Deployment(MIRD)這方法於多重區域無線感測網路中佈放感測器，在這裡網路存活時間是被忽略的。由結果發現，面積和事件發生率λE是關鍵因素，而λE的效應會隨著面積覆蓋率的增加而漸漸變得無關緊要。更進一步的，我們提出多重無線感測網路中的Maximum Information Capacity Deployment(MICD)，用機率的概念來處理能量消耗與網路存活時間的問題，並提出一個快速尋找最佳節點數分割的演算法。由結果分析顯示，面積是最關鍵的影響因素，而 λE的影響並沒這麼明顯。

Currently, most works of Wireless Sensor Networks(WSNs) are mainly dealing with single-area networks that is ine±cient for some special WSNs in which event occurs frequently in some part of the network, and less frequent in others. While the number of sensors is limited, it is necessary to divide this network into multi-area for optimum deployment and multi-area WSNs are therefore formed. Based on this idea, we propose the Maximum Information Rate Deployment(MIRD) to deploy sensors e±ciently in multi-area WSNs in which network lifetime is ignored. It is found that area and event occurring rate ¸E are critical factors in MIRD, and the efect of ¸E becomes irrelevant when area coverage is large. Furthermore, we propose the Maximum Information Capacity Deployment(MICD) in multi-area WSNs to deal with the problem of energy consumption and network lifetime in a probability sense, and provide a searching algorithm for the optimal deployment. The result reveals that area is the critical factor in MICD, while the efect of ¸E is less significant.