使用虛擬網路叢集首推舉策略來強化異質感測網路效能與分析

Abstract

無線感測網路利用許多微小的感測元件來收集資料，在環境偵測、安全監控、軍事用途上等都有優良的功效。將感測網路分群可以有效延長無線網路的壽命，同時也得以提升穩定度。傳統的研究普遍著重於同質性網路，及網路中的所有節點擁有相同的電量以及硬體設備，然而異質性網路的出現是無法避免的，在網路中加上不同類型的節點除了能讓感測到的資料更多元，也可以延長整體網路壽命。有些研究利用預先假設網路中異質節點的總類以及數量來設計相對的函數，得以延長網路壽命，但這些假設反而侷限了擴充性以及彈性。此外，過往的研究大部分著重於電量以及群組領導者的選擇，鮮少有研究著重於叢集領導者的分布等。LEACH-C是少數有研究此方面的論文，然而其論文中假設每個節點都裝置有GPS接收器，雖然有精確的位置但卻額外負擔許多經濟成本。在本篇論文中，我們會提出一個有效節能的分群演算法，其可在不得知每個節點的位置、異質節點的種類以及數量的前提下，讓叢集首能得到相對合理的位置分配，同時也附帶一個極為節省能源的叢集首選擇方式。最後利用ns-2模擬實驗來驗證我們所提出的方法，數據顯示能有效延長網路壽命以及增進封包傳輸量。

Wireless Sensor Networks (WSN) with lots of small sensor nodes can be used as an effective tool for gathering data in various environments. Cluster routing mechanisms provide an efficient way for prolonging the network lifetime and increase its stability. In the past, most researchers work hard on homogeneous WSNs, but the trend of heterogeneity of network nodes is unavoidable. Some researches try to get better performance by previous assumption of initial power and nodes ratio, but these mechanisms are not flexible enough. Besides, previous researches rarely focus on cluster head distribution. LEACH-C can get the best effect on cluster head distribution but the user need to afford the additional cost for the GPS receiver of whole sensors. In this paper, we proposed Virtual Cluster Head Election Scheme, which offers an energy-efficient cluster head election algorithm and takes care about cluster head distribution of the whole network. Simulation results show that our scheme balance the position of cluster head and choose sensor nodes with relative higher energy as cluster heads, and achieved an obvious improvement on the network lifetime and throughput.