無線感測網路下之高效能叢集式資料散播技術

Abstract

一般而言，無線感測網路受限於有限的能力和非充電式電池，所以如何降低各個節點的能源消耗是一個重要的研究議題。現存的資料散播協定採用泛播機制來傳遞興趣與尋找傳送路徑會造成大量的能源消耗。為了減輕這個問題，叢集式資料散播可以被使用來降低能源消耗與延長網路壽命。能源階層式被動叢集(ELPC)技術是基於叢集式資料散播來平衡感測節點的能源消耗與增加網路壽命。但是ELPC仍然採用泛播探索性資料訊息來選擇一條傳送路徑，這會造成額外的能源消耗和延遲。因此我們提出一個高效能叢集式資料散播(ECDD)技術來解決這個問題。在我們的方法中，除了附加控制訊息到興趣中來執行隨選式被動叢集，為了幫助感測節點去選擇下一個到匯集點最少跳躍數的傳輸節點，我們也利用額外的控制訊息來對每一個感測節點設定一個跳躍數。利用這個方法，可以找到一條最短路徑來傳輸感測資料給匯集點。對資料散播而言，避免使用探索性資料訊息可以有降低感測節點的能源消耗與達到低延遲時間的優點。實驗結果顯示，我們的方法在平均耗損能量上可以改善直接擴散(DD)協定達到61.30%，另外也可改善ELPC達到22.33%。此外，我們的方法在平均延遲時間上相對於DD與ELPC，分別可降低57.45%和23.49%的延遲時間。ECDD可運作於需長期性監控與即時性回報的應用，如社區看護系統和災區救援方面。

Generally, the wireless sensor network (WSN) is restricted with limited capability and un-rechargeable battery power. Therefore, how to reduce energy consumption of sensor nodes becomes an important research issue. Existing data dissemination protocols adopted flooding to propagate interests and find forwarding paths in WSNs, which cause large energy consumption. To relieve this problem, cluster-based data dissemination was adopted to reduce energy consumption and prolong network lifetime. The energy level-based passive clustering (ELPC) scheme is based on cluster-based data dissemination to balance energy consumption of nodes and prolong network lifetime. However, ELPC still has to use exploratory data messages to select a forwarding path by flooding, which causes extra energy consumption and delay. Therefore, we propose an efficient cluster-based data dissemination (ECDD) scheme to resolve this problem. In the proposed ECDD, besides piggyback control information into interests to perform on-demand passive clustering, we also use control information to set each node a hop count for assisting a node to select next forwarding node with the least hop count to the sink. In this way, a shortest path to forward sensed data back to the sink can be found. The advantages of avoiding exploratory data messages for data dissemination are reducing energy consumption of sensor nodes and achieving low delay. Simulation results show that ECDD is 61.30% better than DD, a classical approach, and 22.33% better than ELPC in terms of average dissipated energy. Furthermore, our approach is 57.45% and 23.49% better than DD and ELPC, respectively, in terms of average delay. The proposed ECDD is feasible for applications of long-term monitoring and real-time responding, such as a community health care system and rescue in a disaster area.