分散式無線感測網路佈局之研究

Researches on Sensor Node Deployment for Distributed Wireless Sensor Networks

楊承志(CHENG-CHIH YANG)

指導教授 ： 邱茂清 ； 溫志宏

國立中正大學/工學院/電機工程學系/博士(2014年)

摘要

一個優質的無線感測網路，端視該網路是否提供最大的感測覆蓋面(coverage area)、良好的連通力(connectivity)及較長的生命週期。受到廣域環境實際需求與限制，許多感測網路感測器之布局並非預設的，而是屬於不定式佈局，這情境就無法保證無線感測網路品質。因此，改善感測網路布局，提升無線感測網路品質是本論文研究的主要目標。
基於在感測網路中實際的應用，本論文以機率感測覆蓋模式(probabilistic sensing coverage model)表示感測點檢測其鄰近事件發生之功能模式。我們以等位曲線(contour graph)表示多個感測點覆蓋面及覆蓋洞(coverage hole)型態。針對感測器佈局之數量，我們提出一個以期望密度考慮的感測器佈局數量演算法。藉由可移動的感測器(mobile sensor)改善網路品質是本文主要策略，針對此議題，我們提出三種混合式虛擬力演算法(Hybrid Virtual Force Algorithm, HVFA)，藉由收集鄰近感測器的資訊，我們可以決定移動感測器新的最佳位置。模擬結果顯示，在給定的覆蓋門檻θ=0.9下 (coverage threshold parameter θ=0.9)，我們所提出的三種演算法在綜合效能指標上皆優於傳統方法。另外，我們也介紹“粗略位置”考量的分散式多匯集點(multiple sink)布局，我們提出一個多匯集點訓練協定演算法(Multiple-Sink Training Protocol Algorithm, MSTPA)來達成感測節點的歸屬分配，進而改善數據傳送所需功耗。模擬結果顯示分散式多匯集點布局相較於單匯集點布局具有優異的節能效果。

關鍵字

混合式虛擬力演算法 ； 機率感測覆蓋模式 ； 感測器佈局 ； 多匯集點訓練協定演算法

並列摘要

The quality of wireless sensor network (WSN) depends on whether the network offers the largest sensing coverage area, good network connectivity and a prolongation life time capability. Therefore, improvement in sensor deployment to promote the quality of distributed WSN is an important topic. It is also the main objective in this dissertation.
Based on realistic applications in sensor networks, a probabilistic sensing coverage model is used in this dissertation to represent the function model of detecting an adjacent event emergence by a sensing node. We sketch multiple sensors coverage and coverage hole using contour graph concept. As for the sensor deployment number, we propose a sensor deployment number algorithm based on the consideration of expected density. The main strategy in this dissertation is to improve the network quality via the movement of mobile sensors. To consider this topic, we propose three Hybrid Virtual Force Algorithms (HVFAs). We can decide the new optimal positions of mobile sensors by collecting the information of neighbor’s nodes. Simulation results show that, under the coverage threshold parameter θ=0.9, the proposed three algorithms outperform the conventional method in terms of synthesized performance index. In addition, we further introduce distributed multiple sink deployment concept based on coarse-grain location awareness. A Multiple-Sink Training Protocol Algorithm (MSTPA) is proposed to achieve the assignment of sensor nodes, and then, to improve the required power consumption of data delivery. Simulation results show that, compared to the centralized single sink deployment, the distributed multiple sink deployment has significant effect on power saving.

並列關鍵字

HVFA ； probabilistic sensing coverage model ； Sensor deployment ； multiple-sink training protocol algorithm

參考文獻

[1] I.F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, “Wireless sensor networks: a survey,” Computer Networks, vol. 38, no. 4, pp. 393-422, 2002.

[2] Z. Bojkovic and B. Bakmaz, “A survey on wireless sensor networks deployment,” WSEAS Transactions on Communications, vol.7, no. 12, pp. 1172-1181, Dec. 2008.

[3] F. Dressler, Self-Organization in Sensor and Actor Networks. New York, John Wiley and Sons, Inc., 2007.

[4] B. Wang, “Coverage problems in sensor networks: a survey,” ACM Computing Surveys, vol. 43, no. 4, article 32, pp.32.1-32.53, Oct. 2011.

[6] X. Wang, G. Xing, Y. Zhang, C. Lu, R. Pless, and C. Gill, “Integrated coverage and connectivity configuration in wireless sensor networks,” in ACM International Conference on Embedded Networked Sensor Systems, pp. 28-39, Nov. 2003.