自动化博览
自動化博覽
자동화박람
AUTOMAITON PANORAMA
2013年
z2期
46-53
,共8页
无线传感器网络%自组织特征映射网络%最小代价生成树%自组织聚类%动态拓扑特性
無線傳感器網絡%自組織特徵映射網絡%最小代價生成樹%自組織聚類%動態拓撲特性
무선전감기망락%자조직특정영사망락%최소대개생성수%자조직취류%동태탁복특성
Wireless Sensor Networks%Self-Organizing Feature Maps%Minimal Spanning Tree%Self-Organizing Clustering%Dynamic Topology
本文设计并实现了一种结合自组织特征映射网络模型（SOFM）和最小代价生成树算法（MST）的无线传感器网络（WSN）构建方法。该方法分为两个阶段，第一阶段利用自组织特征映射网络模型竞争优化得出无线传感器网络的汇聚节点和汇聚层；第二阶段，采用分布式计算方法，对自组织特征映射网络聚类后得到的各种“类别”的传感器，并行使用最小代价生成树算法进行局部无线传感器网络构建；最终完成全局无线传感器网络构建的全过程。实验研究结果表明：SOFM-MST-WSN构建算法可以快速、有效、优化地构建满足动态拓扑特性和自组织特性要求的无线传感器网络；且计算步骤固定，适合在各种军事国防和日常民用领域中的无线传感器网络应用系统中使用。
本文設計併實現瞭一種結閤自組織特徵映射網絡模型（SOFM）和最小代價生成樹算法（MST）的無線傳感器網絡（WSN）構建方法。該方法分為兩箇階段，第一階段利用自組織特徵映射網絡模型競爭優化得齣無線傳感器網絡的彙聚節點和彙聚層；第二階段，採用分佈式計算方法，對自組織特徵映射網絡聚類後得到的各種“類彆”的傳感器，併行使用最小代價生成樹算法進行跼部無線傳感器網絡構建；最終完成全跼無線傳感器網絡構建的全過程。實驗研究結果錶明：SOFM-MST-WSN構建算法可以快速、有效、優化地構建滿足動態拓撲特性和自組織特性要求的無線傳感器網絡；且計算步驟固定，適閤在各種軍事國防和日常民用領域中的無線傳感器網絡應用繫統中使用。
본문설계병실현료일충결합자조직특정영사망락모형（SOFM）화최소대개생성수산법（MST）적무선전감기망락（WSN）구건방법。해방법분위량개계단，제일계단이용자조직특정영사망락모형경쟁우화득출무선전감기망락적회취절점화회취층；제이계단，채용분포식계산방법，대자조직특정영사망락취류후득도적각충“유별”적전감기，병행사용최소대개생성수산법진행국부무선전감기망락구건；최종완성전국무선전감기망락구건적전과정。실험연구결과표명：SOFM-MST-WSN구건산법가이쾌속、유효、우화지구건만족동태탁복특성화자조직특성요구적무선전감기망락；차계산보취고정，괄합재각충군사국방화일상민용영역중적무선전감기망락응용계통중사용。
An algorithm of Wireless Sensor Networks construction based on Self-Organizing Feature Maps and Minimal Spanning Tree is designed and implemented in this paper. The algorithm consists of two stages. In the first stage, aggregation nodes and convergence layers of the wireless sensor network are competed and optimized by using self-organizing feature maps model. In the second stage, local wireless sensor networks are constructed by adopting distributed computing to all “categories” of sensor-nodes clustered by self-organizing feature maps and using minimal spanning tree algorithm in parallel. Finally the whole construction process of global wireless sensor network is completed. Experimental results show that the SOFM-MST-WSN algorithm can construct the wireless sensor networks quickly, efficiently and optimally. At the same time, the algorithm steps are fixed and the algorithm can meet the requirements of dynamic topology and self-organizing characters of wireless sensor networks. So it is suitable for use in a variety of wireless-sensor-network systems both in military and daily civil application areas.
