【作者】 王洋；

【导师】 林中；

【作者基本信息】 北京邮电大学 ， 机械电子工程， 2010， 硕士

【摘要】 随着物联网时代的到来,无线传感器网络的发展和应用即将掀起新的信息科技浪潮。用于监测突发情况的事件驱动型无线传感器网络,能对所监控的事件快速响应,具有实时性强、灵敏度高等特点,并且网络生存期较长,可实现对目标区域的长时间监控,在很多领域有着广泛的应用前景。本文通过分析LEACH、TEEN等传统网络分簇协议在突发事件监测应用中的不足,如未考虑事件发生的具体位置和事件的严重程度、全网节点成簇造成事件区域外节点不必要的能耗等,综合考虑事件驱动型无线传感器网络的特点和节能的需要,提出了新的基于动态半径的分簇融合算法(EDRCA算法)。算法根据事件发生区域内节点的剩余能量及事件发生时所受刺激的程度选择簇头,在高效监控事件的前提下,最大限度地均衡各节点的能耗；各节点之间建立多级功率梯度,采用分级发射功率实现节点间的相互通信,通过降低发射功率达到节约能量的目的；根据不同簇半径下的成簇效果及网络平均能耗优化确定成簇大小,建立了更适用于突发事件监测的动态网络拓扑结构,提高了各簇成员节点所采集数据的相关程度,为簇头节点的数据融合做好准备；为了进一步减轻簇头负担,采用设定簇头能耗门限的方式实现被触发节点范围内簇的重构,有效地避免了单个节点因能量消耗过快而失效,进而延长了网络寿命。利用NS2网络仿真软件将EDRCA算法和同样适用于突发事件监测的TEEN协议进行对比,实验结果证明,EDRCA算法有效降低了网络能耗,与TEEN相比,网络生存时间可延长20%左右。更多还原

【Abstract】 With the era of Internet of Things, the development and applications of wireless sensor networks will trigger a new wave of information technology. Event-driven wireless sensor networks are used in monitoring emergency events. With the characters of strong real-time and high sensitivity, it can quickly respond to the events. Because of the longer lifetime of the network, it can monitor the target area for a long time. As a result, it has broad application prospects in many areas.LEACH, TEEN and other traditional protocols did not take the location and severity of incidents into account, and also caused a waste of energy outside the region of events. According to the character of event-driven WSNs and the need for saving energy, a new event-driven clustering aggregation algorithm based on dynamic cluster range is proposed.The algorithm selects cluster head according to the residual energy and excited intensity of each node which is in the region of events. Being efficiently monitoring the events, it can well balance the energy consumption of each node. The nodes establish multi-level power gradient with each other, use hierarchical transmission power for communication. This can certainly save the energy of nodes by reducing the transmission power. It optimizes the size of cluster according to the clustering result and the average energy consumption of different cluster ranges. Because it forms better dynamic topology and improves the relevance of data collection, it is good for data aggregation of cluster heads. To further reduce the burden of cluster head, the threshold of cluster head energy consumption is used to re-build the new network topology. This can avoid quickly dying of the nodes due to excessive energy consumption and extend the lifetime of the network.In comparison with TEEN, the results of NS2 simulation show that the EDRCA algorithm can considerably reduces the energy consumption and prolong the lifetime of the network for about 20%.更多还原