基于流体动力学模型的无线传感器网络部署技术研究

发布时间：2018-06-11 13:03

本文选题：节点部署 + 无线传感器网络　；参考：《南京理工大学》2016年博士论文

【摘要】：节点部署技术是无线传感器网络研究中的热点之一。优化的节点部署策略,可以较大程度地增强网络中节点的容错能力和负载均衡,亦可较好地提升网络的性能、延长网络的生命周期,降低网络的部署代价。网络的节点部署问题是无线传感器网络研究中的一个难题,尤其在复杂的部署环境中显得尤为突出。本文应用流体动力学模型对无线传感器网络节点部署技术进行了研究。首先采用基于位置信息的理想流体模型对节点部署进行了研究,为弥补理想流体模型连通性不好的问题,又采用了非守恒粘性流体模型对三维网络的节点部署进行了研究。相比于理想流体模型,部署效果得到了提高。然而,无论是理想流体模型还是粘性流体模型,其本质上都是一种集中式部署策略,这对于网络负载均衡及延长网络生存期是不利的。本文最后提出了基于改进人工鱼群的流体模型分布式节点部署算法,较好地分散了节点功耗,延长了无线传感器网络的生命周期。仿真实验结果证明,该部署策略可以取得较高的覆盖度,较好地完成目标区域的部署工作。本文的主要研究内容以及创新点如下:(1)针对二维移动无线传感器网络,提出了基于位置信息的理想流体模型节点部署算法,把无线传感器网络抽象为理想流体,那么,部署过程中节点的移动则遵循流体微团的物理规则。采用覆盖度及覆盖均匀度两个性能指标,对该部署算法的性能进行了相应评价,并对整个部署过程做了仿真实验。实验结果表明,基于位置信息的理想流体模型节点部署算法,相比于经典虚拟力部署算法,网络部署性能取得了较好效果;(2)针对基于位置信息的理想流体模型在部署过程中网络的连通性的欠缺问题,进一步使用非守恒粘性流体模型对网络部署进行研究,并考虑了实际部署过程中环境的复杂性,特别是地形、地貌的影响(粘滞阻力),提出了基于非守恒粘性流体模型的三维空间节点部署算法,并进行了仿真研究,相比于基于位置信息的理想流体模型节点部署算法,部署效果得到了进一步提高;(3)针对三维水下无线传感器网络,结合鱼群算法,提出了基于改进人工鱼群的流体模型节点自部署算法。该算法适应于水下传感器网络的应用环境,在算法中水下传感器节点被视为人工鱼和粘性流体,事件则被视为人工鱼的食物。网络节点的部署过程,从而就可被视为流体自主移动以完成自部署和人工鱼寻觅食物的过程。针对动态事件设计了具体算法,并进行了仿真研究。该算法是一种分布式部署算法,既分散了节点功耗,又较好地延长了网络生命周期;(4)在苛刻、未知的部署环境中,本文从高精度卫星地图上提取部署区域的地形地貌及高度信息,然后使用部署算法在卫星地图上进行虚拟部署。利用卫星地图进行节点虚拟部署,该方法能够以接近真实环境来评估部署算法的性能。此外,在实际部署之前进行节点虚拟预部署,可以根据虚拟部署结果预先规划部署策略,预估部署代价,为在真实环境中的部署工作提供数据参考,同时也降低了部署代价。本文理论研究与仿真结果较好吻合,从而证实了理论计算与分析的正确性和可靠性,也验证了文中提出的基于流体动力学模型的节点部署方法的优越性和正确性。为节点部署技术的研究及应用,做了有益的探索。
[Abstract]:Node deployment technology is one of the hotspots in the research of wireless sensor networks. The optimized node deployment strategy can greatly enhance the fault tolerance and load balancing of nodes in the network. It can also improve the network performance, prolong the network life cycle and reduce the network deployment cost. The node deployment problem of the network is wireless. A difficult problem in the research of sensor networks is particularly prominent in the complex deployment environment. This paper applies the hydrodynamic model to the node deployment technology of wireless sensor networks. Firstly, the ideal fluid model based on position information is used to study the nodes, in order to make up the ideal fluid model connectivity. A non conserved viscous fluid model is used to study the node deployment of the three-dimensional network. Compared with the ideal fluid model, the deployment effect has been improved. However, both the ideal fluid model and the viscous fluid model are essentially a centralized deployment strategy, which is for network load balancing and extension. The long network lifetime is unfavorable. At the end of this paper, a distributed node deployment algorithm based on the improved artificial fish swarm is proposed, which can better disperse the node power and prolong the life cycle of the wireless sensor network. The simulation experiment results show that the deployment strategy can achieve high coverage and complete the target area better. The main research contents and innovation points of this paper are as follows: (1) aiming at two-dimensional mobile wireless sensor networks, an ideal fluid model node deployment algorithm based on position information is proposed to abstract the wireless sensor network into an ideal fluid. Then, the movement of nodes in the deployment process follows the physical rules of the fluid micromass. The performance of the deployment algorithm is evaluated with two performance indexes, including the coverage degree and the coverage uniformity, and the simulation experiment is made on the whole deployment process. The experimental results show that the deployment algorithm of the ideal fluid model node based on position information has achieved good results compared with the classical virtual force deployment algorithm; (2) the network deployment performance is better than that of the classical virtual force deployment algorithm. For the lack of connectivity of the ideal fluid model based on position information in the process of deployment, the non conserved viscous fluid model is used to further study the network deployment, and the complexity of the environment is considered in the actual deployment process, especially the topography, the viscous resistance, and the non conserved viscous fluid. The 3D space node deployment algorithm is studied and the simulation research is carried out. Compared with the location information based ideal fluid model node deployment algorithm, the deployment effect has been further improved. (3) aiming at the three-dimensional underwater wireless sensor network and the fish swarm algorithm, the self deployment calculation of the fluid model node based on the improved artificial fish swarm is proposed. The algorithm adapts to the application environment of underwater sensor networks. In the algorithm, underwater sensor nodes are regarded as artificial fish and viscous fluid, and events are considered as the food of artificial fish. The deployment process of network nodes can be considered as the process of self deployment and artificial fish seeking for food. This algorithm is a distributed deployment algorithm, which not only disperses the power consumption of the nodes, but also prolongs the life cycle of the network. (4) in the harsh, unknown deployment environment, this paper extracts the topography and height information of the deployment area from the high precision satellite map, and then uses the deployment algorithm. Virtual deployment on satellite maps. Using satellite maps for virtual deployment of nodes, this method can evaluate the performance of the deployment algorithm in close proximity to the real environment. In addition, the virtual deployment of nodes before the actual deployment can pre plan the department strategy according to the virtual deployment results and estimate the deployment cost for the real environment. The deployment work provides reference for data and reduces the cost of deployment. The theoretical research in this paper is in good agreement with the simulation results, which confirms the correctness and reliability of the theoretical calculation and analysis. It also validates the superiority and correctness of the node deployment method based on the hydrodynamic model in the paper. The study and application have made useful exploration.
【学位授予单位】：南京理工大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP212.9;TN929.5

【参考文献】