无线传感器网络的非均匀分簇路由协议研究

发布时间：2018-04-17 00:23

本文选题：无线传感器网络 + 路由协议　；参考：《江南大学》2014年硕士论文

【摘要】：无线传感器网络是由大量的传感器节点构成的多跳、自组织网络，这些传感器节点可随机分布在监测区域，采集温湿度等各类有用数据，在进行数据融合后，传递给汇聚节点，供用户研究分析。无线传感器网络具有价格低廉，，可快速部署，规模大等特点，因而在环境检测、智能家居、军事国防等领域有广阔的应用前景，受到了国内外研究人员的高度关注。路由协议是无线传感器网络的关键技术之一，它对网络的存活时间有至关重要的影响。它的作用是降低单个节点的能量消耗，均衡整个网络节点的能量消耗，延长网络的生命周期。按照网络的拓扑结构，路由协议可划分为平面路由协议与分簇路由协议，论文对分簇路由协议，且着重对非均匀分簇路由协议展开研究。非均匀分簇路由协议的主要思想是划分规模大小不同的簇，解决在均匀分簇的网络中，由多跳通信造成的“热区”问题。论文的主要工作和创新点如下： 1)对经典的分簇路由协议(LEACH)及相关改进算法进行重点研究，掌握分簇路由算法的主要思想和运行步骤；接着对典型的非均匀分簇路由协议(EEUC)及相关改进算法进行深入的剖析，分析协议存在的问题，归纳影响因素；然后对应用群智能算法的分簇路由协议(EBUCP等)进行研究，分析其在簇头优化选举和选择簇间多跳路径方面的优势。 2)论述一种基于非均匀分簇的WSN双簇头路由算法(D-EEUC)，算法改进EEUC算法的候选簇头选举方式，提高最终簇头的选举质量；同时引入双簇头机制，将监测区域分为近区与远区，分区域选举副簇头。近区与远区的副簇头所负责的任务不同，近区副簇头主要负责转发数据，远区副簇头主要负责采集和融合数据。算法考虑节点能量、密度等不同影响因素，选举远区的副簇头，改善在簇规模较大时，簇内通信代价较大的情况。 3)论述一种基于混合蛙跳算法的WSN非均匀分簇路由算法(SFLUCR)，算法改进混合蛙跳算法的局部搜索策略，利用其全局寻优能力好的特点，求得一组全局最优解，其中目标函数需要综合考虑簇的紧凑性，簇的非均匀性以及节点能量等因素。利用EEUC的竞争半径思想，让部分最优解的节点退出竞选，避免簇头过于密集，选举出最终簇头。算法采用簇头轮换的机制，减少了选举次数，只在规定的更新轮次时才重新选举簇头。通过Matlab仿真实验，本文所提的D-EEUC与SFLUCR算法都能够较好地实现非均匀分簇的网络拓扑，满足解决“热区”问题的要求，且与LEACH与EEUC算法比较，本文算法较好地均衡了网络的能量消耗，延长了整个网络的生命周期。
[Abstract]:Wireless sensor network is a multi-hop, self-organizing network composed of a large number of sensor nodes. These sensor nodes can be randomly distributed in the monitoring area, collect various useful data, such as temperature and humidity, and pass them to the convergence node after data fusion.For user research and analysis.Wireless sensor networks (WSN) have many advantages such as low price, rapid deployment and large scale. Therefore, wireless sensor networks have broad application prospects in the fields of environmental detection, smart home, military defense and so on, and have been highly concerned by researchers at home and abroad.Routing protocol is one of the key technologies in wireless sensor networks.Its function is to reduce the energy consumption of a single node, balance the energy consumption of the entire network node, and prolong the network life cycle.According to the topology of the network, the routing protocols can be divided into planar routing protocols and clustering routing protocols.The main idea of non-uniform clustering routing protocol is to divide clusters of different sizes to solve the "hot zone" problem caused by multi-hop communication in uniform clustering networks.The main work and innovation of the thesis are as follows:1) focusing on the classical clustering routing protocol (LEACH) and its related improved algorithms, mastering the main ideas and running steps of the clustering routing algorithm, and then deeply analyzing the typical non-uniform clustering routing protocol (EEUC) and its related improved algorithms.The problems of the protocol are analyzed and the influencing factors are summarized. Then the clustering routing protocol (EBUCP) based on the swarm intelligence algorithm is studied to analyze the advantages of the cluster head optimization election and the selection of multi-hop paths between clusters.2) this paper discusses a WSN double cluster head routing algorithm based on non-uniform clustering. The algorithm improves the candidate cluster head election method of EEUC algorithm, and improves the final cluster head election quality, and introduces the double cluster head mechanism to divide the monitoring region into near and far areas.Sub-regional election of sub-cluster head.The sub-cluster head in the near area is mainly responsible for forwarding data, and the sub-cluster head in the far area is mainly responsible for data collection and fusion.Considering different influence factors such as node energy and density, the algorithm elects the sub-cluster head in the far region to improve the communication cost of the cluster when the cluster size is large.3) A WSN non-uniform clustering routing algorithm based on hybrid leapfrog algorithm is presented. The algorithm improves the local search strategy of the hybrid leapfrog algorithm and obtains a group of global optimal solutions by using its good global optimization ability.The objective function needs to consider the compactness of the cluster, the nonuniformity of the cluster and the energy of the node.Using the idea of competition radius of EEUC, the node of partial optimal solution is withdrawn from the election, so that the cluster head is avoided to be too dense, and the final cluster head is elected.The algorithm adopts the mechanism of cluster head rotation, which reduces the number of elections and elects the cluster head only in the specified update cycle.Through the Matlab simulation experiment, the D-EEUC and SFLUCR algorithms proposed in this paper can realize the network topology of non-uniform clustering well, and meet the requirements of solving the "hot zone" problem, and compared with the LEACH and EEUC algorithms.In this paper, the energy consumption of the network is well balanced and the lifetime of the network is prolonged.
【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP212.9

【共引文献】