Dissertation > Excellent graduate degree dissertation topics show

Research on Sensors’ Layout, Localization and Mobile Nodes’s Path Planning in Wireless Sensor Networks

Author: ZhuJian
Tutor: ZhaoHai
School: Northeastern University
Course: Computer System Architecture
Keywords: wireless sensor networks mobile nodes path planning layout of network selection of reference nodes localization model lifetime
CLC: TP212.9
Type: PhD thesis
Year: 2010
Downloads: 48
Quote: 0
Read: Download Dissertation

Abstract

Wireless sensor networks has a wide application fields, it can be used in military affairs, environment supervision, samrt home and so on. Most of correlation researches are limited to immobile network, and they are uncomfortable for some mobile applications such as inspecting of wild animal’s living and tracing of patients’heartbeat, in these applications, the nodes always mobile at a low speed, and they need mobile nodes in WSNs. The introduced mobile nodes can not only advance the ability of service but also can adjust the structure of WSNs to optimize its performance----energy consumption, latency and lifetime. So, the mobile nodes have an important signification to WSNs, and the path planning of mobile nodes based on WSNs has to be solved firstly.WSN has its characters such as special application environment, restricted resource and so on, so, some well worked navigation devices such as GPS can hardly be used. For example, while in outer space or underground colliery applications, the GPS signal can’t work. WSNs is a kind of denseness and sensor network, it has the terms of path planning. However, due to WSNs’restricted resource, the path planning is not a simple search problem anymore, but a problem which needs to be given attention to the result of planning and network’s performance.The algorithm of path planning is divided into four parts in this paper:network’s layout, reference nodes’ selection, localization model and path searching. The four parts are the necessaries of path planning firstly; there are inner relations among them secondly, the relations can advance the performance of path planning ulteriorly.The main works and contributions of the paper are focused on the following aspects:1. In the research of network’s layout, two strategies are promoted:LG (Layout Guide) and LO (Layout Optimization). In LG, the ratioλbetween sensor radius and communication radius can change the efficiency cover ratio, it can guide people to layout the network according to the givenλ; In LO, the power levels are adjusted to decrease the energy consumption in condition of unchanging the degree of network.2. In the research of reference nodes’selection, CRS is advanced, it takes "multi-turns" mechanism to balance the advantage of planning result and complexity. The result of experiments shows that, the coverage ratio of CRS is 41.7% and 38.3% higher than random selection and SHARP selection.3. In the research of localization model, an algorithm called FTLM is advanced based fuzzy recognition. In FTLM, the unknown node is limited into a small area by setting several swatch nodes, the procedure is simple and doesn’t need larger number of reference nodes, it can save energy and bandwidth. Moreover, the precision and complexity can be adjusted by the set of swatch nodes to satisfy applications’ requirements. It is very suitable for the path planning research in this paper. This result of experiments show that, the complex degree of FTLM is O(n) and its localization precision is 7.3% more than SBL whose degree is O(nm), in which, m is the number of reference nodes.4. In the research of path search, we promote a algorithm based wireless sensor networks which is called PS. In this algorithm, the sensors are considered as the guides, and the paths are constructed by packets’communication. While in path selection, we give attentions to the lifetime of WSNs and the length of the path, the PS algorithm is simple and it has a low computing complex. The result of experiments shows that, the latency of PS algorithm is 4.2s lower than gridding algorithm, and PS can cause a longer lifetime than gridding.5. Finally, a path planning algorithm called WPP based WSN is advanced by banding the above researches together. A lot of experiments are performed to test the inner relations and the algorithm’s performance. The result shows that, the performance of WPP can be adjusted to satisfy the requirements of users by changing the four parameters in WPP.This paper solves the conflicts between the path planning result and the restricted resource in WSNs. Although, the finally planning result maybe not the shortest path, but, we think it is more valuable to get a more lifetime and low planning complexity. The result of this paper is more suitable for the applications which have no block clusters or requires high lifetime. The researches of this paper provide a novel way for the path planning based WSNs.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation components,parts > Transmitter ( converter),the sensor > Sensor applications
© 2012 www.DissertationTopic.Net Mobile