基于WSN的生物创新实验室关键技术研究

发布时间：2017-12-27 03:19

本文关键词：基于WSN的生物创新实验室关键技术研究　出处：《华东师范大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着物联网技术的普及,在校园生活中出现了未来教室、智能讲堂等新颖的管理与教学模式。这种新颖的模式将课堂教育与WSN技术相融合,使教学方式更加丰富多彩,提高了学生的学习兴趣和老师的工作效率。然而该模式中传感器节点的能量有限且不可补充,如何有效地减少节点能耗、充分利用系统中的资源是延长网络运作时间的关键。本文基于WSN设计了生物创新实验室系统,该系统通过在监测区域内部署相应的传感器节点对生物生长环境进行实时监测,使学生更加直观地了解生物的生长方式及其生长过程中的环境参数变化,并且实现了对生物创新实验室的智能化管理。同时针对该系统中传感器节点的能量损耗问题,本文分别从节点数据融合和数据路由传输两个方面出发,设计了基于深度学习的数据融合算法以及基于分层模型的蚁群优化多径路由算法。(1)数据融合算法:该算法将深度学习技术应用于WSN数据融合领域,首先对监测区域内的传感器节点进行聚类划分,然后利用特征提取及分类模块对各聚类内节点进行特征提取、分类,最后由分簇簇首根据分类结果对节点采集的数据进行压缩融合,并发送至聚类融合点进行路由转发。(2)多径路由算法:针对传统蚁群算法的不足之处设计分层模型,改进链路初始化信息素浓度、链路启发值以及链路信息素更新策略,最后综合考虑节点的剩余能量、网络传输延时、分组丢失率及路径相关度等因素,通过自定义目标函数依次选取目标值最大的多条可达路径对数据进行传输。实验结果显示,本文设计的基于深度学习的数据融合算法能够有效地对生物创新实验室监测区域内传感器节点采集的冗余数据进行压缩,减少了网络中待传输的数据量,并且提高了基站接收数据的正确率;与其它路由协议相比,本文设计的基于分层模型的蚁群优化多径路由算法在节点平均能耗、传输时延及数据包丢失率等方面具有明显的优势,确保了该生物创新实验室系统中采集的环境参数信息实时、准确地传输至本地服务器。
[Abstract]:With the popularization of the Internet of things technology, new management and teaching modes, such as future classrooms and intelligent auditorium, have appeared in campus life. This novel model combines the classroom education with the WSN technology, makes the teaching way more colorful, improves the students' interest in learning and the work efficiency of the teachers. However, the energy of sensor nodes is limited and cannot be supplemented. How to effectively reduce node energy consumption and make full use of resources in the system is the key to prolong network operation time. In this paper, based on the design of the WSN biological laboratory system innovation, the system of biological growth environment monitored by the corresponding sensor nodes deployed in the monitoring area, make students more intuitive understanding the changes of environmental parameters and biological growth in the growth process, and realizes the intelligent management of the laboratory of biological innovation. Meanwhile, aiming at the energy loss problem of sensor nodes in the system, this paper designs data fusion algorithm based on deep learning from the two aspects of node data fusion and data routing and transmission respectively, and the ant colony optimization multi-path routing algorithm based on hierarchical model. (1) data fusion algorithm: the algorithm will deep learning technology is applied to WSN data fusion, firstly, clustering of sensor nodes in the monitoring area, and then use the feature extraction and classification module for feature extraction and classification of the nodes in each cluster, finally by clustering according to the classification results of the first node data acquisition compression fusion, and sent to the clustering routing forwarding point. (2) multipath routing algorithm: according to the deficiency of design hierarchical model of traditional ant colony algorithm, improved link initialization pheromone and heuristic value link and link pheromone update strategy, considering the residual energy and network transmission delay, packet loss rate and node path correlation factors, through the defined objective function followed by selecting the target the maximum value of multiple paths for data transmission. The experimental results show that the deep learning data fusion algorithm can effectively to biological innovation laboratory monitoring area, sensor nodes collect redundant data compression based on this design, reduce the amount of data to be transmitted in the network, and to improve the correct rate of the base station receives the data; compared with other routing protocols, multipath routing algorithm has obvious the advantage in the average energy consumption of nodes, transmission delay and packet loss rate and other aspects of ant colony optimization based on hierarchical model designed in this paper, to ensure that the acquisition of the biological innovation laboratory system in the environmental parameter information timely and accurately transmitted to the local server.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP202;TP212.9;TN929.5

【相似文献】