基于ZigBee的农业传感网络与土壤湿度模型的研究

发布时间：2018-01-07 05:28

本文关键词：基于ZigBee的农业传感网络与土壤湿度模型的研究　出处：《复旦大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着物联网(Internet of Things, IoT)概念的兴起和发展,物联网技术很快地渗透到了人类生产和生活的各个领域。其中,农业物联网代表了现代农业技术发展的前沿,被称为未来物联网最重要的发展方向之一。而无线传感器网络(wireless sensor network, WSN)是物联网最重要的支撑技术和实现手段。因此,农业无线传感网络的研究与实践对于农业物联网的推广以及“精准农业”的实现有着重要的现实意义。ZigBee技术是一种标准化的面向无线自动控制的低速率、低功耗和低成本的无线网络组网方案,得到了广泛的认可和应用,是农业无线传感网络重要的实现方法之一。但是,现有的农业无线传感网络依旧存在着网络能耗大和传感器成本高的问题,严重制约了农业物联网的发展。首先,针对农业无线传感网络的能耗问题,本文提出了功率调节优化算法和低功耗路由算法。其中,功率调节优化算法用于优化终端节点的能耗,低功耗路由算法可以延长ZigBee网络中路由器节点的工作时间。本文提出的算法在Z-Stack协议栈上实现。测试表明,本文提出的改进算法可以有效地节省网络能耗,进而延长网络寿命。其次,针对农业无线传感网络的传感器成本问题,本文重点就土壤湿度的传感技术进行了研究,分别用多元线性回归和BP神经网络的方法建立了土壤湿度电阻法标定模型。测试表明,当测试样土的土壤类型和标定样土一致的条件下,可以使用成本低廉的土壤电阻传感器对土壤湿度进行较准确的测定。更进一步地,鉴于土壤湿度在一天之内变化小的特点,结合土壤热运动的规律,本文提出基于支持向量机(Support Vector Machine, SVM)的土壤湿度分类模型,根据该模型,在农业传感节点无土壤湿度传感器的情况下,能对节点所处地点的土壤湿度做出分类,从而对土壤墒情作出判断。最后,在上述研究的基础之上,本文设计并实现了一套基于ZigBee技术的农业无线传感网络的解决方案,该方案具有低成本、低功耗的特点,可以用于实际农业数据的实时采集,该解决方案是利用现代通信和传感器技术对农业现代化转型发展的一种有益探索和尝试。
[Abstract]:With the rise and development of the concept of Internet of things (IoT), Internet of things (IOT) technology has quickly penetrated into various fields of human production and life. The agricultural Internet of things represents the frontier of the development of modern agricultural technology. It is known as one of the most important development directions of the Internet of things in the future. Wireless sensor network is a wireless sensor network. WSNs are the most important supporting technology and implementation means of the Internet of things. Research and practice of Agricultural Wireless Sensor Network for extension of Agricultural Internet of things and Precision Agriculture. ZigBee technology is a standardized low rate wireless automatic control. Low-power and low-cost wireless network networking scheme has been widely recognized and applied. It is one of the most important methods to realize agricultural wireless sensor network. The existing agricultural wireless sensor network still has the problems of high energy consumption and high sensor cost, which seriously restricts the development of agricultural Internet of things. First, the energy consumption of agricultural wireless sensor network. In this paper, a power regulation optimization algorithm and a low power routing algorithm are proposed, in which the power regulation optimization algorithm is used to optimize the energy consumption of terminal nodes. Low-power routing algorithm can extend the working time of router nodes in ZigBee network. The proposed algorithm is implemented on Z-Stack protocol stack. The improved algorithm proposed in this paper can effectively save energy consumption and prolong the network life. Secondly, the sensor cost of agricultural wireless sensor network can be solved. In this paper, the sensing technology of soil moisture was studied, and the calibration model of soil moisture resistance method was established by multivariate linear regression and BP neural network. When the soil type of the tested soil is the same as that of the calibrated soil, the low cost soil resistance sensor can be used to measure soil moisture accurately. In view of the small change of soil moisture in one day and the regularity of soil thermal movement, this paper proposes a support Vector Machine based on support vector machine. According to the model, the soil moisture of the node can be classified when the agricultural sensor node has no soil moisture sensor. Finally, on the basis of the above research, this paper designed and implemented a set of agricultural wireless sensor network based on ZigBee technology, which has low cost. Low power consumption can be used to collect real time agricultural data. This solution is a useful exploration and attempt to make use of modern communication and sensor technology in the transformation and development of agricultural modernization.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：S152.71;TN92;TP212.9

【参考文献】