基于低功耗广域网的海岛水产养殖环境监测系统研制

针对海岛环境中水产养殖区域分散、工作环境恶劣、人工巡检不便等问题,设计了基于低功耗广域物联网的海岛养殖环境监测系统。系统包括集成Arduino和传感器的终端采集节点,通过LoRa技术实现数据汇总和远距离传输的汇聚网关,利用Python与PostgreSQL开发用于数据接收、存储、处理、访问和控制的后台监测系统。通过对网络拓扑复杂度、能耗等方面的评估,表明在海岛环境下部署水产养殖环境监测系统,相比传统Zigbee多跳无线传感网,采用LoRaWAN,其单跳节点覆盖范围更大,而网络复杂度、能耗等更优。测试表明该系统能以较低功耗实现整片区域内远距离数据采集,有效传输养殖区水体环境数据。网络生存期与传输可靠性测试表明,当传感器节点采用3.7 V/4 200 mAh锂电池,上传周期为30 min时,监测网络的有效生存期理论上可达2.4 a;在800 m通信范围内,发射功率为20 mW时,节点丢包率小于3.6%,具有较高的通信可靠性。该研究可为水产养殖生产和物联网应用研究提供有效参考。

Development of island aquaculture environment monitoring system based on low-power wide area networks

The low-power wide area networks (LPWAN) provide efficient and feasible technical means for monitoring the environment of island aquaculture. At present, most successful environment monitoring applications have been deployed in artificially controlled environments such as cities and parks, or with the help of mobile telecommunication networks. Unlike previous researches, this study introduces LPWAN into wild environments such as islands and integrates with existing sensor technologies to build long-range wireless sensing and monitoring systems. Due to the relatively scattered areas, adverse working environments, inconvenience of manual inspection and long period of data collection, several problems exist if traditional wireless sensor network (WSN) is used for the deployment of island aquaculture environmental monitoring system. Firstly, the communication range of the node is limited; secondly, the network needs too many routing nodes and is too complex; thirdly, the overall network power consumption will be rather high. In order to improve the performance of the system, the paper designed a new kind of monitoring system for island aquaculture environment based on LPWAN. The system includes terminal acquisition nodes with integrated Arduino and some sensors, a sink gateway for data aggregations and long-distance transmission via the long range radio (LoRa), and a background database management system for data collection, storage, processing, access and control using Python and PostgreSQL. When the monitoring network is actually deployed, it is considered that some nodes locate in the center of the aquaculture area and the monitoring worker cannot easily reach the locations. Therefore it is difficult to easily place and deploy the terminal sensor nodes. To solve the problem, we also designed and implemented an unmanned ship equipped with terminal sensor nodes to facilitate the deployment of terminal nodes. Through assessment of the complexity of the network topology and energy consumption, it was demonstrated that LoRaWAN is superior to the traditional ZigBee network to deploy a monitoring network in the island aquaculture environment. Our tests also showed that this actual system can effectively achieve the long-distance data acquisition and transmission of the environment information for the island aquaculture with low power consumption. Network lifetime and transmission reliability tests showed that the network has an effective lifetime of 2.4 years when terminal sensor nodes use the 3.7V/4200 mAh battery and the upload period is 30 minutes. Within 800 m for the communication distance, the node packet loss rate is less than 3.6% when the transmission power is 20 mW. Furthermore, it is found that the received signal strength fluctuates significantly within 24 hours. We investigated that at night, the received signal strength decreases significantly and the quality of the communication link reduces. It showed that the island environment of the aquaculture area has a great impact on the communication quality of the LPWAN system. In particular, the temperature changes at night and the wind enhancement can lead to rapid decay of the LoRa signal and the quality of the network communication drops significantly. According to the test in the island environment, the system can effectively transmit data from the terminal sensor nodes and reflect the information of island aquaculture environment.