Research On The Application Of Wireless Sensor Network In Storage Environment Monitoring

Grain storage has an important significance for people’s livelihood、defense and otheraspects of the state. However, there is a serious lack of conditions and technical means forgrain storage of our nations, which leads to a large amount of food deterioration、rot andinsect pests. For the poor drawbacks of the low efficiency and accuracy of artificia lpoint-by-point measurement, the grain storage environment monitoring system based onwireless sensor networks were studied. The system has achieved real-time monitoring of thetemperature、humidity and pest data in the granary by sensors.This paper is based on the national "863" project. The hardware and software of thesystem has been designed. According to the characteristics of the limited energy, high densityand much redundant data of nodes in the warehouse environment monitoring system, networkenergy saving strategies has been researched. The results in this paper are shown as follow.1. The hardware of the node has been designed. The detection node has been configuredwith temperature、humidity and pyroelectric infrared sensor. The actuator node has beenconfigured with PCM1770audio converter chip. The processor, memory, communicationmodule and power module of the node has been designed respectively. The host computeruses the LABVIEW software to achieve regulatory of granary various environmentalparameters, and achieve remote monitoring by remote panel technology (Remote Panel) andGSM technology. Finally the real-timing and intelligence of granary monitoring system canbe given. The larger impact of the error can be effectively avoided for fusion results. Therebythe energy consumption of nodes can be reduced.2. According to the project, we put forward an improved DS evidence combination rulebased on fuzzy theory. Aggregation node sends the determined results to the user and actuatornodes. The actuator nodes play audio to expel pests. Simulation result shows: this fusionalgorithm has a better robustness and lower processing time. In order to reduce theredundancy of the measuring information, a data fusion method based on the trust has beenproposed.3. According to the characteristics of the limited energy of nodes. A energy-based Clusteringalgorithm has been proposed. The ECA has the advantage of uniform cluster head distribution,good stability, can effectively extend the network lifetime. Focusing on the characteristic ofhigh distribution density of Wireless sensor network node, putting forward the method ofcluster node scheduling based on clustering algorithm. Giving the method o f calculatingactive nodes in the cluster from the two aspects of the network coverage and monitoringaccuracy. And determining the final node in a working state based on the energy anddistribution of nodes in the cluster, reducing the data collection of energy consumption in thecluster and wireless communication channel interference.