Research On Sensor-cloud Based Precision Irrigation Control Of Farmland And Provision Optimization Methods

As the underlying supporting technology of the Internet of Things(IoT),wireless sensor networks(WSNs)are widely applied in various agricultural applications.However,due to the limitations of deployment,energy,computing,communication,and scalability,the WSNs also face many bottlenecks in agricultural applications.Thus,the sensor-cloud(SC)technology is regarded as a trend for the future development of traditional WSNs.By combining WSNs with cloud computing technology,and making full use of virtualization technology in the cloud,the limitations of WSNs in deployment,computing,and scalability have been greatly eased.This paper focuses on sensor-cloud technology and conducts in-depth research on its appli-cation in the agriculture fields.We analyze the layered architecture and working pattern of the agricultural sensor-cloud system? we propose a sensor-cloud based precision sprinkler irrigation control system model and give the formal definitions as well as the specific work process of each component in the system.Furthermore,targeting the issue of virtual sensor provisioning(VSP),we propose a similarity of measurements based VSP optimization algorithm to reduce the net-work energy consumption and improve the quality of data service.The main contributions of this article can be summarized as:(1)Based on the development background of sensor-cloud technology,an in-depth analysis of WSNs,cloud computing,and sensor-cloud technologies is introduced.We also in-troduce a layered sensor-cloud system framework for precision agriculture.We give the related definitions of the main responsibilities of each layer,the included entity objects,the work content of each entity,and the overall working pattern.Finally,we analyze the advantages and challenges of agricultural sensor-cloud technology.(2)We propose a sensor-cloud based precise sprinkler irrigation control system model.By extending the traditional sprinkler irrigation system to the physical layer of sensor-cloud system,the remote control of sprinkler irrigation equipment such as sprinklers can be realized.The proposed model further divides the irrigation area according to the geo-graphic location where the irrigation equipment is deployed.The sprinklers in each area are controlled by solenoid valve drivers,enabling precise irrigation control.Experimen-tal results show that the proposed system model has a lower energy consumption level and cost than the traditional WSN system.(3)We propose a measurement similarity-based VSP optimization algorithm.First,we ex-ploit the linear regression model to classify the physical sensors according to the chang-ing trend of historical data,and then the k-means clustering algorithm is performed to analyze each class.The physical sensors with higher similarity are clustered into one cluster.Finally,select one representative sensor from each cluster to minimize the num-ber of redundant nodes.Experimental results show that the proposed scheme has lower energy consumption and higher data accuracy than the benchmark scheme.