A Sensor Cloud Service Model Based On Data Prediction And Reuse

With the development of Cloud computing,Internet of Things and Sensor networks,sensor cloud had been realized as an emerging technology to provide high quality engaged localized services.However,the computing and storage capabilities of the fog node have not been fully utilized.Compared with sensor networks,the feature of sensing data and structure of sensor cloud are more complex,and it is difficult to manage with traditional processing and analysis solutions.Aiming at the characteristics of Sensing-as-a-Service of sensor cloud environment,this paper introduced a data mining schedule into data processing and transmission in the service process in sensor cloud.We proposed a pattern mining and matching scheme that is suitable for sensor cloud environment.Generating prediction data in different tasks to match the existing sensing data with sequential pat-terns provides a scheme for task migration in sensor cloud environment.This paper discusses the parameter setting of the prediction model.The parameters involved in the prediction model can be flexibly adjusted and balanced to achieve a more user-friendly prediction effect.Simulate result shows that our model is effective to reduce energy consumption and improve response time under a high accuracy.Besides,this paper proposed a sensor cloud service solution based on data reuse.A Sensor cloud service scheme is constructed with reuse rate of sensing data as the main performance index.In this service scheme,a heuristic algorithm based on shuffled frog leaping algorithm is introduced to allocate network nodes in response to the needs of users.In addition,considering the uncertainty of service parameters in the process of Sensor cloud service,this paper introduces a user state rotation queue.The reusing of sensor data in Sensor cloud can effectively reduce the energy consumption and energy consumption of sensor nodes,as well as bandwidth usage.The service model and related algorithms proposed in this paper have been verified by simulation experiments,which can effectively reduce system energy consumption and improve response speed on Sensor cloud network.