Research And Implementation Of Data Acquisition Mechanism Based On Compressive Sensing For Vehicular Sensor Network

Smart city can effectively alleviate the increasingly serious urban problems and receive widespread attention.The Vehicular Sensor Network(VSN)achieves efficient coverage of the sensing area due to the mobility of the vehicle and is one of the important methods for data collection in a smart city.However,with the development of smart city,the amount of data that needs to be collected has grown explosively.In the VSN,the rapid movement of the vehicle causes a drastic change in the topology,and data transmission of a large amount of data becomes very difficult.On the other hand,the non-uniform distribution of vehicles in the VSN also leads to data redundancy and data voids in data acquisition.The study found that most of the data collected in the vehicle data acquisition network city data collection has spatial and temporal correlation,and Compressive Sensing(CS)can be used to compress and reconstruct the data.Therefore,this paper studies the urban data acquisition mechanism of vehicular sensor network based on CS.The specific research content is as follows:First,the VSN data acquisition model based on CS is studied.In view of In view of the difficulty of mass data transmission in VSN data acquisition model,this paper analyzes the existing VSN network data acquisition model and CS method and proposes a VSN data acquisition model based on compressed sensing,and discusses The application of CS in this model.Second,the dynamic CS method of VSN.To reduce the reconstruction accuracy caused by the dynamic changes of data characteristics and vehicle distribution,a dynamic compression acquisition method was proposed.This method takes into account the characteristics of data sparsity and dynamic distribution of vehicle nodes in the data acquisition process of VSN.Based on the existing CS methods,it adds measurement feedback control,and adjusts the measurement rate according to the current situation.Third,a priority based on VSN data transmission method.For different compressed data contribution to reconstruction accuracy,this paper proposes a priority-based VSN data transmission method.This method prioritizes all data packets that need to be transmitted.In the next hop selection,both the geographical location and the load degree of the node are taken into consideration,and the forwarding strategy is selected according to the data priority.Fourth,based on the above two studies,we design and implement a prototype system based on compressive sensing for on-board sensor network data acquisition.Through this platform,we validate the vehicle sensing network dynamic compressive sensing method and priority-based vehicle transmission.Influencing the effectiveness of the data transmission method of the network,the platform can achieve high-precision,low-consumption urban data acquisition of the vehicular sensor network.