| With the development of IOT (Internet of Things),the IOT terminals are widely used and developed in many fields. Vehicle mounted IOT, as the core of the future urban traffic,has been developing rapidly. Automobiles are the main shape of the network, though their essence is the vehicle mounted terminals in them. These terminals participated in the traffic of the city, and formed a large network as the nodes of it. The essential problem to prototyping a vehicle mounted IOT is how to effectively utilize huge amount of data generated by these terminals. Based on these demands, this paper purposed using Hadoop distributed computation platform to storage the huge amount of vehicle mounted IOT data,and analyzing the data under this frame.First, based on the data of which, this study analyzed the data's formation process and scale of Vehicle mounted IOT, purposed the basic data structure that suits vehicle mounted IOT to transmit and storage. Combining with the advantages and characteristics of different data manage components,this paper purposed a storage scheme that mainly a combination of HDFS, HBase and Hive, and made detailed designation based on practical storage demands. Furthermore, this paper discussed and analyzed the methods of vehicle mounted IOT data analyze and data mining. Two kinds of methods have been put forward to deal with the data analysis of different size using Hive QL and MapReduce model. Also this paper designed and realized data statistics, driver behavior analysis, quantity of state association rule mining etc. based on the application requirements of Vehicle mounted IOT.According the purposed frame model, this paper has realized the construction of Hadoop and integrated relevant components, set corresponding parameters to improve and analyze the actual storage performance, and then verified the feasibility and effectiveness of the processing method for vehicle mounted IOT data put forward in this paper by experiment results and performance. |
PDF Full Text Request