OceanRoute: Vessel Mobility Data Processing and Analyzing Model Based on MapReduce

作     者：LIU Chao LIU Yingjian GUO Zhongwen JING Wei 

作者机构：Department of Computer Science and Technology Ocean University of China 

出 版 物：《Journal of Ocean University of China》 (中国海洋大学学报（英文版）)

年 卷 期：2018年第17卷第3期

页      面：594-602页

核心收录：

学科分类：080904[工学-电磁场与微波技术] 0908[农学-水产] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 0710[理学-生物学] 0810[工学-信息与通信工程] 0707[理学-海洋科学] 0815[工学-水利工程] 080402[工学-测试计量技术及仪器] 0804[工学-仪器科学与技术] 0824[工学-船舶与海洋工程] 081001[工学-通信与信息系统] 082401[工学-船舶与海洋结构物设计制造] 

基　　金：supported by the National Key R&D Program (No. 2016YFC1401900) the China Postdoctoral Science Foundation (No. 2017M620293) the National Natural Science Foundation of China (Nos. 61379127, 61379128, 61572448) the Fundamental Research Funds for the Central Universities (No. 201713016) Qingdao National Labor for Marine Science and Technology Open Research Project (No. QNLM2016ORP0405) Natural Science Foundation of Shandong (No. ZR2018BF006) 

主　　题：ocean delay tolerant network MapReduce mobility pattern trace similarity vessel data analysis 

摘      要：The network coverage is a big problem in ocean communication, and there is no low-cost solution in the short term. Based on the knowledge of Mobile Delay Tolerant Network(MDTN), the mobility of vessels can create the chances of end-to-end communication. The mobility pattern of vessel is one of the key metrics on ocean MDTN network. Because of the high cost, few experiments have focused on research of vessel mobility pattern for the moment. In this paper, we study the traces of more than 4000 fishing and freight vessels. Firstly, to solve the data noise and sparsity problem, we design two algorithms to filter the noise and complement the missing data based on the vessel s turning feature. Secondly, after studying the traces of vessels, we observe that the vessel s traces are confined by invisible boundary. Thirdly, through defining the distance between traces, we design MR-Similarity algorithm to find the mobility pattern of vessels. Finally, we realize our algorithm on cluster and evaluate the performance and accuracy. Our results can provide the guidelines on design of data routing protocols on ocean MDTN.