C-UTBDS:蜂窝网络大数据驱动城市交通出行特征挖掘

C-UTBDS:Urban Traffic Travel Characteristic Mining Based on Cellular-network Big Data

随着城市化进程不断加快,城市功能分区正从规模增长向质量提升转型发展,优化城市空间布局、深化多模式融合交通运营和打造一体化出行服务平台是城市交通数字化转型的核心需求,挖掘城市交通出行特征并对其进行分析,有助于完善城市立体化交通服务体系、满足多样化出行需求、推进城市用地合理开发利用以及指导城市决策者制定合理的规划措施。蜂窝信令数据(Cellular Signaling Data,CSD)具有覆盖范围广、样本量大、可长期连续监测等优势,蜂窝网络大数据可以较低的成本挖掘大规模人口的起讫(Origin Destination,OD)分布和出行行为模式,是促进未来城市智能交通发展的重要组成部分。基于此,对现有交通信息采集方法、发展制约因素和蜂窝信令数据价值进行了总体概述,梳理了面向蜂窝网络大数据的城市交通出行特征挖掘框架、关键技术研究进展和未来发展方向。首先,依据基于蜂窝信令数据的城市交通出行特征挖掘系统的功能规划与发展需求给出了其架构设计与应用框架。其次,从蜂窝网络出行链构建角度总结了蜂窝移动通信网络结构、出行链特征与提取框架,阐述了针对出行链中噪声数据和轨迹震荡的数据优化方法,以及出行链轨迹与实际路网融合时的路网匹配技术。然后,面对蜂窝网络大数据驱动下的城市空间结构优化与多模式交通发展需求,从人口流动监测、出行模式识别、行为分析与预测等城市交通出行特征挖掘方面详细介绍研究现状。最后,从5G优化定位、多源数据处理与挖掘、细粒度出行模式识别、基于组件的系统模型体系构建等方面指出了未来研究的技术方向和发展趋势。

With the continuous acceleration of urbanization,urban functional zoning has shifted the emphasis from scale growth to quality improvement.Optimizing urban spatial layout,deepening multi-mode integration of traffic operation,and building an integrated travel service platform are the core needs of urban transportation's digital transformation.Urban traffic travel characteristic mining and behavior analysis are helpful in improving the urban multidimensional transportation service system,meeting diversified travel needs,promoting the rational development and utilization of the urban land,and guiding the urban decision makers to formulate reasonable planning measures.Cellular signaling data(CSD)has the advantages of wide coverage,large sample size,and long-term continuous monitoring.Cellular-network big data can analyze the origin-destination(OD)distribution and travel behavior pattern of individuals or large populations at a lower cost,thus being important for promoting the development of future urban intelligent transportation.In this paper,the existing traffic information collection methods,development constraints,and the importance of cellular signaling data are summarized,and the architecture of an intelligent transportation system based on cellular signaling data,key technology research progress,and future development direction are reviewed.First,according to its functional planning and development requirements,the architectural design and application framework of the CSD-based urban traffic big data system(C-UTBDS)are proposed.Second,from the perspective of cellular network travel chain construction,the structure of the cellular mobile communication network,travel chain characteristics,and extraction framework are summarized;the noise data of the travel chain,data optimization methods for track vibration,and the road network matching technology,when the travel chain trajectory is integrated with the actual road network,are expounded.Then,considering the needs of urban spatial structure optimization and multi-mode traffic development driven by cellular-network big data,the research status of urban traffic travel characteristics mining is introduced in detail,including population flow monitoring,travel pattern recognition,behavior analysis and prediction.Finally,the technical direction and development trend of future research are highlighted from the aspects of 5G optimization positioning,multi-source data processing and mining,fine-grained travel pattern recognition,and component-based system model architecture construction.