基于移动轨迹数据的城市基站人流预测研究
Urban Base Station Traffic Forecasting Based on Mobile Trajectory
摘要: 城市通讯基站负载日益增长且不均衡，为优化基站资源调度分配，提高基站服务质量，提出了一种基于时空关联特征的GA-BP神经网络分钟级多步预测方法。本研究基于移动用户轨迹数据，建立基站间负载时空转移概率矩阵，提取影响基站负载的时空因素。针对BP神经网络存在易陷入局部极小值的问题，使用遗传算法对BP神经网络的初始权值和阈值进行优化，构建GA-BP神经网络模型进行分钟级负载预测模型。该模型基于某运营商脱敏后的移动用户轨迹数据进行训练与预测，结果显示，基于时空关联特征的GA-BP模型能有效对基站分钟级多步负载进行预测。
Abstract: Communication base station load is increasing and unbalanced. In order to optimize base station resource scheduling and improve base station service quality, we proposed a model of multi-step base station load prediction in minute-level with spatio-temporal correlation data based on BP neural network optimized by genetic algorithm. By using mobile user trajectory data, we estab-lished the spatio-temporal transition probability matrix between base stations and extracted the spatio-temporal features which affect the base station load. As BP neural network easily got into the local extremes, genetic algorithm was employed to optimize the initial weights and thresholds of BP network model. The algorithm proposed is verified by a mobile user trajectory dataset encrypted by the operator. The results showed that the optimized model with spatio-temporal correlation features can effectively predict the multi-step base station load in minute-level.
文章引用：骆彦彦. 基于移动轨迹数据的城市基站人流预测研究[J]. 计算机科学与应用, 2019, 9(4): 757-768. https://doi.org/10.12677/CSA.2019.94085

参考文献

[1] Leng, Z., Gao, J., Zhang, B., et al. (2013) Short-Term Traffic Flow Forecasting Model of Optimized BP Neural Network Based on Genetic Algorithm. Proceedings of the 32nd Chinese Control Conference, Xi’an, 26-28 July 2013, 8125-8129.
[2] Tsai, T.H., Lee, C.K. and Wei, C.H. (2009) Neural Network Based Temporal Feature Models for Short-Term Railway Passenger Demand Forecasting. Expert Systems with Applications, 36, 3728-3736. [Google Scholar] [CrossRef] 
[3] 丁闪闪, 王维锋, 季锦章, 等. 基于时空关联和BP_Adaboost的分钟级交通参数预测[J]. 公路交通科技, 2016, 33(5): 98-104.
[4] Min, W. and Wynter, L. (2011) Real-Time Road Traffic Prediction with Spa-tio-Temporal Correlations. Transportation Research Part C Emerging Technologies, 19, 606-616. [Google Scholar] [CrossRef] 
[5] Li, S., Wang, L. and Liu, B. (2014) Prediction of Short-Term Traffic Flow Based on PSO-Optimized Chaotic BP Neural Network. International Conference on Computer Sciences & Applications, Wuhan, 14-15 De-cember 2013, 292-295. [Google Scholar] [CrossRef] 
[6] 王凯, 张仕学, 汤仕爽, 等. 人群聚集规模与移动基站接入量之间的关系研究[J]. 软件导刊, 2015, 14(8): 61-63.
[7] 孙莹, 陈夏明, 王海洋, 等. 城市尺度下基站人群时空预测模型[J]. 计算机应用研究, 2016, 33(12): 3521-3526.
[8] 方志祥, 倪雅倩, 张韬, 等. 利用终端位置时空转移概率预测通讯基站服务用户规模[J]. 地球信息科学学报, 2017, 19(6): 772-781.
[9] Meng, H., Long, F., Guo, L., et al. (2016) Cooperating Base Station Location Optimization Using Genetic Algorithm. Chinese Control and Decision Conference, Yinchuan, 28-30 May 2016, 4820-4824. [Google Scholar] [CrossRef] 
[10] Do, T.M.T. and Gatica-Perez, D. (2014) Where and What: Using Smartphones to Predict Next Locations and Applications in Daily Life. Pervasive and Mobile Computing, 12, 79-91. [Google Scholar] [CrossRef] 
[11] Chandra, R., Ong, Y.S. and Goh, C.K. (2017) Co-Evolutionary Multi-Task Learning with Predictive Recurrence for Multi-Step Chaotic Time Series Prediction. Neurocomputing, 243, 21-34. [Google Scholar] [CrossRef] 
[12] 殷礼胜, 何怡刚, 董学平, 等. 交通流量VNNTF神经网络模型多步预测研究[J]. 自动化学报, 2014, 40(9): 2066-2072.
[13] Li, Q. and Chan, M.F. (2017) Predictive Time-Series Modeling Using Artificial Neural Networks for Linac Beam Symmetry: An Empirical Study. Annals of the New York Academy of Sciences, 1387, 84-94. [Google Scholar] [CrossRef] [PubMed]
[14] Zhang, R.D. and Wang, S.Q. (2005) Neural Network Based Multi-Step Predictive Con-trol for Nonlinear Systems. Control and Decision, 20, 332-336.
[15] Qi, Y. and Ishak, S. (2014) A Hidden Markov Model for Short Term Prediction of Traffic Conditions on Freeways. Transportation Research Part C: Emerging Technologies, 43, 95-111. [Google Scholar] [CrossRef] 
[16] 焦恩伟, 吴越. 基于LTE的人群密度监控及预测[J]. 通信技术, 2017, 50(3): 492-495.
[17] 高玉明, 张仁津. 基于遗传算法和BP神经网络492-495的房价预测分析[J]. 计算机工程, 2014, 40(4): 187-189.
[18] 严旭, 李思源, 张征. 基于遗传算法的BP神经网络在城市用水量预测中的应用[J]. 计算机科学, 2016, 43(s2): 547-550.