Design of Spatiotemporal Prediction Algorithms for Wireless Network Traffic
《中华人民共和国国民经济和社会发展第十四个五年规划和2035 年远景目标纲要》提出了中国未来经济和社会发展的宏观要求,强调新质生产力的重要性,并推进6G等前沿技术的研究和应用[1]。MIMO技术作为新时代无线通信的关键技术之一,通过配备多个发射天线和多个接收天线进行数据传输和通信,利用波束赋形、空分复用、空分多址等方式显著提高通信系统的容量、覆盖范围和数据传输速率[2]。随着5G大规模MIMO(5G-mMIMO,5G massive MIMO)技术的成熟与广泛应用,无线网络环境变得更加复杂,信号的传播与接收不再仅仅依赖传统的二维(经纬度)或三维(经纬度、高度)位置,这使得测量报告(MR,Measurement Report)中记录的更多是高维波束空间信息,而非传统的地理坐标。传统的流量预测方法在这种环境下显得不足,因为它们无法充分利用高维波束空间中的复杂信息。 根据最新的调查,5G连接数从2019年的约1300万增长到2023年的17.6亿,增长超过100倍[3][4],截至2024年第一季度全球连接数已接近20亿[5][6],这种显著的增长趋势将在未来的6G网络时代继续进行,随之而来的是设备产生的海量数据对网络带宽、延迟和稳定性提出的更高要求。由于用户的上网行为在时间和空间上具有高度异质性[7][8],不同时间和空间下的流量差异可能非常显著。为了满足这些日益严峻的要求,积极有效地分配无线网络资源变得至关重要。作为网络资源管理系统的一个组成部分,无线网络流量预测面临着严格的准确性和可靠性要求[9],同时也面临着海量连接带来的需求上的紧迫性。 经调查研究,传统的流量预测往往基于历史数据,如经典的LSTM[10]模型、Transformer[11]模型都是从单一的时间维度进行流量预测。随着无线通信技术的发展与演进,近年来,国内外在无线网络流量预测方面的研究主要集中在时空预测模型上,科研人员将单一的时间维度演进为同时考虑时间和空间两个维度的流量预测[12][13],目前的方法主要依赖于物理空间中的二维或三维地理位置信息。然而,针对高维波束空间的流量预测研究尚处于起步阶段。我们的项目正是为了解决这一挑战,旨在开发新的算法和模型,能够在高维波束空间中进行更准确的无线流量预测,从而优化网络资源的分配和用户体验。 参考文献: [1] 中华人民共和国国务院. 中华人民共和国国民经济和社会发展第十四个五年规划和2035 年远景目标纲要[M]. 北京: 人民出版社, 2021. [2] J. An, C. Yuen, L. Dai, M. Di Renzo, M. Debbah and L. Hanzo. Near-Field Communications: Research Advances, Potential, and Challenges[J]. IEEE Wireless Communications, 2024, 31(3), 100-107. [3] Cisco, “Cisco annual internet report (2018–2023),” 2020. Accessed: Mar. 9, 2020. [Online]. Available: https://www.cisco.com/c/en/us/solutions/collateral/executive-perspectives/annual-internet-report/white-paper-c11-741490.html [4] Business Wire, "Global 5G Connections Reach Nearly Two Billion," 2024. Accessed: July 3, 2024. [Online]. Available: https://www.businesswire.com/news/home/20240702919332/en/Global-5G-Connections-Reach-Nearly-Two-Billion [5] 5G Americas, "Global 5G connections surge to 1.76 billion, 66 percent growth year over year as North America leads charge," 2024. Accessed: Sep. 3, 2024. [Online]. Available: https://www.5gamericas.org/global-5g-connections-surge-to-1-76-billion-66-percent-growth-year-over-year-as-north-america-leads-charge/. [6] Advanced Television, "Data: 5G connections near 2bn," 2024. Accessed: July 3, 2024. [Online]. Available: https://advanced-television.com/2024/07/03/data-5g-connections-near-2bn/ [7] Z. Wang, J. Hu, G. Min, Z. Zhao, Z. Chang, and Z. Wang, "Spatial-Temporal Cellular Traffic Prediction for 5G and Beyond: A Graph Neural Networks-Based Approach," IEEE Transactions on Industrial Informatics, vol. 19, no. 4, pp. 5722-5731, Apr. 2023, doi: 10.1109/TII.2022.3182768. [8] B. Gu, J. Zhan, S. Gong, W. Liu, Z. Su and M. Guizani, "A Spatial-Temporal Transformer Network for City-Level Cellular Traffic Analysis and Prediction," in IEEE Transactions on Wireless Communications, vol. 22, no. 12, pp. 9412-9423, Dec. 2023, doi: 10.1109/TWC.2023.3270441. [9] Z. Wang, J. Hu, G. Min, Z. Zhao, and J. Wang, "Data-Augmentation-Based Cellular Traffic Prediction in Edge-Computing-Enabled Smart City," IEEE Transactions on Industrial Informatics, vol. 17, no. 6, pp. 4179-4189, June 2021, doi: 10.1109/TII.2020.3009159. [10] Hochreiter S, Schmidhuber J. "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997. [11] Vaswani A, Shazeer N, Parmar N, et al. "Attention is all you need," ArXiv preprint arXiv:1706.03762, 2017. [12] Chen, X., Chuai, G., Zhang, K., & Gao, W. "Spatial-temporal Cellular Traffic Prediction: A Novel Method Based on Causality and Graph Attention Network," IEEE Wireless Communications and Networking Conference (WCNC), 2023, pp. 1-7, doi:10.1109/WCNC55385.2023.10118616. [13] Y. Yao, B. Gu, Z. Su and M. Guizani, "MVSTGN: A Multi-View Spatial-Temporal Graph Network for Cellular Traffic Prediction," in IEEE Transactions on Mobile Computing, vol. 22, no. 5, pp. 2837-2849, 1 May 2023, doi: 10.1109/TMC.2021.3129796.