[1]佟谣,刘波,齐小刚.基于参数优化VMD和改进BiLSTM的低轨卫星网络业务预测方法[J].智能系统学报,2026,21(3):627-638.[doi:10.11992/tis.202508026]
TONG Yao,LIU Bo,QI Xiaogang.A traffic prediction method for a low earth orbit satellite network based on parameter-optimized VMD and improved BiLSTM[J].CAAI Transactions on Intelligent Systems,2026,21(3):627-638.[doi:10.11992/tis.202508026]
点击复制
TONG Yao,LIU Bo,QI Xiaogang.A traffic prediction method for a low earth orbit satellite network based on parameter-optimized VMD and improved BiLSTM[J].CAAI Transactions on Intelligent Systems,2026,21(3):627-638.[doi:10.11992/tis.202508026]
基于参数优化VMD和改进BiLSTM的低轨卫星网络业务预测方法
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
21
期数:
2026年第3期
页码:
627-638
栏目:
学术论文—机器学习
出版日期:
2026-05-05
- Title:
- A traffic prediction method for a low earth orbit satellite network based on parameter-optimized VMD and improved BiLSTM
- Keywords:
- low earth orbit satellite network; service traffic; traffic prediction; machine learning; variational mode decomposition; improved sparrow search algorithm; self-attention mechanism; bidirectional long short-term memory network
- 分类号:
- TP181
- 文献标志码:
- 2026-3-6
- 摘要:
- 低地球轨道卫星业务预测对于缓解拥塞问题和改善资源调度至关重要。为进一步提高业务预测精度，提出了基于参数优化VMD(variational mode decomposition)和改进BiLSTM的预测方法VPI-TPIB。该方法包含两个核心模型：基于改进麻雀搜索算法参数优化的变分模态分解(variational mode decomposition based on parameter optimization of improved sparrow search algorithm, VPI)模型和基于改进双向长短期记忆网络的业务预测(traffic prediction based on improved bidirectional long short-term memory network, TPIB)模型。在VPI模型中，采用引入Tent混沌映射和高斯变异改进后的麻雀搜索算法对VMD关键参数的选择进行优化，从而有效提升数据分解效果；在TPIB模型中，通过引入自注意力机制改进BiLSTM，实现对分解数据的动态特征权重分配和双向时序建模，提高模型预测精度。实验结果表明，与基准LSTM模型相比，所提方法在两个真实数据集上的MAE分别降低了42.64%和81.59%。
- Abstract:
- Traffic prediction in low earth orbit satellite networks is critical for mitigating congestion and optimizing resource allocation. To further enhance prediction accuracy, a prediction method based on parameter-optimized variational mode decomposition (VMD) and an improved bidirectional long short-term memory (BiLSTM) network is proposed. The method comprises two core models: a parameter-optimized VDM model based on an improved sparrow search algorithm (VPI) and a traffic prediction model based on an improved BiLSTM network (TPIB). In the VPI model, an improved sparrow search algorithm incorporating Tent chaotic mapping and Gaussian mutation is adopted to optimize key parameters for VMD, thereby improving decomposition performance. In the TPIB model, a self-attention mechanism is introduced to enhance BiLSTM, enabling dynamic feature weight allocation for decomposed data and bidirectional temporal modeling, thereby improving prediction accuracy. Experimental results show that, compared with the baseline LSTM model, the proposed method reduces the mean absolute error (MAE) by 42.64% and 81.59% on the two real-world datasets, respectively.
- 参考文献/References:
-
[1] MAO Bomin, ZHOU Xueming, LIU Jiajia, et al. On an intelligent hierarchical routing strategy for ultra-dense free space optical low earth orbit satellite networks[J]. IEEE journal on selected areas in communications, 2024, 42(5): 1219-1230
[2] HUI Mingming, ZHAI Shenghua, WANG Daqing, et al. A review of LEO-satellite communication payloads for integrated communication, navigation, and remote sensing: opportunities, challenges, future directions[J]. IEEE internet of things journal, 2025, 12(12): 18954-18992
[3] PRIYA J C, NANTHAKUMAR G, CHOUDHURY T, et al. 6G-DeFLI: enhanced quality-of-services using distributed hash table and blockchain-enabled federated learning approach in 6G IoT networks[J]. Wireless networks, 2025, 31(1): 361-375
[4] CHEN Chen, SUN Chengxin, LI Huimin, et al. ST-GAGCN-LEO: a spatiotemporal graph attention and gated convolutional network for LEO satellite traffic prediction[J]. IEEE transactions on aerospace and electronic systems, 2025, 61(4): 9669-9685
[5] 朱圳, 刘立芳, 齐小刚. 基于数据挖掘的通信网络故障分类研究[J]. 智能系统学报, 2022, 17(6): 1228-1234 ZHU Zhen, LIU Lifang, QI Xiaogang. Research on communication network fault classification based on data mining[J]. CAAI transactions on intelligent systems, 2022, 17(6): 1228-1234
[6] JIANG Bin, YAN Yingchun, YOU Li, et al. Robust secure transmission for satellite communications[J]. IEEE transactions on aerospace and electronic systems, 2023, 59(2): 1598-1612
[7] ZHANG Yan, WANG Yong, CAO Haotong, et al. Self-similar traffic prediction for LEO satellite networks based on LSTM[J]. IET communications, 2025, 19(1): e12863
[8] 张定华, 胡祎波, 曹国彦, 等. 面向工业网络通信安全的数据流特征分析[J]. 西北工业大学学报, 2020(1): 199-208 ZHANG Dinghua, HU Yibo, CAO Guoyan, et al. Dataflow feature analysis for industrial networks communication security[J]. Journal of Northwestern Polytechnical University, 2020(1): 199-208
[9] XIAO Zhenyu, YANG Junyi, MAO Tianqi, et al. LEO satellite access network (LEO-SAN) toward 6G: challenges and approaches[J]. IEEE wireless communications, 2024, 31(2): 89-96
[10] PENG Liang, YAN Jie, WEI Peng, et al. Spatio-temporal correlation-based incomplete time-series traffic prediction for LEO satellite networks[J]. Frontiers of information technology & electronic engineering, 2025, 26(5): 788-804
[11] TIAN Miao, SUN Chen, WU Shaozhi. An EMD and ARMA-based network traffic prediction approach in SDN-based Internet of vehicles[J]. Wireless networks, 2021: 1-13.
[12] 冯强, 赵建光, 杨茸, 等. 时间序列中非平稳性和波动性的建模及预测[J]. 计算机科学与探索, 2025, 19(5): 1313-1321 FENG Qiang, ZHAO Jianguang, YANG Rong, et al. Modeling and predicting time series with non-stationarity and volatility[J]. Journal of frontiers of computer science and technology, 2025, 19(5): 1313-1321
[13] BI Jing, ZHANG Xiang, YUAN Haitao, et al. A hybrid prediction method for realistic network traffic with temporal convolutional network and LSTM[J]. IEEE transactions on automation science and engineering, 2022, 19(3): 1869-1879
[14] KAVEHMADAVANI F, NGUYEN V D, VU T X, et al. Intelligent traffic steering in beyond 5G open RAN based on LSTM traffic prediction[J]. IEEE transactions on wireless communications, 2023, 22(11): 7727-7742
[15] LI Guohao, TIAN Zhongda. A new method of network traffic prediction based on combination model[J]. Peer-to-peer networking and applications, 2024, 17(3): 1075-1090
[16] HAN Chi, XIONG Wei, YU Ronghuan, et al. LEO satellite network traffic forecasting based on ARIMA-BPNN model[C]//2023 3rd International Conference on Electronic Information Engineering and Computer Communication. Piscataway: IEEE, 2024: 402-406.
[17] DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544
[18] MA Changxi, HU Yanming, XU Xuecai. Hybrid deep learning model with VMD-BiLSTM-GRU networks for short-term traffic flow prediction[J]. Data science and management, 2025, 8(3): 257-269
[19] ZHANG Hong, YANG Gang, YU Hailiang, et al. Kalman filter-based CNN-BiLSTM-ATT model for traffic flow prediction[J]. Computers, materials & continua, 2023, 76(1): 1047-1063.
[20] LI Guohui, DENG Haonan, YANG Hong. Traffic flow prediction model based on improved variational mode decomposition and error correction[J]. Alexandria engineering journal, 2023, 76: 361-389
[21] FU Youfa, LIU Dan, FU Shengwei, et al. Enhanced Aquila optimizer based on tent chaotic mapping and new rules[J]. Scientific reports, 2024, 14: 3013
[22] ALAWIDA M. Enhancing logistic chaotic map for improved cryptographic security in random number generation[J]. Journal of information security and applications, 2024, 80: 103685
[23] 孟学尧, 郭倩倩, 郭海儒. 一种改进萤火虫算法的模糊聚类方法[J]. 小型微型计算机系统, 2021, 42(6): 1165-1170 MENG Xueyao, GUO Qianqian, GUO Hairu. Fuzzy clustering method based on improved firefly algorithm[J]. Journal of Chinese computer systems, 2021, 42(6): 1165-1170
[24] 回立川, 陈雪莲, 孟嗣博. 多策略混合的改进麻雀搜索算法[J]. 计算机工程与应用, 2022, 58(16): 71-83 HUI Lichuan, CHEN Xuelian, MENG Sibo. Improved sparrow search algorithm based on multi-strategy mixing[J]. Computer engineering and applications, 2022, 58(16): 71-83
[25] SUN Shuguang, WANG Zewei, WANG Jingqin, et al. Remaining useful life prediction for circuit breaker based on SM-CFE and SA-BiLSTM[J]. IEEE transactions on instrumentation and measurement, 2023, 72: 3517314
[26] GUO Jielong, LIU Meijun, LUO Peng, et al. Attention-based BILSTM for the degradation trend prediction of lithium battery[J]. Energy reports, 2023, 9: 655-664
[27] ZHU Rong, TANG Jie, HE Xuansen, et al. SE-MAConvLSTM: a deep learning framework for short-term traffic flow prediction combining squeeze-and-excitation network and multi-attention convolutional LSTM network[J]. PLoS one, 2024, 19(12): e0312601
[28] ZHANG Zhiwei, GONG Shuhui, LIU Zhaoyu, et al. A novel hybrid framework based on temporal convolution network and transformer for network traffic prediction[J]. PLoS one, 2023, 18(9): e0288935
[29] MA Xuesen, ZHENG Biao, JIANG Gonghui, et al. Cellular network traffic prediction based on correlation ConvLSTM and self-attention network[J]. IEEE communications letters, 2023, 27(7): 1909-1912
[30] 殷礼胜, 孙双晨, 魏帅康, 等. 基于自适应VMD-Attention-BiLSTM的交通流组合预测模型[J]. 电子测量与仪器学报, 2021, 35(7): 130-139 YIN Lisheng, SUN Shuangchen, WEI Shuaikang, et al. Traffic flow combination prediction model based on adaptive VMD-attention-BiLSTM[J]. Journal of electronic measurement and instrumentation, 2021, 35(7): 130-139
[31] 王瑞, 李哲, 逯静. 基于融合聚类和BKA-VMD-TCN-BiLSTM的短期光伏功率预测[J]. 电子科技大学学报, 2025, 54(4): 592-603 WANG Rui, LI Zhe, LU Jing. Short-term photovoltaic power forecasting based on fusion clustering and BKA-VMD-TCN-BiLSTM[J]. Journal of University of Electronic Science and Technology of China, 2025, 54(4): 592-603
[32] LIU Zhiguo, LI Weijie, FENG Jianxin, et al. Research on satellite network traffic prediction based on improved GRU neural network[J]. Sensors, 2022, 22(22): 8678
[33] CAO Mengxiang, LIU Jin, ZHI Jiamao, et al. TLS-net: a hybrid time series prediction model combining TCN and LSTM for ship-satellite network traffic[C]//2023 7th International Conference on Transportation Information and Safety. Piscataway: IEEE, 2023: 1168-1173.
[34] 唐文杰, 肖一磊, 孔祥宇, 等. CycleLLH: 一种基于周期性整合的新型网络流量预测模型[J]. 计算机学报, 2024, 47(12): 2867-2888 TANG Wenjie, XIAO Yilei, KONG Xiangyu, et al. CycleLLH: a new network traffic prediction model based on cycle integration[J]. Chinese journal of computers, 2024, 47(12): 2867-2888
[35] QI Liangang, WANG Enqiang, XU Tianfang, et al. Improving satellite network efficiency with terminal traffic prediction and SQP-SRA algorithm[J]. Computer communications, 2025, 242: 108293
备注/Memo
收稿日期:2025-8-22。
基金项目:国家自然科学基金项目(62372354, 62373291)；航空科学基金项目(2024Z071081003, 2024M066081001).
作者简介:佟谣，硕士研究生，主要研究方向为卫星网络资源管理。E-mail：23071213248@stu.xidian.edu.cn。;刘波，副教授，主要研究方向为软件定义网络、网络流量优化、卫星网络路由与资源分配。E-mail：lbo.xidian@163.com。;齐小刚，教授，博士生导师，主要研究方向为复杂系统建模与仿真、网络算法设计与应用。主持完成国家自然科学基金项目等30余项，登记软件著作权13项，发表学术论文150余篇。E-mail：xgqi@xidian.edu.cn。
通讯作者:齐小刚. E-mail：xgqi@xidian.edu.cn
更新日期/Last Update:
1900-01-01