基于可解释机器学习模型的5G下行空口关键性能研究

林劼

doi:10.11959/j.issn.1000-0801.2025122

您当前的位置：

首页 >

文章列表页 >

基于可解释机器学习模型的5G下行空口关键性能研究

工程与应用 | 更新时间：2025-06-03

- 基于可解释机器学习模型的5G下行空口关键性能研究
- Study on the key performance of 5G downlink air interface based on an explainable machine learning model
- 电信科学 2025年41卷第5期页码：178-185
- 作者机构：
  
  中国电信股份有限公司福州分公司，福建福州 350001
- 作者简介：
  
  [ "林劼（1998- ），男，中国电信股份有限公司福州分公司工程师，主要研究方向为无线网络优化。" ]
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2025122
  中图分类号： TN915
- 收稿：2025-01-09，
  
  修回：2025-03-09，
  
  纸质出版：2025-05-20
- 稿件说明：
移动端阅览
林劼.基于可解释机器学习模型的5G下行空口关键性能研究[J].电信科学,2025,41(05):178-185.

LIN Jie.Study on the key performance of 5G downlink air interface based on an explainable machine learning model[J].Telecommunications Science,2025,41(05):178-185.
林劼.基于可解释机器学习模型的5G下行空口关键性能研究[J].电信科学,2025,41(05):178-185. DOI： 10.11959/j.issn.1000-0801.2025122.

LIN Jie.Study on the key performance of 5G downlink air interface based on an explainable machine learning model[J].Telecommunications Science,2025,41(05):178-185. DOI： 10.11959/j.issn.1000-0801.2025122.

摘要

小区下行有效吞吐率和用户面时延是5G空口的重要性能指标。为研究并解释其背后的影响因素，使用局部线性森林结合Shapley值法进行预测和解释。实验结果显示，相比传统方法，该方法具有较高的预测准确性。在此基础上通过计算各影响因素对结果的贡献并分析其影响变化趋势，找出问题小区对应的问题因素。最后，通过贡献重要性排序，推演重要因素改变对结果的影响，实现更加全面理解和分析性能指标的目的。

Abstract

The cell downlink effective throughput and user plane latency are important performance indicators of the 5G system’s air interface. To study the impact factors behind them

the integration of Shapley value with local linear forest model was used to make predictions with explanations. The experiment results show that compared to traditional method

this method achieves a better predict precision. Based on this

by calculating each factor’s contribution towards the results and analyzing the corresponding trends

a problematic factor for each problematic cell was found. Finally

by sorting the importance of each factor’s contribution

changes in results were inferred by changes in important factors

thus more comprehensive understanding and analysis of the performance indicators were achieved.

关键词

Keywords

references

工业和信息化部 . 2024年前11个月通信业经济运行情况 [Z ] . 2024 .

MIIT . Economic operation of the telecommunications industry in the first 11 months of 2024 [Z ] . 2024

FRIEDBERG R , TIBSHIRANI J , ATHEY S , et al . Local linear forests [J ] . Journal of Computational and Graphical Statistics , 2021 , 30 ( 2 ): 503 - 517 .

AAS K , JULLUM M , LØLAND A . Explaining individual predictions when features are dependent: more accurate approximations to Shapley values [J ] . Artificial Intelligence , 2021 , 298 : 103502 .

汤建东 , 肖清华 . 5G容量能力分析 [J ] . 邮电设计技术 , 2020 ( 3 ): 50 - 53 .

TANG J D , XIAO Q H . Analysis of 5G system capacity ability [J ] . Designing Techniques of Posts and Telecommunications , 2020 ( 3 ): 50 - 53 .

魏垚 , 熊尚坤 . 5G eMBB空口时延分析与增强技术研究 [J ] . 移动通信 , 2019 , 43 ( 9 ): 42 - 46 .

WEI Y , XIONG S K . Analysis on 5G eMBB uu interface latency and study on enhancement technique [J ] . Mobile Communications , 2019 , 43 ( 9 ): 42 - 46 .

王瑞 , 乔辉 , 高超 , 等 . 5G多频网用户速率优化研究 [J ] . 现代信息科技 , 2025 , 9 ( 1 ): 8 - 12 .

WANG R , QIAO H , GAO C , et al . Research on user rate optimization of 5G multi-frequency network [J ] . Modern Information Technology , 2025 , 9 ( 1 ): 8 - 12 .

CHOI J H , CHOI Y , LEE K S , et al . Explainable model using Shapley additive explanations approach on wound infection after wide soft tissue sarcoma resection: “big data” analysis based on health insurance review and assessment service hub [J ] . Medicina , 2024 , 60 ( 2 ): 327 .

HE Z Y , YANG Y J , FANG R Z , et al . Integration of Shapley additive explanations with random forest model for quantitative precipitation estimation of mesoscale convective systems [J ] . Frontiers in Environmental Science , 2023 , 10 : 1057081 .

CILÍNIO M , PEREIRA M , DUARTE D , et al . Explainable fault analysis in mobile networks: a SHAP-based supervised clustering approach [C ] // Proceedings of the 2023 16th International Conference on Signal Processing and Communication System (ICSPCS) . Piscataway : IEEE Press , 2023 : 1 - 9 .

3GPP. Physical layer procedures for data(Release 15): TS 38.214 [S ] . 2019 . .

BREIMAN L . Random forests [J ] . Machine Learning , 2019 .

TIBSHIRANI J , ATHEY S , FRIEDBERG R , et al . Generalized random forests [EB ] . 2024 .

SHAPLEY LS . A value for n-person games [M ] // Contributions to the theory of games Ⅱ . Princeton : Princeton University Press , 1953 : 307 - 317 .

JULLUM M , OLSEN L H B , REDELMEIER A , et al . Prediction explanation with dependence-aware Shapley values [EB ] . 2024 .

3GPP. Study on self evaluation towards I MT-2020 submission(Release 16): TS 37910 [S ] . 2019 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

SWTA-Shapley：一种高效的联邦学习贡献评估方法

云计算中基于Shapley值改进遗传算法的虚拟机调度模型