大语言模型辅助的室内多源定位智能调度与连续性控制方法

李成

doi:10.11959/j.issn.1000-0801.DXKX260578

Chinese

您当前的位置：

首页 >

文章列表页 >

大语言模型辅助的室内多源定位智能调度与连续性控制方法

研究与开发 | 更新时间：2026-09-30

- 大语言模型辅助的室内多源定位智能调度与连续性控制方法
- Large-language-model-assisted intelligent scheduling and continuity control for indoor multi-source positioning——Li Cheng（China Telecommunications Corporation，Beijing，Postal Code：100033） Niu Rui（China Telecom Unmanned Technology (Jiangsu) Co., Ltd.，Nanjing，Postal Code：210019） Shaozhen（China Telecom Co., Ltd. Shanghai Research Institute，Shanghai，Postal Code：201315）
- 电信科学 2026年
- 作者机构：
  
  中国电信集团有限公司本部、北京、邮编：100033）牛锐（中电信无人科技（江苏）有限公司、南京、邮编：210019）邵震（中国电信股份有限公司上海研究院、上海、邮编：201315
- 作者简介：
  
  李成（1992年-），男，博士研究生，工程师，主要研究方向为5G-A/6G、卫星通信、融合定位。
- 基金信息：
  
  无线通感一体化可靠感知理论及关键技术研究(U25B2040)
- DOI：10.11959/j.issn.1000-0801.DXKX260578
  中图分类号： TN929.5
- 收稿：2026-08-22，
  
  修回：2026-09-14，
  
  录用：2026-09-20，
- 稿件说明：
移动端阅览
李成. 大语言模型辅助的室内多源定位智能调度与连续性控制方法[J/OL]. 电信科学, 2026.

李成. Large-language-model-assisted intelligent scheduling and continuity control for indoor multi-source positioning——Li Cheng（China Telecommunications Corporation，Beijing，Postal Code：100033） Niu Rui（China Telecom Unmanned Technology (Jiangsu) Co., Ltd.，Nanjing，Postal Code：210019） Shaozhen（China Telecom Co., Ltd. Shanghai Research Institute，Shanghai，Postal Code：201315）[J/OL]. Telecommunications Science, 2026.
李成. 大语言模型辅助的室内多源定位智能调度与连续性控制方法[J/OL]. 电信科学, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260578.

李成. Large-language-model-assisted intelligent scheduling and continuity control for indoor multi-source positioning——Li Cheng（China Telecommunications Corporation，Beijing，Postal Code：100033） Niu Rui（China Telecom Unmanned Technology (Jiangsu) Co., Ltd.，Nanjing，Postal Code：210019） Shaozhen（China Telecom Co., Ltd. Shanghai Research Institute，Shanghai，Postal Code：201315）[J/OL]. Telecommunications Science, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260578.

摘要

针对大型复杂室内场景多技术协同难、跨场景切换滞后、定位精度不稳定和服务连续性不足等问题，参考3GPP（3rd Generation Partnership Project

第三代合作伙伴计划）定位框架并进行平台级扩展，提出大语言模型辅助的室内多源定位智能调度与连续性控制方法。方法以蜂窝、Wi-Fi等测量信息为主，并可扩展适配蓝牙/UWB（Ultra Wide Band，超宽带）和惯性传感等多源数据，通过时空对齐、信号语义编码、场景识别、约束化调度和融合解算实现动态源选择。基于40万平方米商业综合体8个楼层实测数据验证，结果表明：场景识别加权准确率达到94.3%，开阔无遮挡LOS（Line-of-Sight，视距）中庭场景定位均方根误差为0.75 m，跨场景定位连续性达到88%，平均定位误差较固定权重融合方法降低49.5%。本方法可提升复杂室内场景下定位服务的连续性、稳定性和跨场景适配能力。

Abstract

To address the difficulties of multi-source positioning coordination

delayed cross-scenario handover

and insufficient service continuity in complex indoor environments such as large commercial complexes

a large language model assisted scheduling and continuity control method for indoor multi-source positioning was proposed inspired by the 3GPP positioning framework and extended at the platform level. Cellular and Wi-Fi measurements were primarily aggregated on the location service platform

with extensible adaptation to Bluetooth/UWB and inertial sensing data. Dynamic source selection was achieved through spatio-temporal alignment

signal semantic encoding

scenario recognition

constrained scheduling

and fusion calculation. The method was evaluated with field data collected from eight floors of a 400 000 m² commercial complex. The results showed that the weighted scenario recognition accuracy reached 94.3%

the root mean square error in open LOS atrium scenarios was 0.75 m

the cross-scenario positioning continuity reached 88%

and the average positioning error was reduced by 49.5% compared with the fixed-weight fusion method. The proposed method improves the continuity

stability

and cross-scenario adaptability of indoor positioning services.

关键词

Keywords

references

LIU H , DARABI H , BANERJEE P , et al . Survey of wireless indoor positioning techniques and systems [J ] . IEEE Transactions on Systems, Man, and Cybernetics , Part C, 2007 , 37 ( 6 ): 1067 - 1080 .

DARDARI D , CLOSAS P , DJURIC P M . Indoor tracking: theory, methods, and technologies [J ] . IEEE Transactions on Vehicular Technology , 2015 , 64 ( 4 ): 1263 - 1278 .

ZAFARI F , GKELIAS A , LEUNG K K . A survey of indoor localization systems and technologies [J ] . IEEE Communications Surveys & Tutorials , 2019 , 21 ( 3 ): 2568 - 2599 .

3GPP. 3GPP TS 23.273 V19.0.0 5G System (5GS) location services (LCS); stage 2 [S ] . 2025 .

3GPP. 3GPP TS 38.305 V19.0.0 NG Radio Access Network (NG-RAN); stage 2 functional specification of user equipment positioning in NG-RAN [S ] . 2025 .

3GPP. 3GPP TS 37.355 V19.0.0 LTE positioning protocol (LPP) [S ] . 2025 .

3GPP. 3GPP TS 38.215 V19.0.0 NR; physical layer measurements [S ] . 2025 .

BREIMAN L . Random forests [J ] . Machine Learning , 2001 , 45 ( 1 ): 5 - 32 .

CHEN T , GUESTRIN C . XGBoost: a scalable tree boosting system [C ] // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . New York : ACM , 2016 : 785 - 794 .

VASWANI A , SHAZEER N , PARMAR N , et al . Attention is all you need [C ] // Advances in Neural Information Processing Systems 30 . Red Hook : Curran Associates , 2017 : 5998 - 6008 .

DEVLIN J , CHANG M W , LEE K , et al . BERT: pre-training of deep bidirectional transformers for language understanding [C ] // Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics . Minneapolis : ACL , 2019 : 4171 - 4186 .

WEI J , WANG X , SCHUURMANS D , et al . Chain-of-thought prompting elicits reasoning in large language models [C ] // Advances in Neural Information Processing Systems 35 . Red Hook : Curran Associates , 2022 : 24824 - 24837 .

HU E J , SHEN Y , WALLIS P , et al . LoRA: low-rank adaptation of large language models [C ] // International Conference on Learning Representations . 2022 .

DETTMERS T , PAGNONI A , HOLTZMAN A , et al . QLoRA: efficient finetuning of quantized LLMs [C ] // Advances in Neural Information Processing Systems 36 . Red Hook : Curran Associates , 2023 : 10088 - 10115 .

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

基于冻结主干双回路进化的多智能体潜空间策略研究

企业AI智能体基础设施的层次化架构：基于云计算实践的分析框架

基于大语言模型与事件演化图谱的诈骗电话识别方法

基于LLM-RFE-XGBoost方法的客户升级投诉预警研究

基于预训练大模型的业务流程实体与关系的自动提取和评估