6G空口人工智能技术专利分析

摘要/Abstract

摘要： 6G空口人工智能技术作为实现6G网络智能内生的关键方向,近年来受到广泛关注。通过分析该技术的专利申请趋势、全球专利布局及主要申请人情况,聚焦信道估计、调制与解调、信道编解码和波束管理4个热点方向,探索技术的演进过程,归纳提出未来需重点关注模型的复杂度、平衡实时性与算力调度、复杂场景下的人工智能融合技术。

关键词: 6G, 人工智能, 空口技术, 信道估计, 调制与解调, 信道编解码, 波束管理

Abstract: 6G air interface artificial intelligence （AI） technology,as a key direction for realizing the endogenous intelligence of 6G networks,has received widespread attention in recent years. The patent application trends,global patent layout,and the profiles of major applicants in the technology domain are analyzed,with a focus on four hot topics：channel estimation,modulation and demodulation,channel encoding and decoding,and beam management. This paper explores the evolution process of the technology,and summarizes and proposes that future focus should be placed on three aspects: complexity of models,balancing real-time performance and computing power scheduling,and AI integration technology in complex scenarios.

Key words: 6G, AI, air interface technology, channel estimation, modulation and demodulation, channel encoding and decoding, beam management

中图分类号:

贺雪莹, 杨雪. 6G空口人工智能技术专利分析[J]. 天津科技, 2025, 52(09): 40-44+50.

HE Xueying, YANG Xue. Patent analysis of 6G air interface artificial intelligence technology[J]. TIANJIN SCIENCE & TECHNOLOGY, 2025, 52(09): 40-44+50.

参考文献

[1] 展望6G！《6G总体愿景与潜在关键技术白皮书》正式发布[EB/OL]. (2021-06-06)[2025-07-02]. https://baijiahao.baidu.com/s?id=1701791060434144066&wfr=spider& for=pc.
[2] 3GPP 6G发展规划:关键时间节点与详细安排[EB/OL]. (2025-02-28)[2025-07-02]. https://baijiahao.baidu.com/s?id=1825384878368687530&wfr=spider&for=pc.
[3] HAN S F,XIE T,CHIH-LIN I,et al.Artificial-intelligence-enabled air interface for 6G:Solutions,challenges,and standardization impacts[J]. IEEE Communications Magazine,2020,58(10):73-79.
[4] 杨刚华,何高宁,陈睿荣,等. 6G无线空口传输技术研究进展与展望[J]. 中国科学:信息科学,2024,54(5):1078-1113.
[5] 周斌. 6G关键驱动与研究挑战[J]. 互联网经济,2020(5):26-35.
[6] 工业和信息化部. “十四五”信息通信行业发展规划[Z/OL]. (2021-11-01)[2025-07-02]. https://www.gov.cn/zhengce/zhengceku/2021-11/16/5651262/files/96989dadf83a4302 895cd17cbeec6600.pdf.
[7] 朱晓丹,黄庆秋. AI赋能6G无线接入网技术研究[J]. 科技创新与应用,2023,13(30):14-16,21.
[8] LIU Y S,TAN Z H,HU H J,et al.Channel estimation for OFDM[J]. IEEE Communications Surveys & Tutorials,2014,16(4):1891-1908.
[9] 李露,李福昌,张犇,等. 基于AI的无线空口技术研究[J]. 邮电设计技术,2024(7):47-51.
[10] WANG Z Q,DU Y,WEI K J,et al.Vision,application scenarios,and key technology trends for 6G mobile communications[J]. Science China Information Sciences,2022,65(5):151301.
[11] CHUN C J,KANG J M,KIM I M.Deep learning-based channel estimation for massive MIMO systems[J]. IEEE Wireless Communications Letters,2019,8(4):1228-1231.
[12] XUE W L,ZHU H W,NIAN Z Y,et al.Channel estimation algorithm based on multi-modal neural networ[C]// 2024 9th International Conference on Signal and Image Processing (ICSIP). Nanjing,China:IEEE,2024:206-210.
[13] INC. D. Processing communications signals using a machine-learning network:US202016856760[P].2020-10-29.
[14] 华中科技大学. 一种基于空间随机采样的智能超表面信道估计方法及系统:202110155767.1[P].2021-06-04.
[15] OY N A S N. Machine learning based channel estimation for an antenna array:US202118696159[P].2024-11-28.
[16] CORPORATION I. Federated learning across UE and RAN:US202117504711[P].2024-09-03.
[17] CO. T H,LTD. Neural network training method and related apparatus:WO2020142103[P].2022-07-07.
[18] 重庆邮电大学. 一种基于轻量级高效LEU-Net的1-比特大规模MIMO信道估计方法:202310563788.6[P].2023-10-20.
[19] 北京航空航天大学,北京遥感设备研究所. 一种临近空间平台到无人机下行多径信道估计方法:202410338643.0[P].2024-04-26.
[20] IMT-2030(6G)推进组. 无线人工智能(AI)技术研究报告[R/OL]. (2022-10-13)[2025-07-02]. https://zhuanlan.zhihu.com/p/573341251.
[21] 诺基亚技术有限公司. 发射器:202110050334.X[P].2021-07-16.
[22] 北京邮电大学. 一种标准调制格式和概率整形调制格式的识别方法:202111636762.7[P].2022-04-12.
[23] NEWMAN E D,MASSENGILL K R. AI-assisted selection of demodulation reference type in 5G and 6G:US202318121779[P].2023-07-20.
[24] 电子科技大学. 一种用于神经网络调制器的抗非理想特性的星座成型方法:202410499088.X[P].2024-08-02.
[25] 查雄,彭华,秦鑫,等. 基于循环神经网络的卫星幅相信号调制识别与解调算法[J]. 电子学报,2019,47(11):2443-2448.
[26] INC. E L. Encoding method and neural network encoder structure usable in wireless communication system:US202018025227[P].2023-10-12.
[27] 高通股份有限公司. 使用神经网络的低开销和高准确性信道状态反馈:202180062161.1[P].2023-05-02.
[28] CO. E S,LTD. ?? ???? ?? ? ?? ?? ??:KR20210051857[P].2022-10-28.
[29] 维沃软件技术有限公司. 信息处理方法、信息处理装置、终端及网络侧设备:202310457754.9[P].2024-10-25.
[30] 大唐移动通信设备有限公司. CSI反馈方法、机器学习模型的训练方法及装置:202310961522.7[P].2025-02-14.
[31] 廖勇,杨植景,李雪. 人工智能在6G空口物理层的潜在应用[J]. 北京邮电大学学报,2022,45(6):21-30.
[32] CORPORATION I. Selection of beamforming directions based on learned performance:WO2016US44735[P].2017-10-26.
[33] 北京小米移动软件有限公司. 传输信息的方法、装置、用户设备及基站:201980000505.9[P].2019-08-02.
[34] 维沃移动通信有限公司. 波束上报方法、波束信息确定方法及相关设备:202011126498.8[P].2022-04-22.
[35] INC. E L,FOUNDATION R U S. Artificial Intelligence-based Beam Management Method in Wireless Communication System and Apparatus Therefor:WO2020KR17558[P].2021-06-10.
[36] 东南大学. 基于波束索引地图的免信道训练大维通信波束对齐方法:202110095713.0[P].2021-05-28.
[37] 重庆邮电大学. 一种基于星地融合架构下低轨卫星网络跳波束优化方法:202210301225.5[P].2022-06-24.
[38] 电子科技大学. 基于强化学习的面向车辆网边缘计算的波束赋形方法:202310814626.5[P].2023-09-22.
[39] 中国电子科技集团有限公司电子科学研究院,北京邮电大学. 基于深度强化学习的RIS联合波束赋形方法及通信系统:202310985821.4[P].2023-12-05.
[40] 北京小米移动软件有限公司. 通信方法、设备和存储介质:202410052316.9[P].2024-02-20.