基于异构计算加速的开源5G架构

doi:10.13190/j.jbupt.2021-092

北京邮电大学学报 ›› 2022, Vol. 45 ›› Issue (1): 63-68.doi: 10.13190/j.jbupt.2021-092

基于异构计算加速的开源5G架构

朱子珅, 章晨宇, 王鲁晗, 路兆铭, 温向明

1. 北京邮电大学信息与通信工程学院, 北京 100876;
2. 北京邮电大学网络体系构建与融合北京市重点实验室, 北京 100876;
3. 北京邮电大学先进信息网络北京实验室, 北京 100876

收稿日期:2021-05-17 出版日期:2022-02-28 发布日期:2021-12-16
通讯作者: 王鲁晗(1989—),男,讲师,邮箱:wluhan@bupt.edu.cn E-mail:wluhan@bupt.edu.cn
作者简介:朱子珅(1998—),男,硕士生
基金资助:
国家重点研发计划项目（2019YFB1803303）；中央高校基本科研业务费项目（2020RC04）

Open Source 5G Architecture Based on Heterogeneous Computing Acceleration

ZHU Zishen, ZHANG Chenyu, WANG Luhan, LU Zhaoming, WEN Xiangming

1. School of Information and Communications Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Beijing Key Laboratory of Network System Architecture and Convergence, Beijing University of Posts and Telecommunications, Beijing 100876, China;
3. Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China

Received:2021-05-17 Online:2022-02-28 Published:2021-12-16

摘要/Abstract

摘要： 针对通用处理器下开源第5代移动通信系统（5G）处理性能差的问题，研究了开源5G的处理特性并提出了基于异构计算加速的架构。该架构将空口协议栈上层与物理层进行分离，并模块化开源5G物理层功能。在保留通用处理器对于上层协议栈流程处理优势的基础上，引入异构硬件对计算资源需求大的物理层功能模块进行加速。提出异构物理层的概念并设计了计算资源适应性分配的算法。其次，首次实现了基于OpenAirInterface平台的异构加速开源5G软件化平台。对信道编解码和调制模块进行实际测试的结果显示，相比于仅在通用处理器下测试结果，异构计算加速架构能够显著地提升物理层模块的处理性能。

关键词: 移动通信, 开源第5代移动通信系统, 异构计算加速, 异构物理层

Abstract: Aiming at the problem of poor processing performance of open source of the fifth generation of mobile communications system(5G) under general processor, the characteristics of open source 5G are studied and the architecture of heterogeneous computing acceleration is proposed. This proposed architecture separates the upper layer and the physical layer of the air interface protocol stack, and modularizes physical layer functions of the open source 5G. Utilizing the advantages of general processor on processing upper protocol stack, heterogeneous hardware is introduced to accelerate physical layer function modules, which is demanding for computing resources. The concept of heterogeneous physical layer and the algorithm of adaptive allocation of computing resources are introduced into the architecture. Then, the heterogeneous accelerated open source 5G software platform is implemented for the first time based on OpenAirInterface platform. The actual tests of channel codec and modulation modules show that the proposed heterogeneous acceleration architecture can significantly improve the processing performance of physical layer compared with general purpose processor architecture.

Key words: mobile communication, open source the fifth generation of mobile communications system architecture, heterogeneous computing acceleration, heterogeneous physical layer

中图分类号:

TN929.5

朱子珅, 章晨宇, 王鲁晗, 路兆铭, 温向明. 基于异构计算加速的开源5G架构[J]. 北京邮电大学学报, 2022, 45(1): 63-68.

ZHU Zishen, ZHANG Chenyu, WANG Luhan, LU Zhaoming, WEN Xiangming. Open Source 5G Architecture Based on Heterogeneous Computing Acceleration[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(1): 63-68.

参考文献

[1] HAN W T, KNOPP R. Open AirInterface:a pipeline structure for 5G[C]//2018 IEEE 23rd International Conference on Digital Signal Processing. Shanghai:IEEE Press, 2018:1-4.
[2] ALAWIEH M, KASPAREK M, FRANKE N, et al. A high performance FPGA-GPU-CPU platform for a real-time locating system[C]//2015 23rd European Signal Processing Conference. Nice:IEEE Press, 2015:1576-1580.
[3] GAVRILOVSKA L, RAKOVIC V, ICHKOV A, et al. Flexible C-RAN:radio technology for 5G[C]//2017 13th International Conference on Advanced Technologies, Systems and Services in Telecommunications. Nis:IEEE Press, 2017:255-264.
[4] MAI S H, LIU Y Q. Implementation of web AR applications with fog radio access networks based on openairinterface platform[C]//2019 5th International Confe-rence on Control, Automation and Robotics. Beijing:IEEE Press, 2019:639-643.
[5] REKOPUTRA N M, HARWAHYU R, SARI R F. Perfor-mance study of OpenAirInterface 5G system on the cloud platform managed by juju orchestration[C]//2019 International Seminar on Research of Information Technology and Intelligent Systems. Yogyakarta:IEEE Press, 2019:211-216.
[6] NAHUM C V, DE NÓVOA MARTINS PINTO L, TAVARES V B, et al. Testbed for 5G connected artificial intelligence on virtualized networks[J]. IEEE Access, 2020, 8:223202-223213.
[7] MANCO J, BAÑOS G G, HÄRRI J, et al. Prototyping V2X applications in large-scale scenarios using OpenAirInterface[C]//2020 IEEE Vehicular Networking Confe-rence. New York:IEEE Press, 2020:1-4.
[8] FIşNE A, ÖZSOY A. Grafik processor accelerated real time software defined radio applications[C]//2017 25th Signal Processing and Communications Applications Conference. Antalya:IEEE Press, 2017:1-4.
[9] HUSSAIN T, KHAN M, REHMAN M U, et al. A high performance software defined radio system architecture and development environment for a wide range of applications[C]//2018 International Conference on Computing, Mathematics and Engineering Technologies. Sukkur:IEEE Press, 2018:1-5.
[10] FANG X, CHEN S M. The design and algorithm mapping of a heterogeneous multi-core processor for SDR[C]//APCCAS 2008-2008 IEEE Asia Pacific Conference on Circuits and Systems. Macao:IEEE Press, 2008:1086-1089.
[11] MAROJEVIC V, REVES X, GELONCH A. Computing resource management for SDR platforms[C]//IEEE 16th International Symposium on Personal, Indoor and Mobile Radio Communications. Berlin:IEEE Press, 2005:685-689.
[12] MAROJEVIC V, BALLESTE X R, GELONCH A. A computing resource management framework for software-defined radios[J]. IEEE Transactions on Computers, 2008, 57(10):1399-1412.

[1]	陈发堂, 韩才君, 张航, 李欣超. 一种时空混合的位置管理方案[J]. 北京邮电大学学报, 2023, 46(3): 67-72.
[2]	章晨宇温向明陈亚文. 蜂窝移动网络低成本高精度授时[J]. 北京邮电大学学报, 2023, 46(1): 103-108.
[3]	廖勇李雪杨植景. 人工智能在6G空口物理层的潜在应用研究进展[J]. 北京邮电大学学报, 2022, 45(6): 22-31.
[4]	牛凯吴泊霖戴金晟王森袁弋非. 面向6G系统的极化编码调制[J]. 北京邮电大学学报, 2022, 45(6): 1-11.
[5]	刘宇, 温向明, 王鲁晗, 路兆铭, 杜科良. 基于VPP的高性能UPF原型[J]. 北京邮电大学学报, 2021, 44(6): 89-95.
[6]	闫实, 彭木根, 王文博. 通信-感知-计算融合：6G愿景与关键技术[J]. 北京邮电大学学报, 2021, 44(4): 1-11.
[7]	亢伟民, 杨鸿文. 星座成形在超奈奎斯特传输中的应用[J]. 北京邮电大学学报, 2021, 44(1): 32-37.
[8]	汪萧萧, 淡振雷, 顾晨伟, 朱卫平, 林敏. 多天线无人机通信系统中的安全波束成形方案[J]. 北京邮电大学学报, 2020, 43(6): 103-109.
[9]	何建华, 赵辉, 徐晓斌, 闫蕾, 王尚广. 基于改进双层LT码的天基物联网数据收集方法[J]. 北京邮电大学学报, 2020, 43(6): 118-125.
[10]	崔春风, 王森, 李可, 董静, 郑智民. 6G愿景、业务及网络关键性能指标[J]. 北京邮电大学学报, 2020, 43(6): 10-17.
[11]	马璐, 刘铭, 李超, 路兆铭, 马欢. 面向6G边缘网络的云边协同计算任务调度算法[J]. 北京邮电大学学报, 2020, 43(6): 66-73.
[12]	张尧祯, 刘昱. 基于SiGe工艺的28 GHz变压器匹配差分 Cascode功率放大器[J]. 北京邮电大学学报, 2020, 43(6): 36-41.
[13]	管婉青, 张海君, 路兆铭. 基于DRL的6G多租户网络切片智能资源分配算法[J]. 北京邮电大学学报, 2020, 43(6): 132-139.
[14]	张平, 许晓东, 韩书君, 牛凯, 许文俊, 兰岳恒. 智简无线网络赋能行业应用[J]. 北京邮电大学学报, 2020, 43(6): 1-9.
[15]	石潇竹, 杜洋帆, 李晓辉, 方聪. 低空小目标检测中的单通道干扰信号重构和抑制算法[J]. 北京邮电大学学报, 2020, 43(5): 143-148.

基于异构计算加速的开源5G架构

Open Source 5G Architecture Based on Heterogeneous Computing Acceleration

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价