用户请求感知的边端缓存与用户推荐联合优化策略

王汝言; 蒋昊; 唐桐; 吴大鹏; 钟艾玲

doi:10.11999/JEIT230898

用户请求感知的边端缓存与用户推荐联合优化策略

doi: 10.11999/JEIT230898

1.
重庆邮电大学通信与信息工程学院重庆 400065
2.
先进网络与智能互联技术重庆市高校重点实验室重庆 400065
3.
泛在感知与互联重庆市重点实验室重庆 400065

基金项目: 国家自然科学基金(62271096, U20A20157)，重庆市自然科学基金(CSTB2023NSCQ-LZX0134)，重庆市高校创新研究群体(CXQT20017)，重邮信通青创团队支持计划(SCIE-QN-2022-04)，重庆市教委科学技术研究项目(KJQN202300632)，重庆市博士后特别资助项目(2022CQBSHTB2057)，重庆市研究生科研创新项目(CYB22250)

详细信息

作者简介:
王汝言：男，教授，研究方向为泛在网络，多媒体信息处理等

蒋昊：男，硕士生，研究方向为边缘缓存

唐桐：男，讲师，研究方向为视频编码传输等

吴大鹏：男，教授，研究方向为泛在无线网络、社会计算等

钟艾玲：女，博士生，研究方向为无线网络优化等

通讯作者:
吴大鹏　wudp@cqupt.edu.cn

中图分类号: TN929.5
计量
- 文章访问数: 242
- HTML全文浏览量: 179
- PDF下载量: 36
- 被引次数: 0
出版历程
- 收稿日期: 2023-08-15
- 修回日期: 2024-01-17
- 网络出版日期: 2024-01-23
- 刊出日期: 2024-07-29

A Joint Optimization Strategy for User Request Perceived Edge Caching and User Recommendation

1.
School of Communications and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
2.
Advanced Network and Intelligent Connection Technology Key Laboratory of Chongqing Education Commission of China, Chongqing 400065, China
3.
Chongqing Key Laboratory of Ubiquitous Sensing and Networking, Chongqing 400065, China

Funds: The National Natural Science Foundation of China (62271096, U20A20157), The Natural Science Foundation of Chongqing, China (CSTB2023NSCQ-LZX0134), The University Innovation Research Group of Chongqing (CXQT20017), The Youth Innovation Group Support Program of ICE Discipline of CQUPT (SCIE-QN-2022-04), The Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN202300632), Chongqing Postdoctoral Special Funding Project (2022CQBSHTB2057), Chongqing Postgraduate Research and Innovation Project (CYB22250)

摘要

摘要: 针对当前边缘缓存场景中普遍存在的用户偏好未知与高度异质问题，该文提出一种用户请求感知的边端缓存与用户推荐联合优化策略。首先，建立点击率(CTR)预测基本模型，引入对比学习方法生成高质量的特征表示，辅助因子分解机(FM)预测用户偏好；然后，基于用户偏好设计一种动态推荐机制，重塑不同用户的内容请求概率，从而影响缓存决策；最后，以用户平均内容获取时延最小化为目标建立边端缓存与用户推荐联合优化问题，将其解耦为边端缓存子问题和用户推荐子问题，分别基于区域贪婪算法和一对一交换匹配算法求解，并通过迭代更新获得收敛优化结果。仿真结果表明，相较于基准模型，引入对比学习方法的预测模型在曲线下面积(AUC)和准确率(ACC)上分别提升1.65%和1.30%，且联合优化算法能够有效降低用户平均内容获取时延，提升系统缓存性能。
- 边缘缓存 /
- 对比学习 /
- 推荐机制 /
- 平均时延
Abstract: Considering the problem of unknown and highly heterogeneous user preference in the current edge caching scenario, a joint optimization strategy of user request perceived edge caching and user recommendation is proposed. Firstly, the basic model of Click Through Rate (CTR) prediction is established, and the contrastive learning method is introduced to generate high-quality feature representation, which could better help Factorization Machine(FM) model to predict user preference. Then, based on the predicted user preference, a dynamic recommendation mechanism is designed to reshape the content request probability of different users, thereby affecting cache decision; Finally, a joint optimization problem of edge caching and user recommendation is established with the goal of minimizing the average user content acquisition delay. It is decoupled into edge caching subproblem and user recommendation subproblem, and solved based on the region greedy algorithm and one-to-one exchange matching algorithm, respectively. The convergence optimization results are obtained through iterative update. The results show that compared with the benchmark model, the contrastive learning method has improved Area Under Curve (AUC) and ACCuracy (ACC) by 1.65% and 1.30%, respectively, and the joint optimization algorithm can effectively reduce the average content acquisition latency of users and improve the system cache performance.
- Edge caching /
- Contrastive learning /
- Recommendation mechanism /
- Average delay

HTML全文

图 1 支持缓存和推荐的无线网络架构图

下载: 全尺寸图片幻灯片

图 2 FM_CL模型框架

下载: 全尺寸图片幻灯片

图 3 FM_CL模型训练表现

下载: 全尺寸图片幻灯片

图 4 不同模型性能对比

下载: 全尺寸图片幻灯片

图 5 用户缓存容量 vs 平均时延($ {R_u} $=6个, $ {t_u} $=0.10)

下载: 全尺寸图片幻灯片

图 6 用户推荐内容数 vs 平均时延($ {C_{{\text{UE}}}} $=40 Mbit, $ {t_u} $=0.10)

下载: 全尺寸图片幻灯片

图 7 UPD容忍度 vs 平均时延($ {C_{{\text{UE}}}} $=40 Mbit, $ {R_u} $=6个)

下载: 全尺寸图片幻灯片

图 8 用户缓存容量 vs 系统缓存命中率($ {R_u} $=6个, $ {t_u} $=0.10)

下载: 全尺寸图片幻灯片

表 1 不同模型引入对比学习方法后性能对比

模型	AUC	ACC	Loss
FM	0.8064	0.7375	0.5264
FM_CL	0.8197	0.7471	0.5088
DeepFM	0.8117	0.7401	0.5203
DeepFM_CL	0.8126	0.7411	0.5196
xDeepFM	0.8123	0.7402	0.5200
xDeepFM_CL	0.8134	0.7414	0.5190
DCN_V2	0.8154	0.7432	0.5155
DCN_V2_CL	0.8171	0.7448	0.5135

下载: 导出CSV

表 2 仿真参数表

参数名	参数值	参数名	参数值
MBS覆盖范围	300 m	D2D带宽	20 MHz
SBS覆盖范围	150 m	无线回程带宽	20 MHz
UE间通信阈值	60 m	MBS发送功率	46 dBm
文件库大小	100	SBS发送功率	30 dBm
内容大小	10 Mbit	UE发送功率	23 dBm
SBS缓存容量	400 Mbit	噪声功率谱密度	174 dBm/Hz
MBS带宽	10 MHz	路径损耗因子	4
SBS带宽	20 MHz	系统参数	0.01

下载: 导出CSV

参考文献(20)

[1]	VAEZI M, AZARI A, KHOSRAVIRAD S R, et al. Cellular, wide-area, and non-terrestrial IoT: A survey on 5G advances and the road toward 6G[J]. IEEE Communications Surveys & Tutorials, 2022, 24(2): 1117–1174. doi: 10.1109/COMST.2022.3151028.
[2]	WU Dapeng, SHI Hang, WANG Honggang, et al. A feature-based learning system for internet of things applications[J]. IEEE Internet of Things Journal, 2019, 6(2): 1928–1937. doi: 10.1109/JIOT.2018.2884485.
[3]	CHENG Guangquan, JIANG Chi, YUE Binglei, et al. AI-driven proactive content caching for 6G[J]. IEEE Wireless Communications, 2023, 30(3): 180–188. doi: 10.1109/MWC.021.2200535.
[4]	FU Yaru, YANG H H, DOAN K N, et al. Effective cache-enabled wireless networks: An artificial intelligence- and recommendation-oriented framework[J]. IEEE Vehicular Technology Magazine, 2021, 16(1): 20–28. doi: 10.1109/MVT.2020.3033934.
[5]	LI Zhidu, BAO Ruili, WU Dapeng, et al. Caching at the edge: A group interest aware approach[C]. 2021-IEEE International Conference on Communications, Montreal, Canada, 2021: 1–6. doi: 10.1109/ICC42927.2021.9500942.
[6]	戚雨龙. 基于用户偏好的D2D缓存技术研究[D]. [硕士论文], 哈尔滨工业大学, 2021. QI Yulong. Research on D2D cache technology based on user preference[D]. [Master dissertation], Harbin Institute of Technology, 2021.
[7]	WU Dapeng, LIU Qianru, WANG Honggang, et al. Socially aware energy-efficient mobile edge collaboration for video distribution[J]. IEEE Transactions on Multimedia, 2017, 19(10): 2197–2209. doi: 10.1109/TMM.2017.2733300.
[8]	CHEN Mingzhe, SAAD W, YIN Changchuan, et al. Echo state networks for proactive caching in cloud-based radio access networks with mobile users[J]. IEEE Transactions on Wireless Communications, 2017, 16(6): 3520–3535. doi: 10.1109/TWC.2017.2683482.
[9]	CHATZIELEFTHERIOU L E, KARALIOPOULOS M, KOUTSOPOULOS I. Caching-aware recommendations: Nudging user preferences towards better caching performance[C]. 2017-IEEE Conference on Computer Communications, Atlanta, USA, 2017: 1–9. doi: 10.1109/INFOCOM.2017.8057031.
[10]	CHATZIELEFTHERIOU L E, KARALIOPOULOS M, and KOUTSOPOULOS I. Jointly optimizing content caching and recommendations in small cell networks[J]. IEEE Transactions on Mobile Computing, 2019, 18(1): 125–138. doi: 10.1109/TMC.2018.2831690.
[11]	FU Yaru, SALAÜN L, YANG Xiaolong, et al. Caching efficiency maximization for device-to-device communication networks: A recommend to cache approach[J]. IEEE Transactions on Wireless Communications, 2021, 20(10): 6580–6594. doi: 10.1109/TWC.2021.3075278.
[12]	FU Yaru, ZHANG Yue, WONG A K Y, et al. Revenue maximization: The interplay between personalized bundle recommendation and wireless content caching[J]. IEEE Transactions on Mobile Computing, 2023, 22(7): 4253–4265. doi: 10.1109/TMC.2022.3142809.
[13]	YU Shuai, DAB B, MOVAHEDI Z, et al. A socially-aware hybrid computation offloading framework for multi-access edge computing[J]. IEEE Transactions on Mobile Computing, 2020, 19(6): 1247–1259. doi: 10.1109/TMC.2019.2908154.
[14]	ANDREWS J G, BACCELLI F, and GANTI R K. A tractable approach to coverage and rate in cellular networks[J]. IEEE Transactions on Communications, 2011, 59(11): 3122–3134. doi: 10.1109/TCOMM.2011.100411.100541.
[15]	WU Dapeng, LI Jifang, HE Peng, et al. Social-aware graph-based collaborative caching in edge-user networks[J]. IEEE Transactions on Vehicular Technology, 2023, 72(6): 7926–7941. doi: 10.1109/TVT.2023.3241959.
[16]	YU Junliang, YIN Hongzhi, XIA Xin, et al. Self-supervised learning for recommender systems: A survey[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(1): 335–355. doi: 10.1109/TKDE.2023.3282907.
[17]	柯智慧. 协作边缘缓存与推荐联合优化策略研究[D]. [硕士论文], 天津大学, 2020. KE Zhihui. Joint optimization of cooperative edge caching and recommendation[D]. [Master dissertation], Tianjin University, 2020.
[18]	ZHANG Tiankui, WANG Yi, LIU Yuanwei, et al. Cache-enabling UAV communications: Network deployment and resource allocation[J]. IEEE Transactions on Wireless Communications, 2020, 19(11): 7470–7483. doi: 10.1109/TWC.2020.3011881.
[19]	HARPER F M and KONSTAN J A. The MovieLens datasets: History and context[J]. ACM Transactions on Interactive Intelligent Systems, 2016, 5(4): 19. doi: 10.1145/2827872.
[20]	WANG Ruoxi, SHIVANNA R, CHENG D, et al. DCN V2: Improved deep & cross network and practical lessons for web-scale learning to rank systems[C]. The Web Conference 2021, Ljubljana, Slovenia, 2021: 1785–1797. doi: 10.1145/3442381.3450078.