- 无标题文档

题名:

 无人机辅助的空地协同DNN推理技术研究    

作者:

 王玲    

学号:

 SZ2216120    

保密级别:

 公开    

语种:

 chi    

学科代码:

 085404    

学科:

 工学 - 电子信息 - 计算机技术    

学生类型:

 硕士    

学位:

 工学硕士    

入学年份:

 2022    

学校:

 南京航空航天大学    

院系:

 计算机科学与技术学院/人工智能学院    

专业:

 电子信息(专业学位)    

研究方向:

 边缘智能    

导师姓名:

 王俊华    

导师单位:

 计算机科学与技术学院/人工智能学院    

完成日期:

 2025-03-06    

答辩日期:

 2025-03-10    

外文题名:

 

Research on Technology of UAV-Assisted Air-Ground Collaborative DNN Inference

    

关键词:

 边缘智能 ; 协同推理 ; 无人机 ; 模型划分 ; 蝙蝠算法 ; 轨迹规划     

外文关键词:

   ; Edge  ; Intelligence ;   ; Collaborative  ; Inference ;   ; Unmanned  ; Aerial  ; Vehicle ;   ; Model  ; Partitioning ; Bat Algorithm ;   ; Trajectory Planning     

摘要:

       近年来,基于深度神经网络(Deep Neural Networks,DNNs)的人工智能(Artificial Intel- ligence,AI)应用,比如图像分类和虚拟/增强现实等,在移动边缘计算(Mobile Edge Com- puting,MEC)中得到了广泛关注。然而,随着DNN的发展及其参数体量的增加,计算需求急剧上升,导致移动设备难以满足实时处理需求,影响服务质量和用户体验。传统方法将数据传输至云服务器计算,但会增加延迟和通信开销。边缘智能(Edge Intelligence,EI)通过在边缘设备部署DNN模型,实现任务在网络边缘处理,从而缓解云服务器的性能瓶颈。然而,在分布式DNN推理中,如何合理划分神经网络层并进行协同推理仍是一个挑战性问题。本文结合无人机(Unmanned Aerial Vehicles,UAVs)与移动边缘计算技术,从系统模型、问题建模、算法设计与实现、实验与分析角度,对无人机辅助的空地协同DNN推理问题进行深入探讨。本文的主要工作如下:

       首先,对动态网络中可靠的空地协同DNN推理策略进行研究。首先提出了一种无人机辅助的DNN协作推理和任务卸载框架,其中无人机作为移动边缘计算节点,辅助移动设备(Mob- ile Devices,MDs)完成DNN任务。为了保证动态卸载网络中的可靠通信,本文提出了一个鲁棒传输问题,并应用条件风险价值(Conditional Value at Risk,CVaR)方法计算了移动设备的最小传输功率。为了加速DNN推理,本文针对协同DNN划分和卸载(Collaborative DNN Parti- tioning and Offlading,CDPO)问题,通过联合优化DNN划分和卸载决策来最小化任务的总推理延迟。受蝙蝠算法的启发,本文设计了一种高效的协同划分与卸载优化(Collaborative Partition and Offloading Optimization,CPOO)算法,将DNN任务划分为独立的子任务并获得卸载决策,从而有效地并行完成DNN推理。此外,本文还探究了传输功率对传输可靠性和任务总推理延迟的影响,针对一个同时考虑传输功率和延迟的主问题,结合次梯度算法和CPOO算法,提出了LaCPOO算法对其进行求解。最后,进行了广泛的仿真实验,表明了该算法在加速DNN推理和保证可靠性上的有效性。

       其次, 对无线供能网络中无人机轨迹与DNN推理联合优化策略进行研究。进一步考虑了无人机的能耗对DNN任务推理的影响,扩展提出了能耗约束的分布式DNN推理模型,并通过无线能量传输(Wireless  Power  Transfer, WPT) 技术为无人机充电以保持其续航能力和计算性能的稳定性。此外,在考虑DNN协同推理与卸载的同时对无人机轨迹进行优化,并设计了一种高效的联合DNN协同划分和卸载与无人机轨迹规划(Collaborative Partition and Offloading and Trajectory Scheduling,CPOTS)算法求解问题,该算法采用块坐标下降(Block Coordinate Descent,BCD)方法,将问题划分为两个子问题,通过CPOO算法求解DNN划分与卸载问题,并利用逐次凸近似(Successive convex approximation,SCA)方法将无人机轨迹规划问题转换为凸问题后,交替求解这两个子问题。最后,通过与其他典型算法对比,实验结果表明该算法在新的约束下实现了更短的推理时延。

 

外摘要要:

 In recent years, artificial intelligence (AI) applications based on deep neural networks (DNNs), such as image classification and virtual/augmented reality, have garnered significant attention in mobile edge computing (MEC). However, with the advancement of DNNs and the increase in parameter volume, computational demands have surged dramatically, making it challenging for mobile devices to meet real-time processing requirements, thereby affecting service quality and user experience. Traditional methods, which involve transmitting data to cloud servers for computation, introduce additional latency and communication overhead. Edge intelligence (EI) addresses this issue by deploying DNN models on edge devices, enabling task processing at the network edge and alleviating the performance bottlenecks of cloud servers. Nevertheless, in distributed DNN inference, how to effectively partition neural network layers and achieve collaborative inference remains a challenging problem. This thesis integrates unmanned aerial vehicles (UAVs) with MEC technology and provides an in-depth exploration of UAV- assisted ground-air collaborative DNN inference from the perspectives of system modeling, problem formulation, algorithm design and implementation, and experimental analysis. The main contributions of this thesis are as follows:

   A study on reliable ground air collaborative DNN inference strategies in dynamic networks. First, a UAV-assisted collaborative DNN inference and task offloading framework is proposed, where UAVs act as mobile edge computing nodes to assist mobile devices (MDs) in completing DNN tasks. To ensure reliable communication in dynamic offloading networks, a robust transmission problem is formulated, and the conditional value at risk (CVaR) method is applied to calculate the minimum transmission power for mobile devices. To accelerate DNN inference, this thesis addresses the Collaborative DNN Partitioning and Offloading (CDPO) problem by jointly optimizing DNN partitioning and offloading decisions to minimize the total inference latency of tasks. Inspired by the bat algorithm, an efficient collaborative partition and offloading optimization (CPOO) algorithm is designed to partition DNN tasks into independent subtasks and determine offloading decisions, thereby effectively parallelizing DNN inference. Additionally, this thesis investigates the impact of transmission power on both transmission reliability and total task inference latency. For a primary problem that jointly considers transmission power and latency, a LaCPOO algorithm is proposed by integrating the subgradient algorithm and the CPOO algorithm to solve it. Extensive simulation experiments demonstrate the effectiveness of the proposed algorithm in accelerating DNN inference while ensuring reliability.     A study on the joint optimization strategy of UAV trajectory and DNN inference in wireless-powered networks. The impact of UAV energy consumption on the execution of DNN inference tasks is further considered, and an energy-constrained distributed DNN inference model is extended. Wire- less Power Transfer (WPT) technology is utilized to recharge UAVs, ensuring sustained operational capability and computational performance stability. Additionally, the research optimizes UAV trajectories while addressing DNN collaborative inference and offloading. An efficient algorithm, collaborative partition and offloading and trajectory scheduling (CPOTS), is designed to solve the problem. The CPOTS algorithm employs the block coordinate descent (BCD) method, dividing the problem into two sub-problems. The DNN partition and offloading issues are resolved using the CPOO algorithm, while the UAV trajectory planning is transformed into a convex problem using the successive convex approximation (SCA) method, with these sub-problems solved alternately. Comparative experiments with other typical algorithms demonstrate that the proposed algorithm achieves shorter inference delays under new constraints.

参考文献:

[1]加快推动移动物联网从“万物互联”向“万物智联”发展[R]. Technical report, 中国电子报, 2024-09-20.

[2]杨旭东. 以标准化引领物联网产业高质量发展助力实体经济与数字经济深度融合——《物联网标准体系建设指南(2024版)》解读[J]. 信息技术与标准化, 2024, (09):4–7.

[3]Sasaki Y. A Survey on IoT Big Data Analytic Systems: Current and Future[J]. IEEE Internet of Things Journal, 2022, 9(2):1024–1036.

[4]Wang S, Li Q. Satellite Computing: Vision and Challenges[J]. IEEE Internet of Things Journal, 2023, 10(24):22514–22529.

[5]许文元. 边缘计算中面向DNN的协同推理优化技术研究[D]. 北京交通大学, 2022.

[6]Tang S, Cui M, Qi L, et al. Edge Intelligence with Distributed Processing of DNNs: A Survey[J]. CMES-Computer Modeling in Engineering & Sciences, 2023, 136(1):5–42.

[7]董裕民, 张静, 谢昌佐, 等. 云边端架构下边缘智能计算关键问题综述:计算优化与计算卸载[J]. 电子与信息学报, 2024, 46(03):765–776.

[8]Chen J, Ran X. Deep learning with edge computing: A review[J]. Proceedings of the IEEE, 2019, 107(8):1655–1674.

[9]何茂霖, 多滨, 胡艳梅, 等. 基于智能超表面的无人机移动边缘计算综述[J]. 无线电通信技术, 2024, 50(02):349–356.

[10]张换然, 申凌峰, 任资卓, 等. 无人机辅助智能边缘网络技术综述[J]. 电讯技术, 2024, 64(02):325–332.

[11]Xiao P, Wang L, Chuan J, et al. Implementation for UAVs Aided Edge Sensing System in Wireless Emergency Communications[C]. Proceedings of 2019 11th International Conference on Wireless Communications and Signal Processing, 2019:1–5.

[12]Feng W, Tang J, Zhao N, et al. NOMA-based UAV-aided networks for emergency communica- tions[J]. China Communications, 2020, 17(11):54–66.

[13]Yang Y, Karimadini M, Xiang C, et al. Wide area surveillance of urban environments using multiple Mini-VTOL UAVs[C]. Proceedings of IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society, 2015:000795–000800.

[14]Zeng Y, Zhang R, Lim T J. Wireless communications with unmanned aerial vehicles: opportunities and challenges[J]. IEEE Communications Magazine, 2016, 54(5):36–42.

[15]Katsigiannis P, Misopolinos L, Liakopoulos V, et al. An autonomous multi-sensor UAV system for reduced-input precision agriculture applications[C]. Proceedings of 2016 24th Mediterranean Conference on Control and Automation, 2016:60–64.

[16]吴雪萌. 无人机辅助通信网络的航迹规划研究与仿真[D]. 北京邮电大学, 2021.

[17]Deng S, Zhao H, Fang W, et al. Edge Intelligence: The Confluence of Edge Computing and Artificial Intelligence[J]. IEEE Internet of Things Journal, 2020, 7(8):7457–7469.

[18]施巍松, 张星洲, 王一帆, 等. 边缘计算:现状与展望[J]. 计算机研究与发展, 2019, 56(01):69–89.

[19]Wang X, Han Y, Leung V C M, et al. Convergence of Edge Computing and Deep Learning: A Comprehensive Survey[J]. IEEE Communications Surveys & Tutorials, 2020, 22(2):869–904.

[20]Svozil D, Kvasnicka V, Pospichal J. Introduction to multi-layer feed-forward neural networks[J]. Chemometrics and intelligent laboratory systems, 1997, 39(1):43–62.

[21]Izhikevich E M. Simple model of spiking neurons[J]. IEEE Transactions on neural networks, 2003, 14(6):1569–1572.

[22]Chauvin Y, Rumelhart D E. Backpropagation: theory, architectures, and applications[M]. Psy- chology press, 2013.

[23]Kang Y, Hauswald J, Gao C, et al. Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge[J]. Acm Sigplan Notices, 2017, 52(1):615–629.

[24]Yang X S. A new metaheuristic bat-inspired algorithm[M]. . Proceedings of Nature inspired cooperative strategies for optimization. Springer, 2010: 65–74.

[25]Ji J, Zhu K, Yi C, et al. Energy Consumption Minimization in UAV-Assisted Mobile-Edge Computing Systems: Joint Resource Allocation and Trajectory Design[J]. IEEE Internet of Things Journal, 2021,8(10):8570–8584.

[26]Hoang L T, Nguyen C T, Pham A T. Deep Reinforcement Learning-Based Online Resource Man- agement for UAV-Assisted Edge Computing With Dual Connectivity[J]. IEEE/ACM Transactions on Networking, 2023, 31(6):2761–2776.

[27]Chai F, Zhang Q, Yao H, et al. Joint Multi-Task Offloading and Resource Allocation for Mobile Edge Computing Systems in Satellite IoT[J]. IEEE Transactions on Vehicular Technology, 2023, 72(6):7783–7795.

[28]Apostolopoulos P A, Fragkos G, Tsiropoulou E E, et al. Data Offloading in UAV-Assisted Multi- Access Edge Computing Systems Under Resource Uncertainty[J]. IEEE Transactions on Mobile Computing, 2023, 22(1):175–190.

[29]Ning Z, Yang Y, Wang X, et al. Dynamic Computation Offloading and Server Deployment for UAV-Enabled Multi-Access Edge Computing[J]. IEEE Transactions on Mobile Computing, 2023, 22(5):2628–2644.

[30]Cheng K, Fang X, Wang X. Energy Efficient Edge Computing and Data Compression Collabo- ration Scheme for UAV-Assisted Network[J]. IEEE Transactions on Vehicular Technology, 2023, 72(12):16395–16408.

[31]Tong S, Liu Y, Misˇic´ J, et al. Joint Task Offloading and Resource Allocation for Fog-Based Intelligent Transportation Systems: A UAV-Enabled Multi-Hop Collaboration Paradigm[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(11):12933–12948.

[32]Wang D, Jia Y, Dong M, et al. Blockchain-Integrated UAV-Assisted Mobile Edge Computing: Trajectory Planning and Resource Allocation[J]. IEEE Transactions on Vehicular Technology, 2024, 73(1):1263–1275.

[33]Song F, Xing H, Wang X, et al. Evolutionary Multi-Objective Reinforcement Learning Based Trajectory Control and Task Offloading in UAV-Assisted Mobile Edge Computing[J]. IEEE Transactions on Mobile Computing, 2023, 22(12):7387–7405.

[34]Han S, Shen H, Philipose M, et al. Mcdnn: An approximation-based execution framework for deep stream processing under resource constraints[C]. Proceedings of the 14th Annual International Conference on Mobile Systems, Applications, and Services, 2016:123–136.

[35]Duan Y, Wu J. Joint optimization of DNN partition and scheduling for mobile cloud computing[C].Proceedings of the 50th International Conference on Parallel Processing, 2021:1–10.

[36]Eshratifar A E, Abrishami M S, Pedram M. JointDNN: An efficient training and inference engine for intelligent mobile cloud computing services[J]. IEEE Transactions on Mobile Computing, 2019, 20(2):565–576.

[37]Li M, Li Y, Tian Y, et al. AppealNet: An efficient and highly-accurate edge/cloud collaborative architecture for DNN inference[C]. Proceedings of 2021 58th ACM/IEEE Design Automation Conference, 2021:409–414.

[38]Ali Z, Jiao L, Baker T, et al. A deep learning approach for energy efficient computational offloading in mobile edge computing[J]. IEEE Access, 2019, 7:149623–149633.

[39]Li E, Zhou Z, Chen X. Edge Intelligence: On-Demand Deep Learning Model Co-Inference with Device-Edge Synergy[C]. Proceedings of the 2018 Workshop on Mobile Edge Communications. Association for Computing Machinery, 2018:31–36.

[40]Tang X, Chen X, Zeng L, et al. Joint multiuser DNN partitioning and computational resource allo- cation for collaborative edge intelligence[J]. IEEE Internet of Things Journal, 2020, 8(12):9511– 9522.

[41]Li C, Xu H, Xu Y, et al. DNN inference acceleration with partitioning and early exiting in edge computing[C]. Proceedings of Wireless Algorithms, Systems, and Applications: 16th International Conference, 2021:465–478.

[42]Jeong H J, Lee H J, Shin C H, et al. IONN: Incremental offloading of neural network computations from mobile devices to edge servers[C]. Proceedings of the ACM symposium on cloud computing, 2018:401–411.

[43]Tian X, Zhu J, Xu T, et al. Mobility-included DNN partition offloading from mobile devices to edge clouds[J]. Sensors, 2021, 21(1):229.

[44]Ren P, Qiao X, Huang Y, et al. Edge-assisted distributed DNN collaborative computing approach for mobile web augmented reality in 5G networks[J]. IEEE Network, 2020, 34(2):254–261.

[45]Mohammed T, Joe-Wong C, Babbar R, et al. Distributed inference acceleration with adaptive DNN partitioning and offloading[C]. Proceedings of IEEE INFOCOM 2020-IEEE Conference on Computer Communications, 2020:854–863.

[46]Ashouri M, Lorig F, Davidsson P, et al. Analyzing distributed deep neural network deployment on edge and cloud nodes in IoT systems[C]. Proceedings of 2020 IEEE International Conference on Edge Computing, 2020:59–66.

[47]Lockhart L, Harvey P, Imai P, et al. Scission: Performance-driven and context-aware cloud-edge distribution of deep neural networks[C]. Proceedings of 2020 IEEE/ACM 13th International Conference on Utility and Cloud Computing, 2020:257–268.

[48]Huang Y, Lin B, Zheng Y, et al. Cost efficient offloading strategy for DNN-based applications in edge-cloud environment[C]. Proceedings of 2019 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Commu- nications, Social Computing & Networking, 2019:331–337.

[49]Huang Y, Qiao X, Tang J, et al. DeepAdapter: A collaborative deep learning framework for the mobile web using context-aware network pruning[C]. Proceedings of IEEE INFOCOM 2020-IEEE Conference on Computer Communications, 2020:834–843.

[50]Hu S, Dong C, Wen W. Enable pipeline processing of DNN co-inference tasks in the mobile-edge cloud[C]. Proceedings of 2021 IEEE 6th International Conference on Computer and Communi- cation Systems, 2021:186–192.

[51]Mao J, Yang Z, Wen W, et al. Mednn: A distributed mobile system with enhanced partition and deployment for large-scale dnns[C]. Proceedings of 2017 IEEE/ACM International Conference on Computer-Aided Design, 2017:751–756.

[52]Zhou L, Wen H, Teodorescu R, et al. Distributing deep neural networks with containerized parti- tions at the edge[C]. Proceedings of 2nd USENIX Workshop on Hot Topics in Edge Computing, 2019:booktitle.

[53]Guo X, Dong C, Wen W. Dynamic computation offloading strategy with dnn partitioning in d2d multi-hop networks[C]. Proceedings of the 2021 9th International Conference on Communications and Broadband Networking, 2021:172–178.

[54]Yang L, Zheng C, Shen X, et al. OfpCNN: On-Demand Fine-Grained Partitioning for CNN Inference Acceleration in Heterogeneous Devices[J]. IEEE Transactions on Parallel and Distributed Systems, 2023, 34(12):3090–3103.

[55]Molchanov P, Tyree S, Karras T, et al. Pruning convolutional neural networks for resource efficient transfer learning[J]. CoRR, 2016..

[56]Li C, Chai L, Jiang K, et al. DNN Partition and Offloading Strategy with Improved Particle Swarm Genetic Algorithm in VEC[J]. IEEE Transactions on Intelligent Vehicles, 2023. 1–11.

[57]Zhang S, Zhang S, Qian Z, et al. DeepSlicing: Collaborative and Adaptive CNN Inference With Low Latency[J]. IEEE Transactions on Parallel and Distributed Systems, 2021, 32(9):2175–2187.

[58]Chen W, Sim M, Sun J, et al. From CVaR to Uncertainty Set: Implications in Joint Chance- Constrained Optimization[J]. Operations Research: The Journal of the Operations Research Society of America, 2010, (2):58.

[59]Zymler S, Kuhn D, Rustem B. Distributionally robust joint chance constraints with second-order moment information[J]. Mathematical Programming, 2013, 137:167–198.

[60]Ebrahim M A, Ebrahim G A, Mohamed H K, et al. A deep learning approach for task offloading in multi-UAV aided mobile edge computing[J]. IEEE Access, 2022, 10:101716–101731.

[61]Lu W, Mo Y, Feng Y, et al. Secure transmission for multi-UAV-assisted mobile edge computing based on reinforcement learning[J]. IEEE Transactions on Network Science and Engineering, 2022, 10(3):1270–1282.

[62]Liu C, Liu K. Toward Reliable DNN-Based Task Partitioning and Offloading in Vehicular Edge Computing[J]. IEEE Transactions on Consumer Electronics, 2024, 70(1):3349–3360.

[63]Wang J, Liu K, Li B, et al. Delay-Sensitive Multi-Period Computation Offloading with Reliability Guarantees in Fog Networks[J]. IEEE Transactions on Mobile Computing, 2020, 19(9):2062– 2075.

[64]Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Commun. ACM, 2017, 60(6):84–90.

[65]Ren W, Qu Y, Qin Z, et al. Energy-Efficient Collaborative DNN Inference in UAV Swarm[C]. Proceedings of 2023 9th International Conference on Big Data Computing and Communications (BigCom). IEEE, 2023:195–202.

[66]Li E, Zeng L, Zhou Z, et al. Edge AI: On-demand accelerating deep neural network inference via edge computing[J]. IEEE Transactions on Wireless Communications, 2019, 19(1):447–457.

[67]Redmon J, Farhadi A. YOLO9000: better, faster, stronger[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2017:7263–7271.

[68]Lin M, Chen Q, Yan S. Network In Network[J]. Computer Science,2013..

[69]Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014..

[70]Hu X, Wong K K, Zhang Y. Wireless-powered edge computing with cooperative UAV: Task, time scheduling and trajectory design[J]. IEEE Transactions on Wireless Communications, 2020, 19(12):8083–8098.

[71]Liu Y, Zhou J, Tian D, et al. Joint communication and computation resource scheduling of a UAV-assisted mobile edge computing system for platooning vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(7):8435–8450.

[72]贾廷宇. 基于强化学习的无线能量收集网络中无人机轨迹优化[D]. 北京交通大学, 2021.

[73]Gu X, Zhang G, Wang M, et al. UAV-aided energy-efficient edge computing networks: Security offloading optimization[J]. IEEE Internet of Things Journal, 2021, 9(6):4245–4258.

[74]Wang J, Zhu K, Dai P, et al. An adaptive Q-value adjustment-based learning model for reli- able vehicle-to-UAV computation offloading[J]. IEEE Transactions on Intelligent Transportation Systems, 2023..

[75]Guo S, Xiao B, Yang Y, et al. Energy-efficient dynamic offloading and resource scheduling in mobile cloud computing[C]. Proceedings of IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications, 2016:1–9.

[76]Liu C, Liu K, Guo S, et al. Adaptive Offloading for Time-Critical Tasks in Heterogeneous Internet of Vehicles[J]. IEEE Internet of Things Journal, 2020, 7(9):7999–8011.

[77]Boyd S, Vandenberghe L. Convex optimization[M]. Cambridge university press, 2004.

[78]Li Y, Ye L, Meng W, et al. Joint Trajectory Optimization and Mobile-Edge Computation Offloading for Multi-UAV-Connected System[C]. Proceedings of ICC 2023-IEEE International Conference on Communications, 2023:5432–5437.

[79]Yang Z, Bi S, Zhang Y J A. Stable online offloading and trajectory control for UAV-enabled MEC with EH devices[C]. Proceedings of 2021 IEEE Global Communications Conference, 2021:01–07.

[80]Jeong S, Simeone O, Kang J. Mobile edge computing via a UAV-mounted cloudlet: Optimization of bit allocation and path planning[J]. IEEE Transactions on Vehicular Technology, 2017, 67(3):2049– 2063.

[81]Ji J, Zhu K, Yi C, et al. Joint resource allocation and trajectory design for UAV-assisted mobile edge computing systems[C]. Proceedings of GLOBECOM 2020-2020 IEEE Global Communications Conference, 2020:1–6.

[82]Ren W, Qu Y, Qin Z, et al. Efficient pipeline collaborative dnn inference in resource-constrained uav swarm[C]. Proceedings of 2024 IEEE Wireless Communications and Networking Conference (WCNC). IEEE, 2024:1–6.

中图分类号:

 TP391    

馆藏号:

 2025-016-0119    

开放日期:

 2025-09-28    

无标题文档

   建议浏览器: 谷歌 火狐 360请用极速模式,双核浏览器请用极速模式