An Overview of Key Technologies on 6G-Enabled Communication and Computing Integration for Energy-Efficiency Optimization

LIU Guangyi; CAI Qing; WANG Xinyao; CHEN Tianjiao; JIN Jing; XUE Yahui; WANG Ailing; WANG Hanning

doi:10.11999/JEIT260399

Article Contents

Article Navigation > Journal of Electronics & Information Technology > 2026 >

LIU Guangyi, CAI Qing, WANG Xinyao, CHEN Tianjiao, JIN Jing, XUE Yahui, WANG Ailing, WANG Hanning. An Overview of Key Technologies on 6G-Enabled Communication and Computing Integration for Energy-Efficiency Optimization[J]. Journal of Electronics & Information Technology. doi: 10.11999/JEIT260399

Citation:

LIU Guangyi, CAI Qing, WANG Xinyao, CHEN Tianjiao, JIN Jing, XUE Yahui, WANG Ailing, WANG Hanning. An Overview of Key Technologies on 6G-Enabled Communication and Computing Integration for Energy-Efficiency Optimization[J]. Journal of Electronics & Information Technology. doi: 10.11999/JEIT260399

Citation:

LIU Guangyi, CAI Qing, WANG Xinyao, CHEN Tianjiao, JIN Jing, XUE Yahui, WANG Ailing, WANG Hanning. An Overview of Key Technologies on 6G-Enabled Communication and Computing Integration for Energy-Efficiency Optimization[J]. Journal of Electronics & Information Technology. doi: 10.11999/JEIT260399

PDF( 3441 KB)

An Overview of Key Technologies on 6G-Enabled Communication and Computing Integration for Energy-Efficiency Optimization

doi: 10.11999/JEIT260399 cstr: 32379.14.JEIT260399

①(China Mobile Research Institute, Beijing 100053, China

Funds: The National Key R&D Program of China (2024YFE0200600)

Accepted Date: 2026-05-15
Rev Recd Date: 2026-05-15

Available Online: 2026-06-02

Abstract

Abstract

Significance Constrained by physical conditions such as size, power consumption, and cost, high energy consumptions have become key bottleneck for the large-scale application of new intelligent terminals. In contrast to Fifth-Generation (5G) networks, Sixth-Generation (6G) will achieve profound architectural enhancement of the RAN, sink computing capabilities toward the RAN side, and enable the RAN to perform part of tasks originally executed by end devices. With the end-edge collaboration, new intelligent terminals are expected to realize lightweight, low-cost and long-endurance evolution, which is of great significance for supporting the large-scale deployment of ubiquitous intelligence in 6G networks. Progress Current advancements in terminal energy consumption optimization with 6G end-edge collaboration are discussed, focusing on three primary offloading modes: local execution, full offloading, and partial offloading. Local execution requires the terminal to process all tasks, leading to high computational energy consumption, while full offloading shifts all tasks to the RAN, reducing terminal energy use but increasing transmission energy costs, particularly in poor channel conditions. Partial offloading combines the advantages of both modes, optimizing energy consumption based on real-time network conditions. For partial offloading, existing research has introduced several optimization techniques to enhance energy efficiency. (1) Feature extraction and filtering: Through semantic encoding and information extraction approaches, feature extraction is performed at the UE to transmit only task-relevant data to the RAN. This reduces the amount of redundant or unnecessary data sent, minimizing transmission energy consumption (2) Model partitioning for offloading: This technique divides a large deep learning model into different layers based on its network structure, with simpler layers processed at the UE and more complex ones offloaded to RAN. By leveraging end-edge collaborative reasoning, this method optimizes energy consumption by balancing the computational load between the terminal and RAN. (3) Model lightweighting: By reducing model complexity through techniques like pruning, quantization, and knowledge distillation, this method lowers computational overhead while maintaining performance. (4) Incremental reasoning: This method focuses on the changes in data or features, performing localized reasoning only on updated portions and reusing historical computations, significantly reducing redundant calculations. The above optimization techniques collectively enhance the performance and energy efficiency of terminal devices within the 6G end-edge collaboration framework. Conclusions This paper provides a comprehensive discussion of terminal energy consumption optimization with 6G end-edge collaboration. It summarizes the functional evolution of enhanced RAN, constructs an end-edge collaborative service framework for communication-computation integration, and establishes a theoretical model including terminal computing energy consumption and transmission energy consumption. The composition and influencing factors of energy consumption under different offloading modes are clarified. Key technologies for energy optimization based on end-edge collaboration are further discussed, including feature extraction and filtering, model partitioning for offloading, model lightweighting, and incremental reasoning. Given the energy consumption fluctuations caused by the dynamic nature of wireless channels, this paper introduces energy optimization mechanisms such as semantic compression, dynamic partitioned offloading, adaptive model pruning, and incremental reasoning to strike a dynamic balance between optimizing energy consumption and maintaining task performance. Taking intelligent robot video understanding as a typical application scenario, a test platform is developed to validate the effectiveness of the proposed optimization mechanisms. This paper also analyzes the challenges currently faced in the research and discusses future research directions. Prospects Although the end-edge collaborative energy-saving technologies have achieved initial progress, they still face many challenges in practical deployment, especially under real network environments, dynamic wireless channels, and large-scale user access. Future research should focus on the trade-off between optimization overhead and system robustness, and further investigate dynamic communication–computation resource substitution modeling in stochastic resource environments, as well as multi-user collaborative strategies and global energy efficiency optimization. Meanwhile, as the technology matures, the standardization and engineering implementation of end-edge collaborative energy-saving frameworks will become crucial for the large-scale adoption of 6G applications. Future studies should therefore promote deeper integration between algorithm design and network architecture, enabling the practical deployment of low-power, high-efficiency intelligent communication systems.
- 6G,
- Radio Access Network (RAN),
- Communication-Computation Integration,
- End-Edge Collaboration,
- Terminal Energy Consumption Optimization

FullText(HTML)

References(53)

References

[1]	TANG Qinqin, YANG Yutian, CUI Jiayi, et al. SeCo4: Co-design of sensing, communication, and computing for intelligent control in industrial cyber-physical systems[J]. IEEE Journal on Selected Areas in Communications, 2025, 43(10): 3362–3378. doi: 10.1109/JSAC.2025.3574623.
[2]	ETSI. 3GPP TR 32.972 Telecommunication management; study on system and functional aspects of energy efficiency in 5G networks[S]. 2024. (查阅网上资料, 未找到出版地和出版者信息, 不确定作者是否正确, 请确认).
[3]	3GPP TR 28.885 Study on energy efficiency and energy saving aspects of 5G advanced networks[S]. 2025. (查阅网上资料, 未找到出版和作者信息, 请确认).
[4]	3GPP TR 28.913 Study on new aspects of Energy Efficiency (EE) for 5G phase 2[S]. 2025. (查阅网上资料, 未找到出版和作者信息, 请确认).
[5]	CAI Qing, ZHOU Yiqing, LIU Ling, et al. Prioritized assignment with task dependency in collaborative mobile edge computing[J]. IEEE Transactions on Mobile Computing, 2024, 23(12): 13505–13521. doi: 10.1109/TMC.2024.3427380.
[6]	LIU Guangyi, WANG Xinyao, CHEN Tianjiao, et al. QoAIS-guaranteed AI service offloading in IoV scenario enabled by 6G native AI network[J]. Journal of Information and Intelligence, 2026, 4(1): 54–66. doi: 10.1016/J.JIIXD.2025.12.003.
[7]	NVIDIA. White paper: NVIDIA ampere GA102 GPU architecture[EB/OL]. https://www.nvidia.com/content/PDF/nvidia-ampere-ga-102-gpu-architecture-whitepaper-v2.pdf. (查阅网上资料,未找到年份信息,请确认).
[8]	JIA Yunjian, HUANG Zhen, LUO Kun, et al. Lightweight joint source-channel coding for semantic communications[J]. IEEE Communications Letters, 2023, 27(12): 3161–3165. doi: 10.1109/LCOMM.2023.3329533.
[9]	YU Xianhua, LI Dong, ZHANG Ning, et al. A novel lightweight joint source-channel coding design in semantic communications[J]. IEEE Internet of Things Journal, 2025, 12(11): 18447–18450. doi: 10.1109/JIOT.2025.3556909.
[10]	LIU Chuanhong, GUO Caili, YANG Yang, et al. Lightweight task-oriented semantic communication empowered by large-scale AI models[J]. IEEE Transactions on Vehicular Technology, 2025, 74(9): 14823–14827. doi: 10.1109/TVT.2025.3563635.
[11]	LIU Yaoqi, HAN Yinhe, LI Hongxin, et al. Computing over space: Status, challenges, and opportunities[J]. Engineering, 2025, 54: 20–25. doi: 10.1016/j.eng.2025.06.005.
[12]	HUANG Kaibin, LAN Qiao, LIU Zhiyan, et al. Semantic data sourcing for 6G edge intelligence[J]. IEEE Communications Magazine, 2023, 61(12): 70–76. doi: 10.1109/MCOM.001.2200962.
[13]	CAI Qing, ZHOU Yiqing, LIU Ling, et al. Query-aware semantic encoder-based resource allocation in task-oriented communications[J]. IEEE Transactions on Mobile Computing, 2025, 24(7): 6413–6429. doi: 10.1109/TMC.2025.3541636.
[14]	WANG Yining, CHEN Mingzhe, LUO Tao, et al. Performance optimization for semantic communications: An attention-based reinforcement learning approach[J]. IEEE Journal on Selected Areas in Communications, 2022, 40(9): 2598–2613. doi: 10.1109/JSAC.2022.3191112.
[15]	STRINATI E C and BARBAROSSA S. 6G networks: Beyond Shannon towards semantic and goal-oriented communications[J]. Computer Networks, 2021, 190: 107930. doi: 10.1016/j.comnet.2021.107930.
[16]	JIN Zhuoran, YUAN Hongbang, MEN Tianyi, et al. RAG-RewardBench: Benchmarking reward models in retrieval augmented generation for preference alignment[C]. Proceedings of the Findings of the Association for Computational Linguistics: ACL 2025, Vienna, Austria, 2025: 17061–17090. doi: 10.18653/v1/2025.findings-acl.877.
[17]	WEN Wen, XIE Renchao, TANG Qinqin, et al. Green orchestra: Joint spatiotemporal task scheduling and hybrid energy coordination in computing power networks[J]. IEEE Transactions on Cognitive Communications and Networking, 2026, 12: 5621–5637. doi: 10.1109/TCCN.2026.3656279.
[18]	CHEN Yuxuan, LI Rongpeng, YU Xiaoxue, et al. Adaptive layer splitting for wireless large language model inference in edge computing: A model-based reinforcement learning approach[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(2): 278–292. doi: 10.1631/FITEE.2025.2400468.
[19]	JEONG H J, LEE H J, SHIN K Y, et al. PerDNN: Offloading deep neural network computations to pervasive edge servers[C]. 2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS), Singapore, Singapore, 2020: 1055–1066. doi: 10.1109/ICDCS47774.2020.00114.
[20]	MOHAMMED T, JOE-WONG C, BABBAR R, et al. Distributed inference acceleration with adaptive DNN partitioning and offloading[C]. IEEE INFOCOM 2020 - IEEE Conference on Computer Communications, Toronto, Canada, 2020: 854–863. doi: 10.1109/INFOCOM41043.2020.9155237.
[21]	LI Mingchu, ZHANG Wenteng, and XIA Dexin. Transformer inference acceleration in edge computing environment[C]. 2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW), Bangalore, India, 2023: 104–109. doi: 10.1109/CCGridW59191.2023.00030.
[22]	LIU Sicong, LUO Hao, LI Xiaochen, et al. AdaKnife: Flexible DNN offloading for inference acceleration on heterogeneous mobile devices[J]. IEEE Transactions on Mobile Computing, 2025, 24(2): 736–748. doi: 10.1109/TMC.2024.3466931.
[23]	LIU Zhicheng, SONG Jinduo, QIU Chao, et al. Hastening stream offloading of inference via multi-exit DNNs in mobile edge computing[J]. IEEE Transactions on Mobile Computing, 2024, 23(1): 535–548. doi: 10.1109/TMC.2022.3218724.
[24]	WANG Hongli, GUO Bin, LIU Jiaqi, et al. Context-aware adaptive surgery: A fast and effective framework for adaptative model partition[J]. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2021, 5(3): 131. doi: 10.1145/3478073.
[25]	ZAWISH M, DAVY S, and ABRAHAM L. Complexity-driven model compression for resource-constrained deep learning on edge[J]. IEEE Transactions on Artificial Intelligence, 2024, 5(8): 3886–3901. doi: 10.1109/TAI.2024.3353157.
[26]	PRADHAN C, LI Ang, SHE Changyang, et al. Computation offloading for IoT in C-RAN: Optimization and deep learning[J]. IEEE Transactions on Communications, 2020, 68(7): 4565–4579. doi: 10.1109/TCOMM.2020.2983142.
[27]	JANKOWSKI M, GÜNDÜZ D, and MIKOLAJCZYK K. Joint device-edge inference over wireless links with pruning[C]. 2020 IEEE 21st International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), Atlanta, USA, 2020: 1–5. doi: 10.1109/SPAWC48557.2020.9154306.
[28]	YU Fang, CUI Li, WANG Pengcheng, et al. EasiEdge: A novel global deep neural networks pruning method for efficient edge computing[J]. IEEE Internet of Things Journal, 2021, 8(3): 1259–1271. doi: 10.1109/JIOT.2020.3034925.
[29]	JIA Yongzhe, LIU Bowen, ZHANG Xuyun, et al. Model pruning-enabled federated split learning for resource-constrained devices in artificial intelligence empowered edge computing environment[J]. ACM Transactions on Sensor Networks, 2024. (查阅网上资料,未找到卷期页码信息,请确认). doi: 10.1145/3687478.
[30]	ZHANG Cui, ZHANG Wenjun, WU Qiong, et al. Distributed deep reinforcement learning-based gradient quantization for federated learning enabled vehicle edge computing[J]. IEEE Internet of Things Journal, 2025, 12(5): 4899–4913. doi: 10.1109/JIOT.2024.3447036.
[31]	ZHANG Mingjin, SHEN Xiaoming, CAO Jiannong, et al. EdgeShard: Efficient LLM inference via collaborative edge computing[J]. IEEE Internet of Things Journal, 2025, 12(10): 13119–13131. doi: 10.1109/JIOT.2024.3524255.
[32]	ZHANG Kangkai, ZHANG Chunhui, LI Shikun, et al. Student network learning via evolutionary knowledge distillation[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022, 32(4): 2251–2263. doi: 10.1109/TCSVT.2021.3090902.
[33]	SUN Lu, LIU Ziqian, WAN Liangtian, et al. Cooperative knowledge-distillation-based tiny DNN for UAV-assisted mobile-edge network[J]. IEEE Internet of Things Journal, 2024, 11(18): 30204–30216. doi: 10.1109/JIOT.2024.3409813.
[34]	TANG Qinqin, XIE Renchao, YU F R, et al. Collective deep reinforcement learning for intelligence sharing in the internet of intelligence-empowered edge computing[J]. IEEE Transactions on Mobile Computing, 2023, 22(11): 6327–6342. doi: 10.1109/TMC.2022.3199812.
[35]	LU Jingfeng, LU Siliang, XU Jiawen, et al. Unknown fault diagnosis of motors based on incremental learning and edge computing[J]. IEEE Internet of Things Journal, 2025, 12(12): 22243–22256. doi: 10.1109/JIOT.2025.3550928.
[36]	WAN Xiaoxue, CEN Lihui, CHEN Xiaofang, et al. Multiview shapelet prototypical network for few-shot fault incremental learning[J]. IEEE Transactions on Industrial Informatics, 2024, 20(10): 11751–11762. doi: 10.1109/TII.2024.3413304.
[37]	YU Wanke and ZHAO Chunhui. Broad convolutional neural network based industrial process fault diagnosis with incremental learning capability[J]. IEEE Transactions on Industrial Electronics, 2020, 67(6): 5081–5091. doi: 10.1109/TIE.2019.2931255.
[38]	CHI Jiancheng, ZHOU Xiaobo, XIAO Fu, et al. Enhancing adaptability and efficiency of task offloading by broad learning in industrial IoT[J]. IEEE Transactions on Network Science and Engineering, 2025, 12(1): 263–276. doi: 10.1109/TNSE.2024.3493053.
[39]	WAN Xiaoxue, CEN Lihui, CHEN Xiaofang, et al. Multiview shapelet prototypical network for few-shot fault incremental learning[J]. IEEE Transactions on Industrial Informatics, 2024, 20(10): 11751–11762. doi: 10.1109/TII.2024.3413304. (查阅网上资料,本条文献与第36条文献重复,请确认).
[40]	JIN Zhuoran, CAO Pengfei, WANG Chenhao, et al. RWKU: Benchmarking real-world knowledge unlearning for large language models[C]. Proceedings of the 38th International Conference on Neural Information Processing Systems, Vancouver, Canada, 2024: 3117.
[41]	DING Ao, QIN Yong, WANG Biao, et al. An elastic expandable fault diagnosis method of three-phase motors using continual learning for class-added sample accumulations[J]. IEEE Transactions on Industrial Electronics, 2024, 71(7): 7896–7905. doi: 10.1109/TIE.2023.3301546.
[42]	SU Yongtao, LIU Yaoqi, ZHOU Yiqing, et al. Broadband LEO satellite communications: Architectures and key technologies[J]. IEEE Wireless Communications, 2019, 26(2): 55–61. doi: 10.1109/MWC.2019.1800299.
[43]	WANG Jialin, ZHANG Jianhua, SUN Yutong, et al. Electromagnetic wave property inspired radio environment knowledge construction and artificial intelligence based verification for 6G digital twin channel[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(2): 260–277. doi: 10.1631/FITEE.2025.2400464.
[44]	YUAN Chunjing, LEI Tong, XUE Ze, et al. Service decoupling for open and intelligent service-based RAN[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(2): 230–245. doi: 10.1631/FITEE.2025.2400248.
[45]	ZHANG Ruichen, LIU Guangyuan, LIU Yinqiu, et al. Toward edge general intelligence with agentic AI and agentification: Concepts, technologies, and future directions[J]. IEEE Communications Surveys & Tutorials, 2026, 28: 4285–4318. doi: 10.1109/COMST.2026.3651702.
[46]	LI Meng, HUANG Yudian, YU F R, et al. Ambient backscatter communication-assisted intelligent resource management for green industrial IoT[J]. IEEE Wireless Communications, 2025, 32(1): 174–181. doi: 10.1109/MWC.001.2400040.
[47]	CHEN Tianjiao, WANG Xiaoyun, HUA Meihui, et al. Incentive-based task offloading for digital twins in 6G native artificial intelligence networks: A learning approach[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(2): 214–229. doi: 10.1631/FITEE.2025.2400240.
[48]	LIU Guangyi, DENG Juan, ZHU Yanhong, et al. 6G autonomous radio access network empowered by artificial intelligence and network digital twin[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(2): 161–213. doi: 10.1631/FITEE.2025.2400569.
[49]	TONG Zhan, SONG Yibing, WANG Jue, et al. VideoMAE: Masked autoencoders are data-efficient learners for self-supervised video pre-training[C]. Proceedings of the 36th International Conference on Neural Information Processing Systems, New Orleans, USA, 2022: 732.
[50]	WANG Limin, HUANG Bingkun, ZHAO Zhiyu, et al. VideoMAE V2: Scaling video masked autoencoders with dual masking[C]. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, Canada, 2023: 14549–14560. doi: 10.1109/CVPR52729.2023.01398.
[51]	FEICHTENHOFER C, FAN Haoqi, MALIK J, et al. SlowFast networks for video recognition[C]. 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea (South), 2019: 6201–6210. doi: 10.1109/ICCV.2019.00630.
[52]	PENG Yifan, LEE J, and WATANABE S. I3D: Transformer architectures with input-dependent dynamic depth for speech recognition[C]. ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, 2023: 1–5. doi: 10.1109/ICASSP49357.2023.10096662.
[53]	SANDLER M, HOWARD A, ZHU Menglong, et al. MobileNetV2: Inverted residuals and linear bottlenecks[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4510–4520. doi: 10.1109/CVPR.2018.00474.