Envisioning Device-to-Device Communications in 6G
文章目录
本次分享的文献
Envisioning Device-to-Device Communications in 6G
Citation:
Zhang S , Liu J , Guo H , et al. Envisioning Device-to-Device Communications in 6G[J]. IEEE Network, 2020, PP(99):1-6.
第六代移动通信展望
未来AI将被大规模运用于6G网络的管理
It is believed that AI will be the most innovative technique that can achieve intelligent automated network operations, management and maintenance in future complex 6G networks.
D2D通信将被纳入6G通信
Driven by AI techniques, device-to-device (D2D) communication will be one of the pieces of the 6G jigsaw puzzle.
6G网络的特点
the future sixth generation (6G) mobile network is expected to be an innately intelligent, highly dynamic, ultradense heterogeneous network that interconnects all things with extremely low-latency and high speed data transmission.
Topic
intelligent D2D communication
缩写词
mobile user equipments (UEs)
personal mobile workstations
Moore’s law 摩尔定律
5G的困局
网络密度增加导致的管理难度增大
通信距离缩短导致网络密度增加
Since higher frequency bands inevitably suffer from high path-loss, the data transmission distance will become much shorter.
Although 5G has benefited greatly from ultradense networks and device-to-device (D2D) communication, further network densification in 6G will encounter many serious problems and challenges, like severe interference, very complex resource management, a vast amount of signaling, prohibitively high cost and energy consumption.
无法满足物联网的需求
it is envisioned that 5G cannot fulfill the requirements of emerging Internet of Everything applications like augmented reality (AR), virtual reality (VR), and mixed reality (MR), which require a convergence of communication, sensing, control, and computing functionalities.
6G相较5G的进步
架构和性能的进步:
Current research trends have shown that 6G network architecture at least has the following characteristics:
- space-air-terrestrial-sea integrated (SATSI),
- artificial intelligence (AI) driven,
- higher frequency bands,
- ultradense heterogeneous,
- green energy, security and privacy.
Space-Air-Terrestrial-Sea Integrated Network: 范围更广
相对于5G的进步：通信网覆盖范围扩大
Relying only on current terrestrial networks cannot fulfill the requirements of extremely broad coverage and ubiquitous connectivity. Therefore, 6G will further integrate sea/underwater networks with respect to 5G space-air-terrestrial networks,
特点：
broad coverage and ubiquitous connectivity
组成：
a space network：satellites
an air network：high altitude platform
low altitude UAV
heterogeneous ultradense terrestrial network
功能：
The SATSI network can make full use of the characteristics of global network coverage, full-frequency radio transmission and full application to achieve seamless high-speed communication in space, air, terrestrial and sea domains.
其中的D2D应用
to support low latency and high speed data transmission.
innately intelligent, highly dynamic networks：管制更加动态与智能
概述：researchers have pointed out that the utilization of AI techniques can help network operators to realize intelligent automated network operation, management and maintenance.
挑战：现有网络管制存在的问题
As 6G network will be much more complex and dynamic than the preceding generations of wireless networks, traditional network management methods will become untenable.
AI如何辅助网络运维和优化：
1.multi-level distributed AI :
global AI, local AI and on-device AI will cooperate and leverage one another’s advantages to maintain the network operation and optimization：
- the local AI center can be embedded into traditional MBSs or MEC servers so as to provide AI proceeding on local network management.
- UEs like smart phones may provide opportunities for on-device data training, which can sense and learn from the local channel pattern, traffic pattern, moving trajectory, etc.,
2.sensing and big data training:
By sensing and network big data training, AI techniques like machine learning (ML) will make complex 6G networks more intelligentand can achieve a high level of efficient network management and optimization, such as network environment sensing, status predicting, proactive configuring, dynamic optimizing, self-healing, etc.
AI的具体应用环境
Intelligent resource allocation, network optimization, small BSs or UAVs deployment, network design.
extremely heterogeneous and ultradense network： 密度更大
网络密度提高带来的挑战：
Note that further network densification will lead to performance degradation due to the severe interference, prohibitively high cost and energy consumption.
Ultradense networks may lead to frequent handover for high speed moving UEs.
D2D communication作用：
to efficiently support a much larger and more diverse set of UEs and applications.
IntellIgent D2D-enhanced mobile edge computing: 传输时延更小
概述：AI will gradually migrate from cloud to network edge, wherein multi-level AI will be employed. The end-to-end transmission time will be greatly reduced due to the extremely high data rate in 6G,
”任务下放，终端合作“模式：
As the capability of single UE is relatively weak, computational intensive tasks may inevitably be allocated to multiple UEs.
AI让资源分配更智能
The key and difficult point to employ D2D-enhanced MEC lies in the optimal management and allocation of network communication and computation resources. We envision that AI techniques will be employed to make the resource allocation and optimization intelligent.
By sensing and network big data training, AI techniques like machine learning (ML) will make complex 6G networks more intelligent and can achieve a high level of efficient network management and optimization, such as network environment sensing, status predicting, proactive configuring, dynamic optimizing, self-healing, etc. AI will gradually migrate from cloud to network edge, wherein multi-level AI will be employed.
移动终端的进步: They can be regarded as personal mobile workstations
硬件系统的大幅提升：
UEs will equipped with powerful high-end processors, extraordinary rich storage resources, plenty of sensors and ultra-long lifetime batteries.
能够在本地处理复杂信息：
smartphones can be regarded as personal mobile workstations that enable on-device AI processing for intelligent networking in the future.
We envision that future UEs will be designed as a type of hand-held mobile computer workstation that can perform intense processes like machine learning, environment sensing, holography processing, AR/VR cloud gaming and photonic computing, while keeping the same thin and portable designs as a phone.
算力更强，从smart phone到AI-dirven smart phone：
from smart mobIle phone to AI-drIven mobIle phone：
Smart phones will have computational capabilities that are on the order of the human brain in the near future. UEs like smart phones may provide opportunities for on-device data training, which can sense and learn from the local channel pattern, traffic pattern, moving trajectory, etc.
本文地址：https://blog.csdn.net/Jinyindao243052/article/details/108785036