Research Area: Machine Learning
User-centric (UC) clustering, an innovative design principle for ultradense networks (UDNs), supports dynamically fluctuating adaptive network topologies. In this article, we introduce the UC-UDN architecture and provide a tutorial on UC clustering design by generalizing the problem under practical constraints. In the context of UC clustering, we briefly present various promising methods that represent constraint options and also provide a pair of case studies focused on design tradeoffs. Finally, the salient future directions of UCUDNs are identified.
Keywords:
Handover
Measurement
Network topology
Computer architecture
Optimization
Reinforcement learning
Author(s) Name: Yan Lin; Rong Zhang; Luxi Yang; Chunguo Li
Journal name: IEEE Vehicular Technology Magazine
Conferrence name:
Publisher name: IEEE
DOI: 10.1109/MVT.2019.2903741
Volume Information: Volume: 14
