Radio Resource Management in Small Cell Networks

Abstract

AbstractUltra‐dense small cell (UDSC) network will play a key role to cope with the capacity issue for 5G cellular mobile systems and beyond because future broadband mobile applications require high‐speed transmission and low latency. Nevertheless, deploying small cell deployment will need to face the challenges of severe interference and excessive energy consumption in very dynamic environments. To address these issues, in this article we introduce a data‐driven resource management (DDRM) framework combined with power control, channel rearrangement, and dynamic antenna clustering (DAC). In particular, an unsupervised learning affinity propagation clustering is applied to UDSC to identify cluster heads, i.e. the cell causing most serious interference to its neighboring cochannel cells. Then, these cluster heads will lower its transmission power. We call this learning‐based interference management approach as affinity propagation power control (APPC). We show that DDRM with APPC can improve energy efficiency and reduce interference for UDSC, while taking account of cell switching on/off, transmission power adjustment, and traffic loads simultaneously.