Performance Investigation on Cell Selection Schemes Associated with Downlink Inter-Cell Interference Coordination in Heterogeneous Networks for LTE-Advanced

Yuya SAITO  Jaturong SANGIAMWONG  Nobuhiko MIKI  Satoshi NAGATA  Tetsushi ABE  Yukihiko OKUMURA  

Publication
IEICE TRANSACTIONS on Communications   Vol.E94-B   No.12   pp.3304-3311
Publication Date: 2011/12/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.E94.B.3304
Print ISSN: 0916-8516
Type of Manuscript: Special Section PAPER (Special Section on Cooperative Communications for Cellular Networks)
Category: 
Keyword: 
LTE-advanced,  heterogeneous network,  cell selection,  reference signal received power,  reference signal received quality,  

Full Text: PDF>>
Buy this Article



Summary: 
In Long-Term Evolution (LTE)-Advanced, a heterogeneous network in which femtocells and picocells overlay macrocells is being extensively discussed in addition to traditional well-planned macrocell deployment to improve further the system throughput. In heterogeneous network deployment, cell selection as well as inter-cell interference coordination (ICIC) is very important to improve the system and cell-edge throughput. Therefore, this paper investigates three cell selection methods associated with ICIC in heterogeneous networks in the LTE-Advanced downlink: Signal-to-interference plus noise power ratio (SINR)-based cell selection, reference signal received power (RSRP)-based cell selection, and reference signal received quality (RSRQ)-based cell selection. The results of simulations (4 picocells and 25 sets of user equipment are uniformly located within 1 macrocell) that assume a full buffer model show that the downlink cell and cell-edge user throughput levels of RSRP-based cell selection are degraded by approximately 2% and 11% compared to those for SINR-based cell selection under the condition of maximizing the cell-edge user throughput due to the impairment of the interference level. Furthermore, it is shown that the downlink cell-edge user throughput of RSRQ-based cell selection is improved by approximately 5%, although overall cell throughput is degraded by approximately 6% compared to that for SINR-based cell selection under the condition of maximizing the cell-edge user throughput.