9th International Conference on Communications and Networking in China

Research Article

Hybrid Bandwidth Allocation for Energy Saving Based on Large-Scale User Behavior in Heterogeneous Networks

  • @INPROCEEDINGS{10.4108/icst.chinacom.2014.256347,
        author={Xing Zhang and Yu Huang and Wenbo Wang and Chih-Lin I and Gang Li and Yami Chen},
        title={Hybrid Bandwidth Allocation for Energy Saving Based on Large-Scale User Behavior in Heterogeneous Networks},
        proceedings={9th International Conference on Communications and Networking in China},
        publisher={IEEE},
        proceedings_a={CHINACOM},
        year={2015},
        month={1},
        keywords={energy saving bandwidth allocation heterogeneous network (hetnet) user behavior},
        doi={10.4108/icst.chinacom.2014.256347}
    }
    
  • Xing Zhang
    Yu Huang
    Wenbo Wang
    Chih-Lin I
    Gang Li
    Yami Chen
    Year: 2015
    Hybrid Bandwidth Allocation for Energy Saving Based on Large-Scale User Behavior in Heterogeneous Networks
    CHINACOM
    IEEE
    DOI: 10.4108/icst.chinacom.2014.256347
Xing Zhang,*, Yu Huang1, Wenbo Wang1, Chih-Lin I2, Gang Li2, Yami Chen2
  • 1: BUPT
  • 2: China Mobile Research Institute
*Contact email: hszhang@bupt.edu.cn

Abstract

One of the main purpose for heterogeneous networks (HetNet) is to reduce the power consumption for various spectrum allocation schemes. Large-scale user behavior in a two-tier HetNet is considered for the energy-efficient spectrum allocation, which characterizes the macro BSs that guarantee traffic requirements in non-hotspot regions and micro BSs that guarantee traffic requirements in hotspot regions. Two hybrid bandwidth allocation schemes are presented for the shared/orthogonal frequency band and the high/low frequency band, where the large-scale user behavior is taken into account. We present closed-form formulas of bandwidth allocation schemes which establish the quantitative relationship between the large-scale user behavior and bandwidth allocation strategy. Both theoretical and simulation results show that the proposed schemes are effective in guaranteeing both the demands of the large-scale user behavior and keeping low power consumption.