Traffic-Aware Network Planning and Green Operation with BS Sleeping and Cell Zooming

LANGUAGE English

SOURCE   IEICE Trans. Commun, Vol.E97-B, NO.11, PP. 2337-2346, November 2014

Published Date:2014-11

ABSTRACT

The traffic load of cellular networks varies in
both time and spatial domains, causing many base stations
(BS) to be under-utilized. Assisted by cell zooming, dynamic
BS sleep control is considered as an effective way to improve
energy efficiency during low traffic hours. Therefore, how
densely the BSs should be deployed with cell zooming and BS
sleeping is an important issue. In this paper, we explore the
energy-optimal cellular network planning problem with dynamic
BS sleeping and cell zooming for the cases in which traffic is
uniformly distributed in space but time-varying. To guarantee
the quality of multi-class services, an approximation method
based on Erlang formula is proposed. Extensive simulations
under our predefined scenarios show that about half of energy
consumption can be saved through dynamic BS sleeping and
power control. Surprisingly, the energy-optimal BS density we
obtained is larger than the one without considering BS sleeping.
In other words, deploying more BSs may help to save energy if
dynamic BS sleeping is executed.
key words: Base station (BS) sleeping, cell zooming, traffic-
aware, network planning, energy efficiency