Cell Zooming for Cost-Efficient Green Cellular Networks

I NTRODUCTION

Wireless cellular networks have been growing

rapidly in the last few decades. The subscriber

number and traffic volume in cellular networks

have explosively increased. Network operators are

always trying their best to satisfy user require-

ments cost-efficiently. In each cell, the base station

(BS) transmits common control signals and data

signals to mobile users (MUs), and the cell size is

defined as the area in which MUs can receive con-

trol signals from the BS. At the stage of network

planning, cell size and capacity are usually fixed

based on the estimation of peak traffic load.

However, traffic load in cellular networks can

have significant spatial and temporal fluctuations

due to user mobility and bursty nature of many

data applications [1]. For example, for a cellular

network in a city, the traffic load in the daytime is relatively heavy in office areas and light in res-

idential areas, while the opposite things happen

in the evening. If the capacity is planned based

on the peak traffic load for each cell, there are

always some cells under light load, while others

are under heavy load. In this case, any static cell

deployment will not be optimal as traffic load

fluctuates. Traffic load fluctuations can be even

more serious as the next generation cellular net-works move towards smaller cells such as micro-

cells, pico-cells, and femto-cells, which make the cell deployment even harder.

On the other hand, traffic load fluctuations

can also contribute to cellular networks if we

know the discipline of the variation. For exam-ple, in some parts of a cellular network, traffic load increases to be higher than the planned capacity, then some MUs will be unable to get services. In the meantime, traffic load in the neighboring cells is light. Load balancing schemes can be used to satisfy user requirements as far as possible [2, references therein].Energy consumption has become one of the most important issues in the world, as the car-bon emissions of energy sources have great negative impact on the environment, and the price of energy is also increasing. Network operators are considering how to reduce the energy consumption and design green cellular networks. The large number of BSs contribute a major portion of the energy consumption of cellular networks. When a BS is in its working mode, the energy consumption of processing circuits and air conditioner takes up about 60percent of the total consumption [3]. There-fore, by merely controlling the transmit power of radio equipments, the effect of energy saving is marginal. However, most of the efforts for energy saving in cellular networks still focus on reducing the transmit power of BSs and MUs.To save the energy of the whole network, the phenomenon of traffic load fluctuation implies that some BSs can be switched off when the traffic load is light. There have been many switching on/off schemes proposed in both academia and industry [4–7].In this article, we propose a new concept of cell zooming , which adaptively adjusts the cell size according to traffic conditions. Cell zooming has the potential to balance the traffic load and

reduce the energy consumption. An example of cell zooming is illustrated in Fig. 1. It is a cellular

network with five cells. One central cell is sur-

rounded by four neighboring cells. BSs are locat-ed at the respective center of the cells, denoted by hollow squares; MUs are randomly distributed in the cells, denoted by solid dots. When some MUs move into the central cell and make it con-gested, the central cell can zoom in to reduce the cell size and therefore release from the conges-tion (Fig. 1b). On the contrary, if some MUs move out of the central cell and cause the neigh-boring cells congested, the neighboring cells can zoom in and the central cell zooms out to avoid any possible coverage hole. If the neighboring cells are designed to have high capacity, and

IEEE Communications Magazine ? November 2010A BSTRACT

Cell size in cellular networks is in general

fixed based on the estimated traffic load. Howev-

er, the traffic load can have significant spatial

and temporal fluctuations, which bring both chal-

lenges and opportunities to the planning and

operating of cellular networks. This article intro-

duces a concept of cell zooming, which adaptively

adjusts the cell size according to traffic load, user

requirements and channel conditions. The imple-

mentation issues of cell zooming are then pre-

sented. Finally a usage case of cell zooming for

energy saving is investigated. Centralized and dis-

tributed cell zooming algorithms are developed,

and simulation results show that the proposed

algorithms can greatly reduce the energy con-

sumption, which leads to green cellular networks.

Zhisheng Niu, Yiqun Wu, Jie Gong, and Zexi Yang, Tsinghua University

Cell Zooming for Cost-Efficient Green Cellular Networks

75

IEEE Communications Magazine ? November 2010

77

IEEE Communications Magazine ? November 2010

?Step 1: Initialize all the L j to be 0, and all the elements in matrix X to be 0.

?Step 2: For each MU i , find the set of BSs who can serve MU i without violating the bandwidth constraints, which means b ij ≤B ~

j

. If the set is empty, MU

is blocked.

Otherwise, associate MU i with a BS has the highest w ij in the set. Update X after each association.

?Step 3: Sort all the BSs by the ratio of to B ~

j by increasing order. All the BSs with the ratio 0 will zoom in to zero and work in sleep mode in the following serving period.For other BSs, find the BS j with the small-est ratio, and re-association the MUs in to other BSs in the network. If no MU is blocked, undate X and go to Step 3. Other-wise, output X and end the procedure.

D ISTRIBUTED A LGORITHM

To reduce the information exchange and signaling overhead, we also propose a distributed cell Figure 5.Traffic distribution in the tested cellular network layout.

x -axis (m)

500

500

1000

Medium

Low load

High load

1500

2000 2500

1000 1500 2000 2500 3000

79

本文来源：https://www.bwwdw.com/article/sgjq.html