![](https://csdnimg.cn/release/download_crawler_static/15548571/bg1.jpg)
Interference coordination strategy based on
Nash bargaining for small-cell networks
ISSN 1751-8628
Received on 6th December 2014
Revised on 2nd March 2015
Accepted on 7th April 2015
doi: 10.1049/iet-com.2014.1186
www.ietdl.org
Guanding Yu
1
, Yang Xu
1
, Rui Yin
2
, Fengzhong Qu
3
✉
1
Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, People’s Republic of China
2
College of Information and Electronic Engineering, Zhejiang Gongshang University, Hangzhou 310018, People’s Republic of China
3
Department of Ocean Science and Engineering, Zhejiang University, Hangzhou 310027, People’s Republic of China
✉ E-mail: jimqufz@zju.edu.cn
Abstract: In this study, a distributed scheme based on the Nash bargaining model is designed to coordinate co-channel
interference for small-cell networks. The authors consider a scenario that resource blocks can be r eused among
different small cells. Different to ex isting works where resource allocation i s conducted at the base stations, t hey
propose the scheme where user initialises resource bargaining request to the serving base stati on once its quality-of-
service cannot be satisfied because of t he severe co-channel interference from other users. Since the gene ral
bargaining problem is a non-linear integer optimisation, the genetic algorithm is utilised to solve it. They also develop
a low-complexity bargai ning mode l which only takes in to account the strongest co-channel interference. Simulation
results show that the proposed distributed scheme can effectively reduce the outage probability of users and improve
the system throu ghput. In addition, the proposed low -complexity bargaining solution can achieve a close performance
to the genetic algorithm-based solution.
1 Introduction
Recently, exponentially increased mobile users with diverse
quality-of-service (QoS) requirement have caused tremendous data
traffic growth in wireless communication systems, which will
continue to increase greatly. To fulfil such demand, three
alternative ways have been proposed: improving the spectral
efficiency, finding more usable frequency spectrum and increasing
the network density. It has been revealed that network
densification can bring more performance improvement than the
other two [1]. Therefore a small cell technique has been widely
investigated as a promising solution for the next-generation
cellular networks [2, 3].
One of the important issues in small-cell networks is the
coordination of co-channel interference caused by spectrum reuse,
which generally includes inter-tier interference between macro
base stations (MBSs) and small cell base stations (SBSs) and
intra-tier interference among SBSs. Owing to geographically
randomness and unplanned deployment, the SBS may generate
great interference to its neighbour cells using the same spectrum
resource, which results in the degradation of users’ QoS, especially
in high dense small-cell networks.
To avoid or alleviate interference, several mechanisms have been
developed in the literature. The enhanced inter-cell interference
coordination has been proposed by the 3rd generation partnership
project (3GPP), and some related methods have been designed in
[4, 5]. In [6], a beam subset selection strategy has been proposed
to reduce interference for a two-tier femtocell network. A
distributed signal-to-interference-plus-noise ratio (SINR)
adaptation method to alleviate inter-tier interference has been
developed in [7].
Another effective method to cope with the interference issue is
resource allocation or resource block (RB) allocation in long-term
evolution (LTE) systems. Generally, RB allocation can be realised
in both centralised and distributed ways. In centralised algorithms,
the MBS has to collect all the channel state information (CSI) to
design the resource allocation strategy. In [8], an energy-efficient
resource allocation algorithm has been developed, taking into
account the interference constraint. A graph-based interference
coordination scheme has been proposed to maximise the system
throughput while guaranteeing that interference is under control, as
well as proportional rate fairness among SBSs [9]. In [10], a
fine-scale physical RB allocation algorithm has been proposed for
effective interference management.
In general, centralised algorithms require tremendous signalling
overhead and complex computation, especially in high dense
networks [11]. Thus, distributed resource allocation algorithms are
much more applicable for small-cell networks, which also have the
merits of strong robustness and good expandability. In [12], a
distributed resource management framework has been proposed for
SBS to opportunistically determine its available resources to
control interference. An adaptive and distributed interference
coordination algorithm has been proposed, which decomposes the
multi-cell resource allocation problem into distributed single-cell
problems [13].
Recently, game theory has been utilised to design interference
coordination strategies for small-cell networks. The Stackelberg
game model has been applied to design a price-based power
allocation algorithm in two-tier femtocell networks [14], and the
repeated game has been used to develop the energy-efficient power
control mechanism [15]. In [16], the authors have proposed a
distributed non-cooperative game to realise sub-channel
assignment, adaptive modulation and power control in multi-cell
orthogonal frequency division multiple access (OFDMA)
networks. In [
17], sequential game and Nash bargaining model
have been used to design a cooperation framework for mobile
operators and fixed-line operators.
In most of the work mentioned above, resource allocation
algorithm is performed at base stations. However, base stations
usually do not know the interference and service condition of
users, and tremendous signalling overhead will be required to
obtain this information periodically. Therefore, in this paper, we
propose the scheme where resource bargaining is initialled at the
user side. Once the QoS of the user cannot be satisfied because of
the severe co-channel interference from other SBSs, a resource
allocation request will be sent to the serving base station to
bargain for more resources or reallocating the RBs. We utilise the
Nash bargaining model to design the distributed RB allocation
IET Communications
Research Article
IET Commun., 2015, Vol. 9, Iss. 13, pp. 1583–1590
1583
&
The Institution of Engineering and Technology 2015