On the Impact of Network Parameters on the Efficiency of Spectrum Allocation in
Cognitive Radio Networks
Abdelbaset S. Hamza
Dept of Electronics & Communication Engineering
Cairo University
Giza, Egypt
bhamza@ieee.org
Abstract—In this work, we study the effect of various
network parameters on the efficiency of the spectrum allocation
problem in cognitive radio networks. In this poster, we present
the results for the impact of the transmission range of the
primary users on the solution quality of the spectrum allocation
problem. A Genetic Algorithm (GA) is implemented and used
to solve the spectrum allocation problem under two utilization
functions, namely, the Mean-Reward (MR), and the Max-
Proportional-Fair (MPF). Simulation results are presented and
their implications are discussed.
Keywords-Wireless networks, spectrum allocation, open spec-
trum, cognitive radio, genetic algorithm.
I. I NTRODUCTION
Efficient utilization of open spectrum in cognitive radio
networks requires appropriate allocation of idle spectrum
frequency bands that are not utilized by licensed users (pri-
mary users) among coexisting cognitive radios (secondary
users) while minimizing interference [1]. This problem is
known as resource allocation in cognitive radio networks,
and is shown to be NP-hard [2]. Accordingly, several heuris-
tics were proposed to solve the resource allocation problem
based on game theory [3], pricing and auction mechanisms
[4], vertex labeling [2], and evolutionary algorithms [1].
The objective of our work is to investigate the impact
of the various design parameters of cognitive networks on
the efficiency of the spectrum allocation under different uti-
lization functions. A genetic algorithm (GA) is implemented
and used to solve the spectrum allocation problem under two
utilization functions, namely, the Mean-Reward (MR), and
the Max-Proportional-Fair functions(MPF). In this poster,
we present the GA solutions under different values of the
transmission range used by the primary users.
II. SYSTEM MODEL AND PROBLEM STATEMENT
Figure 1 shows a sample cognitive radio network. A
typical cognitive network consists of a set of primary users
each is assigned a channel selected from a pool of
orthogonal, non-overlapping spectrum bands that differ in
bandwidth and transmission range. Each channel is asso-
ciated with transmission range
(, ), which is assumed
here to be the same for all channels for simplicity in analysis.
S
4
S
1
S
5
S
3
S
2
P
1
P
4
P
5
P
3
P
2
Primary User
Secondary User
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Available Spectrum
Unavailable Spectrum
Figure 1. Structure of a simple cognitive radio network.
There are coexisting secondary users that are planned
to utilize these idle channels occupied by primary users in
order to provide their services. A secondary user can be
a wireless access point. It is assumed that each secondary
user can utilize multiple channels at one time, but limited
to the radio interface constraint. The following are the key
components in the used model [2]:
∙ Channel reward: = {
,
}
×
, a matrix represent-
ing the channel reward:
,
represents the maximum
reward that can be acquired by user using channel
.
∙ Conflict free channel assignment: Let
= {
,
∣
,
∈ {0, 1}}
×
, a matrix that
represents the assignment:
,
=1 if channel is
assigned to user .
∙ User Reward: ℜ = {
=
∑
−1
=0
,
⋅
,
}
×1
represents the reward vector that each user gets for a
given channel assignment. In our context, the reward is
the coverage area of the secondary user.
We consider two utilization functions [ (ℜ)], namely,
Mean-Reward (MR), and Max-Proportional-Fair (MPF) [2].
=
1
−1
∑
=0
=
1
−1
∑
=0
−1
∑
=0
, ⋅ , (1)
2011 Eighth International Conference on Information Technology: New Generations
978-0-7695-4367-3/11 $26.00 © 2011 IEEE
DOI 10.1109/ITNG.2011.194
1074