Professional Documents
Culture Documents
Abstract—Cognitive radio (CR) is a promising wireless based on constraining the total and transmit powers of each
paradigm that provides efficient spectral usage. Orthogonal subchannel. Another power allocation algorithm with partial
frequency division multiplexing (OFDM) is a potential technology channel state information is proposed in [8] where the WF
providing many advanced functionalities in terms of power
and rate control for cognitive radio networks (CRNs). Power process is run only once, and then, the final power allocation
allocation for CRNs is a crucial task for better interference vector is directly calculated to maximize the SUs throughput.
management. In this paper, a subcarrier assignment scheme In order to maximize the downlink capacity of CRNs, a
and a novel power allocation algorithm using geometric water suboptimal power allocation algorithm based on linear water-
filling is presented for OFDM based CRNs. This algorithm is filling (LWF) scheme is proposed in [9] that considers inter-
optimized such a way to maximize the sum rate of secondary
users by allocating power more efficiently, while constraining the ference and transmit power constraints. Another suboptimal
1) total transmit power, 2) individual subchannel transmit power power allocation scheme is proposed in [10] that reduces the
as well as 3) individual subcarrier peak power of secondary users, computational complexity of determining the water level to
for a given interference level to the primary users. Numerical allocate the power for each subcarrier under a peak power
results show that this algorithm provides better utilization of constraint. For weighted sum rate maximization in CRNs, a
power resources thus maximizes the sum rate than the existing
algorithms. power limited multilevel WF algorithm has been proposed
Key words: Cognitive radio, power allocation, OFDM, geo- in [11] where subcarrier power constraints were added to the
metric water filling. traditional problem. With the total power constraint and the
power constraint on each subchannel, an iterative partitioned
I. INTRODUCTION water-filling (IPWF) algorithm was proposed in [12] to realize
Spectrum scarcity is becoming a crucial challenge with ever optimal power allocation in OFDM based CRNs. However,
increasing demand for services in the wireless communication in the above studies, they did not consider individual peak
paradigm. While there is constant push for new spectrum, power constraints for the subcarriers. To the best of author
Federal Communication Commissions (FCC) observes that knowledge, none of the papers in the literature yet considers
current licensed spectrum is significantly underutilized [1]. individual peak power constraints for subcarriers together with
Cognitive radio (CR), first coined by Joseph Mitola [2], total power constraint and the power constraint on each sub
has been proposed as a solution for efficiently utilizing radio channel.
resources of the exsiting spectrum. CRs, with its ability The main contributions of this paper are: the power alloca-
to smartly interact with the surrounding environment, are tion problem in OFDM-based CRNs is modeled considering
amenable to allow the coexistence licensed (primary) users both individual subcarrier peak power constraints and sub-
(PU) and unlicensed (secondary) users (SU) sharing the same channel power constraints with total power constraint. In this
bandwidth opportunistically without causing harmful interfer- work, a more efficient algorithm, named as iterative partitioned
ence to each other. weighted geometric water-filling with individual peak power
However, appropriate power, modulation and data rate man- constraints is proposed along with subcarrier assignment. The
agement impose unique challenges in opportunistic spectrum proposed algorithm is shown to maximize the sum rate by pro-
access in cognitive radio networks (CRNs). Popular orthogonal viding better utilization of power resources than the existing
frequency division multiplexing (OFDM) better manages these algorithms. Also power allocation is performed considering
issues and is a potential technology for CRNs too [3]–[5]. The the weight factor of each channel.
subcarrier configurability of OFDM allows SUs to flexibly fill The organization of the rest of this paper is as follows:
the spectral holes left by PUs without causing unacceptable section II provides the idea behind the geometric water-
interference. Also, the fast Fourier transform (FFT) component filling approach. Section III describes the system model and
of OFDM systems provides an additional tool for opportunistic formulates the problem along with the proposed solutions.
spectrum detection. Simulation results are depicted in section IV and finally section
Power allocation problem in OFDM based CRNs has been V concludes the paper.
widely studied under different settings. A grouped power
allocation scheme is presented in [6] where at the first stage, II. G EOMETRIC WATER - FILLING M ETHOD
power is assigned to some groups considering interference The conventional water-filling (CWF) [13] problem can
constraints with the PU band. At the second stage, the re- be described as follows: given 𝑃𝑇 > 0, is the total signal
maining power is allocated to others with water-filling (WF) power (or volume of the water); the allocated power and the
algorithm. A low-complexity algorithm using incrementing propagation path gain for the 𝑖𝑡ℎ channel are given as 𝑃𝑖 and
or decrementing power is proposed in [7]. This algorithm is ℎ𝑖 respectively where, 𝑖 = 1...𝑁 ; and 𝑁 is the total number
𝑖=1
ℎ𝑛 ℎ𝑖
D
Fading
40
Power
20
power resources in OFDM based CRNs, is investigated. The
20
10
allocation is done according to total power constraint, indi-
0 0 vidual subchannel power constraint, and individual subcarrier
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 IPWF GWFPP IGPP
Subcarriers peak power constraint. Numerical examples are provided to
illustrate the effectiveness of the algorithm. The proposed al-
Fig. 3: Power Allocation comparison for three schemes: (a)
gorithm allows better utilization of available power resources,
IPWF; (b) GWFPP; (c) IGPP; (d) achieved sum rate for three
thus maximizes the sum rate than IPWF and GWFPP. The
schemes. algorithm allocated more power to wireless channels that have
less fading, saving power resources.
R EFERENCES
[1] Federal Communications Commission, “Spectrum policy task force,” pp.
02 – 135, Nov 2002.
[2] J. Mitola, “Cognitive radio: an integrated agent architecture for software
defined radio,” PhD thesis, KTH Royal Institute of Technology, Stock-
holm, Sweden, 2000.
[3] T. Weiss and F. Jondral, “Spectrum pooling: An innovative strategy for
enhancement of spectrum efficiency,” IEEE Comm. Mag., vol. 42, no. 3,
pp. 8 – 14, Mar 2004.
[4] U. Berthold and F. Jondral, “Guidelines for designing OFDM overlay
systems,” in Proc. 1st IEEE Int. Symp. New Frontiers in DySPAN, pp.
626 – 629, Nov 2005.
[5] H. Tang, “Some physical layer issues of wide-band cognitive radio
systems,” in Proc. 1st IEEE Int. Symp. New Frontiers in DySPAN, pp.
151 – 159, Nov 2005.
[6] E. Hosseini and A. Falahati, “Power allocation grouping scheme con-
sidering constraints in two separate stages for OFDM-based cognitive
radio system,” in Proc. IEEE Int. Conf. Electrical Information and
Communication Technology (EICT), pp. 1 – 6, Feb 2014.
[7] Q. Qi, A. Minturn, and Y. Yang, “An efficient water-filling algorithm
for power allocation in OFDM-based cognitive radio systems,” in Proc.
IEEE International Conference on Systems and Informatics (ICSAI), pp.
Fig. 4: Power allocation by IGPP considering weight factors 2069 – 2073, May 2012.
[8] X. Zhou, B. Wu, P.-H. Ho, and X. Ling, “An efficient power allocation
Subchannel-1 Subchannel-2
40 20 algorithm for OFDM based underlay cognitive radio networks,” in Proc.
IEEE Global Telecommunications Conference (GLOBECOM), pp. 1 –
Allocated Power
Allocated Power
IPWF
30
GWFPP 15
5, Dec 2011.
IGPP [9] W. Jian, Y. Longxiang, and L. Xu, “Resource allocation based on
20 IPWF linear waterfilling algorithm in CR systems,” in Proc. IEEE Wireless
10 GWFPP
10 IGPP Communications, Networking and Mobile Computing (WiCOM), pp. 1
– 4, Sept. 2011.
0 5 [10] C.-H. Chen and C.-L. Wang, “An efficient power allocation algorithm for
1 2 3 4 1 2 3
multiuser OFDM-based cognitive radio systems,” in Proc. IEEE Wireless
Subcarriers Subcarriers
Subchannel-3 Subchannel-4
Communications and Networking Conference (WCNC), pp. 1 – 6, April
25 15 2010.
[11] Z. Tang, G. Wei, and Y. Zhu, “Weighted sum rate maximization for
Allocated Power
Allocated Power
20 IPWF IPWF
GWFPP 10 GWFPP OFDM-based cognitive radio systems,” Telecommun Syst, vol. 42, pp.
15 IGPP IGPP 77 – 84, Oct. 2009.
10 [12] P. Wang, M. Zhao, L. Xiao, S. Zhou, and J. Wang, “Power allocation in
5
5
OFDM-based cognitive radio systems,” in Proc. IEEE Global Telecom-
munications Conference (GLOBECOM), pp. 4061 – 4065, Nov 2007.
0
1 2 3 4
0
1 2 3 4
[13] D. P. Palomar and J. R. Fonollosa, “Practical algorithms for a family of
Subcarriers Subcarriers water-filling solutions,” IEEE Transaction on Signal Processing, vol. 53,
no. 2, pp. 686 – 695, Feb 2005.
[14] P. He, L. Zhao, S. Zhou, and Z. Niu, “Water-Filling: A geometric ap-
Fig. 5: Optimal power allocation vs subcarriers for each proach and its application to solve generalized radio resource allocation
subchannel using IPW, GWFPP and IGPP problems,” in IEEE transactions on Wireless Communications, pp. 3637
– 3646, July 2013.
subcarriers according to their noise power. Thus, the subcarrier [15] Q. Zhao and B.M.Sadler, “A survey of dynamic spectrum access: signal
processing, networking, and regulatory policy,” IEEE Signal Processing
that has highest noise power, comes first and the subcarrier that Magazine, vol. 55, no. 5, pp. 2294 – 2309, May 2007.
has lowest noise power, comes last for each subchannel. From [16] D. Tse and P. Viswanath, Fundamentals of Wireless Communications.
Fig. 5, we found that, for each subchannel, the algorithm IGPP Cambridge University Press, May.