Ddtt Yhgfdf Yte

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. 9, NO. 6, JUNE 2010

2055

Bandwidth Exchange: An Energy Conserving
Incentive Mechanism for Cooperation
Dan Zhang, Ryoichi Shinkuma, Member, IEEE, and Narayan B. Mandayam, Fellow, IEEE

Abstract—Cooperative forwarding in wireless networks has
shown to yield rate and diversity gains, but it incurs energy costs
borne by the cooperatingnodes. In this paper we consider an
incentive mechanism called Bandwidth Exchange (BE) where the
nodes flexibly exchange the transmission bandwidth as a means
of providing incentive for forwarding data, without increasing
either the total bandwidth required or the total transmit power.
The advent of cognitive radios and multicarrier systems such as
Orthogonal Frequency Division Multiple Access(OFDMA) with
the ability to flexibly delegate and employ a number of subcarriers makes this approach particularly appealing compared
to other incentive mechanisms that are often based on abstract
notions of credit and shared understanding of worth. We consider
a �� -node wireless network over a fading channel and use a Nash
Bargaining Solution (NBS) mechanism to study the benefits of BE
interms of rate and coverage gains. We also propose two heuristic
algorithms based on simple probabilistic rules for forwarding
and study the tradeoffs in terms of performance among these
approaches. Our results reveal that bandwidth exchange based
forwarding can provide transmit power savings in OFDMA
networks of at least 3dB compared to noncooperation.
Index Terms—Cognitive radio, incentivemechanism, bandwidth exchange, Nash bargaining, OFDMA.

C

I. I NTRODUCTION

OOPERATIVE forwarding is an essential technique to
enhance connectivity and throughput for wireless networks. However, forwarding always incurs some sort of cost
– a real cost like power, and/or an opportunity cost like delay.
Recent work in [1] has shown that even in the absence of
such costs, cooperation amongnodes in a wireless network
is not guaranteed and may require incentives. Current studies on cooperative forwarding mechanisms largely fall into
four categories: reputation based mechanisms [2]–[6], credit
based incentives [7], [8], network assisted pricing mechanisms
[9], [10] and mechanisms based on forwarding games [11]–
[14]. These prior techniques often mimic the operation of
a complexeconomy and their efficient operation requires

Manuscript received August 20, 2009; revised January 19, 2010; accepted
March 27, 2010. The associate editor coordinating the review of this paper
and approving it for publication was C. Xiao.
D. Zhang and N. B. Mandayam are with WINLAB, Department of
Electrical and Computer Engineering, Rutgers University, Piscataway, NJ,
08854 USA (e-mail:{bacholic, narayan}@winlab.rutgers.edu).
R. Shinkuma is with the Graduate School of Informatics, Kyoto University Yoshida-honmachi, Sakyo-ku, Kyoto, 606-8501, Japan (e-mail:
shinkuma@i.kyoto-u.ac.jp). He was a visiting scholar at WINLAB, Rutgers
University, from Sep. 2008 to Aug. 2009.
This paper has been submitted in part for presentation at IEEE PIMRC
2009.
This work is supported in part bythe NSF (no. CCF-0634973 and CNS0721826). It is also supported in part by the Japan Society for the Promotion
of Science (JSPS) under Grant-in-Aid for Encouragement of Young Scientists
(B) (no. 21760288).
Digital Object Identifier 10.1109/TWC.2010.06.091263

such enablers as a stable currency, a system of credit or
a shared understanding of what things are worth. In real
economies, theseenablers are achieved over long periods
of time, and even with experience, the overall functioning
of such economies is difficult to predict, a lesson we have
learned frequently and with some pain. The main contribution
of this paper is to circumvent some of these difficulties by
exploring the incentive induced from exchanging a fraction of
individually preassigned bandwidth among nodes, referred...