IEEE BE-BTECH NS2 PROJECT@ DREAMWEB TECHNO SOLUTION

Improving the Network Lifetime of MANETs
through Cooperative MAC Protocol Design
Xiaoyan Wang, Member, IEEE and Jie Li, Senior Member, IEEE
Abstract—Cooperative communication, which utilizes nearby terminals to relay the overhearing information to achieve the diversity
gains, has a great potential to improve the transmitting efficiency in wireless networks. To deal with the complicated medium access
interactions induced by relaying and leverage the benefits of such cooperation, an efficient Cooperative Medium Access Control
(CMAC) protocol is needed. In this paper, we propose a novel cross-layer distributed energy-adaptive location-based CMAC protocol,
namely DEL-CMAC, for Mobile Ad-hoc NETworks (MANETs). The design objective of DEL-CMAC is to improve the performance of the
MANETs in terms of network lifetime and energy efficiency. A practical energy consumption model is utilized in this paper, which takes
the energy consumption on both transceiver circuitry and transmit amplifier into account. A distributed utility-based best relay selection
strategy is incorporated, which selects the best relay based on location information and residual energy. Furthermore, with the purpose
of enhancing the spatial reuse, an innovative network allocation vector setting is provided to deal with the varying transmitting power of
the source and relay terminals. We show that the proposed DEL-CMAC significantly prolongs the network lifetime under various
circumstances even for high circuitry energy consumption cases by comprehensive simulation study.
Index Terms—Network lifetime, cooperative communication, medium access control protocol, relay selection
Ç
1 INTRODUCTION
A Mobile Ad-hoc NETwork (MANET) is a self-config-
ured network of mobile terminals connected by wire-
less links. Mobile terminals such as cell phones, portable
gaming devices, personal digital assistants, (PDAs) and tab-
lets all have wireless networking capabilities. By participat-
ing in MANETs, these terminals may reach the Internet
when they are not in the range of Wi-Fi access points or cel-
lular base stations, or communicate with each other when
no networking infrastructure is available. MANETs can also
be utilized in the disaster rescue and recovery described in
[26]. One primary issue with continuous participation in
MANETs is the network lifetime, because the aforemen-
tioned wireless terminals are battery powered, and energy
is a scarce resource.
Cooperative communication (CC) [2] is a promising
technique for conserving the energy consumption in
MANETs. The broadcast nature of the wireless medium
(the so-called wireless broadcast advantage) is exploited in
cooperative fashion. The wireless transmission between a
pair of terminals can be received and processed at other
terminals for performance gain, rather than be considered
as an interference traditionally. CC can provide gains in
terms of the required transmitting power due to the spatial
diversity achieved via user cooperation. However, if we
take into account the extra processing and receiving
energy consumption required for cooperation, CC is not
always energy efficient compared to direct transmission.
There is a tradeoff between the gains in transmitting power
and the losses in extra energy consumption overhead.
CC has been researched extensively from the informa-
tion theoretic perspective [1], [2], [3], [4], [5] and on the
issues of relay selection [19], [21], [22], [23], [24]. Recently,
the work on CC with regard to cross-layer design by con-
sidering cooperation in both physical layer and MAC layer
attracts more and more attention. Without considering the
MAC layer interactions and signaling overhead due to
cooperation, the performance gain through physical layer
cooperation may not improve end-to-end performance.
Cooperative MAC (CMAC) protocol considering the
practical aspect of CC is vital. Liu et al. have proposed a
CMAC protocols named CoopMAC [7] to exploit the
multi-rate capability and aimed at mitigating the through-
put bottleneck caused by the low data rate nodes, so that
the throughput can be increased. With the similar goal,
Zhu and Cao [8] have proposed a CMAC protocolor wire-
less ad hoc network. However, beneficial cooperation con-
sidering signaling overhead is not addressed in [7] and
[8]. A busy-tone-based cross-layer CMAC protocol [9] has
been designed to use busy tones to help avoiding colli-
sions in the cooperative scenario at the cost on transmit-
ting power, spectrum, and implementation complexity. A
reactive network coding aware CMAC protocol has been
proposed by Wang et al. [10], in which the relay node can
forward the data for the source node, while delivering its
own data simultaneously. But the network lifetime is not
addressed in [10]. A distributed CMAC protocol [11] has
been proposed to improve the lifetime of wireless sensor
networks, but it is based on the assumption that every
X. Wang is with the Information Systems Architecture Science Research
Division, National Institute of Informatics, 2-1-2 Hitotsubashi, Chiyoda-
ku, Tokyo, 101-8430, Japan. E-mail: wangxy@nii.ac.jp.
J. Li is with the Faculty of Engineering, Information and Systems, Univer-
sity of Tsukuba, Tsukuba Science City, Ibaraki 305-8573, Japan.
E-mail: lijie@cs.tsukuba.ac.jp.
Manuscript received 16 Nov. 2012; revised 3 Apr. 2013; accepted 4 Apr. 2013.
Date of publication 10 Apr. 2013; date of current version 6 Mar. 2015.
Recommended for acceptance by W. Jia.
For information on obtaining reprints of this article, please send e-mail to:
reprints@ieee.org, and reference the Digital Object Identifier below.
Digital Object Identifier no. 10.1109/TPDS.2013.110
1010 IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 26, NO. 4, APRIL 2015
1045-9219 ß 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

node can connect to the base station within one hop,
which is impractical for most applications. A CMAC pro-
tocol for vehicular networks, particularly for gateway
downloading scenarios, has been designed by Zhang et al.
[12]. A drawback in [12] is that it can only be utilized in
the scenario that all the vehicles are interested in the same
information. Moreover, Moh and Yu [13] have designed a
CMAC protocol named CD-MAC which lets the relay
transmit simultaneously with the source using space-time
coding technique. Shan et al. [14] have explored a concept
of cooperation region, whereby beneficial cooperative
transmissions can be identified. However, energy con-
sumption is not evaluated for both of them.
The existing CMAC protocols mainly focus on the
throughput enhancement while failing to investigate the
energy efficiency or network lifetime. While the works on
energy efficiency and network lifetime generally fixate on
physical layer [20] or network layer [19]. Our work focuses
on the MAC layer, and is distinguished from previous pro-
tocols by considering a practical energy model (i.e., energy
consumption on both transceiver circuitry and transmit
amplifier), with the goal to enhance energy efficiency and
extend network lifetime. The tradeoff between the gains
promised by cooperation and extra overhead is taken into
consideration in the proposed protocol. In addition, in the
previous works, very little attention has been paid to the
impact brought by varying transmitting power in CC on
the interference ranges, since constant transmitting power is
generally used. The interference ranges alteration in both
space and time will significantly affect the overall network
performance. We also address the issue of effective coordi-
nation over multiple concurrent cooperative connections
with dynamical transmitting power in this paper.
In this paper, we propose a novel distributed energy-
adaptive location-based CMAC protocol, namely DEL-
CMAC, for MANETs. DEL-CMAC is designed based on the
IEEE 802.11 distributed coordination function (DCF), which
is a widely used standard protocol for most of wireless net-
works. DEL-CMAC comprises a relay-involved handshak-
ing process, a cross-layer power allocation scheme, a
distributed utility-based best relay selection strategy, and
an innovative Network Allocation Vector (NAV) setting.
From the perspective of information theory, higher diversity
gain can be obtained by increasing the number of relay ter-
minals. From a MAC layer point of view, however, more
relays lead to the enlarged interference ranges and addi-
tional control frame overheads. We employ single relay ter-
minal in this paper to reduce the additional communication
overhead. DEL-CMAC initiates the cooperation proactively,
and utilizes the decode and forward (DF) protocol [1] in the
physical layer. We summarize our contributions as follows.
We propose DEL-CMAC that focuses on the net-
work lifetime extension, which is a less explored
aspects in the related work. By considering the
overheads and interference due to cooperation, as
well as the energy consumption on both transceiver
circuitry and transmit amplifier, DEL-CMAC can
significantly prolong the network lifetime.
A distributed energy-aware location-based best
relay selection strategy is incorporated, which is
more reasonable for MANETs comparing with the
existing schemes based on channel condition.
For a desired outage probability requirement, a
cross-layer optimal transmitting power allocation
scheme is designed to conserve the energy while
maintaining certain throughput level.
To deal with the presence of relay terminals and
dynamic transmitting power, we provide an innova-
tive NAV setting to avoid the collisions and enhance
the spatial reuse.
Extensive simulation results reveal that DEL-CMAC
can significantly extend the network lifetime under
various scenarios at the cost of relatively low
throughput and delay degradation, compared with
IEEE standard DCF and throughput-aimed scheme
CoopMAC [7].
The remainder of the paper is organized as follows. We
present preliminaries and model in Section 2. In Section 3,
we describe the proposed DEL-CMAC protocol. In Sec-
tion 4, we further elaborate the detail of the DEL-CMAC,
including the best relay selection strategy, the cross-layer
power allocation scheme and the NAV setting. Simulation
results and discussions are addressed in Section 5. Con-
clusions are drawn in Section 6.
2 MODELS AND PRELIMINARIES
In this section, we present the employed system and energy
models, and the background knowledge about DCF and
decode and forward protocol.
2.1 System and Energy Models
As shown in Fig. 1, a multi-hop MANET with randomly
deployed mobile terminals is considered, where all termi-
nals have the capability to relay. To come up with a rea-
sonable system model, we assume that data connections
among terminals are randomly generated and the routes
are established by running Ad hoc On-demand Distance
Vector (AODV) [16], which is a widely used conventional
routing protocol for MANETs. There are two types of relay
terminals in our network, i.e., routing relay terminals and
cooperative relay terminals. In the system model, AODV
builds the route in a proactive manner by selecting the
routing relay terminals firstly. When a route is established,
DEL-CMAC initiates the cooperation in a hop-by-hop
manner by selecting the cooperative relay terminals. In
this paper, the source and destination terminals are referred
to the terminals at MAC layer, and the relay terminals indi-
cate the cooperative relay terminals. For convenience, we
use term source, relay and destination in the remainder of
Fig. 1. Multi-hop MANET scenario.
WANG AND LI: IMPROVING THE NETWORK LIFETIME OF MANETS THROUGH COOPERATIVE MAC PROTOCOL DESIGN 1011

the paper to denote the source terminal, relay terminal and
destination terminal respectively. It is reasonable to assume
that the energy is consumed both on transmitting and
receiving the data, similar energy consumption model is
used in previous work, e.g., [28]. To transmit a packet, the
energy cost is Ct ¼ ðP þ P0
ÞT. And to receive a packet, the
energy cost is Cr ¼ P0
T. P refers to the power consump-
tion at transmit amplifier (also denotes as transmitting
power in this paper), and P0
refers to the power consump-
tion at transceiver circuitry. To study the effect of energy
consumption on transceiver circuitry, the cases P0
=P ¼
0:5; 1; 2 are generally examined. Low P0
=P ratio indicates
that the energy consumption on transmit amplifier
accounts for great proportion of the total energy consump-
tion. And high P0
=P ratio indicates the high circuitry
energy consumption case.
2.2 DCF
The basic operations of the proposed DEL-CMAC are
based on the IEEE 802.11 DCF [18]. In DCF, after a trans-
mitting terminal senses an idle channel for a duration of
Distributed InterFrame Space (DIFS), it backs off for a
time period that chosen from 0 to its Contention Window
(CW). After the backoff timer expires, the well-known
RTS-CTS-DATA-ACK procedure is carried out (see Fig. 2).
Any terminal overhearing either the RTS or the CTS
extracts the information contained in the MAC frame
header, and sets its NAV to imply the time period during
which the channel is busy.
2.3 Decode and Forward
DEL-CMAC utilizes the DF protocol [1] with Maximum-
Ratio-Combiner (MRC) in the physical layer. Due to the
limited space, the details of DF are presented at Appendix
A, which can be found on the Computer Society Digital
Library at http://doi.ieeecomputersociety.org/10.1109/
TPDS.2013.110.
3 THE PROPOSED DEL-CMAC PROTOCOL
In this section, with the objective of prolonging the network
lifetime and increasing the energy efficiency, we present a
novel CMAC protocol, namely DEL-CMAC, for multi-hop
MANETs. When cooperative relaying is involved, the
channel reservation needs to be extended in both space and
time in order to coordinate transmissions at the relay. To
deal with the relaying and dynamic transmitting power,
besides the conventional control frames RTS, CTS and
ACK, additional control frames are required. DEL-CMAC
introduces two new control frames to facilitate the coopera-
tion, i.e., Eager-To-Help (ETH) and Interference-Indicator (II).
The ETH frame is used for selecting the best relay in a dis-
tributed and lightweight manner, which is sent by the win-
ning relay to inform the source, destination and lost relays.
In this paper, the best relay is defined as the relay that has
the maximum residual energy and requires the minimum
transmitting power among the capable relay candidates.
The II frame is utilized to reconfirm the interference range
of allocated transmitting power at the winning relay, in
order to enhance the spatial reuse. Among all the frames,
RTS, CTS, ETH and ACK are transmitted by fixed power.
And the transmitting power for the II frame and data
packet are dynamically allocated. We denote the time dura-
tions for the transmission of RTS, CTS, ETH, ACK and II
frames by TRTS, TCTS, TETH, TACK and TII, respectively.
3.1 Protocol Description
The frame exchanging process of DEL-CMAC is shown
in Fig. 3. Similar to the IEEE 802.11 DCF protocol, the
RTS/CTS handshake is used to reserve the channel at
first. As we know, the cooperative transmission is not
necessary in the case that the transmitting power is small
[6], because the additional overhead for coordinating the
relaying overtakes the energy saving from diversity gain.
Those inefficient cases are avoided by introducing a
transmitting power threshold Lp. In DEL-CMAC, upon
receiving the RTS frame, the destination computes the
required transmitting power for the direct transmission
PD
s (given in Section 4.2). There are two cases depending
on the calculated PD
s .
Case (i): PD
s Lp. The destination sends a CTS frame
with flag field (FLAG_P) equal to 0, which implies
that the direct transmission is adequate. Thus, when
the transmitting power for the direct transmission is
sufficiently low, DEL-CMAC is reduced to the DCF
protocol and thus has backward compatibility with
the legacy 802.11 standard.
Case (ii): PD
s Lp. FLAG_P in the CTS frame is set
to 1, which indicates that the cooperative relaying
is desired. All the terminals having overheard RTS
and CTS, and not interfere with other ongoing
transmissions are considered as the relay candi-
dates. After the relay candidates check if they are
Fig. 2. IEEE 802.11 DCF.
Fig. 3. The frame exchanging process of DEL-CMAC.

able to reduce the energy consumption (given in
the Eq. (1)), the capable relay candidates contend
for relaying by sending ETH after a utility-based
backoff (utility function is provided in Section 4.1).
Notice that there may exist the case that two relay
candidates hidden with each other (outside the
transmission range). However, they can still sense
the message sent from each other (within the sens-
ing range which is set at 1:9 times of the transmis-
sion range in the simulator by default). The case
that multiple ETH frames collide due to hidden
would not exist. After SIFS (short interframe
space), the winning relay broadcasts the II message
to reconfirm the interference range of the allocated
transmitting power at relay, which is used in the
NAV setting (see Section 4.3). After the above con-
trol frame exchanging, the source and relay cooper-
atively send the same data frames to the
destination in two consecutive time intervals using
the allocated transmitting power (see Section 4.2).
Finally, the destination sends an ACK back to the
source if it decodes the message successfully.
The flow charts of the terminals are given in the
Appendix B, available in the online supplemental mate-
rial. The detailed protocol operations are provided from
the perspective of different terminals:
3.1.1 Operations at the Source
i. When a source wants to initiate the data transmis-
sion with payload length L bytes, it first senses the
channel to check if it is idle. If the channel is idle for
DIFS, the source chooses a random backoff timer
between 0 and CW. When the backoff counter
reaches zero, the source sends out a RTS to reserve
the channel. Notice that different from DCF, the loca-
tion information of the source is carried in the RTS,
which is used in the optimal power allocation.
ii. If the source does not receive a CTS within
TRTS þ TCTS þ SIFS, a retransmission process will be
performed. Otherwise, in the case that FLAG_P of
CTS is 0, the DEL-CMAC is reduced to DCF protocol,
and we omit its operations in the following. In the
case that FLAG_P is 1, the source waits for another
Tmax
Backoff þ TETH þ SIFS, where Tmax
Backoff is the maxi-
mum backoff time for the relay (given in Section 4.1).
If ETH is not received, which means that no capable
relay exist, the source sends the data by direct trans-
mission with data rate R.
iii. If both CTS and ETH are received, after waiting for
TII þ SIFS, the source initiates a cooperative trans-
mission with data rate 2R using the optimal trans-
mitting power PC
s which is piggybacked in the ETH.
Notice that in order to maintain the end-to-end
throughput, doubled data rate is employed in the
cooperative transmission mode. We assume that the
terminal can support two transmission rates by dif-
ferent coding and modulation schemes.
iv. If an ACK is not received after 16ðL þ LhÞ=2R þ
TACK þ 2SIFS, where Lh is the header length (in
bytes), the source would perform a random backoff
same as DCF. Otherwise, the transmission process
succeeds and the source handles the next packet in
the buffer if any. Notice that the unit for L and Lh
is byte, and the unit for data rate is bits per second,
thus the transmission time for one data frame is
8ðL þ LhÞ=2R.
3.1.2 Operations at the Destination
i. Upon receiving the RTS, the destination sends a
CTS back after SIFS. The CTS contains the location
information of the destination, the FLAG_P, and
the transmitting power for the direct transmission
PD
s (in the form of dB m, occupying 4 bytes),
which is used for the possible relay contention.
ii. In the case that FLAG_P is 1, if the destination
has not heard any ETH within Tmax
Backoff þ TCTS þ
TETH þ SIFS, it assumes that the direct transmis-
sion will be performed and waits for the data
packet from the source.
iii. Otherwise, the destination waits for the data pack-
ets from the source and winning relay. If the desti-
nation can decode the combined signals correctly,
it sends back an ACK. Otherwise, it just lets the
source timeout and retransmit.
3.1.3 Operations at the Relay
i. Any terminal that receives both RTS and CTS (with
FLAG_P equals 1) and does not interfere with other
transmissions in its vicinity can be regarded as a
relay candidate. Upon receiving the CTS, each relay
candidate checks whether it is able to reduce the
total energy consumption by
À
2PD
s À PC
s À PC
r À 2P0
Á
Â ðL þ LhÞ=2R
À
À
PC
r þ P0
Á
Â TII À ðP þ 3P0
Þ Â TETH 0:
(1)
PC
s and PC
r refer to the transmitting power in the coop-
erative transmission mode for source and relay (given
in Section 4.2), PD
s and P refer to the transmitting
power in the direct transmission mode for source and
the fixed transmitting power respectively. Term
ð2PD
s À PC
s À PC
r À 2P0
Þ Â ðL þ LhÞ=2R denotes the
saved energy consumption in transmitting the data by
CC, term ðPC
r þ P0
Þ Â TII and ðP þ 3P0
Þ Â TETH
denotes the additional energy consumption on control
overhead. By Eq. (1), the relay checks whether CC can
reduce the total energy consumption both on transmit-
ting and receiving, compared to direct transmission.
Every capable relay candidate (satisfies Eq. (1)), starts
a backoff timer after SIFS interval. The specific utility-
based backoff scheme is given in Section 4.1.
ii. Intuitively, the backoff at a better relay expires ear-
lier, hence the best relay will send out an ETH first.
The lost relays give up contention when sensing the
ETH. The ETH contains the optimal transmitting
power PC
s for the source (in the form of dBm, occu-
pying 4 bytes).
iii. After SIFS, the winning relay broadcasts the II
message using power PC
r . II message is used to

reconfirm the interference range of the relay with
the objective to enhance the spatial reuse. The detail
of the NAV setting will be explained in Section 4.3.
Then, the winning relay waits for the data packet
from the source to arrive. Upon receiving the data
packet, the winning relay forwards it to the destina-
tion with data rate 2R using transmitting power PC
r .
3.2 Further Discussions
Compared with the IEEE 802.11 DCF, the proposed DEL-
CMAC has additional control message overhead in the case
that the FLAG_P is 1 and the capable relay candidates are
absent. The duration of this overhead is a constant time
equal to Tmax
Backoff þ TETH þ SIFS. It is undesirable but inevi-
table if we try to coordinate multiple connections with coop-
erative relaying and choose the best relay in a distributed
fashion. However, notice that the probability of no capable
relay candidate exists is quite small given a general node
density deployment (addressed in Section 4.3). In addition,
this overhead duration is relatively short comparing with
the payload transmission duration. From the simulation
results provided in Section 5, we observe that the perfor-
mance of throughput and delay only decreases by around
5 percent, which may be acceptable when considering the
significant increase in the network lifetime.
Another issue in DEL-CMAC is the hidden terminal
problem due to the terminal mobility. Consider a situation
as follows. After the exchanges of RTS, CTS and ETH, a ter-
minal located outside the transmission range originally
moves into the range. Due to the lack of NAV setting, this
terminal may interfere with the ongoing transmission, lead-
ing to a collision. However, the hidden terminal issue
caused by the terminal mobility is not unique for our DEL-
CMAC, and it already exists in the original IEEE 802.11
DCF. In this paper, we consider the probability that hidden
terminal issue occurs is considerably low, and we leave it as
our future work.
4 DETAIL AND SUPPLEMENT OF DEL-CMAC
In this section, we elaborate the detail and the supplement
of the proposed DEL-CMAC. Specifically, we address the
optimal power allocation scheme, the utility-based best
relay selection strategy, and the NAV setting in the follow-
ing sections.
4.1 Utility-Based Best Relay Selection
Selecting the best relay distributed and efficiently affects
the performance of the CMAC protocol significantly. The
existing relay selection schemes that incorporated into the
CMAC protocols, largely depend on the instantaneous
channel condition, which based on the assumption that
the channel condition is invariant during one transmit ses-
sion. For MANETS that deployed in heavily built-up
urban environments or heavy traffic environments, this
assumption is hard to guarantee [25]. This implies that the
“best” selected relay terminal according to channel condi-
tion during the route construction or handshaking period,
may not be the best one in the actual data transmission
period. Selecting the best relay terminal based on the
instantaneous location instead of instantaneous channel
condition may be more reasonable for MANETs. In this
paper, we propose a distributed energy-aware location-
based best relay selection strategy which is incorporated
into the control frame exchanging period in DEL-CMAC.
The location information of individual wireless devices
can be obtained through GPS or other localization algo-
rithms [27]. The required location information of source
and destination is carried by RTS and CTS frames. Thus
no additional communication overheads are involved.
DEL-CMAC chooses the best relay based on a utility-based
backoff, which depends on the required transmitting
power to meet certain outage probability (related to indi-
vidual location) and the residual energy of individual ter-
minals. It is carried out in a distributed, lightweight and
energy-efficient fashion, in which the backoff of the relay
that has the minimum utility value expires first.
We define the Backoff Utility function for relay r as
BUr ¼ t min
E
Er
; d

Â
PC
r
PD
s =2
; (2)
where Er is the current residual energy of relay r, PC
r is the
transmitting power at relay r in cooperative mode, and PD
s is
the transmitting power at source s in direct mode (both
obtained through the equations in Section 4.2). The parame-
ters in Eq. (2) include the energy consumption threshold d,
the constant unit time t, and the initial energy E. Intuitively,
the terminal with high residual energy and low transmitting
power (i.e., small BUr value), has a comparatively short
backoff time. The terminal whose backoff expires first will be
selected as the winning relay. The threshold d is to restrict the
maximum backoff time within an acceptable range. Since
when the residual energy is very low, E=Er will be extremely
large, leading to a very long backoff time that we should pre-
vent. The term PC
r is strictly upper bounded by PD
s =2, i.e.,
the term
PC
r
PD
s =2
is always less than 1. Thus, BUr is upper
bounded by the maximum backoff time Tmax
Backoff which is
equal to t Á d.
We observe that there is a tradeoff between the probabil-
ity of collision (due to extremely close utility value) and the
time spent in the relay selection process. The value of t can-
not be made too large to postpone the time to find the best
relay, or too small to raise the probability of collision. In our
simulation, t is set to 0:1 ms. However, setting t properly
can only depress the collision probability but cannot avoid
the collision completely. Incorporate the collision free relay
selection strategies [14], [29] into our utility-based backoff
scheme is our future work.
Different from the existing best relay selection schemes,
the proposed strategy utilizes the location information and
takes the residual energy into considerations. Besides, it is
completely distributed and every terminal makes the deci-
sion independently. Using the proposed relay selection
strategy, the energy consumption rate among the terminals
can be balanced, and the total energy consumption can be
reduced.
4.2 Optimal Power Allocation
Optimal power allocation is indispensable for a cross-layer
CMAC protocol that aims at increasing energy efficiency. In
this section, we address the power allocation for CC and

direct transmission under the given outage probability. We
start with deriving the transmitting power at source in the
direct transmission mode, which is calculated by the desti-
nation after it receives the RTS. Then, under the same out-
age probability and end-to-end data rate, the optimal
transmitting power at source and relay in the cooperative
transmission mode is calculated by individual relay candi-
dates after the RTS/CTS handshake.
4.2.1 Direct Transmission
In order to meet a desired outage probability PO
D, the mini-
mum transmitting power in the direct transmission mode is
given as
PD
s ¼ À
ð2R
À 1ÞN0da
sd
ln
À
1 À PO
D
Á ; (3)
where R is the transmission rate, dsd is the distance between
the source and the destination, a is the path loss exponent, hsd
is the channel fading gain and N0 is variance of the noise com-
ponent. Due to the space constraint, we provide the derivation
of Eq. (3) in the Appendix C, available in the online supple-
mental material.
4.2.2 Cooperative Transmission
The optimal power allocation for cooperative transmission
exists when the transmitting power at source PC
s equals the
transmitting power at relay PC
r . And PCÃ
s is the solution to
the following equation
da
sdG
À
da
sd þ da
rd
Á
À da
rdG
À
da
sr þ da
sd
Á
þ
À
1 À PO
C
ÁÀ
da
rd À da
sd
Á
¼ 0;
(4)
where
GðdÞ ¼ exp À
ð22R
À 1ÞN0d
PC
s

:
Due to the space constraint, the derivation and solution
of Eq. (4) are presented in the Appendix D, available in the
online supplemental material.
4.3 Spatial Reuse Enhancement
As the involvement of relaying and varying transmitting
power, the interference ranges in DEL-CMAC are changing
during one transmit session. In order to avoid the interfer-
ence and conserve the energy, delicate NAV setting is
required. NAV limits the use of physical carrier sensing,
thus conserves the energy consumption. The terminals
listening on the wireless medium read the duration field in
the MAC frame header, and set their NAV on how long
they must defer from accessing the medium. Taking IEEE
802.11 DCF for instance, the NAV is set using RTS/CTS
frames (see Fig. 2). No medium access is permitted during
the blocked NAV durations.
Comparing with the simple NAV setting in DCF, the
setting in DEL-CMAC needs to be considerably modified.
The presence of relays will enlarge the interference
ranges and the dynamic transmitting power makes the
interference ranges vary during one transmit session.
Impropriate NAV setting induces energy waste and colli-
sions. Specifically, setting the NAV duration too short
will wake up the terminal too soon, which results in
energy waste due to medium sensing. On the other hand,
setting it too long will reduce the spatial efficiency, which
results to the performance degradation in terms of
throughput and delay. Thus, effective NAV setting is nec-
essary and critical. Unfortunately, most of the previous
works does not address the NAV setting issue in CC [9],
[13], not to mention the one with varying transmitting
power. In this paper, we divide the transmission ranges
for the source, destination and relay to five different
regions (see Fig. 4). Since different transmitting power
lead to different transmission ranges, there exist two
ranges for the relay. As shown in Fig. 4, the solid circle
denotes the transmission range for fixed transmitting
power (with radius r1), and the dashed circle denotes the
transmission range for the allocated transmitting power
(with radius r2). Notice that it is not necessary to consider
the transmission range with allocated transmitting power
at the source, since all the terminals lie inside the solid
circle of the source will interfere with the ACK. Thus,
they must defer accessing the medium until the very end
of the whole session. In the following, we address the
specific NAV setting for our DEL-CMAC from the per-
spective of different regions by Fig. 5.
Fig. 4. An illustration for the NAV setting ranges.
Fig. 5. NAV setting for DEL-CMAC.

4.3.1 Region 1 (The Terminals That can Receive Both
the RTS and CTS)
The terminals in this region are the relay candidates.
According to our DEL-CMAC, they contend for the winning
relay after the RTS/CTS exchange. Upon receiving the ETH,
all the lost relays should keep silence until the whole trans-
mit session is finished. Notice that for the sake of the relay
selection, the terminals cannot set their NAVs as soon as
they receive the RTS as in the IEEE 802.11 DCF. All the
neighboring terminals have to wait until the end of the CTS
and then make their decisions. Thus, the NAV duration in
region 1 is TII þ TACK þ 16ðL þ LhÞ=2R þ 4SIFS.
4.3.2 Region 2 (The Terminals That can Receive the
RTS But Not the CTS)
Those terminals set their NAV durations until the end of the
ACK, which is Tmax
Backoff þ TETH þ TII þ TACK þ 16ðL þ LhÞ=
2R þ 5SIFS.
CTS but not the RTS)
The same as the terminals in region 2, they set their NAV
until the end of the ACK.
4.3.4 Region 4 (The Terminals That can Receive the II)
As we mentioned before, according to different transmit-
ting power, there exist two transmission ranges at the
relay. One is the transmission range for the ETH message
with fixed transmitting power (large solid circle with
radius r1 in Fig. 4), the other is the transmission range for
the II message and data with allocated transmitting power
(small dashed circle with radius r2 in Fig. 4). The terminals
in region 4 fall inside the small transmission range at the
relay, they should defer the medium access until the end
of the data transmissions (two phases). Recall that in
802.11 DCF, the nodes outside the transmission ranges of
source and destination do not set NAV, they use physical
carrier sensing to avoid the possible collision. Thus, same
as the setting in 802.11 DCF, the NAV duration for nodes
in region 4 ends before the ACK frame. The NAV duration
for them is 16ðL þ LhÞ=2R þ 2SIFS.
ETH But Not the II)
The terminals in this region fall inside the large transmis-
sion range at the relay but outside the small one. Those ter-
minals have a relatively short NAV duration comparing to
the terminals in region 4, which is only 8ðL þ LhÞ=2R. Since
when the source finishes its data transmission, the terminals
in region 5 and the relay may not interfere with each other.
By utilizing II frame, the nodes in this region may initiate
their transmission in advance given they are outside the
interference range of the destination.
5 PERFORMANCE EVALUATION
n this section, we evaluate DEL-CMAC via extensive sim-
ulations comparing with IEEE 802.11 DCF and CoopMAC
[7]. Since the purpose of our scheme is to prolong the net-
work lifetime and increasing the energy efficiency, the
evaluation metrics in this paper are the transmitting
power, total energy consumption, network lifetime,
aggregated throughput and average delay. The transmit-
ting power denotes the power consumed at transmit
amplifier (without the power consumed at transmit cir-
cuitry). The total energy consumption is the summation
of the transmitting (including both transmit amplifier and
circuitry) and receiving energy cost at the source, destina-
tion and relay. The lifetime is defined as the duration
from the network initialization to the time that the first
terminal runs out of power. To validate the performance
improvements in DEL-CMAC, we utilize both the single-
hop scenario and the multi-hop multi-connection sce-
nario. The simulation is carried out in QualNet network
simulator [15]. The initial energy of all the terminals are
set to 1 J. The propagation channel of two-ray path loss
model is adopted. Constant data rate with 1 Mbps is used
in DEL-CMAC and DCF, while adapted data rates with
1; 2; 5:5 Mbps are used in CoopMAC. The fixed transmit-
ting power used for control frames is set to 10 dBm and,
the fixed transmitting power used for data frame in Coop-
MAC is set to 15 dBm due to the high data rate (the trans-
mitting power for the data frames in DEL-CMAC and
DCF is dynamically allocated). The simulation settings
and parameters are listed in Table 1.
5.1 Single-Hop Scenarios
We first compare our DEL-CMAC with the IEEE 802.11 DCF
in a single-hop scenario that only consists of three terminals
(one source, one destination and one relay), to show the dif-
ferences between cooperative and non-cooperative commu-
nication on energy consumption. As shown in Fig 6, the
distance between source and destination changes from 5 to
30 m, and angles ffSDR and ffDSR keep at arccosð2=3Þ.
Fig. 7 shows the variance of the transmitting power to
satisfy different outage probability requirements, when the
distance between source and destination is 20 m. It is
Fig. 6. An illustration of the single-hop scenario.
TABLE 1
Simulation Parameters

straightforward that high outage probability requirement
leads to high cost in terms of transmitting power. We
observe that for the required data rate and outage probabil-
ity, the transmitting power for cooperative transmission is
far less than the one for direct transmission. Since the proba-
bility of success 99.9 percent is acceptable for most of the
wireless network applications, the simulation study in the
remainder of this paper are all based on the outage proba-
bility 0.1 percent.
To validate the efficiency of the cooperative communi-
cation, only investigating transmitting power is not
enough. The processing and receiving energy consump-
tion at relay and destination should also be taken into
account. In Fig. 8, we compare the total energy consump-
tion in one transmit session. The ratio of the energy con-
sumption on transceiver circuitry to transmit amplifier,
i.e., P0
=P, at 0:5; 1; 2 are investigated. We plot the energy
consumption at different distances for different P0
=P
ratios. For short distance, the direct transmission is more
efficient in all cases, since the overhead for control frames
dominates the energy saving from cooperative diversity.
Notice that due to the transmitting power threshold check
and the energy consumption check by Eq. (1), inefficient
cases for cooperative transmission (i.e., when the direct
channel condition is good or the distance is short) are
ruled out in DEL-CMAC. The energy consumption of
DCF, however, is dramatically increased as the distance
raises (around 7 meters for case P0
=P ¼ 0:5 and 11 meters
for case P0
=P ¼ 2). While the energy cost by DEL-CMAC
remains in the same level even for farther distances. The
energy consumption of DEL-CMAC is significantly below
DCF for medium to long distances, considering the circuit
energy consumption at both sender and receiver.
5.2 Multi-Hop Multi-Connection Scenarios
Next, we illustrate the performance of DEL-CMAC in a
realistic multi-hop multi-connection scenario along with
IEEE 802.11 DCF and CoopMAC. This complex scenario
takes the interference and collision caused by different
connections into account. As shown in Fig. 9, terminals
are randomly placed in a square area of 200 Â 200 m2
.
The dashed lines indicate that all the terminals belong to
the same subnet. The 5 solid lines indicate that 5 Con-
stant Bit Rate (CBR) connections, in which sources
(nodes 1; 11; 21; 31; 41) transmit UDP-based traffic at 1
packet per 100 milliseconds to the destinations (nodes
20; 30; 40; 50; 10) through multi-hop. The data payload
length is set to 1;024 bytes (unless stated otherwise).
AODV [16] routing protocol is used to establish the rout-
ing paths, which is widely used in MANETs. Other rout-
ing protocols as DSR or energy aware routing protocol
can also be used, the performance of the proposed MAC
layer scheme is independent of network layer schemes.
We vary the number of terminals in the area from 20 to
60 while keeping the number of CBR to 5. In Fig. 10, we
compare the network lifetime of DEL-CMAC with IEEE
802.11 DCF and CoopMAC in a static network. It is clear
that our DEL-CMAC always outperforms DCF and
Fig. 7. Transmitting power versus outage probability.
Fig. 8. Energy consumption versus s-d distance.
Fig. 9. A snapshot of the multi-hop network.
Fig. 10. Network lifetime versus the node density in a static environment
(with 95 percent confidence interval).

CoopMAC in all cases. CoopMAC [7] is designed to
increasing the throughput, in which fixed transmitting
power and adapted data rates are utilized. It is reasonable
that the network lifetime of CoopMAC is the shortest, due
to the lacking of power control and the additional control
overhead for cooperative communication. The perfor-
mance gain of DEL-CMAC over DCF and CoopMAC
raises as the number of terminals increases. The reason
can be explained from the following two aspects. First, if
the node density is low, some terminals have to play the
role as the source and cooperative relay alternately. This
additional relay energy cost is expected to impact the per-
formance negatively. The growing availability of relay
candidates results in balanced energy consumption. To be
more specific, if the node density is high enough, the ter-
minals having their own data to send or serving as rout-
ing relay, are rarely selected as the cooperative relay for
other connections. Because their residual energy is lower
than the others. Second, the higher the node density is,
the higher the probability that relay candidates are
located in the ideal positions for the existing source-desti-
nation pairs. Thus, high node density leads to a transmit-
ting power reduction for both source and relay by our
optimal power allocation scheme. To be specific, at least
2:2 and 3:9 times lifetime improvements for case
P0
=P ¼ 0:5, and 1:4 and 2:4 times lifetime improvements
for cases P0
=P ¼ 2, can be obtained by DEL-CMAC over
DCF and CoopMAC, respectively.
Next, we study the impact of mobility on the network
lifetime. We incorporate the random waypoint model [17]
in our simulation. Each terminal selects a random posi-
tion, moves towards it in a straight line at a constant
speed that is randomly picked from a range, and pauses
at that destination. The terminal repeats this process
throughout the simulation. We set the maximum speed at
10 mps and the pause time at 10 s. From Fig. 11, we can
observe the similar result as in static scenario. Our DEL-
CMAC prolongs the network lifetime by at least 2 and 4:8
times for case P0
=P ¼ 0:5, and 1:3 and 2:8 times for case
P0
=P ¼ 2, compared with DCF and CoopMAC respec-
tively. Thus, we conclude that our DEL-CMAC performs
well both in stationary and mobile scenarios in terms of
network lifetime.
An examination on the relationship between the net-
work lifetime and the data payload size for DEL-CMAC is
provided in Fig. 12. We observe that when the node den-
sity is low and payload size is small, the transmitting
power saved by cooperative transmission is greatly can-
celed out by the overhead entailed by the cooperative
relaying, e.g., only 1:26 times lifetime enhancement for 20
terminals and 128 bytes payload. However, as the node
density and payload size raise, the lifetime gain that our
DEL-CMAC can achieve becomes more and more signifi-
cant, e.g., 2:87 times enhancement for 50 terminals and
1;024 bytes payload. Thus, we conclude that DEL-CMAC
is more suitable for the networks with large payload size
and fairly high node density.
Finally, Figs. 13 and 14 depict the aggregated through-
put and average delay for the three schemes both in static
and mobile environments. The CoopMAC outperforms
the two others in both throughput and delay due to the
utilization of multiple data rates. And the performance of
CoopMAC decreases considerably in the mobile scenario,
Fig. 11. Network lifetime versus the node density in mobile environment
(with 95 percent confidence interval).
Fig. 12. Network lifetime versus data payload size (P0
=P ¼ 0:5 and with
95 percent confidence interval).
Fig. 13. Throughput performance versus the node density (with 95 per-
cent confidence interval).
Fig. 14. Delay performance versus the node density (with 95 percent
confidence interval).

since the table-based proactive relay selection may not
adapt to moving networks. For DEL-CMAC, the through-
put of the network decreases by at most 7.89 percent in
static environment and, 4.04 percent in mobile environ-
ment, compared to DCF. And the delay increases by at
most 5.61 and 3.93 percent in static and mobile environ-
ments, respectively. These results are expected since the
additional control frame overhead is required to coordi-
nate the cooperative transmission. Besides, the utility-
based backoff used for choosing the best relay, and the
enlarged interference range by relaying also affect the
throughput and delay negatively. However, comparing to
the network lifetime gain, we consider that the perfor-
mance reduction around 5 percent in throughput and
delay is acceptable. Notice that CoopMAC enhances the
throughput and delay at the cost of considerably network
lifetime degradation. The proposed scheme is particularly
suitable for the MANETs in which the network lifetime is
the primary requirement, e.g., the MANETs utilized in the
disaster rescue. For instance, when an earthquake hits
someplace, to search for survivors, a large number of
small robots, equipped with various sensors and a camera,
can be deployed all over the rubble. The robots form a
MANET to forward the sensing data to the base stations
and to coordinate their movements so that the entire area
is searched. In order to maximize the chance of finding
survivors, the network lifetime is much more important
than the throughput and delay in this kind of applications.
6 CONCLUSION
In this paper, we have proposed a novel distributed energy-
adaptive location-based cooperative MAC protocol for
MANETs. By introducing DEL-CMAC, both energy advan-
tage and location advantage can be exploited thus the net-
work lifetime is extended significantly. We have also
proposed an effective relay selection strategy to choose the
best relay terminal and a cross-layer optimal power alloca-
tion scheme to set the transmitting power. Moreover, we
have enhanced the spatial reuse to minimize the interfer-
ence among different connections by using novel NAV set-
tings. We have demonstrated that DEL-CMAC can
significantly prolong the network lifetime comparing with
the IEEE 802.11 DCF and CoopMAC, at relatively low
throughput and delay degradation cost.
As a future work, we will investigate our DEL-CMAC
for larger scale network size and with high mobility. We
will also consider to develop an effective cross-layer coop-
erative diversity-aware routing algorithm together with
our DEL-CMAC to conserve energy while minimizing the
throughput and delay degradation.
ACKNOWLEDGMENTS
The authors would like to thank Prof. Yuguang Fang, the
editor and the anonymous reviewers, whose insightful com-
ments helped us to improve the quality of the paper. This
work is partially supported by Grant-in-Aid for Scientific
Research of Japan Society for Promotion of Science (JSPS)
and Collaboration Research Grant of National Institute of
Informatics (NII), Japan. Jie Li is the corresponding author.
REFERENCES
[1] J.N. Laneman, D.N.C. Tse, and G.W. Wornell, “Cooperative
Diversity in Wireless Networks: Efficient Protocols and Outage
Behaviour,” IEEE Trans. Information Theory, vol. 50, no. 12,
pp. 3062-3080, Dec. 2004.
[2] A. Sendonaris, E. Erkip, and B. Aazhang, “User Cooperation
Diversity-Part I: System Description,” IEEE Trans. Comm., vol. 51,
no. 11, pp. 1927-1938, Nov. 2003.
[3] J.N. Laneman and G.W. Wornell, “Distributed Space-Time-Coded
Protocols for Exploiting Cooperative Diversity in Wireless
Networks,” IEEE Trans. Information Theory, vol. 49, no. 10,
pp. 2415-2425, Oct. 2003.
[4] J. Wu, M. Cardei, F. Dai, and S. Yang, “Extended Dominating Set
and Its Applications in Ad Hoc Networks Using Cooperative
Communication,” IEEE Trans. Parallel and Distributed Systems,
vol. 17, no. 8, pp. 851-864, Aug. 2006.
[5] Y. Zhu, M. Huang, S. Chen, and Y. Wang, “Energy-Efficient
Topology Control in Cooperative Ad Hoc Networks,” IEEE Trans.
Parallel and Distributed Systems, vol. 23, no. 8, pp. 1480-1491, Aug.
2011.
[6] S. Cui, A.J. Goldsmith, and A. Bahai, “Energy-Efficiency of MIMO
and Cooperative MIMO in Sensor Networks,” IEEE J. Selected
Areas in Comm., vol. 22, no. 6, pp. 1089-1098, Aug. 2004.
[7] P. Liu, Z. Tao, S. Narayanan, T. Korakis, and S.S. Panwar,
“CoopMAC: A Cooperative MAC for Wireless LANs,” IEEE J.
Selected Areas in Comm., vol. 25, no. 2, pp. 340-354, Feb. 2007.
[8] H. Zhu and G. Cao, “rDCF: A Relay-Enabled Medium Access
Control Protocol for Wireless Ad Hoc Networks,” IEEE Trans.
Mobile Computing, vol. 5, no. 9, pp. 1201-1214, Sept. 2006.
[9] H. Shan, P. Wang, W. Zhuang, and Z. Wang, “Cross-Layer Coop-
erative Triple Busy Tone Multiple Access for Wireless Networks,”
Proc. IEEE GLOBECOM, pp. 1-5, Dec. 2008.
[10] X. Wang, J. Li, and M. Guizani, “NCAC-MAC: Network Coding
Aware Cooperateive Medium Access Control for Wireless
Networks,” Proc. IEEE Wireless Comm. and Networking Conf.
(WCNC ’12), pp. 1646-1651, Apr. 2012.
[11] C. Zhai, J. Liu, L. Zheng, and H. Xu, “Lifetime Maximization via a
New Cooperative MAC Protocol in Wireless Sensor Networks,”
Proc. IEEE GLOBECOM, pp. 1-6, Dec. 2009.
[12] J. Zhang, Q. Zhang, and W. Jia, “VC-MAC: A Cooperative MAC
Protocol in Vehicular Networks,” IEEE Trans. Vehicular Technol-
ogy, vol. 58, no. 3, pp. 1561-1571, Mar. 2009.
[13] S. Moh and C. Yu, “A Cooperative Diversity-Based Robust MAC
Protocol in Wireless Ad Hoc Networks,” IEEE Trans. Parallel and
Distributed Systems, vol. 22, no. 3, pp. 353-363, Mar. 2011.
[14] H. Shan, H. Cheng, and W. Zhuang, “Cross-Layer Cooperative
MAC Protocol in Distributed Wireless Networks,” IEEE Trans.
Wireless Comm., vol. 10, no. 8, pp. 2603-2615, Aug. 2011.
[15] http://www.scalable-networks.com/products/qualnet/, 2013.
[16] C.E. Perkins and E. Royer, “Ad-hoc On-demand Distance Vector
Routing,” Proc. IEEE Workshop Mobile Computing Systems and
Applications, 1999.
[17] J. Broch, D.A. Maltzn, D.B. Johnsonn, Y. Hun, and J. Jetcheva, “A
Performance Comparision of Multi-hop Wireless Ad Hoc Net-
work Routing Protocols,” Proc. ACM MobiCom, Oct. 1998.
[18] Wireless LAN Medium Access Control (MAC) and Physical Layer
(PHY) Spec, IEEE 802.11 Standard, 1999.
[19] T. Himsoon, W.P. Siriwongpairat, Z. Han, and K.J.R. Liu,
“Lifetime Maximization via Cooperative Nodes and Relay
Deployment in Wireless Networks,” IEEE J. Selected Areas in
Comm., vol. 25, no. 2, pp. 307-317, Feb. 2007.
[20] A.K. Sadek, W. Yi, and K.J.R. Liu, “On the Energy Efficiency
of Cooperative Communications in Wireless Sensor
Networks,” ACM Trans. Sensor Networks, vol. 6, no. 1, article 5,
Dec. 2009.
[21] S. Kadloor and R. Adve, “Relay Selection and Power Allocation in
Cooperative Cellular Networks,” IEEE Trans. Wireless Comm.,
vol. 9, no. 5, pp. 1675-1685, May 2010.
[22] A. Ibrahim, A.K. Sadek, W. Su, and K.J.R. Liu, “Cooperative Com-
munications with Relay-Selection: When to Cooperate and Whom
to Cooperate with?” IEEE Trans. Wireless Comm., vol. 7, no. 7,
pp. 2814-2827, July 2008.
[23] E. Beres and R. Adve, “Optimal Relay-Subset Selection and
Time-Allocation in Decode-and-Forward Cooperative
Networks,” IEEE Trans. Wireless Comm., vol. 7, no. 7, pp. 2145-
2156, July 2010.

[24] L. Chen, L. Libman, and J. Leneutre, “Conﬂicts and Incentives in
Wireless Cooperative Relaying: A Distributed Market Pricing
Framework,” IEEE Trans. Parallel and Distributed Systems, vol. 22,
no. 5, pp. 758-772, May 2011.
[25] A.J. Goldsmith, Wireless Communications. Cambridge Univ. Press,
2005.
[26] C.A. Patterson, R.R. Muntz, and C.M. Pancake, “Challenges in
Location-Aware Computing,” Pervasive Computing, IEEE, vol. 2,
no. 2, pp. 80-89, Apr. 2003.
[27] C. Savarese, J. Rabaey, and J. Beutel, “Locationing in Distributed
Ad-hoc Wireless Sensor Networks,” Proc. Int’l Conf. Acoustics,
Speech, and Signal Processing, 2001.
[28] M. Gokturk and O. Gurbuz, “Cooperation in Wireless Sensor Net-
works: Design and Performance Analysis of a MAC Protocol,”
Proc. IEEE Int’l Conf. Comm. (ICC ’08), pp. 4284-4289, May 2008.
[29] X. Qin and R. Berry, “Opportunistic Splitting Algorithms for
Wireless Networks,” Proc. IEEE INFOCOM, pp. 1662-1672, Mar.
2004.
Xiaoyan Wang (S’11-M’14)) received the BE
degree from Beihang University, China, in 2004,
and the ME in 2010 and Ph.D. in 2013 from the
University of Tsukuba, Japan. He is currently
working as the assistant professor with the Infor-
mation Systems Architecture Science Research
Division at National Institute of Informatics,
Japan. His research interests include wireless
communications and networks, with emphasis
on cognitive radio networks, game theoretical
approaches, and cooperative communications.
Jie Li (M’94-SM’04) received the BE degree in
computer science from Zhejiang University,
Hangzhou, China, the ME degree in electronic
engineering and communication systems from
the China Academy of Posts and Telecommuni-
cations, Beijing, China. He received the DrEng
degree from the University of Electro-Communi-
cations, Tokyo, Japan. He has been with the Fac-
ulty of Engineering, Information and Systems,
University of Tsukuba, Japan, where he is a pro-
fessor. His research interests are in mobile dis-
tributed computing and networking, network security, mobile cloud
computing, OS, modeling, and performance evaluation of information
systems. He has served as a secretary for Study Group on System Eval-
uation of IPSJ and on several editorial boards for IPSJ Journal and so
on, and on Steering Committees of the SIG of System EVAluation (EVA)
of IPSJ, the SIG of DataBase System (DBS) of IPSJ, and the SIG of
MoBiLe computing and ubiquitous communications of IPSJ. He has
been a co-chair of several international symposia and workshops. He
has also served on the program committees for several international
conferences such as IEEE INFOCOM, IEEE ICDCS, IEEE ICC, IEEE
GLOBECOM, and IEEE MASS. He is a senior member of the IEEE and
ACM, and a member of Information Processing Society of Japan (IPSJ).
For more information on this or any other computing topic,
please visit our Digital Library at www.computer.org/publications/dlib.

IEEE BE-BTECH NS2 PROJECT@ DREAMWEB TECHNO SOLUTION

More Related Content

What's hot (20)

Viewers also liked (7)

Similar to IEEE BE-BTECH NS2 PROJECT@ DREAMWEB TECHNO SOLUTION (20)

Recently uploaded (20)

IEEE BE-BTECH NS2 PROJECT@ DREAMWEB TECHNO SOLUTION