Communications and Network, 2013, 5, 512-516
http://dx.doi.org/10.4236/cn.2013.53B2094 Published Online September 2013 (http://www.scirp.org/journal/cn)
Performance Analysis of Wireless Sensor Network Based
on OPNET*
Yi Sun, Yue Sun, Peng Xu, Haocheng Liu
School of Electrical and Electronic Engineering, North China Electric Power University, Beijing, China
Received May 2013
ABSTRACT
The wireless sensor network is a new hot spot in the field of wireless network. Compared with other network technolo-
gy, wireless sensor network has less communication protocols, mainly including Wireless HART, ISA100.11a,
IEEE1451, and ZigBee/IEEE802.15.4. IEEE802.15.4 was designed for short-range, low-complexity, low-power, low-
rate and low-cost wir eless network which provides two-way wireless communication technology [1]. In this paper, the
performance of IEEE 802.15.4 is analyzed based on OPNET simulator. The simulation result indicates the influence of
ACK mechanism and different network load on the system performance, i.e. end-to-end delay, packet reception ratio
and throughout of node, which provides an important theoretical basis for the construction of actual network.
Keywords: Wi re less Sensor Networ k; OPNET; IEEE8 0 2.15.4; Performance
1. Introduction
As wireless com mun ication, integrated circu its, sensor and
micro electromechanical system, i.e. MEMS, are rapidly
developing, it is possible to the mass production of tiny
sensor nodes with the functions of wireless communica-
tion, data acquisition and processing and collaboration
[2]. Wireless sensor network is an intelligent private
network composed of sensor nodes with specific func-
tions which exchange information by self-organizing wire-
less communication and accomplish a specific function
together [3]. IEEE802.15.4 agreement with low-rate aims
at meeting the need of low-power and low-cost in the
industrial automation, intelligent household, medical ap-
plication, etc. Therefore, the research of the performance
of IEEE802.15.4 is crucial for the design of wireless
sensor net wo rk .
OPNET employs object modeling method and graphi-
cal editor, providing comprehensive simulation develop-
ment environment for the communication network and
modeling of a distributed system. Based on event driven
mechanism and compared with the time driven, compu-
tational efficiency has been greatly improved [4]. OP-
NET provides three modeling mechanism: the bottom is
the process model, implementing the algorithm agree-
ment; the middle is the node model, employing process
model to implement the corresponding equipment func-
tion; the top is the network model, employing node mod-
el to construct reality network topology structure. Three
layer models completely correspond to the actual proto-
col, equipment and network, which fully reflect the net-
work related features.
The purpose of this paper is to investigate the perfor-
mance of IEEE802.15.4 in different network require-
ments. We use OPNET to analyze different scenarios, i.e.
whether to adopt an ACK mechanism and different net-
work load. According to the simulation results we ana-
lyze the performance of different parameters, including
different mode and load condition. Based on the simula-
tion we can get a suitable solution for the network per-
formance.
The rest of this paper is arranged as following. A brief
description of IEEE802.15.4 is given in Section 2. Sec-
tion 3 introduces the selection of simulation topology.
The simulation parameter configuration is given in Sec-
tion 4. Simulating results are depicted and analyzed in
Section 5 and then Section 6 gives the conclusion.
2. The IEEE 802.15.4 MAC Protocol
IEEE802.15/TG4 formulated the IEEE802.15.4 for low-
rate wireless personal area network, i.e. LR-WPAN. The
standard gives priority to low-power, low-rate and low-
cost, aiming at providing unified standard for individuals
or families. The characteristics of LR-WPAN network
are similar to the sensor network, and it is regarded as
*
The work was supported by a grant from the National Science and
Technology M ajor Project of China (No.2010ZX03006-005-01).

Y. SUN ET AL.
513
sensor communication standard by many research insti-
tutions. The group devotes to the standard of the physical
layer of WPAN network, i.e. PHY, and media access
layer, i.e. MAC, aiming at providing communication
standards to communicate with wireless communication
device in the personal operating space, i.e. POS. POS
generally refers to the scope of 10 meters near to users
and users can be stationary or moving within this scope.
The agreement is designed with the following features
[5-7]:
• Data rates of 250 kbps, 40 kbps and 200 kbps.
• Support network topology: the star and peer to peer.
• Two addressing modes: 16-bit short and 64-bit IEEE
addressing.
• Support optional time slot allocation.
• 16 channels in the 2.45 GHz ISM band, 10 channels
in the 915 MHz ISM band and one channel in the 868
MHz band.
• Support carrier sense multiple access with collision
avoidance, CSMA-CA.
• Handshake protoc ol for trans fer reliability.
• Power management to ensure low power consump-
tion.
• Support the link quality indicator.
The architecture of IEEE802.15.4 is shown in Figure
1. The structure includes the physical layer, the network
layer and the application layer. The physical layer con-
tains the control mechanism of the radio frequency tran-
sceiver along with its low-level and the access mechan-
ism that the MAC sub-layer provides access to the phys-
ical channel. The network layer provides the network con-
figuration and message routing. The applicatio n layer pro-
vides predetermined functions of equipment. Of these,
the physical and medium access control layers have com-
pleted the standardized.
Figure 1. 802.15.4 System structure.
3. The Selection of Simulation Topology
In different application environment, IEEE802.15.4 pro-
vides two types of network topology: the star and peer to
peer topology [7-9]. In this paper, mesh network and a
cluster of tree network which is based on a peer to peer
topology, as Figu res 2 and 3 show, are used as the simu-
lation scenarios. A lthough in practice the topology that a
specific network adopt may differ, this two topologies
are two of the most basic situatio n and have enough been
used in our experiment to conclude some basic properties
of IEEE802.15.4. In a mesh network, there is only one
PAN coordinator. It allows any device to make message
route other device in multiple hops and can provide reli-
able multi-path routing. Cluster tre e network is a specific
peer-to-peer network. Most of the equipment in the net-
work is FFD, and RFD can only be connected to the end
node on a brand. Any FFD can act as a coordinator
which coordinate other devices and provide synchroniza-
tion services. But at the same time there can be only one
FFD as a coordinator in the entire network.
4. Build Simulation Scenario
The following three metrics are defined to measure the
performance of IEEE 802.15.4.
Figure 2. Mesh typology.
Figure 3. Cluster tree network.

Y. SUN ET AL.
514
Throughout of the node: the number of forwarding
data bits of each node in the network.
Delay: the needed time of a packet transmitted from one
device to another in the network, including transmission
delay, propagation delay, processing delay and queuing
delay.
Packet reception ratio: the ratio of all sent packets to
all packets successfully received in MAC layer of the
coordinator.
A simulation scenario is built to test the performance.
In this scenario, 15 nodes are distributed randomly in a
50 × 50 m2 area. The source packets obey Poisson dis-
tribution. The message of application layer is 1024 bits.
All the wireless transmission models are based on two-
ray-ground. What’s more, the direct propagation path and
the ground reflection path are both considered in the ex-
periment, so it’s more accurate than Free-Space model.
The value of BO (Beacon Order) and SO (Superframe
Order) are equal in the model, and there’s no inactive
part in Superframe.
The simulation scenario parameters are listed in Table
1.
5. Experiment and Simulation Analysis
Two experiments are designed to investigate the perfor-
mance of IEEE 802.15.4.
Experiment 1: The effects of ACK (ACKnowledge
Character) on the system performance.
This experiment is designed to compare the change of
packet reception ratio and delay between two scenarios:
one with ACK, and another without. This will provide a
theoretical bas is for that whet her the ACK is needed when
building a network. When a coordinator wishes to trans-
fer data to a device in a beacon-enabled network, it indi-
cates in the network beacon that the data message is
pending. The device periodically listens to the network
beacon, and if a message is pending, transmits a MAC
command requesting this data, using slotted CSMA-CA.
The device acknowledged the successful reception of the
data by transmitting an acknowledgement frame. The
process is shown in Figure 4, and the experiment results
are shown in Figures 5 and 6.
Table 1. Parameters in Simulation Scenario [10-11].
Parameters Values
packet size (bit) 1024
Min backoff exponent 3
channel sensing duration (s) 0.1
packet reception-power threshold (W) 0.5
traffic type poisson
routing protocol AODV
simulation time (s) 1200
It can be seen that when packet interval time is less
than 0.8 s, ACK has little effect on the packet reception
ratio, while the delay is much larger than the non-ACK
Figure 4. Pending Data from Coordinator to Device at
Beacon-enabled M od el.
Figure 5. Influence of ACK on Reception Ratio.
Figure 6. Influence of ACK on Delay.
Coordinator End Device
Data Packet
ACK
(optional)
Data packet
Beacon
ACK
(optional)

Y. SUN ET AL.
515
network. Once the interval time is larger than 0.8 s, the
packet reception ratio of scenario with ACK is 30 percent
higher than the non-ACK one, while the delay between
two scenarios are nearly the same. The reason is that
when the sending interval time is very small, there will
be many packets in the network which are transmitted at
the same time. That will cause a higher collision ratio so
the packet reception ratio is very low, besides, the net-
work with ACK has more packets to transmit, so its de-
lay is much larger. As the increasing of sending interval
time, the probability of data collision is becoming lowe r,
so that the effic ien cy of ACK is reflected. Therefore, ACK
is appropriate for the network which packet interval time
is larger. For the network with giant flow, ACK would
have the opposite effect.
Experiment 2: The effect of node number on system
performance.
This experiment is designed to trace the change of node
throughout and delay in scenario with different node num-
bers, wishing to analyze how to deploy the density of
nodes in di fferent application scenario.
As shown in the Figures 7 and 8, when the node
number is between 15 and 20, the throughput of node
Figure 7. Throughout of Node.
Figure 8. Average peer-to-peer delay.
increases along with the node numbers. However, due to
the network congestion, once the node number is larger
than 20, the average throughout decreases. Since the in-
crease of node numbers results in higher collision ratio,
the delay also accordingly increases.
6. Conclusion
IEEE 802.15.4 was designed to meet the need for a low-
cost and low-power wireless network. It has features such
as low transmission rate, small communication range, easy
to install, etc. In this paper, a scenario is built based on
OPNET and two experiments are taken for investigating
the performance of IEEE802.15.4. By using OPNET, we
have simulated the influence of ACK on network per-
formance in different packet intervals. According to the
results, for the application environments with different
requirement in time delay and reliability, the option of
packet sending ratio and validation mechanism is flexible.
In addition, by simulating the influence on the system
throughput and delay in different network load condi-
tions, it proves that network throughput has a largest
threshold, and with the increase of network scale, net-
work could reach conges tion and lead to the deter ioration
of the network performance. Therefore, we should com-
prehensively investigate the network load impact on sys-
tem performance when forming a network.
REFERENCES
[1] L. Xu, Z. D. Chen and C. Huang, “The Application of
Game Theory in Wireless Sensor Network,” Science
Press, 2012.
[2] J. X. Liang, “Research on Energy Efficiency Technolo-
gies of MAC Protocols in WSN Based on IEEE 802.15.4,”
Master Thesis, Xidi an University, Xi’an, 2010.
[3] R. Shorey, A. Ananda, M. C. Chan and W. T. Ooi, “Mo-
bile Wireless and Sensor Networks Technology Applica-
tion and Future Directions,” China Machine Press, 2010.
[4] W. B. Wang and J. W. Zhang, “OPNET Modeler and
Simulation,” Posts & Telecom Press, 2003.
[5] M Gribaudo, D Manini, A Nordio and C. F. Chiasserini,
“Analysis of IEEE 802.15.4 Sensor Networks for Event
Detection,” IEEE Global Telecommunications Confe-
rence, 2009, pp. 1-6.
[6] B Xia, Q Fu, D Li and L. Zhang. “Performance Evalua-
tion and Channel Modeling,” 16th Asia-Pacific Confe-
rence on Communications, 2010, pp. 497-502.
[7] Wireless Medium Access Control (MAC) and Physical
Layer (PHY) Specifications for Low Rate Wireless Per-
sonal Area Networks (LRWPANs), IEEE Std. 802.15.4,
2006.
[8] P. D. Marco, P. Park, C. Fischione and K. H. Johansson,
“Analytical Modelling of IEEE 802.15.4 for Multi-hop
Networks with Heterogeneous Traffic and Hidden Ter-
minals,” IEEE Global Telecommunications Conference,

Y. SUN ET AL.
516
2010, pp. 1-6.
[9] M. Gribaudo, D. Manini, A. Nordio and C. F. Chiasserini,
“Transient Analysis of IEEE 802.15.4 Sensor Networks,”
IEEE Transactions on Wireless Communications, April
2011, pp. 1165-1175.
[10] Y. Zhang, S. Wang, Z. Liu, W. Zhou and D. Liu, “Per-
formance Analysis of Wireless Sensor Network Based on
NS-2,” IEEE International Conference on Systems and
Informatics, 2012, pp. 1445-1448.
[11] C. Ramassamy, I. HFoucha and P. Hunel, “Impact of
Transmission Range in 802.15.4 with Usual Routing
Protocols,” Wireless Communications and Mobile Com-
puting Conference (IWCMC), 2012, pp. 728-733.