Content uploaded by Khaled Elleithy
Author content
All content in this area was uploaded by Khaled Elleithy
Content may be subject to copyright.
Content uploaded by Khaled Elleithy
Author content
All content in this area was uploaded by Khaled Elleithy
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
10.5121/ijcnc.2010.2311 140
I
NTERFERENCE OF
802.11
B
WLAN
AND
B
LUETOOTH
:
A
NALYSIS AND
P
ERFORMANCE
E
VALUATION
Anil Mathew, Nithin Chandrababu, Khaled Elleithy, Syed Rizvi
Department of Computer Science and Engineering, University of Bridgeport,
Bridgeport, CT 06604
{
amathew, nchandra, elleithy, srizvi}@bridgeport.edu
A
BSTRACT
IEEE 802.11 and Bluetooth, these two operating in the unlicensed 2.4 GHz frequency band are becoming
more and more popular in the mobile computing world. The number of devices equipped with IEEE
802.11 and Bluetooth is growing drastically. Result is the number of co-located devices, say within 10
meters, grown to a limit, so that it may causes interference issues in the 2.4 GHz radio frequency
spectrum. Bluetooth supports both voice synchronous connection oriented (SCO) data and asynchronous
connection less (ACL) packets. In this paper, we investigate the interference issues of 2.4 GHz frequency
band. In addition, this paper presents a new Bluetooth voice packet Synchronous Connection Oriented
with Repeated Transmission (SCORT) scheme to optimize the performance of 2.4 GHz frequency band by
minimizing the interference between Bluetooth and 802.11 wireless networks. For the sake of
experimental verifications, we provide a comprehensive simulation results using Matlab Simulink.
K
EYWORDS
Bluetooth, 802.11 Wireless Networks, Mobile Nodes, Interference.
1.
I
NTRODUCTION
The growth of wireless networks has transformed our daily life into such a situation that we
cannot think of a life without devices like computers, mobile phones etc. The wireless networks
that interconnect these devices are adding more and more nodes into it each minute [2]. These
devices communicate with each other using many popular standards developed by IEEE and
such other groups [1].
The most popular among these communication standards are IEEE 802.11 or Wi-Fi and the
Bluetooth. Almost 75% of the devices in the mobile computing world are equipped with either
one of these or both. These technologies use the radio frequency for communication [3, 4]. The
Bluetooth operates in 2.4GHz ISM band. Unfortunately, IEEE 802.11 also operates in the same
2.4GHz ISM band that causes significant interference. There are different versions of IEEE
802.11 like 802.11a, 802.11b, 802.11g, and 802.11n to name a few. In this paper we consider
802.11b which operates in the 2.4GHz ISM band as shown in Fig. 1. When a node using IEEE
802.11b as a wireless standard wants to send a packet through the network, it uses the carrier
sense protocol running at the medium access control (MAC) layer to determine whether the
medium is occupied or idle [5].
If it finds the medium idle (i.e., none of the other stations sensing any RF energy in the
channel), it issues Clear to Send (CTS) request packet [6] to the destination node. If destination
node wants to communicate, it sends back a small Ready to Send (RTS) packet to the sending
node. When a sending node receives a positive acknowledgement from the destination node
(i.e., RTS packet), both sending and receiving nodes can start communicating with each other
by exchanging the regular data packets. Using the same technique, while another co-located
IEEE 802.11b network tries to send the packet, it will postpone the transmission [14].
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
141
This technique provides a good resolution for mutual interference between co-located IEEE
802.11 networks. However, when it comes to a co-located Bluetooth and IEEE 802.11 network,
they just do not communicate with each other. There is a definite chance of collision when they
use the same channel at a particular time.
A Bluetooth device may haphazardly begin transmitting packets while an IEEE 802.11 device is
sending a frame [8, 9]. This may results in interference which forces the IEEE 802.11 station to
retransmit the frame when it realizes that the destination station is not going to send back an
acknowledgment. This lack of coordination is the basis for interference between Bluetooth and
802.11 [10, 15].
The objective of this paper is to build a simulation model and study the impact of interference
between IEEE 802.11b and Bluetooth. We also investigate about a new Bluetooth voice packet
to reduce interference which is proposed by IEEE working group on co-existence.
This paper is organized as follows: Section 2 presents related work. Section 3 presents the
simulation model with a brief discussion. In Section 4, we present “SCORT” the new voice
packet. Testing of the model and results are presented in Section 5. Finally we conclude in
Section 6.
2.
RELATED
WORK
Vilovic et al. [16] analyzed the wireless network performance with 802.11 devices and
Bluetooth devices co-existing. Jayaparvathy et al. [17] did a delay performance analysis of
802.11 networks. Garetto et al. analyze the performance of 802.11 WLANs that employ the
Distributed Coordination Function (DCF). They consider contending stations within radio
proximity and investigate the case in which stations operate under non-saturated conditions
[18].
Bluetooth device can send both voice and data packets through a radio channel with a data rate
of 1Mbps. Bluetooth are a short range personal area network (PAN) [12]. Its operating range is
normally no more than 10 meters. Transmitting power of a Bluetooth Tx is very low. It is just 1
mW. Bluetooth uses Gaussian Frequency Shift Keying (GFSK) modulation technique [11].
Bluetooth also uses Frequency Hopping Spread
Spectrum (FHSS) technique to reduce
Figure 1: 2.4 GHz ISM Spectrum
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
142
interference from other devices operating in the same frequency spectrum. Interference
in Bluetooth system can be recovered or sometimes avoided using various coexistence
techniques [12, 13, 14, and 15]. Figure 2 represents the utilization of time slot in
Bluetooth. In this paper, we consider synchronous connection oriented with Repeated
Transmission (SCORT) to reduce the effect of interference in Bluetooth synchronous
connection oriented (SCO) voice links.
A time division multiplexing technique divides the channel into slices of 625 µs slots as shown
in Figure 5. A new hop frequency is used for each slot. Bluetooth supports both voice and data
transmission. Bluetooth voice transmission is called SCO where as data transmission is refereed
as asynchronous connection less (ACL). Bluetooth SCO link is established between a master
device and a slave device in the Piconet as shown in Figure 3. SCO link uses reserved slots to
communicate. Bluetooth master device use these reserved slots to maintain the communication.
Bluetooth establishes an ACL link to transmit data. Unlike SCO, ACL links can be established
between one master device and up to seven slave devices. ACL packets are transmitted in the
free slots after SCO transmission. An ACL packet can be occupied up to one, three or five slots.
All ACL packets other than Broadcast from master are acknowledged.
2.1. Synchronous Connection Oriented (SCO) Link
Bluetooth voice transmission is done by SCO. The SCO link is a symmetric point to
point voice link for sending and receiving voice packets at regular intervals of time. The
Figure 2. Bluetooth time Slot
Figure 3. Bluetooth SCO & ACL
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
143
SCO packets are transmitted in only every sixth slot. This period of time is equal to 3.75
ms. The return path of transmission from the slave to master takes place on the next
slot. Bluetooth can support a maximum of up to three voice calls at the same time. In
Figure 4,
T1
,
T2
, and
T3
are the transmit slots for each SCO master link. Slots
R1
,
R2
,
and
R3
are the return path for the slaves.
A master device initializes and controls the SCO link. Up to a maximum of three SCO links can
be maintained by a master device at the same time. When a master device sends SCO a packet
in a slot, the slave device sends back in the following slot. This shows that the transmission of
packet is symmetric (i.e., data rate is same in both directions). The length of the Bluetooth-SCO
packet is always one slot. There is no acknowledgement for SCO packets. SCO packet
transmission happens always in reserved slots at regular time intervals, every two, four or six
slots.
There are different types of SCO voice packets like HV1, HV2, and HV3. HV1 carries 10 data
bytes and is transmitted every 2 slots, HV2 carries 20 data bytes and is transmitted every 4 slots
and HV3 carries 30 data bytes and is transmitted every 6 slots .The data rate of HV1, HV2, HV3
packets are 64Kbps. HV1 and HV2 uses 1/3 and 2/3 rate forward error correcting (FEC)
mechanisms, respectively. There is no FEC in HV3.
2.2. Asynchronous Connection Less (ACL) Link
Bluetooth data transmission is called ACL which is different from SCO transmission in many
respects. In data transmission there is no margin for error allowed. If an error occurs, those
packets must be transmitted again. Different techniques can be used to implement it. In the case
of Bluetooth ACL transmission, the system will wait for acknowledgement from the receiver. It
will send the packets repeatedly till an acknowledgement is received. The receiver will check
the packet and verify the cyclic redundancy code (CRC) to make sure that the packet is received
correctly. In ACL-Tx the throughput (in bps) must be checked. The bit error rate (BER) does
not matter much. The throughput may go down if there is a significant amount of retransmission
is required. The receiver sets the Automatic Repeat Request Number (ARQN) bit in the header
part of the packet. It will then send it to transmitter in the return path packet. That is how
receiver sends an ACK. By checking the ARQN, transmitter senses if the transmission was
successful. If the value of ARQN is 1, it means a successful transmission, and if ARQN is 0 it
means a failed transmission. In the case of a one way communication (master-to-slave), the
slave sends back a dummy packet in the next slot. NULL packet or dummy packet does not
have any payload. Figure 5 shows the DM1 packet being transmitted in the first slot, and the
slave replying with a NULL packet containing the ACK in the immediately following slot. The
master then transmits again in the next slot.
Figure 4. Bluetooth SCO voice slot
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
144
3.
B
LUETOOTH
S
IMULATION
M
ODEL
Figure 6 shows the simulation model of the network in MATLAB Simulink. This model
simulates Bluetooth in full duplex communication mode. We use two similar devices each with
a transmitter and receiver. One of them should be set as master and the other as the slave. Other
than two Bluetooth devices, we also have an 802.11b packet generating block as an interference
source, error reading meters and instrumentation.
Figure 5. Asynchronous Connection Less (ACL) link
Figure 6. Bluetooth Interference Simulation Model
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
145
3.1. Transmitter Design
The transmitter mode is shown in Figure 7. Figure7 shows the state flow diagram of the data
transmission. The transmitted model processes both data and voice input. It also performs
Header Error Control (HEC) using FEC. Buffering and modulation is also done at the
transmission side as shown in Figure 7. Frequency hopping is the transmission modulation
technique. When the “ACL_packets” is entered, the transition to “Transmit_blank_packet” will
happen. The “Enable_Audio=0" & "Get_blank_Packet=1" actions activates to disable audio and
to generate a new data packet. When the next slot is about to transmit, the transmitter checks the
status of ARQN bit returned from the receiving device.
If it is in "Transmit_blank_Packet" ARQN is one, it stays in the state and transmits another new
packet. If ARQN is zero, it shifts to the "Re_Transmit_Packet". This simulation model uses
frame based processing. It can transmit samples having high number of frames in each step of
the simulation. This technique enables quick simulation of digital systems. In this particular
model, a top sample rate of 100 MHz is used.
Figure 8 shows the state flow diagram of the data transmission. When the “ACL_packets” is
entered the transition to “Transmit_blank_packet” will happen. The “Enable_Audio = 0" &
"Get_blank_Packet = 1" actions activates to disable audio and to generate a new data packet.
When the next slot is about to transmit, the transmitter will check the status of ARQN bit
returned from the receiving device. If it is in "Transmit_blank_Packet" ARQN is one, it stays in
the state and transmits another new packet. If ARQN is zero, it shifts to the
"Re_Transmit_Packet". If the transmitter is in “Re_Transmit_Packet", and ARQN is one, it
shifts to “Transmit_blank_Packet" else it will not shift and will stay in "Re_Transmit_Packet".
Figure 7. Bluetooth device having both Transmitter & Receiver
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
146
3.2. Receiver Design
The state flow diagram of receiver is shown in Figure 9. It can be seen in Figure 9 that the
receiver waits a new packet all the time. When it senses the arrival of a packet it will register the
arrival. It will also make sure the decoder is enabled. The above mentioned sequence of events
is triggered because of the detection of an arriving packet. The receiver has to make a number of
decisions to make sure whether the received packet is correct or incorrect.
A DM1 packet will be checked for integrity. The receiver performs a HEC. The address is also
verified. The receiver makes sure the packet is new and is not a duplicate. It also
checks the
CRC.
If all these checks are correct, the packet will be accepted else the packet will be rejected. This
happens in the case of a repeated packet arriving or in the case of its CRC failing. This flow
diagram is implemented in State flow semantics as shown in Figure 8. This image captured
during a simulation, illustrates the animation provided with State flow which highlights the
decision path (in bold) through the flow chart.
3.3. Channel and Interferer Modeling
The 802.11b channel bandwidth is approximately 22 MHz. The Simulation model has a block
which produces signals in this bandwidth. This block can be configured to specify mean packet
rate, packet length, power, and frequency location in the ISM band. This block is then
connected to the channel where the distance between the interference source and Bluetooth
system can be varied. Figure 10 shows the addition of 802.11b interference into the channel. We
use this model in our experimental verifications to determine the behavior of added interference.
4.
C
OEXISTENCE
S
OLUTION
-
SCORT
V
OICE
T
RANSMISSION
The coexistence task group working on coexistence has suggested the use of a special voice
packet to fight interference. The synchronous connection-oriented with Repeated Transmission
(SCORT) packet achieves more robust transmission by replacing bit-level redundancy with
packet-level redundancy. The state flow diagram of SCORT is presented in Figure 11. It works
Figure 8. Transmitter state flow diagram Figure 9. Receiver state flow diagram
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
147
by repeating the transmission of the same packet three times in one SCO interval. SCORT does
not have any error correction. SCORT is transmitted every second time slot. As the same packet
is being transmitted three times in a row, only one voice link will be there which is a full duplex
link. If interference destroys the transmission during first slot, there are still three other slots or
opportunities to communicate the packet. This provides an improvement for frame-error rate
(FER) in an interference scenario. It does not affect the BER of the payload.
Figure 10. 802.11b Interference Source added
Figure 11. SCORT State Flow Diagram
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
148
5.
E
XPERIMENTS
A
ND
R
ESULTS
Using the above models, we perform a series of tests to evaluate the performance of a Bluetooth
system under interference. We used DM1 packet type to check the performance of ACL
transmission. Packet types HV1, HV2 and HV3 are used to evaluate SCO performance. Finally
we used SCORT packet type to compare its performance with HV1, HV2 and HV3.
Figure 12 shows the Bluetooth system throughput. The throughput (in kbps) has been evaluated
by varying the distance between the device and the interference source. It should be noted in
Figure 12 that a consistent values of throughput is achieved with respect to a constant increase
in the distance between the Bluetooth devices. From Figure 12, we can see that the throughput
of a Bluetooth system is about 128 kbps without 802.11b interference source.
Figure 13 shows the reduction in the throughput when 802.11b interfering source come closer to
the Bluetooth system. Figure 14 demonstrates the BER performance with respect to Eb/No. It
should be noted in Figure 14 that the BER decreases linearly over the values of Eb/No.
However, the BER divergence in Figure 14 is very rapid and acceptable for a maximum value
of Eb/No.
For Figure 15, we measured the difference in FER when using a SCORT voice packet rather
than the regular HV1, HV2, and HV3 packets. From Figure 15, we can see that when using
SCORT packets, there is a considerable reduction in the FER.
6.
C
ONCLUSIONS
This paper presented a model for the interference of both IEEE 802.11 b and Bluetooth. Our
analysis shows that interference increases significantly with respect to an increase in the number
of participating devices. Techniques such as SCORT are a big leap in the future for such
networks. Our simulation results suggest that the use of SCORT packets can minimize the effect
Figure 12. Bluetooth system throughput with respect to the distance between Bluetooth devices
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
149
of interference between 802.11 wireless network and Bluetooth devices. In the future, it is
expected that the wireless industry will mature in such a way that smooth data and voice
transmission will be achieved with a minimal interference.
Figure 13. Bluetooth Master and Slave device throughput in the presence of 802.11b
Figure 14: BER
versus Eb/No Figure 15: BER
versus large values of Eb/No
International journal of Computer Networks & Communications (IJCNC),Vol.2, No.3, May 2010.
150
R
EFERENCES
[1] Golmie, N. (2004). Bluetooth dynamic scheduling and interference mitigation, Mobile Networks
and Applications, Vol. 9, Issue 1, pp. 21-31.
[2] Golmie, N.; Dyck, R. E.; Soltanian, A. (2001). Interference of Bluetooth and IEEE 802.11:
simulation modeling and performance evaluation, Proceedings of the 4th ACM International
Workshop on Modeling, Analysis and Simulation of Wireless and Mobile Systems, pp. 11- 18.
[3] Golmie, N.; Dyck, R. E.; Soltanian, A. ; Tonnerre, A. ; Rébala, O. (2003). Interference evaluation
of Bluetooth and IEEE 802.11b systems, Journal of Wireless Networks, Vol. 9, Issue 3, pp. 201 –
211.
[4] Song, M.; Shetty, S.; Gopalpet, D. (2007). Coexistence of IEEE 802.11b and Bluetooth: an
integrated performance analysis, Mobile Networks and Applications, Vol. 12, Issue 5, pp. 450 –
459.
[5] Wadhwa, M.; Song, M. (2006). Performance of IEEE 802.11b devices in the presence of adaptive
frequency hopping enabled Bluetooth devices, Proceedings of the 4th IEEE International
Conference on Information Technology: Research and Education, pp. 74 - 48.
[6] Subramanian, V.; Ramakrishnan, K.; Kalyanaraman, S.; Lusheng, J. (2006). Impact of interference
and capture effects in 802.11 wireless networks on TCP, Proceedings of the 2nd International
Workshop on Wireless Traffic Measurements and Modeling.
[7] Mathew, A.; Chandrababu, N.; Elleithy, K.; Rizvi, S. (2009). IEEE 802.11 & Bluetooth
interference: simulation and coexistence, 7th Annual Conference on Communication Networks and
Services Research, pp. 217 - 223.
[8] Matthew, B. (2009). Wi-Fi (IEEE 802.11b) and Bluetooth coexistence issues and solutions for the
2.4 GHz ISM band, Texas Instruments.
[9] Cordeiro, C.; Abhyankar, S.; Toshiwal, R.; Agarwal, R. (2003). A novel architecture and
coexistence method to provide global access to/from Bluetooth WPANs by IEEE 802.11 WLANs,
Proceedings of IEEE International Perform Computer Communication, pp 23-30.
[10] Chek, M.; Kwok, Y. (2007). Design and evaluation of practical coexistence management schemes
for Bluetooth and IEEE 802.11b systems, International Journal of Computer and
Telecommunications Networking, Vol.51, No.8, pp. 2086 – 2103.
[11] Santivanez, C.; Ramanathan, R.; Partridge, C.; Krishnan, R.; Condell, M.; Polit, S. (2006).
Opportunistic spectrum access: challenges, architecture, protocols, Proceedings of the 2nd Annual
International Workshop on Wireless Internet, Vol. 220, Article No. 13.
[12] Kamerman, A. (2000). Coexistence between Bluetooth and IEEE 802.11 CCK: solutions to avoid
mutual interference, IEEE P802.11 Working Group Contribution, IEEE P802.11-00/162r0.
[13] Pasolini, G. (2004). Analytical investigation on the coexistence of Bluetooth Piconets, IEEE
Communications Letters, Vol. 8, No. 3, pp. 144-146.
[14] Shellhammer, S. (2001). Collaborative coexistence mechanism: TDMA of Bluetooth and 802.11,
IEEE 802.15-01/025r0.
[15] Lansford, J.; Nevo, R. (2000). MEHTA: A method for coexistence between co-located 802.11b
and Bluetooth systems, IEEE 802.15-00/036r0.
[16] Vilovic, I.; Zovko-Cihlar, B. (2004). Performance analysis of wireless network using Bluetooth
and IEEE 802.11 devices, In Proceedings of the 46th International Symposium in Electronics in
Marine, pp. 235 - 240.
[17] Jayaparvathy, R.; Rajesh, S.; Anand, S.; Srikanth, S. (2003). Delay performance analysis of 802.11
communications, In Proceedings of 9th Asia-Pacific Conference on Communications, pp. 223-226.
[18] Garetto, M.; Chiasserini, C.-F. (2005). Performance analysis of 802.11 WLANs under sporadic
traffic, Lecture Notes in Computer Science, pp.1343-1347.
