Lab 09 Mobile WLAN
Lab 09 Mobile WLAN
OBJECTIVES
This lab simulates mobility in wireless local area networks. Well study the effect of mobility
on the TCP performance. In addition, the lab examines how the request to send (RTS) and
clear to send (CTS) frames are utilized in avoiding the hidden node problem usually induced
by mobility in WLANs.
OVERVIEW
One of the requirements of the IEEE 802.11 standard is to handle mobile stations in wireless
local area networks (WLANs). Mobile stations are defined as the stations that access the LAN
while in motion. IEEE 802.11 handles station mobility within the MAC sublayer, and hence
such mobility is hidden from the higher layers in the network. However, the disconnection
and reconnection events induced by mobility in a WLAN significantly affect the performance
of higher-layer protocols such as TCP. For example, TCP interprets disconnection due to
mobility as congestion, and hence, it multiplicatively decreases its congestion window size.
After reconnection, TCP takes an unnecessarily longer time to recover the congestion window to a size that matches the available bandwidth.
IEEE 802.11 utilizes the request to send (RTS) and clear to send (CTS) frames in various
circumstances to further minimize collisions. RTS and CTS are especially useful in solving
the hidden node problem in WLANs that have mobile stations. Exchanging the RTS and CTS
between the sender and the receiver informs nearby stations that a transmission is about to
begin. Duration information in RTS/CTS frames are used to set the network allocation vector
(NAV) in all stations that are within the reception range of the RTS/CTS frames. This way, the
problem of a hidden sender can be solved because any station that sees the CTS frame knows
that it is close to the receiver and, therefore, cannot transmit for the period of time indicated
in the NAV. If transmitted data frames are short, sending RTS/CTS frames is not recommended, since it adds overhead inefficiency. Therefore, a threshold is defined to use RTS/CTS
only on frames longer than a specified length.
In this lab, we will simulate a wireless LAN with mobile workstations and server. The workstations will run an FTP application to upload files to the server. We will study the effect of
node mobility on the performance of the TCP connection for the FTP session. We will also
study the role of the RTS and CTS frames in avoiding the hidden node problem usually
induced by mobility in wireless LANs.
89
PRE-LAB ACTIVITIES
& Read Section 4.4 from Computer Networks: A Systems Approach, 5th Edition.
: Go to www.net-seal.net and play the following animation:
PROCEDURE
Create a New Project
1. Start OPNET IT Guru Academic Edition Choose New from the File menu.
2. Select Project and click OK Name the project <your initials>_MobileWLAN, and the
scenario Mobile_noRTSCTS Click OK.
3. In the Startup Wizard: Initial Topology dialog box, make sure that Create Empty Scenario is
selected Click Next Select Campus from the Network Scale list Click Next Make
sure that Kilometer is the unit chosen for the Size and then assign 2 and 1 to the X Span
and Y Span, respectively Click Next twice Click OK.
90
1. The Object Palette dialog box should now be on the top of your project space. If it is not there,
open it by clicking
Select wireless_lan from the pull-down menu on the object palette.
2. Add the following objects from the palette to the project workspace: Application Config,
Profile Config, two wlan_wkstn (mob), and one wlan_server (mob).
a. To add an object from a palette, click its icon in the object palette Move your mouse
to the workspace Click to drop the object in the desired location Right-click to
finish creating objects of that type.
3. Close the palette.
4. Arrange and rename the objects you added as shown:
5. Position the workstations and the server according to the x and y positions shown in the
following table:
a. To position an object: Right-click on the object Advanced Edit Attributes Edit
the x position and y position attributes.
Node
x position
y position
Client A
FTP_Server
Client B
1.25
1.5
1.75
0.5
0.5
0.5
LAB 9
Mobile Wireless Network
91
2. Select both ClientA and ClientB in the network simultaneously Right-click on one of
them Edit Attributes Check the Apply Changes to Selected Objects check box:
a. Expand the Application: Supported Profiles hierarchy Set rows to 1 Set Profile
Name to FTP_Profile.
b. Edit the Application: Destination Preferences attribute as follows: Set rows to 1
Set Symbolic Name to FTP Server Edit Actual Name Set rows to 1 In the new
row, assign FTP_Server to the Name column as shown Click OK.
92
LAB 9
Mobile Wireless Network
93
6. In the Edit Trajectory Information dialog box, verify that the trajectory info matches the
values shown in the following figure:
Note: The trajectory makes ClientA start moving after 2 min from the beginning of the simulation. ClientA waits at X Pos 0.5 for 1 min and at X Pos 0.25 for 20 s.
1.
2.
3.
4.
Click on
and the Configure Simulation window should appear.
Assign 10.0 minutes to the Duration attribute.
Assign 256 to the Seed attribute.
Click OK, and then Save your project.
94
LAB 9
Mobile Wireless Network
3. Click OK to run both simulations. Depending on the speed of your processor, this task
may take several seconds to complete.
4. After the simulation of both scenarios completes, click Close and Save your project.
95
96
LAB 9
Mobile Wireless Network
5. Go back to the Compare Results dialog box Expand the TCP Connection hierarchy
for the ClientA Select the Traffic Received (bytes) statistic Select sample_sum to
replace As Is as shown in the following figure Click Show.
6. Repeat the above step for the Traffic Received (bytes) by ClientB.
7. The resulting graphs should resemble the following graphs.
97
FURTHER READINGS
ANSI/IEEE Standard 802.11, 1999 Edition: Wireless LAN Medium Access Control (MAC) and
Physical Layer (PHY) Specifications.
Transmission Control Protocol: IETF RFC number 793 (www.ietf.org/rfc.html).
EXERCISES
1. Explain how Load and Congestion Window Size are affected by the mobility of ClientA.
2. Explain how enabling RTS/CTS helps in avoiding the hidden node problem and hence
explain the effect of RTS/CTS frames on the network performance.
3. The graphs show that the server terminates the FTP session with ClientA earlier if
RTS/CTS is enabled. However, the server terminates the FTP session with ClientB later if
RTS/CTS is enabled. Explain why.
4. Create a new scenario as a duplicate of the Mobile_noRTSCTS scenario. Name the new
scenario twoMobiles_noRTSCTS. Create a second new scenario as a duplicate of the
Mobile_RTSCTS scenario. Name the second new scenario twoMobiles_RTSCTS. In
both new scenarios, edit the attribute of the FTP_Server, and assign <your initials>_
left_trajectory to its trajectory attribute. Run the simulation for all scenarios and create
the graphs for the Load (bits/sec), Congestion Window Size (bytes), and Traffic
Received (bytes) statistic results, as we did in this lab. Analyze the graphs explaining
the effect of the server mobility on the network performance.
LAB REPORT
Prepare a report that follows the guidelines explained in the Introduction Lab. The report
should include the answers to the preceding exercises as well as the graphs you generated
from the simulation scenarios. Discuss the results you obtained and compare these results
with your expectations. Mention any anomalies or unexplained behaviors.
98