Lab Assignments: To Be Completed by August 10
In these lab assignments, you are to design, run, and analyze various
simulation tasks. Your submission should consist of a brief (2-3 pages)
report that describes your simulation setup, presents the major
results, and analyzes the findings: did they make sense to you, were
there any surprising insights, etc. I do not need/want to see printouts
of simulation scripts or raw data sets. And obviously, to allow me to
read the reports, the descriptions should be written in English. While
I am not going to mark the grammar and spelling, it will be up to you
to write the reports well enough that I can understand your
presentation.
Also note that the assignments are meant to be individual assignments.
The course has a TA you can ask for help, and you can discuss the basic
approach for each task with your classmates, but I expect that each
student designs and runs his/her own simulations, collects their own
results, and submits an individually written report. However, the lab
assignments follow somewhat closely the experiments we discuss in
class, so you are welcome to use the scripts and other material I am
posting on the NS2 webpage for the course.
Finally, I have to be able to read and print the reports. So please use
a popular word processor such as Word (or a format that can be read by
Word). Alternatively, you can submit the report in PDF format.
Lab Assignment 1: ZigBee
In class, we experiment with determining the maximum throughput for
IEEE 802.11 MACs. NS2 also supports ZigBee as an alternative MAC
protocol. Information on how to set up and run NS2 simulations
with ZigBee, including sample scripts, can be found on this website.
Write scripts that experimentally determine the maximum throughput for
ZigBee and write a (brief) report that summarizes your findings. Your
report should plot achievable throughput versus number of stations
(using a fixed packet size) and also throughput versus packet size (for
a fixed number of stations).
Lab Assignment 2: OLSR
OLSR is a proactive routing protocol for MANETs, standardized by the IETF as RFC 3626. You can download a version of NS2 that implements OLSR from the University of Murchia or from INRIA
in France. In class, we experiment with AODV and DSDV as routing
protocols and the conclusion is that in basically all cases AODV (the
reactive protocol) is superior to DSDV (a proactive protocol). Yet
apparently the IETF believes that there are scenarios where proactive
protocols are superior/more appropriate. Design simulation scenarios
that capture such cases, run experiments with both AODV and OLSR, and
report the differences in performance. The report should discuss both
protocol overheads and protocol performance (packet delivery ratio,
packet latency).
Lab Assignment 3: HTTP/FTP/... Traffic
Create a wired-cum-wireless scenario in which a basestation connects a
wired node to a MANET with 30 nodes. Establish an application-level
data flow between one of the wireless nodes and the fixed node (using
FTPor HTTPor any other application layer protocol of your choice). Also
establish a controlled number of background CBR flows among the mobile
nodes in the MANET. Run experiments to determine the impact of the
MANET background traffic on your application flow performance. Some
items/criteria to consider could be throughput, packet loss, etc. as a
function of the background traffic (number of flows, level of mobility, or offered
load: total number of packets/second multiplied by size).