UWA Logo
  Faculty Home | School Home    
           
Home
About the School
Contact and People
Future Undergraduate Students
Prospective Postgraduates
Current Students
Current Postgraduates
Research
IT News
Awards
Industry Links and Prizes
School and IT Information
Other
Internal Information

Projects

The project for this unit requires you to use your knowledge of networking protocols to evaluate and compare two ad-hoc routing protocols, namely AODV and DSR. Much of the research to date concentrates on comparing these protocols in terms of their throughput (sometimes referred to as reliability), latency, routing overhead and energy efficiency.

Your task is to evaluate the two protocols with respect to the sizes of networks within which they are optimal. For example consider a two node network. One would expect that both AODV and DSR would operate in a similar manner. Now consider a two-hop network of three nodes. As the network grows metrics such as those mentioned above will begin to diverge for the two protocols. You are required to simulate both protocols under "different sized networks" and under "different mobilities". The purpose is to find the hop-count (if it exists) where AODV and/or DSR are no longer effective routing protocols.

This project is purposely left under-specified. Intuitively you may first consider to use the throughput as a good metric to evaluate the effectiveness of the protocol in different sized networks. However, you must also think about the design of the protocols and comment on the reasons why a particular protocol behaves the way it does. It is not enough to simply state the optimal network sizes; you must make an attempt to explain the trends. You will also find that as you vary the mobility from low to high, there will be some additional trends that may need some explaining.

Whether it sounds like it or not, this project is intended to be fun! To get you started, here are some parameters which you may choose to in your simulations:

  1. In order to keep the traffic conditions manageable, it is sufficient to only have 10 source/destination pairs for a 1500x300 network of 50 nodes. As you grow the grid and the number of nodes within the network, do not change the number of source/destination pairs. This is to keep your trace files manageable.
  2. The source should be generating traffic at 2 packets per second (period of 0.5s). This will ensure that the number of packets generated does not exceed the network bandwidth capacity.
  3. The mobility should be simulated by generating scenario scripts mentioned in the ns-2 lecture. The Marc Greis tutorial on running such scripts is available from the unit homepage. Think about the "pause" times you will want to use. If unsure, check the available literature, such as that provided in the two labsheets.

Please note that any questions about the project should be directed to the help419 forum. This will allow all students to benefit from the provided answers .

Here are some useful links :

Top of Page