Analyzing Impact of Mobility Models on Routing Protocols using ns-2, Assignments of Computer Science

A homework assignment for a university-level computer science course, where students are required to use the ns-2 network simulator to study the effect of different mobility models on routing protocols. Students must construct simulation scenarios for three mobility models: random way point (rwp), random point group mobility model (rpgm), and manhattan mobility model (mh). They need to compute statistics such as average speed, spatial distribution, link holding time, and correlation of link breakage events. Students will use routing protocols like dsr, aodv, and dsdv and analyze packet loss ratio and routing overhead.

Typology: Assignments

Pre 2010

Uploaded on 08/17/2009

koofers-user-hfb
koofers-user-hfb 🇺🇸

9 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
COSC6397 Homework Assignment 4
October 4, 2004
Due date: Oct. 20, 2004; Maximum group size = 2
In this project, we use ns-2 to study the impact of mobility models on
the performance of routing protocols.
Mobility models: Download the mobility generator from IMPORTANT
web site http://nile.usc.edu/important/. Construct simulation scenarios
for random way point model (RWP), random point group mobility model
(RPGM) and Manhattan mobility model (MH) using the parameters listed
in Table 1. Compute the following statistics from the resulting scenario file:
Average speed among all users over time.
Spatial distribution (if applicable). In a 2-D plane, at time tcompute
the density of nodes for each 50mx50m square. Vary t(e.g., t = 0, t
= 100s) and compare if the spatial distribution changes over time.
Distribution of link holding time defined as the total time that a link
exists. Draw histogram of the link hold time for each mobile scenario
(e.g., use Matlab) and compute the mean and variance. Explain why
the shapes for different mobility models are different.
(Bonus) Correlation of link breakage events. Consider two links A-B
and C-D, record the time that the link is established and broken. Find
out if there is correlation between them in particular when the links
share common nodes.
[Hint: you can either instrument the mobility generator to get these statistics
or write your own program using the output scenario files from the genera-
tor.]
Assume the wireless transmission range is 250m, i.e., if two nodes are
separated by no more than 250m, there exists a bi-directional link between
1
pf3

Partial preview of the text

Download Analyzing Impact of Mobility Models on Routing Protocols using ns-2 and more Assignments Computer Science in PDF only on Docsity!

COSC6397 Homework Assignment 4

October 4, 2004

Due date: Oct. 20, 2004; Maximum group size = 2 In this project, we use ns-2 to study the impact of mobility models on the performance of routing protocols.

Mobility models: Download the mobility generator from IMPORTANT web site http://nile.usc.edu/important/. Construct simulation scenarios for random way point model (RWP), random point group mobility model (RPGM) and Manhattan mobility model (MH) using the parameters listed in Table 1. Compute the following statistics from the resulting scenario file:

  • Average speed among all users over time.
  • Spatial distribution (if applicable). In a 2-D plane, at time t compute the density of nodes for each 50mx50m square. Vary t (e.g., t = 0, t = 100s) and compare if the spatial distribution changes over time.
  • Distribution of link holding time defined as the total time that a link exists. Draw histogram of the link hold time for each mobile scenario (e.g., use Matlab) and compute the mean and variance. Explain why the shapes for different mobility models are different.
  • (Bonus) Correlation of link breakage events. Consider two links A-B and C-D, record the time that the link is established and broken. Find out if there is correlation between them in particular when the links share common nodes.

[Hint: you can either instrument the mobility generator to get these statistics or write your own program using the output scenario files from the genera- tor.] Assume the wireless transmission range is 250m, i.e., if two nodes are separated by no more than 250m, there exists a bi-directional link between

Table 1: Parameters for Different Mobility Model Mobility model Parameter Value RWP num. of nodes 75 dimension 1000mx1000m maximum speed 20 meter/second pause time 0 RPGM num. of nodes 75 num. of groups 5 dimension 1000mx1000m num. of nodes in each group 15 speed deviation 0. angle deviation 0. group leader trace file generated using the above RWP model† MH num. of nodes 75 maximum speed 20 meter/second minimum speed 5 meter/second acceleration 10% of maximum speed map /man/mapset

†Refer to the description of checkpoint files in the user manual.

them. The above statistics reflect the degree of mobility in a mobile sce- nario. For example, the shorter the link holding time, the faster the relative movement tends to be. Next, we investigate their interaction with routing protocols.

Routing protocols: use DSR, AODV and DSDV as routing protocols respectively. Among 75 nodes, pick 10 source and destination pairs, each running a constant bit rate (CBR) source of packet size 500Byte and trans- mitting at 10Kbps. Generate statistics of packet loss ratio (computed as the number of application layer packets sent/application layer packets de- livered), routing overhead (e.g., number of RREQ, RREP messages etc.) For RPGM case, pick 6 pairs within the same group and 4 pairs from dif- ferent groups. (Bonus: vary the ratio between intra-group and inter-group source-destination pairs for RPGM and show how routing overhead may change.)

Appendix

Suggested Settings For mobile scenario, it is necessary to run the sim- ulation for at least 1000s to get representative statistics. In addition, each