ComponentBased Routing for Mobile Ad Hoc Networks

1
Component-Based Routing for Mobile Ad Hoc
Networks

• Chunyue Liu, Tarek Saadawi Myung Lee

CUNY, City College
2
Outline

• Motivation
• Objective
• Component analysis
• Current work
• Future work

3
Motivation

• Existing research on evaluating MANET routing
  performance is limited to predefined scenario
• The reasons of simulation results on MANET
  routing are not well understood and interpreted
• The shared similarities and specific properties
  of existing MANET routing protocols are never
  analyzed under a generalized structure.

4
Objective

• Develop a generalized component-based routing
  protocol that can cover most existing MANET
  routing protocols
• Analyze existing MANET routing protocols at
  component level
• Optimize component-based routing performance

5
Component analysis

• Route information representation
• Route information initialization
• Route discovery
• Route information management

6
Our Focuses

• How route information initialization affects the
  performance of on-demand MANET routing protocols
• How to improve MANET routing performance by
  introducing multipath mechanism.

7
Route information initialization
8
Argument

• Pure on-demand is not the most efficient way to
  do ad hoc routing, some initial determination of
  necessary route information when a network is
  first deployed or a new routing domain is
  configured, may improve the total performance of
  pure on-demand routing.

9
Definition

• We define route initialization as a mechanism
  that allows source nodes to know of necessary
  route information before the transmission of real
  data packets.
• Two ways of initialization A-initialization, all
  nodes doing initialization S-initialization,
  only source nodes doing initialization. For the
  situations without initialization, we call them
  No-initialization.

10
Simulation Model

• OPNET Modeler
• The MAC layer uses the IEEE 802.11 based wireless
  radio with wireless LAN range of 250m
• 50 nodes, randomly distributed in 1500300m2
• Random waypoint mobility model
• Every result is an average of five runs of
  simulation with different simulation seeds

11
Performance Metrics

• Packet delivery rate, the ratio of the total data
  packets successfully delivered to destinations to
  those generated by sources.
• Average end-to-end delay of data packets, this
  includes all possible delays caused by buffering
  during route discovery, queuing delays at
  interface queues, retransmission delays at the
  MAC, and propagation and transfer times.
• Normalized routing overhead, the number of
  routing packets transmitted per data packet
  successfully delivered to destinations.

12
Results for DSR

• Fixed mobility, various duration of
  initialization time

     a) Average packet delivery rate vs. the
duration of initialization time b) Normalized
message overhead vs. the duration of
initialization c) End-to-end delay vs. the
duration of initialization
13
Results for DSR (Cont.)

• Fixed duration of initialization time, various
  mobility

      a) Average packet delivery rate vs.
pause time b)
Normalized message overhead vs. pause time
c) Average end-to-end delay vs. pause
time
14
Results for AODV

• Fixed duration of initialization time, various
  mobility

     a) Average packet delivery rate vs.
pause time b)
Normalized message overhead vs. pause time
c) Average end-to-end delay vs. pause
time
15
Summary

• S-initialization is always more efficient than
  A-initialization under the same traffic pattern.
• An optimal value of the initialization duration
  exists for DSR to get the best average end-to-end
  delay, while AODV does not have this feature.
• For lower mobility ad hoc network (pause time is
  greater than 100 seconds), the initialization
  brings more performance improvement to DSR than
  to AODV
• For high mobility ad hoc network, we think
  initialization should not be applied to either
  DSR or AODV

16
Future work

• Identify quality factors of a route that may
  affect MANET routing performance
• Improve routing performance by using multiple
  paths

17
Technical Approaches

• Operating conditions network size, node density,
  mobility, link capacity, traffic patterns,
  fraction of unidirection links
• Route quality factors bandwidth, freshness,
  security, energy
• Performance metrics End-to-end data throughput,
  End-to-end delay, Average delivery rate, Average
  message overhead, Energy consumption

18
Route quality factors

• Why route quality factors
• What factors
• number of routes
• distance (hop count or distance count)
• bandwidth (high or low)
• freshness (good or stale)
• security (secure or not)
• energy (node power high or low)

19
Route Freshness

• We define Route Freshness Index (RFI) as the
  probability that the route is valid, that is to
  say, all the links on the route are active.
• Link Freshness Index (LFI) is defined to be the
  probability that the link is active.
• RFI
• Maintenance of LFI and RFI

20
Multipath routing

• Multipath routing addresses the following
  components
• Route information representation
• Multiple routes discovery
• Route cache management
• Route selection strategies

21
Multiple routes discovery

• How local discovery is put together to determine
  multiple paths
• How to rank multiple paths
• How to handle route reply storm

22
Route cache management

• What kind of route information will be filled
  into cache
• Methods to update of route quality factor
  information and how often
• Methods to handle node join and leave
• Performance effects of above schemes

23
Route selection strategies

• Shortest path
• Throughput
• Reliability
• Delay, jitter
• Freshness
• Energy conservation
• f(a,b,c,d)

24
THANK YOU!