Multicasting%20in%20Ad%20Hoc%20Networks

1
Multicasting in Ad Hoc Networks

• Dewan Tanvir Ahmed
• University of Ottawa
• Email dahmed_at_discover.uottawa.ca

2
Recap

• Multicasting
• Group communication
• One-to-many
• In Battle field
• Many-to-many
• Rescue team communication
• Why not using existing multicast protocol
• Resource constraints
• Frequent tree reorganization
• signaling overhead
• loss of datagram
• Protocol design
• robustness vs. efficiency

3
Multicasting in MANET

• Structure
• Tree-based
• Shared multicast tree
• Vulnerable to high mobility, load and large group
• Mesh-based
• Quick reconfigurable
• Excessive message overhead
• Focusing on
• Position Based
• Energy
• Life time improvement
• Minimizing TEC
• Reliability
• QoS, etc.

4
Multicast Routing Protocols
MAODV
AMRoute
OLAM
L-REMiT
WARM
AMRIS
MZR
MCEDAR
CAMP
?
S-REMiT
STMP
NSMP
PUMA
PAST-DM
ODMRP
DCMP
G-REMiT
ADMR
PBM
FGMP
DDR
MANSI
5
MAODV (Royer and Perkins, 1999)

• Each multicast group has a group leader
• 1st node joining a group becomes Group Leader
• Responsible for maintaining group SN (sequence
  number)
• SN ensures freshness of routing information
• A node on becoming a group leader
• Broadcasts a Group Hello message

6
MAODV (Royer and Perkins, 1999)
Group Join Process
Broadcast - RREQ
Multicast Activation
Broadcast Group Hello
Only GM Responds
L
7
MAODV (Royer and Perkins, 1999)
Leaving a Multicast Group
Non leaf Node Must remain as a Tree member
L
Leaf Node Send a Prune
Again Leaf Node Remove himself from MT
8
MAODV (Royer and Perkins, 1999)

• Observation
• Similar to unicast AODV
• Leader helps in tree maintenance
• No alternate path as it forms a tree
• Excessive use of RREQ
• lead to multicast tree instability

9
ODMRP (Bae, Lee, Su, Gerla, 2000)
Join Reply
Join Request
Forwarding Group
Broadcast
Multicast RT
b
Y, Z
s
b, c
s
c
s
X
s
a, W
s
d, e
s
Sender
e
a
d
10
ODMRP (Bae, Lee, Su, Gerla, 2000)
Robustness
11
ODMRP (Bae, Lee, Su, Gerla, 2000)

• Observation
• Sender Forms and Maintains the multicast group
• Dont need to be built on top of a unicast
  routing protocol
• Richer connectivity
• May have multiple routes for one particular
  destination
• Helps in case of topology changes and node
  failures
• soft state
• Member nodes are refreshed as needed by source
• Do not send explicit leave message
• Periodic Broadcast of Join Request
• Control overhead of route refreshes gt
  Scalability issue.

12
PAST-DM (Chao Prasant, 2004)

• Progressively Adapted Sub-Tree in Dynamic Mesh
• Build virtual mesh spanning all members
• Use unicast tunneling Concept

13
PAST-DM (Chao Prasant, 2004)
Initial Virtual Physical Topology
D
Adapt Virtual Topology
B
A
Same Initial Topology
C
Physical Topology Changed
D
D
D
B
A
B
C
B
A
A
C
C
14
PAST-DM (Chao Prasant, 2004)
Dynamic Mesh Creation
Group Join Request - ERS
One of the Group member Respond
Not Blind Flooding
Send Virtual Link State Packet
15
PAST-DM (Chao Prasant, 2004)
Data Delivery Tree
A
C
A
B
C
D
E
B
D
E
16
PAST-DM (Chao Prasant, 2004)

• Pros
• Easy to join to a group
• Cons
• Link State Table exchange with neighbors (no
  flooding)
• Mandatory GROUP_LEAVE message
• Hard to prevent
• different unicast tunnels from sharing same
  physical links
• To Construct Data Delivery tree
• Whole topology information is required
• i.e. Decision is local but information is global

17
PBM (Martin et al., 2003)

• Position Based Multicast
• Forwarding Decision
• Based on Geographical Position
• Sender has the knowledge (assumptions)
• Position of destination(s)
• Position of neighbor(s)
• Its own position
• No Maintenance of distribution structure
  (Tree/Mesh)
• Resorts flooding

• Two conflicting minimization goals
• Length of path to individual destination
• Total hops to forward to all destination

18
PBM (Martin et al., 2003)
Current Forwarding Node
Two conflicting Minimization goal
Potential Forwarding Node
Total hops to forward to all destination
Length of path to individual destination
19
PBM (Martin et al., 2003)
Greedy Multicast Forwarding
K Forwarding node N Set of all neighbors of
K W Set of all subsets of W Z Set of all
destinations d(x,y) distance between x and y
Minimize the expression
Remaining distance to all destinations
20
PBM (Martin et al., 2003)
Perimeter Multicast Forwarding
No progress for destination(s)
Greedy Multicast
Perimeter Multicast
Like FACE Traverse the boundaries of the gaps in
the network until Greedy can be resumed
21
PBM (Martin et al., 2003)

• Observation
• Static Environment
• Performs well
• Dynamic Environment
• Create routing loops
• Packet loss

22
L-REMiT (Bin and Gupta, 2003)

• Energy Optimization
• Total Energy Consumption (TEC)
• Network/System Lifetime (NL/SL)

Minimum Energy Multicast Tree
Maximum Lifetime Multicast Tree
TEC 12 EU/P NL 480/12 40 P
TEC 810 18 EU/P NL 480/10
48 P
23
L-REMiT (Bin and Gupta, 2003)
Energy Cost of a node

Life Time of a node
Life Time of MT
Bottleneck Node
24
L-REMiT (Bin and Gupta, 2003)
Lifetime - Refining Energy efficiency of
Multicast Tree
gain LT(Tnew) LT(Told)
25
L-REMiT (Bin and Gupta, 2003)
Bottleneck Node
Save 9
2
8
4.75
4
3.3
2.5
1
2.75
9
2.25
6
10
1
2.4
1.5
4.3
2.3
3
7
5
26
Conclusion

• State of the Art
• MAODV
• Low overhead
• Low latency
• ODPRP
• Backup paths
• Scalability issues
• Holes at Energy Saving

27
References

• S. Corson and J. Macker, Mobile ad hoc
  networking (MANET) Routing protocol performance
  issues and evaluation considerations, RFC 2501,
  January 1999.
• E. Royer, and C. E. Perkins, Multicast operation
  of the ad hoc on-demand distance vector routing
  protocol, MobiCom, Aug. 1999, pages 207-218.
• Sung-Ju Lee, William Su, and Mario Gerla,
  "On-demand multicast routing protocol (ODMRP) for
  ad hoc networks", Internet Draft,
  draft-ietfmanet-odmrp-02.txt, 2000, work in
  progress.
• C. E. Perkins and E. M. Royer. Ad-hoc On-Demand
  Distance Vector Routing. In Proceedings of 2nd
  IEEE Wksp. Mobile Comp. Sys. and Apps., pages
  90--100, Feb. 1999.
• C. Gui and P. Mohapatra, Efficient Overlay
  Multicast for Mobile Ad Hoc Networks, Proc. IEEE
  WCNC03, New Orleans, LA, Mar., 2003.
• Mauve, M., Füßler, H., Widmer, J., Lang, T.,
  "Poster Position-Based Multicast Routing for
  Mobile Ad-Hoc Networks", In Proceedings of Fourth
  ACM International Symposium on Mobile Ad Hoc
  Networking and Computing MobiHoc 2003. Hrsg.
• B. Wang and S. K. S. Gupta. S-REMiT S-REMiT A
  Distributed Algorithm for Source-based Energy
  Efficient Multicasting in Wireless Ad Hoc
  Networks , In Proceedings of IEEE GlobleCOM, San
  Francisco, CA, Dec. 2003, pp. 3519-3524

28
References

• B. Wang and S. K. S. Gupta, "G-REMiT An
  Algorithm for Building Energy Efficient of
  Multicast Trees in Wireless Ad Hoc Networks", In
  Proceedings of IEEE International Symposium on
  Network Computing and Applications (NCA),
  Cambridge, MA, April 2003, pp. 265-272.
• Bin Wang, Sandeep K. S. Gupta. "On Maximizing
  Lifetime of Multicast Trees in Wireless Ad hoc
  Networks," International Conference on Parallel
  Processing (ICPP'03), 2003.
• J. J. Garcia-Luna-Aceves, and E. L. Madruga, "The
  core-assisted mesh protocol," IEEE Journal on
  Selected Area in Communications, Special Issue on
  Ad-Hoc Networks, Vol. 17, No. 8, Aug. 1999.
• C. W. Wu, Y. C. Tay, and C-K. Toh, "Ad hoc
  multicast routing protocol utilizing increasing
  id-numbers (AMRIS) Functional Specification,"
  Internet draft, IETF, Nov. 1998.
• C. Gui and P. Mohapatra, "Efficient Overlay
  Multicast for Mobile Ad Hoc Networks," Proc. IEEE
  WCNC'03, New Orleans, LA, Mar., 2003.

29
Thank You!
30
Questions and Answers

• Mention two differences between MAODV and ODMRP?
• AODV uses a shared bi-directional multicast tree
  while ODMRP maintains a mesh topology rooted from
  each source.
• ODMRP broadcasts the reply back to the source
  while MAODV unicast the reply.
• MAODV does not activate a multicast route
  immediately while ODMRP does.
• What algorithms are used in Position Based
  Multicasting (PBM)? When it switches one to
  other?
• Greedy multicast (GM)
• Perimeter multicast PM)
• When there is no progress for one or more
  destinations, it switches GM to PM for
• these destination(s) and continues PM until GM
  can be resumed.
• What are the two conflicting goals in designing
  multicast tree in terms of energy?
• Minimum Energy Multicast Tree
• Optimizes (minimize) total energy consumption of
  the multicast tree
• Maximum Lifetime Multicast Tree
• Optimizes (maximizes) lifetime of the multicast
  tree.