• Geo-routing: GPSR, GeRaF, etc– Motion assisted routing– Direction Forwarding
Where do we stand?
• OLSR and TBRPF can dramatically reduce the “state” sent out on update messages
• They are very effective in “dense” networks.
• However, the state still grows with O(N)
• Neither of the above schemes can handle large scale nets from 10’s to thousands of nodes
• What to do?
The previous schemes reduce control traffic O/H but do not significantly reduce routing table size
Solution: use hierarchical routing to reduce table size
In the process, reduce also control traffic O/H
Proposed hierarchical schemes include:– Hierarchical State Routing (HSR)– Fisheye State Routing (FSR)– Landmark Routing – Zone routing (hybrid scheme)
Hierarchical Routing
Hierarchical State Routing (HSR)
• Loose hierarchical routing in Internet• Main challenge in ad hoc nets: maintain/update the hierarchical partitions in the face of mobility
• Solution: distinguish between “physical” partitions and “logical” grouping– physical partitions are based on geographical proximity
– logical grouping is based on functional affinity between nodes (e.g., tanks of same battalion, students of same class)
• Physical partitions enable reduction of routing overhead
• Logical groupings enable efficient location management strategies using Home Agent concepts
HSR - physical multilevel partitions
Level = 0
5
1
7
6
11
4
23
10
98
1
23
4 Level = 1
1 3Level = 2
DestID
1
6
7
<1-2->
<1-4->
<1-3>
Path
5-1
5-1-6
5-7
5-1-6
5-7
5-7
HSR table at node 5:
HID(5): <1-1-5>
HID(6): <3-2-6>(MAC addresses)
Hierarchical addresses
HSR - logical partitions and location management
• Logical (IP like) type address <subnet,host>– Each subnet corresponds to a particular user group (e.g., tank battalion in the battlefield, search team in a search and rescue operation, etc)
– logical subnet spans several physical clusters
– Nodes in same subnet tend to have common mobility characteristic (i.e., locality)
– logical address is totally distinct from MAC address
HSR - logical partitions and location management (cont’d)• Each subnetwork has at least one Home Agent to manage membership
• Each member of the subnet registers its own hierarchical address with Home Agent – periodical/event driven registration; stale addresses are timed out by Home Agent
• Home Agent hierarchical addresses propagated via routing tables; or queried at a Name Server
• After the source learns the destination’s hierarchical address, it uses it in future packets
• Example: Landmark Routing
Scope of Fisheye
1
2
3
4
5
67
8
9
9
10
11
12
14 1516 17
18 19
20
21
2223
2425
26
27
28
29
30
31
3234
35
36
Hop=1
Hop=2
Hop>2
13
1
2
3
4
5
67
8
9
9
10
11
12
14 1516 17
18 19
20
21
2223
2425
26
27
28
29
30
31
3234
35
36
Hop=1
Hop=2
Hop>2
13
1
2
3
4
5
67
8
9
9
10
11
12
14 1516 17
18 19
20
21
2223
2425
26
27
28
29
30
31
3234
35
36
Hop=1
Hop=2
Hop>2
13
Fisheye State Routing (FSR)
Fisheye State Routing (FSR)
• Topology data base at each node - similar to link state (e.g., OSPF)
• Routing information is periodically exchanged with neighbors only ( “Global” State Routing)– similar to distance vector, but exchange entire topology matrix
• Routing update frequency decreases with distance to destination – Higher frequency updates within a close zone and lower frequency updates to a remote zone
– Highly accurate routing information about the immediate neighborhood of a node; progressively less detail for areas further away from the node
Message Reduction in FSR TC (Topology Control) message
0
5
1
2
4
3
0:{1}1:{0,2,3}2:{5,1,4}3:{1,4}4:{5,2,3}5:{2,4}
101122
LST HOP
0:{1}1:{0,2,3}2:{5,1,4}3:{1,4}4:{5,2,3}5:{2,4}
212012
LST HOP
0:{1}1:{0,2,3}2:{5,1,4}3:{1,4}4:{5,2,3}5:{2,4}
321101
LST HOP
Optimized “Fisheye” Link State Routing (OFLSR)
• Based on Optimized Link State Routing (OLSR)
• Borrows idea from Fisheye State Routing (FSR)
• Different frequencies for propagating the Topology Control (TC) message of OLSR to different scopes (e.g. different hops away)
scope 1
scope 2
scope 3
scope 4
scope width
Scalability Property of OFLSR
• Scalability to Node Mobility– 100 nodes, 1600mX1600m field, 367m Tx range
1.Node address = {subnet ID, Host ID}2.Look up local routing table to locate dest fail
3.Look up landmark table to find destination subnet LM1
4.Send a packet toward LM1
Link Overhead of LANMAR• Dramatic O/H reduction from linear to O(N) to O (sqrtN)
LANMAR: Local Scope Optimization
• Goal: find local routing scope size that minimizes routing overhead– size of landmark distance vector: O ( N / G)– size of local Link State topology map: O ( m * d ) N: total # of nodes; d: avg # of one-hop neighbors (degree)
)2
(dh
NlmH Ο=
10),2( 1 ≤≤Ο= + ααdhlocalH
locallm HHH +=
H (
Ro
uti
ng
ov
erh
ea
d)
h (scope size)
*h
Total O/H
Landmark O/H
Local route O/H
LANMAR enhances MANET routing schemes
We compare:
(a) MANET routing schemes: DSDV, OLSR and FSR; and
(b) same MANET schemes, BUT with LANMAR overlay on top
Delivery Ratio
• DSDV and FSR decrease quickly when number of nodes increases• OLSR generates excessive control packets, cannot exceed 400 nodes
OLSR
DSDV
FSR
LANMAR-DSDV
LANMAR-OLSR
LANMAR-FSR
Mobile Backbone Overlay• Landmark Overlay provides routing scalability
• However the network is still flat - paths have many hops poor TCP and QoS performance!!
• Solution: Mobile Backbone Overlay (MBO)• MBO is a physical overlay – ie long links
• MBO provides performance scalability• LANMAR extends “transparently” to the MBO
Backbone NodeBackbone Node
Logical SubnetLogical Subnet
LandmarkLandmark
sourcesource
destdest..
UAVUAV
Landmark routing concept extends transparently to the multilevel backbone
Fast BB links are “advertised” and immediately used When BB link fails, the many hop alternate path is chosen
Backbone Network and LANMAR• Why a Backbone “physical” hierarchy?
– To improve coverage, scalability and reduce hop delays
• Backbone deployment– automatic placement: Relocate backbone nodes from dense to sparse regions (using repulsive forces)
• Key result: LANMAR automatically adjusts to Backbone
• Combines low routing O/H (LANMARK logical hierarchy) + low hop distance and high bandwidth (Backbone physical hierarchy)
Extending Landmark to Hierarchical Network
• Backbone nodes are independently elected
• All nodes (including backbone nodes) are running the original LANMAR
• In addition, backbone nodes re-broadcast landmark information via higher level links
• Backbone Routes preferred by landmark (they are typically shorter)
Extending Landmark (cont)• If backbone node is lost, Landmark routing “fills the gap” while a replacement backbone node is elected
• Advantages– Seamless integration of “flat” ad hoc landmark routing with the backbone environment provides instant backup in case of failures
– Easy deployment, simple changes to ordinary ground nodes
– Remove limitations of strictly hierarchical routing
Variable number of Backbone Nodes
• Decrease of average end-to-end delay while increasing # of backbone nodes
0
10
20
30
40
50
60
0 9 18 27 36
# of backbone nodes
Average end-end delay(sec)
Hierarchical Landmark
Flat Landmark
Variable number of Backbone Nodes
• Increase of delivery fraction while increasing # of backbone nodes
0.5
0.6
0.7
0.8
0.9
1
0 9 18 27 36
# of backbone nodes
Delivery Fraction
Hierarchical Landmark
Flat Landmark
Variable Speed with 1000 nodes
Delivery fraction while increasing mobility speed
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
0 2 4 6 8 10
Mobility speed(m/sec)
Delivery fraction
Hierarchical Landmark
Flat Landmark
AODV
LANMAR implementation in IPv6 LINUX environment
• Use IPv6’s Group ID to distinguish groups• Support many more members in each group
(than IPv4)
• A packet to remote destination is routed to corresponding Landmark based on IPv6 address lookup
