Mobile Wireless Networking CS691 004 Spring 2005

1
Mobile Wireless Networking CS691 004 (Spring
2005)

• Ad Hoc On-Demand Routing

Xiaoyan Hong UA CS Department hxy_at_cs.ua.edu
2
Ad Hoc Routing Protocols

• Wireless multihop routing challenges
• Overview of proposed routing schemes
• Reactive (on demand) wireless routing

3
Wireless multihop routing challenges

• topology dynamics
• mobility, unreliable wireless connection (fading,
  external interferences), node join/leave
• limited resources
• bandwidth, computational power, energy supply
• application demands
• low latency for real time applications,
  multimedia applications
• need to scale to large numbers (100s to 1000's)

4
Wireless multihop routing challenges (contd)

• Routing concerns
• Routing overhead
• computational overhead
• table storage overhead
• Communication overhead
• In Manet, communication overhead is more critical
• routing packet size
• number of routing packets

5
Proposed ad hoc Routing Approaches

• Conventional wired-type schemes (global routing,
  proactive)
• Distance Vector Link State
• Hierarchical (global routing) schemes
• Fisheye, Hierarchical State Routing, Landmark
  Routing
• On- Demand, reactive routing
• Source routing backward learning
• Location Assisted routing (Geo-routing)
• DREAM, LAR etc

6
Conventional wired routing limitations

• Distance Vector (eg, Bellman-Ford, DSDV)
• routing control O/H linearly increasing with net
  size
• convergence problems (count to infinity)
  potential loops
• Link State (eg, OSPF)
• link update flooding O/H caused by frequent
  topology changes
• CONVENTIONAL ROUTING DOES NOT SCALE TO SIZE AND
  MOBILITY

7
Distance Vector
0
Routing table at node 5
1
3
2
4
Tables grow linearly with nodes Control O/H
grows with mobility and size
5
8
Link State Routing

• At node 5, based on the link state packets,
  topology table is constructed
• Dijkstras Algorithm can then be used for the
  shortest path

0
1
0,2,3
1,4
3
2
1,4,5
4
2,3,5
5
2,4
9
On-Demand Routing Protocols

• Routes are established on demand as requested
  by the source
• Only the active routes are maintained by each
  node
• Channel/Memory overhead is minimized
• Two leading methods for route discovery source
  routing and backward learning (similar to LAN
  interconnection routing)

10
Reactive (on-demand) routing

• Routing only when needed

• Advantages
• eliminate periodic updates
• reduce routing table size
• Disadvantages
• high flood-search overhead with
• mobility, distributed traffic
• high route acquisition latency

0
1
3
2
4
5
11
Existing On-Demand Protocols

• Dynamic Source Routing (DSR)
• Associativity-Based Routing (ABR)
• Ad-hoc On-demand Distance Vector (AODV)
• Temporarily Ordered Routing Algorithm (TORA)
• Zone Routing Protocol (ZRP)
• Signal Stability Based Adaptive Routing (SSA)
• On Demand Multicast Routing Protocol (ODMRP)

12
Dynamic Source Routing (DSR)

• Forwarding source route driven instead of
  hop-by-hop route table driven
• ltSourcegt ltIDn1, IDn2,, IDnmgt ltDestinationgt
• No periodic routing update message is sent
• Nodes ignore topology changes not affecting
  active routes with packets in the pipe
• The first path discovered is selected as the
  route
• Two main phases
• Route Discovery
• Route Maintenance

13
DSR - Route Discovery

• To establish a route, the source floods a Route
  Request message with a unique request ID
• Packet has route record ltRREQ, RREQid, IDsrc,
  IDn1, IDn2,, IDnmgt
• Duplicate detect ltinitiator, RREQidgt list
• Loop detect own id appear in route record?
• Reply?
• Append its own ID to the route record for
  re-broadcast
• Route Reply message containing path information
  is sent back to the source either by
• the destination, or
• intermediate nodes that have a route to the
  destination
• Reverse the order of the route record, and
  include it in Route Reply.
• Unicast, source routing
• Each node maintains a Route Cache which records
  routes it has learned and overheard over time

14
DSR - Route Maintenance

• Route maintenance performed only while route is
  in use
• Error detection
• Monitors the validity of existing routes by
  passively listening to data packets transmitted
  at neighboring nodes
• Lower level acknowledgements
• When problem detected, send Route Error packet to
  original sender to perform new route discovery
• Host detects the error and the host it was
  attempting
• Route Error is sent back to the sender the packet
  original src

15
DSR - Optimization

• Full use of route cache
• Cache all the possible ltsource, destinationgt
  pairs from packet route records of data or reply
  packets
• Update the cache at anytime.
• E.g
• when A-gtB-gtC-gtD is discovered,
• A leaned A-gtB, A-gtB-gtC
• B learned B-gtC, B-gtC-gtD
• Use cache to avoid broadcasting requests
• Append an existing cache entry to the accumulated
  route record in request, and send reply.
• Will have multiple replies (with different route
  lengths)
• Delay in sending reply and monitor other
  transmission of reply
• Possible loops BgtA-gtB-gtC-gtD, if A reply from
  cache, B doesnt
• A doesnt reply.
• Expending ring search

16
Ad hoc On-Demand Distance Vector Routing (AODV)

• Primary Objectives
• Provide unicast, broadcast, and multicast
  capability
• Initiate forward route discovery only on demand
• Disseminate changes in local connectivity to
  those neighboring nodes likely to need the
  information
• Characteristics
• On-demand route creation
• Effect of topology changes is localized
• Control traffic is minimized
• Two dimensional routing metric ltSeq, HopCountgt
• Storage of routes in Route Table

17
Route Table

• Fields
• Destination IP Address
• Destination Sequence Number
• HopCount
• Next Hop IP Address
• Precursor Nodes
• Expiration Time
• Each time a route entry is used to transmit data,
  the expiration time is updated to
• current_time active_route_timeout

18
Unicast Route Discovery

• Source broadcasts Route Request (RREQ)

• ltFlags, Bcast_ID, HopCnt, Src_Addr, Src_Seq,
  Dst_Addr, Dst_Seqgt
• Node can reply to RREQ if
• It is the destination, or
• It has a fresh enough route to the
  destination
• Otherwise it rebroadcasts the request
• Nodes create reverse route entry
• Record Src IP Addr / Broadcast ID to prevent
  multiple rebroadcasts

Source
Destination
Route Request Propagation
19
Forward Path Setup

• Destination, or intermediate node with current
  route to destination, unicasts Route Reply (RREP)
  to source
• ltFlags, HopCnt, Dst_Addr, Dst_Seq, Src_Addr,
  Lifetimegt
• Nodes along path create forward route
• Source begins sending data when it receives
  first RREP

Source
Destination
Forward Path Formation
20
Path Maintenance
3
3
3
1
1
Destination
Destination
2
2
Source
Source
4
4

• Movement of nodes not along active path does not
  trigger protocol action
• If source node moves, it can reinitiate route
  discovery
• When destination or intermediate node moves,
  upstream node of break broadcasts Route Error
  (RERR) message
• RERR contains list of all destinations no longer
  reachable due to link break
• RERR propagated until node with no precursors for
  destination is reached

21
Optimizations

• Hello messages
• Maintain local connectivity to neighbors.
• Update lifetime of the active route
• Detect link losses
• Local repair
• Upstream host detecting the link break locally
  broadcasts a RREQ.
• TTL is set to smaller than the hops to the
  original sender.
• Data packet buffered locally.
• If no RREP received, send RERR to the original
  sender.
• Expending ring search.

22
Evaluation

• Metrics
• Packet delivery fraction
• Average end-to-end delay
• Normalized routing load
• Network conditions
• Varying mobility
• Varying of communication sources (fixed packet
  rate)
• Varying offered load ( of sources fixed,
  increasing packet rate)

23
Observations

• Routing load
• DSR lower than AODV ( in terms of of packets)
  due to aggressive caching (though more replies
  and errors.
• AODV is dominated by RREQ packets
• Counting the MAC overhead
• DSR and AODV similar! (RREP and RERR are unicast
  in DSR)
• Packet delivery and delay
• DSR better if less stressful traffic load.
• Caching only provide significant benefits to a
  certain extent! Stale caches are chosen in high
  loads unnecessary bandwidth consumption and
  pollution of caches in other nodes.
• Delays are due to buffering (route discovery
  latency) and congestion
• Only hop counts are considered in route selection.

24
Observations (conts)

• Network load
• Network capacity is poor
• upper bound 1/(n1) fraction, n of neighbors
• Wasted bandwidth from dropped packets
• Routing packets
• MAC overhead

25
Outline

• Help from geographical Information
• LAR
• DREAM
• Comparisons

26
Location-Aided Routing (LAR)

• Ko and Vaidya (Texas A M)
• Location assisted (requires GPS)
• On-demand
• No periodic messages
• LAR works like DSR except it limits the flooded
  area of Route Requests using location information

27
Route Discovery using flooding
28
Expected Zone

• Known to sender the location of the destination L
  at a previous time t0, and the average speed of
  dest v
• An estimated region for destination to appear at
  t1.
• If no such knowledge, the expected zone is the
  entire ad hoc field.
• The more info about the mobility of the dest, the
  smaller the zone.

29
Request Zone

• Nodes within the zone forwards route requests.
• Include Expected Zone (at least src and dest)
• Tradeoffs between discovery overhead and latency.

30
Determine Request Zone

• Scheme 1
• The source specifies a rectangular request zone
  which includes the source and the expected zone
  where the destination may reside. A smallest
  rectangular.
• Source includes the four corners in the request.
• Nodes within the request zone propagate Route
  Requests. Otherwise, discard the request
• The destination includes its current location and
  time in route reply, and maybe average speed.
• Size of the zone affected by the speed and the
  time.

31
Scheme 1
32
Determine Request Zone (contd)

• Scheme 2
• The source specifies the distance (Dist_s)
  between itself and the destination and dests
  location
• Nodes forward Route Requests if their distances
  to the destination is less than or equal to the
  distance indicated by the packet using Dist_i.
  Otherwise, discard the request.
• Dist_i Delta gt Dist_j

33
Scheme 2
34
Differences of Scheme1 and Scheme2

• Assume Dist_k gt Dist_j
• In scheme1, I will not forward, but J,K,L
  forward.
• In scheme2, I, J will forward and K will not
  forward.
• So I and K act differently

35
of routing pkts per data pkts

• 30 nodes, vary speed

36
of routing pkts per data pkts

• Average speed 4.5 units/sec, vary of nodes

37
Reading list for this lecture

• DSR
• David B. Johnson and David A. Maltz. Dynamic
  Source Routing in Ad Hoc Wireless Networks. In
  Mobile Computing, edited by Tomasz Imielinski and
  Hank Korth, Chapter 5, pages 153-181, Kluwer
  Academic Publishers, 1996.
• DSR Internet Draft
• AODV
• Charles E. Perkins and Elizabeth M. Royer. Ad-Hoc
  On Demand Distance Vector Routing. In WMSCA '99,
  New Orleans, Louisiana.
• AODV Internet Draft
• LAR Young-Bae Ko, Nitin H. Vaidya
  Location-Aided Routing (LAR) in Mobile Ad Hoc
  Networks. Mobicom 1998
• DREAM Stefano Basagni, Imrich Chlamtac, Violet
  R. Syrotiuk, Barry A. Woodward A Distance
  Routing Effect Algorithm for Mobility (DREAM).
  Mobicom 1998.
• Comparison
• Josh Broch, David A. Maltz, David B. Johnson,
  Yih-Chun Hu, and Jorjeta Jetcheva. A Performance
  Comparison of Multi-Hop Wireless Ad Hoc Network
  Routing Protocols. In Proceedings of the Fourth
  Annual ACM/IEEE International Conference on
  Mobile Computing and Networking, ACM, Dallas, TX,
  October 1998.
• Samir R. Das, Charles E. Perkins, and Elizabeth
  M. Royer. "Performance Comparison of Two
  On-demand Routing Protocols for Ad Hoc Networks."
  Proceedings of the IEEE Conference on Computer
  Communications (INFOCOM), Tel Aviv, Israel, March
  2000, p. 3-12.
• Tracy Camp, Jeff Boleng, Brad Williams, William
  Navidi, Lucas Wilcox Performance Comparison of
  Two Location Based Routing Protocols for Ad Hoc
  Networks, Infocom 2002.

38
The Selective Intermediate Nodes Scheme for Ad
Hoc On-Demand Routing Protocols

• Yunjung Yi, Mario Gerla and Taek Jin Kwon
• Computer Science Department
• University of California, Los Angeles, CA 90095
• http//www.cs.ucla.edu/NRL

39
Outline

• Introduction
• Description Selective Intermediate Nodes
• Simulation Studies
• Conclusion

40
Introduction

• Malfunction of reactive routing in the stressed
  network
• Lost routing information due to the congestion
  (e.g., route reply, route query) triggers
  successive recovery routing control packets
  (e.g., re-issued route query)
• Lack of scalability with the increase of the
  offered load and the geographic network density
• Our solution
• Selective Intermediate Nodes
• Only non-saturated nodes become intermediate
  nodes based on accurate measurement of local load
• Eliminate misbehaviors of reactive routing in the
  stressed network

41
Selective Intermediate Nodes
All nodes relay RREQs

• Upon reception of RREQ (Route Queries), a node
  relays the packet if
• The local load level is IDLE
• The local load level is MODERATE and It is the CH
  or gateway
• O.W. it drops RREQs
• ? implicit Call Admission Control

Only CH and Gateways relay
No relay in this area
42
Selective Intermediate Nodes (2)

• Underlying Cluster Platform
• Based on Passive Clustering
• Use on-going traffic to exchange cluster-related
  information
• ?No extra background clustering overhead
• ?effective with reactive routing schemes
• Provide a gateway selection mechanism to reduce
  the redundant gateways
• Local Load Measurement
• Based on monitored channel utilization
• Passive monitoring without exchanging extra
  messages
• Exploit the advantage of shared medium
• Each node can monitor the transmission of
  neighbors

43
Local Load Level Measurement

• Each node monitors the channel
• Channel Status IDLE, TRANSMITTING, RECEIVING and
  COLLISION
• Record time spending at each status on
  idle_time, trans_time, recv_time and
  coll_time respectively.
• Measuring the channel utilization periodically
  (INTERVAL)

Channel utilization
Transmission Rate
Collision Probability
44
Local Load Level Measurement (2)

• Based on the channel utilization, the local load
  is calculated
• Simple Algorithm (SIMPLE)
• Fair Share Algorithm(FS)
• Each node aggressively reduces its transmission
  rate

Maximum Channel Utilization(predefined)

• Li ? IDLE if Ui lt CHANNEL_IDLE_THRESHOLD
• Li ?SATURATED if (Ui gt ?max ?) or (Qi gt
  QUEUE_THRESHOLD)
• Otherwise, Li ? MODERATE

Local Load Level
Average Queue Size

• Li ? IDLE if Ui lt CHANNEL_IDLE_THRESHOLD
• Li ?SATURATED if (Ui gt ?max ?) or (Qi gt
  QUEUE_THRESHOLD)
• or (ANi gt 2 and TXi gt TRANSopt)
• Otherwise, Li ? MODERATE

Average Number of Active Neighbors
45
Reactive Routing Protocols with SIN

• Modify RREQ (Route Request) handling
• When a new RREQ comes in with valid TTL
  (time-to-live) (i.e., TTL gt 0)
• Based on local load level (Li)
• IDLE each node follows routing protocol
• Relays RREQ packet if necessary
• MODERATE only CH and gateways forward RREQ
  packets if necessary
• SATURATED every node stops forwarding
• Reduce the burden to the congested area
• Improve the quality of finally established path
  (detour non-saturated path instead of the
  shortest jamming one)
• Implicit call rejection if no path found

46
Simulation Study - Environment

• Environment
• Global Mobile Simulation (GloMoSim)
• Protocol Stack
• UDP, AODV/DSR, IEEE 802.11 DCF, two-ray radio
  propagation model, random-way point mobility
• Parameters
• Bandwidth 2 Mbits/sec
• Mobility 20 m/s without pause or 2-20 m/s with
  10 seconds pause time
• Transmission range 250 m
• Measurement Interval 1 second
• CHANNEL_IDLE_THRESHOLD 0.2
• QUEUE_THRESHOLD 0.8 MAX_QUEUE
• ?max 0.8 (the maximum channel utilization)
• Based on IEEE 802.11 DCF performance study

47
Throughput (AODV/DSR) with Increasing Offered Load
AODV FS AODV SIMPLE AODV PC AODV
100 nodes in 2000 x 600 m2 CBR with 4 512
bytes/sec Increase the number of CBR sessions
from 10 t0 100 20 m/s without pause time
DSR FS DSR SIMPLE DSR PC DSR
48
Normalized CTRL OH (AODV/DSR) with Increasing
Offered Load
AODV AODV PC AODV SIMPLE AODV FS
100 nodes in 2000 x 600 m2 CBR with 4 512
bytes/sec Increase the number of CBR sessions
from 10 t0 100 20 m/s without pause time
DSR DSR PC DSR SIMPLE DSR FS
49
Throughput (DSR) with Increasing Density
DSR FS DSR SIMPLE DSR PC DSR
Various number of nodes from 50 to 200 in 2000 x
600 m2 30 pairs CBR with 4 512 bytes/sec 2 20
m/s with 10 pause time mobility
50
Conclusion

• FS improves the scalability over SIMPLE algorithm
• Each node should control its transmission rate to
  improve overall performance
• Indicate that the uncontrolled routing overhead
  interferes with the scalability for the reactive
  routing protocols
• Provide the correct measurement of channel load
  based on local-network information

51
Reading list for this lecture

• ( (3) The Selective Intermediate Nodes Scheme for
  Ad Hoc On-Demand Routing Protocols - Yunjung
  Yi,Taek Jin Kwon and Mario Gerla, Proceedings of
  ICC 2002, New York City, New York, April 2002.