Mostafa Ammar

1
Message Ferrying and Other Short
StoriesMobility-Assisted Data Delivery in
Wireless Networks

• Mostafa Ammar
• College of Computing
• Georgia Institute of Technology
• Atlanta, GA

Message Ferry Project Group Members Ellen
Zegura, Wenrui Zhao, Hyewon Jun, Jeonghwa Yang,
Yang Chen, Shashi Merugu Funding NSF and DARPA
2
Traditional Wired Networks
endsystem (source)
endsystem (destination)
router

• separation between endsystems and routers
• routers responsible for finding stable path

3
Traditional Mobile Ad-hoc Wireless Networks
(MANET)
node (source)
node (destination)
node endsystem router

• no separation between endsystems and routers
• nodes responsible for finding stable path

4
Traditional Mobile Ad-hoc Wireless Networks
(MANET)
node (source)
node (destination)

• nodes may move
• routing layer responsible for reconstructing
  (repairing) stable paths when movement occurs

5
The Traditional MANET Wireless Paradigm

• The Network is Connected
• There exists a (possibly multi-hop) path from any
  source to any destination
• The path exists for a long-enough period of time
  to allow meaningful communication
• If the path is disrupted it can be repaired in
  short order
• Looks like the Internet above the network layer

6
A Brief History of Wireless Nets

• Wireless networks are as old as the Internet
  itself
• DARPA PRnet
• Initial motivation for some protocol functions
  (e.g., IP-layer fragmentation)
• PRnet -gt SURANet -gt Mobile Ad-hoc Net (MANET)
• Latest MANET wave coincided with 802.11
  activities
• Most wireless today is base-station oriented (not
  mobile, nor ad-hoc)
• My conclusion attempt to emulate wired net model
  for MANET has led to failure to achieve wide
  deployment

7
The Rise of Sparse Disconnected Networks
8
Sparse Wireless Networks

• Disconnected
• By Necessity
• By Design (e.g. for power considerations)
• Mobile
• With enough mobility to allow for some
  connectivity over time
• Data paths may not exist at any one point in time
  but do exist over time

9
Mobility-Assisted Data DeliveryA New
Communication Paradigm

• Mobility used for connectivity
• New Forwarding Paradigm
• Store
• Carry for a while
• 
  forward
• Special nodes Transport entities that are not
  sources or destinations

10
Data Applications

• Nicely suitable for Message-Switching
• Delay tolerance but can work at multiple time
  scale
• (a.k.a. Delay Tolerant Networks )

11
Epidemic Routing

• Vahdat and Becker
• Utilize physical motion of devices to transport
  data
• Store-carry-forward paradigm
• Nodes buffer and carry data when disconnected
• Nodes exchange data when met
• data is replicated throughout the network
• Robust to disconnections
• Scalability and resource usage problems

12
Epidemic Routing
13
Epidemic Routing
14
Epidemic Routing
15
Epidemic Routing
message is delivered
16
The Trouble with ER

• Potentially high-failure rate
• Message duplication consumes nodal resources
• Some mobility patterns can cause disconnection
• Can be improved with contact probability
  information - Levine et al

17
Other Original Systems

• ZebraNet and SWIM
• Data MULE and Smart-Tags
• Vehicle-to-Vehicle Communication
• Message Ferrying
• DakNet

18
SWIM
19
Vehicles on Highways Networks
Source
Destination
20
Vehicles on Highways Networks
Source
Destination
21
Vehicles on Highways Networks
Source
Destination
22
DakNet (Pentland, Fletcher, and Hasson)
23
Message Ferrying (MF) _at_ GT

• Zhao and Ammar
• Exploit non-randomness in device movement to
  deliver data
• A set of nodes called ferries responsible for
  carrying data for all nodes in the network
• Store-carry-forward paradigm to accommodate
  disconnections
• Ferries act as a moving communication
  infrastructure for the network

24
Message Ferrying System (cont.)
MF
25
Putting It All Together

• Common Features
• Disconnection
• Store, carry and forward
• Other dimensions where they may differ
• Special Nodes?
• Source/Destination Mobile?
• Potential for controlling mobility for data
  transport purposes?
• Data Communication Pattern

26
More on Message Ferrying
27
MF Variations

• Ferry Mobility
• Task-oriented, e.g., bus movement
• Messaging-oriented, e.g., robot movement
• Regular Node Mobility
• Stationary
• Mobile task-oriented or messaging-oriented
• Number of ferries and level of coordination
• Level of regular node coordination
• Ferry designation
• Switching roles as ferry or regular node

28
Our Work So Far

• Ferry Route Design Problem
• Single Ferry FTDCS 2003
• Multiple Ferries INFOCOM 2005
• MF with Mobile Nodes MobiHoc 2004
• MF as a power-savings device PerCom 2005
• Other Work
• Ferry Election/Replacement WCNC 2005
• Non-Proactive MF Routing for mobile nodes
  Mobihoc 2006
• The V3 Architecture V2V Video Streaming PerCom
  2005
• Multipoint Communication in DTN/MF SIGCOMM DTN
  Workshop 2005
• Throw-Box (Relay) deployment MASS 2006
• Power Management Schemes in DTNs SECON 2005,
  CHANTS 2006

29
Outline

• Ferry Route Design Problem
• Single Ferry FTDCS 2003
• Multiple Ferries INFOCOM 2005
• MF with Mobile Nodes MobiHoc 2004
• MF as a power-savings device PerCom 2005

30
Ferry Route Design
31
Route Design - Stationary Nodes

• Ferry route problem
• Given node locations and expected traffic between
  nodes, find ferry route such that ferry visits
  all nodes, meets traffic requirements and
  minimizes average message delay
• Solution
• A generalization of Traveling Salesman Problem

32
Numerical Results

• Experiment settings
• n nodes in 4km x 4km area
• A single ferry moving at speed 20m/s
• Node distributions
• Random uniform node distribution (UN)
• Random clustered node distribution (CN)
• Traffic models
• Uniform traffic
• Non-uniform traffic

33
Impact of Network Size

• MF provides reasonable performance
• For 40 nodes, each node can send at 10Kbps with
  1070s delay.

34
Outline

• Ferry Route Design Problem
• Single Ferry FTDCS 2003
• Multiple Ferries INFOCOM 2005
• MF with Mobile Nodes MobiHoc 2004
• MF as a power-savings device PerCom 2005

35
Multiple Ferry Route Design

• Why multiple ferries?
• Data transport capacity
• Fault tolerance
• Multiple ferries introduce new problems
• Which ferry serves which node?
• Interaction between ferries
• Tradeoff between number of ferries and
  performance improvement

36
Multiple Ferry Route Design Problem

• Networks with n stationary nodes and m ferries
• Ferries move at a constant speed and follow
  periodic routes
• Bandwidth requirements are known
• e.g., node A sends to node B at 10kbps
• Problem find optimal ferry routes such that
  bandwidth requirements are met and average delay
  is minimized
• NP-hard problem

37
Algorithms

• Single Route Algorithm (SIRA)
• All ferries follow the same route
• No interaction between ferries
• Multiple Route Algorithm (MURA)
• Ferries can follow different routes
• No interaction between ferries
• Node Relaying Algorithm (NRA)
• Nodes relay data between ferries
• Ferry Relaying Algorithm (FRA)
• Data exchange between ferries

38
Algorithms

• Single Route Algorithm (SIRA)
• All ferries follow the same route
• No interaction between ferries
• Multiple Route Algorithm (MURA)
• Ferries can follow different routes
• No interaction between ferries
• Node Relaying Algorithm (NRA)
• Nodes relay data between ferries
• Ferry Relaying Algorithm (FRA)
• Data exchange between ferries

39
Algorithms

• Single Route Algorithm (SIRA)
• All ferries follow the same route
• No interaction between ferries
• Multiple Route Algorithm (MURA)
• Ferries can follow different routes
• No interaction between ferries
• Node Relaying Algorithm (NRA)
• Nodes relay data between ferries
• Ferry Relaying Algorithm (FRA)
• Data exchange between ferries

40
Algorithms

• Single Route Algorithm (SIRA)
• All ferries follow the same route
• No interaction between ferries
• Multiple Route Algorithm (MURA)
• Ferries can follow different routes
• No interaction between ferries
• Node Relaying Algorithm (NRA)
• Nodes relay data between ferries
• Ferry Relaying Algorithm (FRA)
• Data exchange between ferries

41
Simulation Results

• Settings
• 5km x 5km area, 40 nodes, ferry speed 10m/s
• Radio range 100m, data rate 10mbps
• Nodes send to random destinations
• Comparison of algorithms
• All algorithms achieve similar delay when number
  of ferries is small or traffic load is high
• MURA performs best

42
Impact of Number of Ferries

• The use of more ferries reduces message delay and
  improves total transport capacity
• Scalability can be achieved by adding more ferries

43
Outline

• Ferry Route Design Problem
• Single Ferry FTDCS 2003
• Multiple Ferries INFOCOM 2005
• MF with Mobile Nodes MobiHoc 2004
• MF as a power-savings device PerCom 2005

44
MF for Networks with Mobile Nodes

• Nodes are mobile and limited in resources, e.g.,
  buffer, energy
• Single ferry is used
• Not limited in buffer or energy
• Data communication in messages
• Application layer data unit
• Message timeout

45
Four Approaches

• Non-Proactive ( Messaging-Specific) mobility
• Ferrying without Epidemic Routing
• Ferrying with Epidemic Routing
• Proactive Routing Schemes
• Node-Initiated MF
• Nodes move to meet ferry
• Ferry-Initiated MF
• Ferry moves to meet nodes

46
Four Approaches

• Non-Proactive ( Messaging-Specific) mobility
• Ferrying without Epidemic Routing
• Ferrying with Epidemic Routing
• Proactive Routing Schemes
• Node-Initiated MF
• Nodes move to meet ferry
• Ferry-Initiated MF
• Ferry moves to meet nodes

47
Four Approaches

• Non-Proactive ( Messaging-Specific) mobility
• Ferrying without Epidemic Routing
• Ferrying with Epidemic Routing
• Proactive Routing Schemes
• Node-Initiated MF
• Nodes move to meet ferry
• Ferry-Initiated MF
• Ferry moves to meet nodes

48
Ferrying w/ Epidemic Routing
49
Ferrying w/ Epidemic Routing
50
Ferrying w/ Epidemic Routing
51
Ferrying w/ Epidemic Routing
52
Without Proactive Mobility

• Ferrying can
• Replace epidemic routing reduced resource usage
• augment epidemic routing providing improved
  performance
• Potentially low successful delivery rates

53
Four Approaches

• Non-Proactive ( Messaging-Specific) mobility
• Ferrying without Epidemic Routing
• Ferrying with Epidemic Routing
• Proactive Routing Schemes
• Node-Initiated MF
• Nodes move to meet ferry
• Ferry-Initiated MF
• Ferry moves to meet nodes

54
Node-Initiated Message Ferrying
Meet the ferry?
OK
If no, keep working
Working
55
Node-Initiated Message Ferrying
Go to Ferry
56
Node-Initiated Message Ferrying
Send/Recv
Go to Work
57
Node-Initiated Message Ferrying
Go to Work
58
Node Trajectory Control

• Whether node should move to meet the ferry
• Goal minimize message drops and reduce proactive
  movement
• Go to ferry if
• Work-time percentage gt threshold
• and
• Estimated message drop percentage gt threshold

59
Simulations

• Ns simulations using 802.11 MAC and default
  energy model
• 40 nodes in 5km x 5km area
• 25 random (source, destination) pairs
• Node mobility
• random-waypoint with max speed 5m/s
• Message timeout 8000 sec
• Single ferry with speed 15m/s
• Rectangle ferry route

60
Performance Metrics

• Message delivery rate
• Message Delay
• Number of delivered messages per unit energy
• Only count transmission energy in regular nodes

61
Message Delivery Rate
FIMF
NIMF
F w/ER
ER
62
Message Delay
FIMF
NIMF
F w/ER
ER
63
Impact of Node Mobility Pattern
64
Impact of Node Mobility Pattern
65
Impact of Node Mobility Pattern
66
Outline

• Ferry Route Design Problem
• Single Ferry FTDCS 2003
• Multiple Ferries INFOCOM 2005
• MF with Mobile Nodes MobiHoc 2004
• MF as a power-savings device PerCom 2005

67
Trading Energy for Latency Using MF

• Consider connected MANETS -- exemplifies use
  of MF techniques in conneted networks.
• Motivation
• Sleep/Wake Cycling is primary technique for
  energy savings in MANETs connected nodes
• Synchronization is difficult and leads to
  inefficiencies in data delivery
• Idea Use the predictability of MF to improve
  power savings techniques

68
Framework of Power Management Mechanisms
Searching
Communicating
Sleeping
69
Power Management Modes
Ferry location in terms of the radio range of a
node
In
Ferry
Out
Out
Node
Wake-up interval
Beacon period
Active window
Sleeping
Communicating
Searching
70
Movement Scenarios for the Sleeping Time
Estimation
Node
Ferry
71
Performance Evaluation

• Ns-2 simulation with 802.11 MAC protocol
• 50 nodes in 2000 m x 500 m and one ferry
• Node buffer to store 700 messages
• Benchmarks Dynamic source routing (DSR) with a
  synchronous wake-up mechanism

• DSRCAM continuous aware mode
• DSR-x wake-up interval of x seconds

72
Impact of Traffic Load on Stationary Nodes

• DSR with large wake-up intervals and MF
• Low energy consumption and high delivery delay

73
Impact of Node Mobility

• The delivery rate in DSR decreases as the speed
  of nodes increases because of more topology
  changes.

74
Where Does MF Fit?

• Consider the space of wireless mobile networks
• Two Important Dimensions
• Relative Mobility
• Density

75
Some Terminology

• A Space Path A multi-hop path where all the
  links are active at the same time
• A Space/Time Path A multi-hop path that exists
  over time
• NOTE S path is a special case of S/T path
• See
• http//www.cc.gatech.edu/fac/Mostafa.Ammar/papers/
  STroute.ps

76
Example
A Space Paths Network
77
Example
A No Path Network
78
Example
A Space Time Path
79
Example
A Hybrid Network
80
The Mobile Wireless Space
High
Relative Mobility
Hybrid Environments
Low
Low
Node Density
High
81
Mapping Solutions to Space
High
Mobility
Low
Low
Node Density
High
82
Concluding Remarks

• Mobility-Assisted Data Delivery
• FINALLY! A realistic mobile wireless network
  paradigm
• Everything looks familiar but this is a truly
  different environment
• Techniques developed have wide applicability
• Fertile Ground for both networking problems and
  novel application paradigms

83
Questions?
84
V2V Video Streaming

• Motivation
• Driver assistance and safety applications
• Business or entertainment applications
• Basic Approaches
• Infrastructure-based approach
• Vehicle to vehicle (V2V) approach
• V2V network
• Special mobile ad hoc networks
• Mobile nodes are installed on a vehicle
• Ample power from the engine
• Computing and storage capability

85
Video streaming in V2V networks

• It is possible!
• Wireless technology
• Up to 54Mbps bandwidth
• 1Mbps under critical conditions
• Video compression technique
• 320240 screen, 64k color, 15fps
• Around 100200kbps data rate
• Power and computing resources

86
Video streaming in V2V networks

• But it is difficult!
• Highly dynamic networks
• Vehicle speed 40mph60mph
• Highly partitioned networks.
• Limited radio range
• Low penetration ratio
• Mobile and transient sources
• Each vehicle provides a portion of video
• Out of order delivery

87
A sample scenario
Deadline
88
V3 Overview

• Service Procedure
• Trigger the video source
• Flooding-based forwarding
• Continuous trigger
• Forward video data
• Store-carry-and-forward
• Sample operation

89
(No Transcript)
90
Performance evaluation

• Road network model
• I-75 in north-west Atlanta
• Two-way traffic
• Traffic model
• Traffic software integrated system (TSIS)
• Four kinds of transportation traffic
• Sparse, medium, dense, and congested
• Performance metrics
• Trigger delay
• Service delay

91
Highway Network
I-285
258
Receiver
256
255
254
I-75
252
250
I-85
Destination region
92
Video trigger latency
93
Service delay
94
Impact of Network Load