Ad Hoc and Wireless Sensor Network

1
Ad Hoc and Wireless Sensor Network

• Wireless Computer Communications
• Min-Xiou Chen

2
Mobile Ad Hoc Networks (MANET)

• Formed by wireless hosts which may be mobile
• Without (necessarily) using a pre-existing
  infrastructure
• Routes between nodes may potentially contain
  multiple hops
• May need to traverse multiple links to reach a
  destination
• Mobility causes route changes

3
Variations

• Fully Symmetric Environment
• all nodes have identical capabilities and
  responsibilities
• Asymmetric Capabilities
• transmission ranges and radios may differ
• battery life at different nodes may differ
• processing capacity may be different at different
  nodes
• speed of movement
• Asymmetric Responsibilities
• only some nodes may route packets
• some nodes may act as leaders of nearby nodes
  (e.g., cluster head)

4
Variations (Cont.)

• Traffic characteristics may differ in different
  ad hoc networks
• bit rate
• timeliness constraints
• reliability requirements
• unicast / multicast / geocast
• host-based addressing / content-based addressing
  / capability-based addressing
• May co-exist (and co-operate) with an
  infrastructure-based network

5
Variations (Cont.)

• Mobility patterns may be different
• people sitting at an airport lounge
• New York taxi cabs
• kids playing
• military movements
• personal area network
• Mobility characteristics
• Speed
• Predictability
• direction of movement
• pattern of movement
• uniformity (or lack thereof) of mobility
  characteristics among different nodes

6
Why Ad Hoc Networks ?

• Ease of deployment
• Speed of deployment
• Decreased dependence on infrastructure

7
Many Applications

• Personal area networking
• cell phone, laptop, ear phone, wrist watch
• Military environments
• soldiers, tanks, planes
• Civilian environments
• taxi cab network
• meeting rooms
• sports stadiums
• boats, small aircraft
• Emergency operations
• search-and-rescue
• policing and fire fighting

8
Challenges

• Limited wireless transmission range
• Broadcast nature of the wireless medium
• Hidden terminal problem
• Packet losses due to transmission errors
• Mobility-induced route changes
• Mobility-induced packet losses
• Battery constraints
• Potentially frequent network partitions
• Ease of snooping on wireless transmissions
  (security hazard)

9
Flooding for Data Delivery
S broadcasts data packet P to all its neighbors
Each node receiving P forwards P to its neighbors
S
D
Node D does not forward the packet
Sequence numbers used to avoid the possibility of
forwarding the same packet more than once
10
Flooding for Data Delivery Advantages

• Simplicity
• May be more efficient than other protocols when
  rate of information transmission is low enough
  that the overhead of explicit route
  discovery/maintenance incurred by other protocols
  is relatively higher
• this scenario may occur, for instance, when nodes
  transmit small data packets relatively
  infrequently, and many topology changes occur
  between consecutive packet transmissions
• Potentially higher reliability of data delivery
• Because packets may be delivered to the
  destination on multiple paths

11
Flooding for Data Delivery Disadvantages

• Potentially, very high overhead
• Data packets may be delivered to too many nodes
  who do not need to receive them
• Potentially lower reliability of data delivery
• Flooding uses broadcasting -- hard to implement
  reliable broadcast delivery without significantly
  increasing overhead
• Broadcasting in IEEE 802.11 MAC is unreliable
• In our example, nodes J and K may transmit to
  node D simultaneously, resulting in loss of the
  packet
• in this case, destination would not receive the
  packet at all

12
Why is Routing in MANET different ?

• Host mobility
• link failure/repair due to mobility may have
  different characteristics than those due to other
  causes
• Rate of link failure/repair may be high when
  nodes move fast
• New performance criteria may be used
• route stability despite mobility
• energy consumption

13
Changes in network

• Number of nodes
• Topology
• Connectivity
• Mobility
• Battery power
• Heterogeneousness

14
Solution models

• Proactive protocols
• Reactive protocols
• Position-aware protocols
• Clustering protocols
• Cooperation between protocols

15
Proactive protocols

• DSDV, STAR, WRP, ...
• Propagate routing information before the routes
  are needed
• Possible to have multiple routes for one
  destination
• Lots of signaling traffic
• All of the routes may never be used

16
Table-driven, Proactive

• Uses periodic route updates to maintain routing
  tables
• Can either be link state or distance vector
• Mobility is treated as link change
• Drawbacks Inefficient if there is few demands
  for routes, and instability if there is high
  mobility

17
Reactive protocols

• AODV, DSR, TORA, ...
• New nodes are added when needed
• Introduce extra latency
• Disappearing routes take time to expire
• May actually cause more signaling if expiration
  times are shortened

18
On demand, Reactive

• No periodic route updates
• Find route when needed by the source node
• Caching will help to improve the performance
• Advantages power and bandwidth efficient
• Disadvantages sometimes longer delay

19
Position-aware protocols

• GPSR, GRA, ABR, ...
• Additional location-dependent metrics
• Geographic positioning (GPSR, GRA)
• Associativity (ABR)
• Signal stability (SSA)
• Reduce network-wide signaling

20
Hierarchical / Clustering protocols

• ZRP, CBR, FSR, LANMAR, ...
• Avoid drastic network-wide changes
• May use hybrid design
• Small clusters converge fast

21
Cooperation between different protocols

• What if a network comes into contact with another
  one?

22
Hybrid of Proactive and Reactive

• Compromised Method
• Reduce disadvantages and promote advantages
• On theory better but IMHO its a bit too
  complicated and smart, the rule of the real
  world is The simpler the better.

23
Considerations of routing protocol design

• Can be adapted to ever changing topology
• Low delay
• Low power consumption
• High throughput (bandwidth)
• High QoS

24
Flooding of Control Packets

• Many protocols perform (potentially limited)
  flooding of control packets, instead of data
  packets
• The control packets are used to discover routes
• Discovered routes are subsequently used to send
  data packet(s)
• Overhead of control packet flooding is amortized
  over data packets transmitted between consecutive
  control packet floods

25
DSDV Destination Sequence Distance Vector Routing

• IBM 1996, simulated but not implemented
• Uses modified Bellman-Ford Algorithm, distance
  vector based, table driven
• Route settling time and route may not converge
• Sequence number derived from dest. node is used
  to keep the routing table up-to-date

26
DSDV Example

• MH4 FW TABLE
• Dest. Next_Hop Metric Seq._No. Install
  Stable_Data
• MH4 ADV. RT. TABLE
• Dest. Metric Seq._No.
• MH1 move to the vicinity of MH5 and trigger a
  broadcast
• Routing Update, with increased sequence number in
  routing table

MH4
MH3
MH5
MH2
MH1
MH1
27
DSR Dynamic Source Routing

• CMU 1996, simulated and implemented in 1999
• An extension of IP, it uses options field in IP
• Two phases routing discovery and routing
  maintenance
• Caching and other features are used

28
DSR

• When node S wants to send a packet to node D, but
  does not know a route to D, node S initiates a
  route discovery
• Source node S floods Route Request (RREQ)
• Each node appends own identifier when forwarding
  RREQ

29
Flooding RREQ
4
6
9
1
S
3
D
10
5
7
2
8
30
Route Discovery in DSR

• Destination D on receiving the first RREQ, sends
  a Route Reply (RREP)
• RREP is sent on a route obtained by reversing the
  route appended to received RREQ
• RREP includes the route from S to D on which RREQ
  was received by node D

31
Send Back RREP
4
6
9
1
S
3
D
10
5
7
2
8
32
Route Reply in DSR

• RREP can be sent only if links are guaranteed to
  be bi-directional
• To ensure this, RREQ should be forwarded only if
  it received on a bi-directional link
• If unidirectional (asymmetric) links are allowed,
  then RREP may need a route discovery for S from
  node D
• If IEEE 802.11 MAC is used to send data, then
  links have to be bi-directional (since Ack is
  used)

33
DSR

• Node S on receiving RREP, caches the route
  included in the RREP
• When node S sends a data packet to D, the entire
  route is included in the packet header
• hence the name source routing
• Intermediate nodes use the source route included
  in a packet to determine to whom a packet should
  be forwarded

34
DSR Routing Maintenance

• Topology A-gtB-gtCxD E
• Suppose A wants to send a packet to E, it first
  sent it to B according to the source routing
  table, and must receive a receipt from B to
  confirm the success of sending.
• B also needs a receipt from C.
• If the link from C to D is broken, C sends back
  link error message to B, and B to A.
• A should search its cache to see if there is
  another route available to E, otherwise, it needs
  to initiate a new route discovery session.

35
DSR Advantages

• Routes maintained only between nodes who need to
  communicate
• reduces overhead of route maintenance
• Route caching can further reduce route discovery
  overhead
• A single route discovery may yield many routes to
  the destination, due to intermediate nodes
  replying from local caches

36
DSR Disadvantages

• Packet header size grows with route length due to
  source routing
• Flood of route requests may potentially reach all
  nodes in the network
• Care must be taken to avoid collisions between
  route requests propagated by neighboring nodes
• insertion of random delays before forwarding RREQ
• Increased contention if too many route replies
  come back due to nodes replying using their local
  cache
• Route Reply Storm problem
• Reply storm may be eased by preventing a node
  from sending RREP if it hears another RREP with a
  shorter route

37
Ad hoc On-Demand Distance Vector (AODV)

• ??mobile nodes?????????????????????
• ?????mobile nodes???active communication
  ????????????
• ??mobile nodes?link breakages????????????,????????
  ,????????????

38
AODV routing protocol

• 1. ?mobile node???packet?routing table
  ????????????Destination node
• ??Route Requests (RREQs) ?????destination
  node?????
• 2. ?RREQs??????????destination node,????Route
  Replies (RREPs)?????RREQs?mobile node?
• 3. ?????????????????????,???Route Errors (RERRs)
• 4. ??Hello Message??????

39
Route Requests (RREQs)

• ??routing table,????route entry,??Route Requests
• ??RREQs??ID,?mobile node??RREQs??,???????,????,???
  packet??
• ??RREQs??????
• ?????????
• ??RREQs?mobile nodes????????,???,?????? fresh
  enough???????destination node,????,???routing
  table,????????
• Fresh enough route
• ??????route entry
• sequence number???,??sequence number???????RREQ???
  sequence number

40
Route Replies (RREPs)

• ??RREQ????,??RREQ?????destination address???
• ??RREQ?????address sequence???routing table
• ????unicast?????Route Reply (RREP)?destination
  node?source node
• ??mobile node??RREP
• ??RREP?????address sequence ???routing table,
• source node?routing table????destination
  node?entry,
• data packet????

41
Route Errors (RRERs)

• mobile node????????RRER
• mobile node?????active route??????hop
• link break
• mobile node????data packet,?????active route????
• ??Local Repair????link break
• ??????????RREQs,???next hop
• ???RREQs????node?
• ??hop count?????
• ??????,???????,hop count??,?????
• ???RREQs????????MN3???,???????????????destination
  node

42
Hello Message

• ???????????
• ????node?local connectivity
• ?node?????mobile nodes?????Hello
  Messages?????mobile nodes?????next
  hop????????(????)
• ??????Hello Messages?
• ????Hello Message????mobile node??

43
ZRP Zone Routing Protocol

• Cornell 1998, simulated only
• Zone based and hybrid of proactive and reactive
• Proactive intra-zone and reactive inter-zone
• Problem How to decide the appropriate radius of
  the zone

44
Multiprotocol ad hoc networks

• Enable using different routing strategies for
  different scenarios
• Require translation of some kind
• All of the signaling may be translated
• Gateway may translate only what is needed

45
Multiprotocol ad hoc networks (cont.)

• Alternative have the nodes negotiate a common
  protocol
• More difficult to ensure loop-freeness
• May cause significant signaling overhead

46
Trade-Off

• Latency of route discovery
• Proactive protocols may have lower latency since
  routes are maintained at all times
• Reactive protocols may have higher latency
  because a route from X to Y will be found only
  when X attempts to send to Y
• Overhead of route discovery/maintenance
• Reactive protocols may have lower overhead since
  routes are determined only if needed
• Proactive protocols can (but not necessarily)
  result in higher overhead due to continuous route
  updating
• Which approach achieves a better trade-off
  depends on the traffic and mobility patterns

47
Possible improvements to existing protocols

• New routing metrics
• Hierarchical routing
• Dynamic clustering
• Help from link layer

48
Summary

• Different routing protocols are suitable for
  different scenarios
• Hybrid protocols look promising
• There may be need for multiprotocol ad hoc
  networks
• Link layer and position information may offer
  some help

49
Mobile Ad Hoc QoS Routing Protocols

• Why we need QoS?
• Services in ad hoc network
• Best Effort
• Limited bandwidth
• QoS is needed, because there is demand for
  wireless video or audio real time delivery.

50
New QoS Metrics in Ad Hoc Networks

• Power consumption
• Network lifetime, or time to network partition

51
Pros and Cons of Intserv

• Intserv
• Pros
• Guarantee complicated QoS demand
• Cons
• Keeping flow state info will cost a massive
  storage
• RSVP signaling packet will contend for bandwidth
  with data packet
• Every mobile node must perform processing of
  admission control, classification, and scheduling

52
Pros and Cons of Diffserv

• Diffserv
• Pros
• Lightweight in interior routers
• Scalable
• Stateless
• Cons
• Ambiguous boundary
• SLA (Service Level Agreement) doesn't fit to MANET

53
Challenges

• Link breakage by mobility
• Limited bandwidth and power
• Broadcast characteristic of radio transmission

54
Research on Ad Hoc QoS

• QoS models, the whole architecture
• Resource Reservation Signaling, the controller
• QoS routing find the route with enough resource
• QoS MAC - basis

55
Some theory for QoS routing

• Suppose we have a multi-hop link S-I1-I2--In-D
• Link delay Dsi1 Di1i2 Din-1in Dind
• Link bandwidth min BWsi1, BWi1i2, , BWin-1in,
  BWind
• Link cost Csi1 Ci1i2 Cin-1in Cind
• Concave and additive metrics bandwidth is
  concave, cost, delay, and jitter is additive
• Wang et al. proved that if QoS contains at least
  two additive metrics, QoS routing is NP-complete
  problem

56
Three difficulties of QoS routing

• Overhead is too high
• Maintaining the precise link state information is
  very difficult
• The reserved source may no longer available
  because of path breakage

57
Current Works

• CEDAR
• Ticket based probing
• QoS routing based on bandwidth calculation

58
Wireless Sensor network

• ?????object tracking?????
• ????????????????
• ??, ??, ??, ??, ??, ??
• ???????
• ?????????
• ?????????????

59
Sensor network?????

• ??
• ??????, ??, ????
• ????????
• ????????
• ??
• ????????/??, ???, ????, ???
• ?????????
• ??
• ????/???????????????/??/????
• ??
• ??????
• ??????
• ????

60
Sensor network???

• ??????????
• ??????
• ????
• ??????????

61
Wireless sensor network????

• ????
• ????
• ?????
• ?????????node
• IP/ID? ??????
• ???

62
???????

• ??
• ???????/???????
• ????
• ???/??/????
• ??
• ???????
• ??sleep/active?????
• Packet transfer delay ?, Packer loss ?
• ????
• ??tree?????, ??data aggregation
• ?????????(sink node)
• ????????

63
????

• Multi-hop????????????
• ????
• NTP
• tree???/?????
• GPS
• ?????, ????
• GPS????????
• ???????
• ?????????
• ????

64
?????

• Sensor node???????
• ?????????node
• GPS???
• ?????????????????????
• ????
• ????
• ?????

65
?????????node

• ??
• ???node??????????, ?????????????node? ???IP??
• ????
• ??? ????????ID
• ?? ????, ????
• ??????
• ?sink????, ????ID,?????
• Node??????????ID
• ?? ??????ID??

66
???

• ??, ????? -gt ????
• Node???
• ?????
• Error detect
• Error handle
• ??????
• ????, ?????
• ???? ??????

67
???(Cont)

• Network??????
• ??????node
• ??probe ask
• ??hello msg
• ??logical tree???
• ?????node?????node????????

68
Wireless sensor network

• What is sensor network?
• Network of small/simple sensors
• Sensors transmit sensor data to a data collection
  node.
• either hop by hop or sent directly
• General Assumptions
• Sensors
• transmit data in wireless connection. But
  transmission range is limited.
• power-and computation-capability limited
  (microsensors)
• Network
• Sensors are distributed randomly in an area.

69
Wireless sensor network

• Characteristics
• Individual sensors are trivial. The entire
  network is essential.
• Individual sensors may prone to fail, out of
  power (simple devices)
• A sensor network should present reliability, long
  life

70
Scenario
Deployment
Target field
71
Scenario
Sensors
Transmission range
Base station -Request data to sensor
network -Receive data from sensors
72
Scenario
Sensors
Data transmission
Base station -Request data to sensor
network -Receive data from sensors
73
Applications

• Battle field
• Detection of enemies
• Environmental monitoring
• Information collecting
• temperature, image, sound etc.
• Disaster (e.g. 9/11)
• Information collecting
• Object tracking

74
Challenge

• Sensor life
• Sensor is energy-limited.
• When battery is out, it dies.
• Problems caused by sensor death
• Sensor network cant acquire sensing data in that
  area.
• Routing path through the sensor cant be
  available.
• It may cause partitioning of network.
• It may encourage energy consumption of other
  sensors on alternative routing paths.
• What is critical in terms of energy consumption.
• Wireless transmission
• Energy consumption is proportional to d2 (square
  of transmission distance).
• Trade off between hop counts and transmission
  distance.
• Even if distance decreases, times of
  transmission increases.

75
Goal

• To maximize the life of network
• Network life is duration from its deployment
  until it cant cover the entire area.
• Even if some sensors die, sensor network need to
  go on working.

76
Research Issue

• Power saving routing
• To route sensor data to destination by consuming
  minimal power
• Effective distribution of energy consumption is
  important
• Aggregation Tree Structure

77
The existing approach

• 1)Direct to base station
• Each sensor transmit data directly to base
  station no matter how long distance is.
• 2)MTE (Minimum transmission energy) protocol,
  Timothy Jason Shepard, MIT
• Chose a closest next hop node to minimize the
  power for data transmission

78
Overview
1)Direct to base station
2) MTE
Base station
Base station
79
Problem
1)Direct to base station
2) MTE
Nodes which are distant from BS die more likely
because of long transmission range.
Nodes which are close to BS die more likely
because of many forwarding times
80
Result

• Result
• Nodes in specific area intensively die.
• Interruption of transmission of data
• Sensor network suffer from partitioning
• Losing coverage in the specific area
• Sensor network lose whole sensing information in
  the specific area.
• Even if many other sensors survive, the sensor
  network doesnt work well.
• Solution
• Effective distribution of energy consumption
• To uniformly distribute power consumption over
  the network sensors

81
New schemes

• LEACH
• (Low-Energy Adaptive Clustering Hierarchy)
• Wendi R.H., Anantha C., Hari B., MIT
• Sensors are organized into clusters. Head of the
  cluster will be represented by members in turn.
  Only cluster head will participate data relaying
• PEGASIS
• (Power-efficient Gathering in Sensor Information
  System)
• Stephanie Lindsey, Cauligi S. R., The Areospace
  Corporation
• A optimization of LEACH, it uses a greed
  algorithm to form cluster by assuming each node
  have a global view of the network. Each node
  communicate only to a close neighbor and take
  turn to send data to data collection node.

82
Basic concept
Base station
83
Basic concept
Some nodes become randomly cluster heads.(The red
ones) Each cluster head forms cluster. (The
dotted circle)
Base station
84
Basic concept
Each node in the cluster will forward the data to
head and the head relays the data to data
collection node.
Base station
85
Basic concept
Some objects may die because of overuse, but it
doesnt occur intensively in specific are. The
algorithm can achieve relative fair energy
consuming rate among a cluster.
Base station
86
In every cycle, different nodes become heads
which form clusters.
Base station
87
Details about algorithm

• 1. Advertisement phase
• 2. Selection phase
• 3. Schedule creation
• 4. Data transmission

88
1. Advertisement Phase

• Each node determines whether or not to become a
  cluster head for each round. (determined a
  priori)
• After it decides to be cluster head, it will
  broadcast advertisement message using CSMA MAC
  protocol.

89
2. Set up phase

• After non-cluster-head receives some
  advertisement messages, it decides which cluster
  to join based on received signal strength of the
  advertisement.
• To minimize the transmission cost
• After it decides which cluster to join, it
  transmits join message back to the cluster head
  using a CSMA MAC protocol.

90
3. Schedule creation

• After cluster head node receives all the
  messages, it creates a TDMA schedule telling each
  node when it can transmit.

91
4. Data transmission

• Once the clusters are created and TDMA schedule
  is fixed, data transmission can begin.
• Each non-cluster-head-node transmit data to
  cluster head on allocated time slot.
• Each non-cluster-head-node can be turned off
  except in allocated time slot to minimize energy
  consumption.

92
Protection of interference

• CDMA
• In the process of 3) Schedule allocation, a
  cluster head can allocate code to each
  non-cluster-head node.
• In the process of 4) Data transmission,
  non-cluster-head-node will transmit converted
  signal with assigned code.
• It prevent interference of signal to use code.

93
Problems

• Transmission cost of advertisement message which
  cluster head issues.
• LEACH use broadcast.
• Cause heavy traffic
• PEGASIS make it more efficient.

94
Efficient Location Tracking Using Sensor Networks

• H. T. Kung and D. Vlah
• Division of Engineering and Applied Sciences
  Harvard University
• Proceedings of 2003 IEEE Wireless Communications
• and Networking Conference (WCNC)

95
Introduction

• Movement Locality
• tracking moving object
• STUN method
• DAB method

96
STUNScalable Trackong Using Networked sensor

• hierarchy to connect the sensors
• using the querying point as the root
• record information
• presence of the objects

97

• Goalefficient quering and message-pruning

98

• The weight represent the frequency of object
  movement between a pair of adjacent sensors

99
Performance Metrics

• Communication Cost
• Delayheight of hierarchy tree
• A good tracking method
• low communication cost
• low delay

100
(No Transcript)
101
DABDrain-and-Balance

• Use event rate information
• a subset of sensor
• merged into balanced subtree
• Applies to multi-dimensional sensor graphs
• Goal
• constructing desirable message-pruning hierarchy
  trees
• communication cost and the query delay are low

102
DAB algorithm

• Initialize T to be an empty graph
• For each draining threshold hi ? H in the
  increasing order of i
• perform a DAB step
• Draining
• Balancing

103
(No Transcript)
104
Comparison to Huffman tree

• Achieves the minimal cost for given set of event
  rates associated with sensor
• Undesirable for the message-pruning purpose
• dont concern with tree balancing

105
(No Transcript)
106
Results for 1D Sensor Graph
107
(No Transcript)
108
Conclusion

• STUN method
• DAB method for building STUN hierarchies
• DAB method is useful in large-scale sensor
  tracking systems

109
Efficient In-Network Moving Object Tracking
inWireless Sensor Networks

• Chih-Yu Lin, Wen-Chih Peng, and Yu-Chee Tseng
• Department of Computer Science and Information
  Engineering
• National Chiao Tung University

IEEE Trans. on Mobile Computing, Vol. 5, No. 8,
Aug. 2006, pp. 1044-56. (SCI)
110
Introduction(Cont)

• Sink
• Update
• updates of an objects location are initiated
  when the object moves from one sensor to another
• Drawback objects move frequently
• Query
• when there is a need to find the location of an
  interested object
• A naive way flood

111
Introduction(Cont)

• DAB
• the first approach where query messages are not
  required to be flooded and update messages are
  not always transmitted to the sink.
• drawbacks
• a logical tree
• Not take the query cost into consideration.
• Our tree
• first stage aims at reducing the update cost
• Deviation-Avoidance Tree (DAT)
• Zone-based Deviation-Avoidance Tree (Z-DAT).
• second stage aims at further reducing the query
  cost
• Query Cost Reduction(QCR) algorithm

112

Preliminaries

• Sink
• Sensors locations are already known
• nearest-sensor model
• Neighbors
• Multiple objects
• Event rate(frequency)
• the communication range

113
Preliminaries(Cont)

• Graph G (VG, EG)
• VG representing sensors
• EG representing links between neighboring sensors
• wG(a, b)sum of event rates from a to b and b to
  a (weight)

114
Preliminaries(Cont)

• a logical weighted tree T will be constructed
  from G.

115
Preliminaries(Cont)

• When an object o moves from the sensing range of
  a to that of b
• dep(o, a, b)
• arv(o, b, a)

116
Preliminaries(Cont)

• Summary of notations.

117
Tree Construction Algorithms

• Algorithm DAT (Deviation-Avoidance Tree)
• the avage update cost

118
Tree Construction Algorithms(Algorithm DAT )

• three observations
• distT (u, lca(u, v)) minimal value
  distG(u,lca(u, v))
• Expect distT (u, sink) distG(u, sink) for each
  u ? VG
• If not u deviates from its shortest path to the
  sink
• If true for each u ? VG T is a
  deviation-avoidance tree.
• wT (u, v), u ? v minimal value is 1
• distT (u,lca(u,v)) distT (v,lca(u,v)) can be
  minimized.
• highest-weight-first principle.

119
(No Transcript)
120
Tree Construction Algorithms (Algorithm DAT)

• Initially, DAT treats each node as a singleton
  subtree.
• Then we will gradually include more links to
  connect these subtrees together.
• In the end, all subtrees will be connected into
  one tree T.

121
Tree Construction Algorithms (Algorithm DAT)

• The DAT growing step
• (u, v) will be included into T only if u and v
  are currently located in different subtrees.
• (u, v) will be included into T only if at least
  one of u and v is currently the root of its
  temporary subtree and the other is on a shortest
  path in G from the former node to the sink.

122
Tree Construction Algorithms (Algorithm DAT)
123
Algorithm Z-DAT (Zone-based Deviation-Avoidance
Tree)

• based on the following locality concept.
• Assume that u is vs parent in T. for any
• edge (x, y) ? EG such that x ? Subtree(v) and
• y ? Subtree(v), arrival/departure events between
  
• x and y will cause a message to be transmitted
  on (p(v), v), thus increasing the value of

124
Algorithm Z-DAT (Zone-based Deviation-Avoidance
Tree)
125
Algorithm Z-DAT (Zone-based Deviation-Avoidance
Tree)

• The algorithm builds T in an iterative manner
  based on two parameters, ? and ?, where ? is a
  power of 2 and ? is a positive integer.

126
Algorithm QCR (Query Cost Reduction)

• QCR is designed to reduce the total update and
  query cost by adjusting the object tracking tree
  obtained by DAT/Z-DAT.
• Eq 3.

127
Algorithm QCR (Query Cost Reduction)

• two observations on Q(T)
• because distT (p(v), sink) is always smaller than
  distT (v, sink), Eq. 3 indicates that placing a
  node as a leaf can save the query cost instead of
  placing it as a non-leaf.
• the second term in Eq. 3 implies that the value
  of distT (p(v), sink) should be made as small as
  possible.

128
Algorithm QCR (Query Cost Reduction)

• C(T) - C(T) Q(T) - Q(T) U(T) - U(T)

129
Algorithm QCR (Query Cost Reduction)
130
Simulation Results

• Update cost environment
• a sensing field of size 256 x 256
• 4096 sensors are deployed in the sensing field
• Regularly or randomly deployed
• Event rates are generated based on a model
  similar to the city mobility model
• C positive constant dtotal number of
  levels

131
Simulation Results (Update Cost)
132
(No Transcript)
133
Simulation Results(Update Cost, modify (?, ?)
Sinks are located at the center of the network.
134
Simulation Results(Query Cost)
(C 1.0)
135
Simulation Results(Total Cost)
136
Conclusion

• Made a logical tree reduce the total
  communication cost incurred by object tracking.
• Use DAT and to reduce update cost.
• Use Z-DAT to reduce zone communication cost.
• Use QCR to improve Z-DAT to reduce query cost.