Tracking issues in the Wireless sensor Network

1

• Tracking issues in the Wireless sensor Network
• 
  
• 
  
• 
  
  
• 
  
  
• 
  
  

2
Outline

• Introduction.
• Localization.

3
Tracking

• Tracking
• Specify targets past trajectory.
• Predict future position of target.
• Collaborative tracking
• Individual or a small group of sensors to track
  target in neighboring region.
• Hand-off to the next group of sensors where the
  target is heading to next.
• Multiple target tracking.

4
Tracking Approaches

• Outdoors
• 1.GPS.
• Indoors
• 1.Active Badge (cellular proximity,
  infrared badges, central server).
• 2.Active BAT ( ultrasound-based,
  more accurate location Based more
  accurate location identification).
• 3.Cricket (ultrasound emitters and
  object receivers, objects self- localize).
• 4. RADAR (IEEE802.11 based, uses
  signal strength and S/N ratio to deduce 2D
  position of wireless devices indoors).
• 5. Laser Range Finder.
• 6. Electro Motive Force Method.
• 7. Smart Floor method.
• 8. Motion Star

5
Location Technique

• Triangulation
• Lateration
• 1. Direct
• 2. Time-of-Flight(e.g.GPS,Activ
  e Bat location system,Cricket,Bluesoft etc.)
• 3. Attenuation. (e.g. Spot on
  Ad hoc location system).
• Angulations (e.g. ominidirectional Ranging)
• Scene Analysis (e.g. RADAR location system)
• Proximity
• Monitoring wireless cellular access points. (e.g.
  Active Badge system)
• Observing automatic ID systems. (e.g. E-Toll
  system, EPC etc)
• Detecting physical contact. (e.g. Capacitive
  field detection, Smart Floor method)

6
Triangulation
y
(x3, y3)
c
(x,y)
b
a
(x1,y1)
(x2,y2)
x
(0,0)
(x-x1)2 (y-y1)2 a2
(x-x2)2 (y-y2)2 b2
(x-x3)2 (y-y3)2 c2
7
Angulations
8
Distance Measurement Technologies Comparisons

• Ultrasonic time-of-flight
• Common frequencies 25 40KHz, range few meters
  (or tens of meters), avg. case accuracy 2-5 cm,
  lobe-shaped beam angle in most of the cases
• Wide-band ultrasonic transducers also available,
  mostly in prototype phases
• Acoustic ToF
• Range tens of meters, accuracy 10cm
• RF Time-of-flight
• Ubinet UWB claims 6 inches
• Acoustic angle of arrival
• Average accuracy 5 degrees (e.g. acoustic
  beam former, MIT Cricket)
• Received Signal Strength Indicator
• Motes Accuracy 2-3 m, Range 10m
• 802.11 Accuracy 30m
• Laser Time-of-Flight Range Measurement
• Range 200, accuracy 2cm very directional
• RFID and Infrared Sensors many different
  technologies
• Mostly used as a proximity metric

9
Possible Implementations/ Computation Models
2. Locally Centralized Some of unknown nodes
compute

• Centralized
• Only one node computes

3. (Fully) Distributed Every unknown node computes

• Each approach may be appropriate for a different
  application.
• Centralized approaches require routing and
  leader election.
• Fully distributed approach does not have this
  requirement.

10
GPS(Global Positioning System)

• For outdoor use, we have the Global Positioning
  System (GPS).
• GPS basics
• GPS determines the distance by measuring the time
  it takes a signal to propagate from satellite to
  receiver
• Need to have very good synchronization of clocks
• Satellite clock is atomic
• Need to know satellite location
• Receive signal from three satellites to determine
  location
• Need a fourth satellite to estimate elevation and
  for accuracy
• Satellite GPS accuracy is getting reasonable
  (10-20 meters)
• BTW, there is intended noise
• Why? Dont want weapons

11
GPS Requirement

• GPS Constellation
• 24 satellites (Space Vehicles or SVs)
• 20,200km altitude (12 hour orbit period)
• 6 orbital planes (55 inclination)
• 4 satellites in each plane
• GPS Satellite Details
• Manufactured by Rockwell International,
• later by Lockheed MS
• 1900 lbs (in orbit)
• 2.2m body, 7m with solar panels
• 7-10 year expected lifetime

12
GPS problems

• GPS doesnt work indoors because the satellite
  signal is weak or reflected which means lowers
  accuracy.
• Indoor location systems is an active research
  area.
• Ideal location sensor in indoor environments have
  the following properties
• Provide fine-grain spatial information at a high
  update rate.
• Unobtrusive, cheap, scalable and robust

13
Cricket System Architecture

• Deploy actively transmitting beacons on walls
  and/or ceilings, and attach listeners to host
  devices (handhelds, laptops, etc.)
• A beacon is a small device attached to some
  location within the geographic space it
  advertises.
• Configure beacons with space Identifiers, and
  optionally with position coordinates

14
How Cricket work?

• Each beacon periodically broadcasts its space
  identifier and position coordinates on a radio
  frequency (RF) channel.
• Each beacon also broadcasts an ultrasonic pulse
  at
• The same time as the RF message Listeners that
  have line-of-sight connectivity to the beacon and
  are within the ultrasonic range will receive this
  pulse.
• RF travels about 10 6 times faster than
  ultrasound, the listener calculate Time
  difference of arrival between the start of the RF
  message from a beacon and the corresponding
  ultrasonic pulse.

15
RADAR

• RADAR attempts to use common off-the-shelf
  components. For example, 802.11b base stations.
  Basically, RADAR makes use of WLAN technology.
• RADAR assumes that the access points (AP)s
  provide overlapping coverage over area of
  interest.
• The user carries a mobile device which helps in
  determining location e.g. laptop, palmtop, badge.
• Practical signal strength model.
• Radio Propagation Model

16
RADAR Approach

• It is a RF based Indoor location tracking system.
• It provides the information based on various base
  station range overlapping areas and their signal
  measurement.
• It combines empirical measurements with signal
  propagation modeling to determine user location.
• RADAR uses signal strength information gathered
  at multiple receiver locations to triangulate the
  users coordinates.
• Triangulation is done using both
  empirically-determined and theoretically computed
  signal strength information.

17
Active Badge

• First indoor badge system
• Based on infrared technology
• Each locatable wears a badge
• Emits a unique ID periodically
• Server collects data from fixed sensors (base
  stations)
• System provides symbolic absolute location
  information
• Sunlight and fluorescent light interfere with
  infrared
• Infrared limits cell sizes to small- or
  medium-sized room.
• Users wear infrared badges Badge emits GUID every
  10 seconds

18
Active Bat system

• Based on ultrasound
• Locatable carry Active Bat tags
• Request/Response protocol
• Controller sends request via short-range
  radio
• Bat replies with ultrasonic pulse
• Controller resets ceiling sensors via
  wired network
• Ceiling sensor measures distance using
  time from reset to ultrasonic pulse arrival
• Estimated distance

19
Active Bat System

• Radio transceiver, controlling logic and an
  ultrasonic transducer.
• Periodically transmits a radio message containing
  a single identifier (corresponds to a Bat unit).
• Placed at known points on the ceiling of the
  rooms to be instrumented.
• Receivers are connected by a wired daisy-chain
  network.
• Receivers monitor the incoming ultrasound and
  record the time of arrival for any bat signal.

20
Acoustic Target Tracking Operation
21
Acoustic Target Tracking
22
Motion Star

• Virtual reality and motion capture
• Fixed antenna generates
• Axial DC magnetic-field pulses
• Receiving antennas measure
• Field pulse in three orthogonal
• Axes (combined with earth magnetic field)
• Pro Accurate resolution of 1mm, 1ms,
  and 0.1
• Cons implementation costs, object
  tethered to control unit, sensors
• must remain within 1-3m of transmitter, sensitive
  to metallic objects

Motion Star Wireless (Magnetic pulse transmitting
antennas receiving antennas and Controller)
23
Smart Floor method

• Embedded Pressure Sensors
• Capture Footfalls.
• Data used for position
• Tracking and pedestrian recognition
• Unobtrusive system
• Does not require people to carry any device or
  tag
• Poor scalability and high incremental cost
• Many users in one room?

24
Spot-On

• Implement ad hoc Lateration with low-cost tags
• Ad-hoc location sensing is a fusion of concepts
  from object location tracking and the theories of
  ad-hoc networking
• Spot-On tags use radio signal strength
  information (RSSI) as a distance estimator to
  perform ad-hoc Lateration.

25
E911

• FCC is requiring wireless phone providers to
  locate any phone that makes an E911 call
• Different approaches
• proximity
• angulations with phased antenna arrays
• GPS-enabled handsets
• Leads to numerous new consumer services

26
Easy Living

• Keeps track of devices etc. in a room
• Uses real time 3D cameras for vision positioning.
• Monitoring from the Internet to control lights,
  audio video, watch television.
• lots of processing power used to analyze frames
  captured, difficult to maintain accuracy, since
  vision struggles with analysis accuracy

27
Comparison
28
Modern applications

• Physical security
• Detecting intruders
• Medical
• Patients in a hospital
• Habitat monitoring
• Wildlife, plants
• Environmental
• Tracking forest fires, pollution
• Smart buildings
• Air traffic control
• Surveillance

Required in most applications Location of the
sensor
29
Target tracking problem

• Problem statement
• A varying number of targets
• Arise at random in space and time
• Move with continuous motions
• Persist for a random time and possibly disappear
• Positions of targets are sampled at random
  intervals
• Measurements are noisy and
• Detection probability lt 1.0
• False alarms
• Goal detect, alert, and track for each target

30
Tracking Challenges

• Data dissemination and storage
• Localization.
• Resource allocation and control
• Operating under uncertainty
• Real-time constraints
• Data fusion (measurement interpretation)
• Multiple target disambiguation
• Track modeling, continuity and prediction
• Target identification and classification

31
Conclusion

• Which one of these approaches is better?
• Difficult to compare error rate.
• RF is not robust, ultrasound systems are
  better but only if ceiling
  mounted.
• Lots of start-up cost with Active Bats same
  with Cricket.
• RADAR is relatively inexpensive in terms of
  hardware but extremely time-consuming to do
  calibration.
• RADAR needs network cards.

32
References

• P. Bahl, V. Padmanabhan, "RADAR An In-Building
  RF-based User Location and Tracking System" IEEE
  INFOCOM 2000, vol. 2, pp. 775-784.
• Nissanka B. Priyantha, Anit Chakraborty and Hari
  Balakrishnan, " The Cricket Location-Support
  System " Proc. 6th ACM MOBICOM, A ugust 2000, pp.
  32-43.
• Andy Hopper, Pete Steggles, Andy Ward, Paul
  Webster, " The Anatomy of a Context-Aware Applica
  tion " Proceedings of the 5th Annual ACM/IEEE
  International Conference on Mobile Computing and
  Networking (Mobicom '99), Seattle, Washington,
  USA, August 1999.
• Special Notes Special thanks goes to MIT for a
  presentation that has great pictures.
• Location Sensing Techniques Jerey Hightower and
  Gaetano Borriello UW-CSE-01-07-01 University of
  Washington, Computer Science and Engineering