Sensor Deployment and Target Localization Based on Virtual Forces

Document Sample
Sensor Deployment and Target Localization Based on Virtual Forces Powered By Docstoc
					Sensor Deployment and Target
Localization Based
on Virtual Forces

             Y. Zou and K. Chakrabarty

    IEEE Infocom 2003 Conference, pp. 1293-1303,.
  ACM Transactions on Embedded Computing Systems,
          vol. 3, pp. 61-91, February 2004.




                                                    1
Outline
   Introduction
   Virtual Force Algorithm (VFA)
   Target localization
   Simulation
   Conclusion



                                    2
Introduction
   This paper focus on : sensor placement
    strategies that maximize the coverage
       Virtual Force Algorithm (VFA)
   2 sensor detection models
       Binary sensor detection model
            Low detection accuracy
       Probabilistic sensor detection model
            High detection accuracy

                                               3
Virtual Force Algorithm (VFA)
   System assumptions
       All sensor nodes are able to communicate with the
        cluster head.
       The cluster head is responsible for executing the
        VFA algorithm and managing the one-time
        movement of sensors to the desired locations.
       Sensors only send a yes/no notification message
        to the cluster head when a target is detected.
            The cluster head intelligently queries a subset of sensors
             to gather more detailed target information.

                                                                      4
Binary sensor detection model
• n * m grid sensor field
• k sensors randomly deployed
• the detection range of sensor : r
• sensor Si is deployed at point (xi, yi)
• For any point P at (x, y), the Euclidean distance
between Si and P


the coverage Cxy(si)
of a grid point P by
sensor Si
                                                      5
      Probabilistic sensor detection model




• Range detection error : re
(re < r)
• a = d(si, P) − (r−re)



                                             6
Virtual forces
   Obstacles exert Repulsive (negative)
    forces on a sensor
   Preferential coverage exert Attractive
    (positive) forces on a sensor




                                             7
Virtual forces(Cont1.)
   The total force action
    on sensor Si be denoted
    by Fi.
   Fi is a vector whose
    orientation (angle) is
    determined by the
    vector sum of all the
    forces acting on Si.
   Let the force exerted on
    Si by another sensor Sj
    be denoted by Fij .

                               8
Virtual forces(Cont2.)
   dij is the Euclidean distance between sensor
    si and sj
   dth is the threshold on the distance between
    si and sj
       The threshold distance dth controls how close
        sensors get to each other
   αij is the orientation (angle) of a line segment
    from si to sj
   wA (wR) is a measure of the attractive
    (repulsive) force

                                                        9
10
Minimize “wasted overlap”  dij ~ 2r




                                       11
Cth be the desired coverage threshold for all grid points




                                                            12
13
Data structure of the VFA algorithm




                                      14
VFA
algorithm




            15
Target localization
   Detection probability table
   Score-based ranking
   Selection of sensors to query




                                    16
Detection probability table




                              17
Si detect a target at gird point P(x,y)
Si doesn’t detect a target at gird point P(x,y)



                                                  18
19
Score-based ranking




                      Srep(t) : the set of sensors that
                      have reported the detection of
                      an object

                      Srep,xy(t) : the set of sensors
                      that can detect point P(x, y) and
                      have also reported the detection
                      of an object.



                                                     20
Selection of sensors to query
   Select sensors based
    on a score-based
    ranking.
   The sensors selected
    correspond to the
    ones that have the
    shortest distance to
    those grid points
    with the highest
    scores.

                                21
Generate the probability table




                                 22
the
pseudocode of
the localization
algorithm




              23
Simulation
   Randomly deploy 20 sensors are placed
    in the sensor field.
   detection radius as 5 units (r = 5),
       Range detection error as 3 units (re = 3)
        for the probabilistic detection model.
   The sensor field is 50 by 50 in
    dimension.
   Simulation tool : MATLAB

                                                    24
Binary Sensor Detection Model




                                25
Probabilistic Sensor Detection Model




                                       26
Sensor Field with a Preferential Area
and an Obstacle




                                        27
Probability-based Target Localization




                                        28
Conclusion
   Propose a virtual force algorithm (VFA)
    as a sensor deployment strategy to
    enhance the coverage.




                                              29
Related references
   Y. Zou and K. Chakrabarty, "A distributed coverage- and
    connectivity-centric technique for selecting active nodes in
    wireless sensor networks", IEEE Transactions on Computers, vol.
    54, pp. 978-991, August 2005.
   Guiling Wang; Guohong Cao; La Porta, T.; “Movement-assisted
    sensor deployment”, IEEE INFOCOM 2004. Volume 4, 7-11
    March 2004 Page(s):2469 - 2479 vol.4
   Wu, J.; Yang, S, “SMART: a scan-based movement-assisted
    sensor deployment method in wireless sensor networks”.;
    IEEE INFOCOM 2005. Proceedings IEEE
    Volume 4, 13-17 March 2005 Page(s):2313 - 2324 vol. 4




                                                                 30

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:5
posted:12/1/2011
language:English
pages:30