Top Banner
- 1 - - 1 - Intentional Mobility in Wireless Sensor Networks Deployment, Dispatch, and Applications Dr. You-Chiun Wang ( 王王王 ) Department of Computer Science, National Chiao-Tung University 2010/10/22
20

Intentional Mobility in Wireless Sensor Networks Deployment, Dispatch, and Applications

Jan 02, 2016

Download

Documents

blaine-maxwell

Intentional Mobility in Wireless Sensor Networks Deployment, Dispatch, and Applications. Dr. You-Chiun Wang ( 王友群 ) Department of Computer Science, National Chiao-Tung University 2010/10/22. - 1 -. Wireless Sensor Networks. SENSROS ARE STATIC!. “ Mobile” Sensor Networks. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 1 -- 1 -

Intentional Mobility in Wireless Sensor Networks

Deployment, Dispatch, and Applications

Dr. You-Chiun Wang (王友群 )

Department of Computer Science,National Chiao-Tung University

2010/10/22

Page 2: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 2 -

Wireless Sensor Networks

WSN

SENSROS ARE STATIC!

Page 3: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 3 -

“Mobile” Sensor Networks

Some sensor nodes can move around (e.g., robots).

Purpose: automatic deployment, network repairing, and sensor dispatch

Page 4: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 4 -

Topics

Automatic Deployment Mobile Sensor Dispatch Systems & Applications

iMouse System VSN (Vehicular Sensor Network) System

Page 5: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 5 -- 5 -

Automatic Deployment

WSN

SENSROS ARE MOBILE!

Page 6: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 6 -- 6 -

In a “Perfect” World

WSN

Page 7: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 7 -- 7 -

In a “Real” World

WSN

WSN

network partition

partial coverage

Page 8: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 8 -- 8 -

Can Sensors Reorganize a WSN “by Themselves”?

network partition

partial coverage

WSN

Page 9: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 9 -

Question Given a sensing field AA possibly with obstacles,

how can we make mobile sensors automatically deploy a network in an efficient way? Use the smallest number of sensors. Sensors can consume the minimum energy to reorganize

the network.

- 9 -

Page 10: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 10 -

Overview of Solutions

We first calculate the locations to place sensors and then dispatch mobile sensors to these locations. Placement solution should use fewer sensors. Dispatch solution should move sensors so that they can

remain the maximum energy after movement.

placement

Energy

dispatch

Page 11: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 11 -- 11 -

Placement Algorithm Partition a sensing field AA into sub-regions and then

place sensors in each region: Single-row regions

A belt-like area between obstacles whose width is NOT larger than , where rmin= min(rs, rc).

We can deploy a sequence of sensors to satisfy both coverage and connectivity.

Multi-row regionsWe need multiple rows of sensors to cover such areas.Note: obstacles may exist in such regions.

min3r

Page 12: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 12 -- 12 -

Step 1: Partition the Sensing Field

From AA, we first identify all single-row regions. Expand the obstacles’ perimeters outwardly and AA’s’s

boundaries inwardly by a distance of . If the expansion overlaps with obstacles, we take a

projection to obtain single-row regions.

The remaining regions are multi-row regions.

min3r3rmin

Obstacle

Obstacleexpansion

obstacle

obstacle

min3r

min3r

min3r

min3r

min3r

min3r

cut-off area O

obstacle

123

4 56

obstacle

obstacle

h

a

c

g

d

e

f

b

Page 13: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 13 -- 13 -

Step 2: Place Sensors in a Single-Row Region

Place sensors along the bisector of region.

obstacle

obstacle

obstacle

obstacle

single-row regions bisectors sensor placements (rc = rs case)

minwidth 3r

triangulationmidpoint

bisector

minwidth 3r

Page 14: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 14 -- 14 -

Step 3: Place Sensors in a Multi-Row Region

Place sensors row by row. A row of sensors guarantee coverage and connectivity. Adjacent rows guarantee continuous coverage.

rc

rc2

rs23

rs23

rs

rs3

rc

rs

rs + rcrs -22

4

Case 1: 3c sr r Case 2 : 3c sr r

Page 15: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 15 -

Step 4: Handle the Boundary Case

Three unsolved problems Some areas near the boundaries

are NOT covered. Connectivity between adjacent

rows needs to be maintained. Connectivity to neighboring

regions should be maintained.uncovered areas22

4cr

sr connectivity

obstacle

obstacle

Solutions Sequentially place sensors along

the boundaries. Not all boundaries should be

placed with sensors.

Page 16: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 16 -- 16 -

Dispatch Algorithm (1/5)

Find a maximum-weight maximum matching in a weighted complete bipartite graph. Sensors vs. locations

We should take care of the obstacles inside the sensing field.

AA

II

Page 17: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 17 -

Dispatch Algorithm (2/5)

II

BD

C

E

Run sensor placement algorithm on II to get the target locations.

L={(x1, y1), (x2, y2), (x3, y3), (x4, y4)}1 2

3 4 Compute energy cost

1 1 2 2

3 3 4 4

1 1 2 2

3 3 4 4

1 1 2 2

3 3 4 4

1

( , ( , )) 9 ( , ( , )) 12

( ,( , )) 8 ( , ( , )) 11

( ,( , )) 11 ( , ( , )) 11

( ,( , )) 9 ( , ( , )) 9

( ,( , )) 10 ( , ( , )) 6

( ,( , )) 11 ( , ( , )) 8

( ,( ,

c A x y c A x y

c A x y c A x y

c B x y c B x y

c B x y c B x y

c C x y c C x y

c C x y c C x y

c D x

1 2 2

3 3 4 4

1 1 2 2

3 3 4 4

)) 14 ( , ( , )) 13

( ,( , )) 12 ( , ( , )) 10

( ,( , )) 33 ( , ( , )) 35

( ,( , )) 30 ( , ( , )) 31

y c D x y

c D x y c D x y

c E x y c E x y

c E x y c E x y

A

Page 18: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 18 -- 18 -

Dispatch Algorithm (3/5)

Construct the weighted complete bipartite graph.

A

B

C

D

E

1

2

3

4

Sensors Locations

• Weights of edges: w(si,lj) = 40 – c(si,lj) - objective function: remaining energy - all sensors have initial energy of 40

A B C D E

1 31 29 30 26 7

2 28 29 36 27 5

3 32 31 29 38 10

4 29 31 32 30 9

A 1: needs 9 energyweight (A,1) = 40 – 9 = 31

Page 19: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 19 -- 19 -

Dispatch Algorithm (4/5)

Find the maximum-weight maximum matching.Hungarian Method: finds the optimal solution in O(n3).

A

B

C

D

E

1

2

3

4

Sensors Locations

A B C D E

1 31 29 30 26 7

2 28 29 36 27 5

3 32 31 29 38 10

4 29 31 32 30 9

Page 20: Intentional Mobility in  Wireless Sensor Networks Deployment, Dispatch, and Applications

- 20 -- 20 -

Dispatch Algorithm (5/5) Sensors are dispatched to the matched

locations.

A

B

C

D

E

1

2

3

4

II

A

BD

C

E

1 2

3 4

Does not moveSensors Locations