Sensor Networks Sensing and Actuation in Pervasive Computing Part II

1
Sensor Networks (Sensing and Actuation in
Pervasive Computing)Part II

• By
• Abu Zafar Abbasi
• PhD Fellow
• FAST-NU, Karachi
• July 18, 2007

2
References

• A Survey on Sensor Networks
• Ian F. Akyildiz, Weilian Su, Yogesh
  Sankarasubramaniam
• IEEE Communication Magazine, p 102-114, August
  2002
• Transforming Agriculture through Pervasive
  Wireless Sensor Networks
• Tim Wark, Peter Corke et al
• IEEE Pervasive Computing, p 50-57, Apil-June 2007
• Wireless Sensor Networks for Battlefield
  Surveillance
• Tatiana Bokareva, Wen Hu, et al
• Land Warfare Conference, October 2006

3
Sensor Networks Revisited (1)

• A sensor network is usually wireless network
  consisting of spatially distributed autonomous
  devices using sensors to cooperatively monitor
  physical or environmental conditions, such as
  temperature, sound, pressure, motion or
  pollutants, at different locations

4
Sensor Networks Revisited (2)

• Features
• Untethered (Wireless )
• Autonomous
• Self-Organizing
• Local processing and storage
• Low power
• Low Cost
• Operate Unattended
• Operate in high volumetric densities

5
Sensor Networks Revisited (3)

• Actuation
• An actuator is a device that produces linear or
  rotary motion from a source of power under the
  action of a source of control
• Actuation is not involved at each smart sensor
  node
• May be deployed at specific locations for
  controlling the phenomenon
• Examples are
• Electric, hydraulic, magnetic etc

6
Sensor Networks Revisited (4)

• Advancement in wireless communication and
  electronics (nanotechnology), MEMS enables
  paractical uses of Sensor Nets
• June 2007 Hitachi introduces RF ID chips of size
  0.05X0.05 mm (like powder)
• May 2006 HP introduces spot chips smaller than a
  grain of rice with 1MB ram, wireless, processor,
  modem, capacitors array, loop antenna)
• Active Companies are
• Cross Bow,
• Dust Networks
• Ember
• Moteiv
• Arch Rock etc

7
Transforming Agriculture through Pervasive
Wireless Sensor Networks

• IEEE Pervasive Computing
• Apr-Jun 2007

8
Motivation

• Optimal, profitable, and sustainable use of land
  and water resources
• Lands have different seasonal patterns,
  hydrological and nutrient cycles, and geology
• Labor problems- less farmers, aged farmers
• Cattle farming
• Traditionally use of fences and musters (30 of
  cost)
• Sensor networks can improve productivity by
  increasing situational awareness of the state of
  pastures and animals

9
Sensor Network platform

• Robust hardware platform to tolerate harsh
  conditions and long term deployment
• Use of devices Fleck-1, Fleck-2, Fleck-3
• Radio (433 MHz, 915 MHz)
• Solar battery, charging circuit
• Sensors and interfaces
• TinyOS

10
Fleck-1

• Fleck -1
• Atmega 128 microcontroller
• NRF903 radio transceiver
• Temperature sensor
• Connected to
• Soil moisture sensors
• Camera modules

11
A solar powered node
12
Pasture Assessment

• Why?
• As weather patterns change, crops mature, and
  cattle graze pastures for food, farmers must
  decide when to irrigate pastures, apply
  fertilizers, or move cattle to another pasture
• How?
• Use of commercially available ECH2O sensors that
  measure surrounding soil volumetric content
• Automatically takes readings at 1 min interval
• Data is aggregated at base for up to date
  moisture profile for whole pasture

13
Soil Moisture
14
Cattle sensing

• Grazing habits, interactions with surroundings,
  cattles behavior
• A previous work ZebraNet
• GPS position hopped in a peer-to-peer fasion when
  other animals come in range

15
Cattle Sensing

• Communication
• Can we reliably send and receive packets between
  mobile cattle nodes (for example, multihop
  routing of behavior states)?
• Can we extract valuable information from each
  nodes position and inertial information so as to
  determine animal and herd state?

16
Cattle Sensing

• Communication Analysis
• Testing of a typical peer-to-peer in a group of
  13 cows placed in 100X600 pastures
• All collars pinged each other at 1 minute
  containing animals GPS position and time

17
Animal behavior and Position(1)
day
night
18
Animal behavior and Position(2)
Behavior of grazing cattle for four days
19
Actuation

• Control animals in landscape
• Automatically controlled gates
• Feeding stations
• Stimuli to animals to influence their motion
• Water through
• Sprinklers

20
Wireless Sensor Networks for Battle field
Surveillance

• Land Warfare Conference
• October 2006

21
Motivation

• Detect and classify multiple targets (e.g
  vehicles, troops)
• Use of inexpensive off-the-shelf wirless devices
• Acoustic and magnetic signals for tracking

22
Sensor Network Architecture

• Large number of low power sensor nodes
• Few powerful cluster-heads micro servers
• Crossbows MicaZ mote (acoustic and magnetic)

23

• System tracks the target based on spatial
  differences of signal strengths from different
  sensor nodes

24
Target signature matching
25
Questions?