FIND: Future Internet Network Design

Download Report

Transcript FIND: Future Internet Network Design

Wireless Sensor
Networks
Nov 1, 2006
Jeon Bokgyun ([email protected])
Reference
Tutorial, Wireless sensor networks,
mobicom 2002
1. Introduction – Deborah
2. Sensor Node Platforms & Energy Issues –
Mani
3. Time & Space Problems in Sensor
Networks – Mani
4. Sensor Network Protocols – Deborah
5. Collaborative Signal Processing – Akbar
6. Discussion - All
2/26
[email protected]
Contents
Introduction
Features of sensor network
Sensor hardware platform
Energy issue
Time synchronization
Node localization
Sensor coverage
Conclusion
3/26
[email protected]
What is the sensor?
Sensor: a transducer that converts a
physical, chemical, or biological parameter
into an electrical signal
Actuator: a transducer that accepts an
electrical signal and converts it into a
physical, chemical, or biological action
Transducer: a device converting energy
from one domain into another. The device
may either be a sensor or an actuator.
4/26
[email protected]
Sensor Network Architecture
Internet,
Satellite, etc
Sink
Sink
Task
Manager
Tens of thousand nodes

Densely deployed
5/26
[email protected]
Sensor Network Applications
Seismic monitoring
Contaminant transport
Ecosystem monitoring
Transportation and urban monitoring
Infant monitoring
Personalized adv.
Etc.
6/26
[email protected]
Sensor Network Features
Densely deployed and prone to failure
The topology changes very frequently
May leverage broadcasting than point-topoint communications
May operate in aggregate fashion
Sensor nodes are limited in power,
computational capacities, and memory
May not have global ID like IP address
Need tight integration with sensing tasks
7/26
[email protected]
Capabilities
Sensor Node HW Platform
StarGate
MK - II
iBadge
MICA Mote
Size, Power Consumption, Cost
8/26
[email protected]
Sensor Node HW Platform







COTS dust prototypes (Kris Pister et al.)
weC Mote (~30 produced, 1998)
Rene Mote (850+ produced, 1999-2000)
Dot (1000 produced, 2000)
Mica node ( 5000+ produced, 2001)
Mica2 (2002)
MicaZ, Telos (2004)
9/26
[email protected]
Sensor Node SW Platform
TinyOS

Programming concepts for resource constrained
networked embedded devices
SOS

Enabling dynamic embedded software.
10/26
[email protected]
Contents
Introduction
Features of sensor network
Sensor hardware platform
Energy issue
Time synchronization
Node localization
Sensor coverage
Conclusion
11/26
[email protected]
Where does the energy go?
Mobilizer
Location Finding System
Processor
Transceiver
Sensor ADC
Memory
Power Unit
ADC : Analog to Digital Converter
12/26
[email protected]
Energy Observation
Communication >> computation (at short
range)
Radio RX power May dominate (at short
range)
Energy Spent in idle RX dominates lifetime
energy consumption
13/26
[email protected]
Radio Energy Management
Short range links



Shutdown based
Turn off sender and receiver
Topology management schemes exploit this
Long range links



Scaling based
Slow down transmissions
Energy-aware packet schedulers exploit this
14/26
[email protected]
Motivation for Time Synchronization
 Most applications require some synchronization
accuracy




Fire and flood tracking
Animal movement
Vehicle movement
Gunshot detection
15/26
[email protected]
Synchronization in Sensor Network
Network time protocol (NTP) for Internet
clock synchronization
Difference: for sensor networks





Time synchronization requirements more
stringent (µs instead of ms)
Power limitations contain resources
May not have easy access to synchronized
global clocks
NTP assumes that pairs of nodes are constantly
connected and experience consistent
communication delays
Often, local synchronization sufficient
16/26
[email protected]
Network Time Protocol (NTP)
 Primary servers (S1)
synchronize to
national time
standards

S1
Satellite, radio, modem
 Secondary servers (S2,
…) synchronize to
primary servers and
other secondary
servers

Hierarchical subnet
S1
S1
S2
S1
S2
S2
S2
S2
S2
S3
S3
S4
S3
}
S1
S2
S3
S3
17/26
S1
Primary
Secondary
[email protected]
RBS (Reference Broadcast Sync.)
NIC
NIC
Sender
Sender
Receiver
Receiver 1
Critical Path
Time
Receiver 2
Critical Path
Traditional critical path:
From the time the sender
reads its clock, to when the
receiver reads its clock
RBS: Only sensitive to the
differences in receive time
and propagation delay
18/26
[email protected]
RBS (cont.)
Receiver to receiver synchronization
Two stage


Transmitter broadcast clock time
Receivers exchange observations
Assumptions


Propagation delay is zero
No clock skew
RBS outperforms NTP

11usec precision over 19.2K radios
19/26
[email protected]
Why is Localization Important?
Very fundamental component for many
other services





GPS does not work everywhere
Smart Systems – devices need to know where
they are
Geographic routing & coverage problems
People and asset tracking
Need spatial reference when monitoring spatial
phenomena
20/26
[email protected]
Techniques for Location Sensing
Measure proximity to “landmarks”
Dead reckoning: position relative to an
initialization point
Measure direction of landmarks
Measure distance to landmarks
Measure difference in distances to two
landmarks
21/26
[email protected]
Solving over multiple hops
Interative Multilateration
Problems
Error accumulation
May get stuck!!!
Unknown node
(known position)
Beacon node
(known position)
22/26
[email protected]
Solving over multiple hops
Collaborative Multilateration
1
1
1
3
3
5
4
2
3
5
5
4
4
2
2
23/26
[email protected]
Sensor coverage
How well can the field be observed?
As the measure of QoS of a sensor
network.
Example usage

Commander
 Weakest path : what path is the enemy likely
to take?
24/26
[email protected]
Worst-case Coverage
 Voronoi Diagram


25/26
Path of Maximal Breach
of Surveillance in the
sensor field lies on the
Voronoi diagram lines.
When adding node, the
next node is deployed
along the edge closest
to the original nodes.
[email protected]
Conclusion
In the future, this wide range of
application areas will make sensor
networks an integral part of our lives.
Sensor network has various constraints.
Briefly, introduce energy issue, timing
synchronization, node localization, and
sensing coverage.
26/26
[email protected]