Chapter 16

download report

Transcript Chapter 16

Chapter 16:
Underwater Sensor Networks
Wireless Sensor Networks
Akyildiz/Vuran
1
WHY UNDERWATER SENSOR NETWORKS?
I.F. Akyildiz, D. Pompili, T. Melodia, “Underwater Acoustic Sensor Networks:
Research Challenges”, Ad Hoc Networks (Elsevier) Journal, March 2005
 Traditional approach for ocean-bottom or ocean column
monitoring:
 Deploy UW instruments to record data during monitoring
mission and then recover them after several months and
evaluate the collected data!!
Wireless Sensor Networks
Akyildiz/Vuran
2
WHY UNDERWATER SENSOR
NETWORKS?
 Problems:
 No real-time monitoring
 No on-line system configuration (no tuning or reconfiguring the
instruments)
 No failure detection
 Limited storage capacity
 DEPLOY UNDERWATER SENSOR NETWORKS TO ENABLE
 REAL-TIME MONITORING
 REMOTE CONFIGURATION
 INTERACTIONS WITH ON-SHORE HUMAN OPERATORS!!!
Wireless Sensor Networks
Akyildiz/Vuran
3
Applications
 Ocean Sampling Networks
 Pollution Monitoring and other environmental monitoring

(chemical,
biological)
 Monitoring of ocean currents and winds
 Detecting climate change
 Understanding effects of human activities on marine ecosystems
 Tracking fishes or micro-organisms
Disaster Prevention
 Sensor networks that measure seismic activity from remote
locations and provide “tsunami” warnings to coastal areas
Wireless Sensor Networks
Akyildiz/Vuran
4
Applications
 Assisted Navigation
 Locate dangerous rocks or shoals in shallow waters,
mooring positions, submerged wrecks
 Distributed Tactical Surveillance
 Monitor areas for surveillance, reconnaissance, targeting
and intrusion detection
 Mine Reconnaissance
Wireless Sensor Networks
Akyildiz/Vuran
5
Typical UW Sensor Tasks
 MEASURE
 Quality of the water
 Light radiation and harmful
algal blooms
 Temperature
 Density
 Salinity
 Acidity
 Chemicals
 Conductivity
 Turbidity
 Hydrogen
 pH Oxygen
 Sulfide Oxidation
Wireless Sensor Networks
Akyildiz/Vuran
6
UNDERWATER SENSOR NETWORKS
2D Architecture for
OCEAN BOTTOM MONITORING
(for environmental monitoring or monitoring of UW plates in tectonics)
Wireless Sensor Networks
Akyildiz/Vuran
7
UNDERWATER SENSOR NETWORKS
3D STATIC Architecture for Ocean Column Monitoring
(for surveillance applications or monitoring of ocean phenomena (ocean bio-geo-chemical processes,
water streams, pollution)
Wireless Sensor Networks
Akyildiz/Vuran
8
UNDERWATER SENSOR NETWORKS
3D DYNAMIC Architecture using Autonomous Underwater Vehicles
(AUVs)
Wireless Sensor Networks
Akyildiz/Vuran
9
UNDERWATER SENSOR NETWORKS
3D Dynamic Architecture using AUVs




Inexpensive AUV submarines equipped with multiple UW sensors that can reach any
depth in the ocean; e.g., Odyssey class AUVs
Drifters
 Use local currents to move and take measurements at preset depths
Gliders
 Battery powered and move 25cm/s (0.5 knots)
 On the surface they can be detected by GPS
 Onshore operators can interact with Gliders
 Depth 200m-1500m with lifespans (few weeks to several months)
 Integration of AUVs with UWSNs is UNEXPLORED !!
Wireless Sensor Networks
Akyildiz/Vuran
10
Ocean Sampling Sensors
Spread Spectrum Modem
http://www.dspcomm.com/
Precision Marine Geodetic
Systems
Acoustic Transponders
http://www.link-quest.com
http://www.link-quest.com
Wireless Sensor Networks
Akyildiz/Vuran
11
Ocean Sampling Sensors
Point measurements in upper
water column 10 and 25 mi off
Moss Landing
http://www.mbari.org/aosn/
Drift buoy: Path followed by
surface currents
Surface station
http://www.mbari.org/aosn/
http://www.link-quest.com
Wireless Sensor Networks
Akyildiz/Vuran
12
Autonomuos Underwater Vehicles (AUVs)


CARIBOU by Bluefin Robotics Corporation
Equipped with state-of-the-art sensors (side-scan sonar and sub-bottom profiler),
and can collect high-quality data for:
 Archaeological remote sensing
 Multi-static acoustic modeling
 Fisheries resource studies and
 Development of concurrent mapping and localization techniques.
Wireless Sensor Networks
Akyildiz/Vuran
13
Autonomous Underwater Vehicles (AUVs)
Solar recharged AUV
Phantom HD2 ROV
http://www.mbari.org/aosn
http://www.link-quest.com
Wireless Sensor Networks
Akyildiz/Vuran
14
Terrestrial vs. Underwater Sensors
Terrestrial Wireless Sensor
Mica Mote MPR300CB
Speed
4 MHz
Flash
128K bytes
Radio Frequency
916MHz or 433MHz
(ISM Bands)
Underwater Acoustic
Modem
Short-range
Medium-range
27- 45 kHz
54-89 kHz
Acoustic Frequency
Data Rate
Transmit Power
7 kbit/s
Data Rate
40 kbits/s (max)
Receive Power
1W
0.75 W
Transmit Power
0.75 mW
Sleep Power
8 mW
Radio Range
100 feet
Radio Range
1000 feet
Power
2 x AA batteries
Wireless Sensor Networks
Akyildiz/Vuran
14 kbit/s
6W
1W
12 mW
3000 feet
15
Terrestrial vs Underwater Sensor
Networks
 Cost
 Deployment
 Power
Wireless Sensor Networks
Akyildiz/Vuran
16
Terrestrial vs Underwater Sensor
Networks
 Available bandwidth is severely limited
 UW channel is severely impaired (in particular due to multi-path and




fading)
Very long (5 orders of magnitude higher than in RF terrestrial
channels) and extremely variable propagation delays
Very high bit error rates and temporary losses of connectivity
(SHADOW ZONES)
Battery power is limited and usually batteries cannot be recharged; no
solar energy!!
Very prone to failures because of fouling, corrosion, etc.
Wireless Sensor Networks
Akyildiz/Vuran
17
Acoustic vs Radio and Optic Waves
 Radio Waves
 Propagate at long distances through conductive sea water only at
extra low frequencies (30-300Hz)
  large antennae and high transmission power
 (e.g., Berkeley MICA motes  transmission range of 120 cm in UW
at 433Mhz)
 Optical Waves
 No high such attenuation but scattering problem
 High precision in pointing the narrow laser beams
 Links in Underwater Networks 
 Acoustic wireless communications
Wireless Sensor Networks
Akyildiz/Vuran
18
Effects on Acoustic Communication
 Path Loss
 Noise
 Multipath
 Doppler Spread
 High and Variable Propagation Delays
 UW Acoustic Channel very limited and dependent on both
range and frequency
Wireless Sensor Networks
Akyildiz/Vuran
19
UW Acoustic Communication Links


Classified according to their range
Available bandwidth for different ranges in UW-A Channels
Range [km]
Bandwidth [kHz]
Very Long
1000
<1
Long
10-100
2-5
Medium
1-10
~ 10
Short
0.1-1
20-50
Very Short
< 0.1
>100

Can also be classified as VERTICAL or HORIZONTAL LINKS according to
the direction of the sound ray with respect to the ocean bottom
Wireless Sensor Networks
Akyildiz/Vuran
20
Factors Influencing Acoustic
Communications
 Transmission (Path) Loss
 Attenuation
 Provoked by absorption due to conversion of acoustic energy
into heat, also by scattering, reverberation, refraction, and
dispersion
 Increases with varying frequency and distance
 Geometric Spreading
 Spreading of sound energy as a result of the expansion of the
wave-fronts;
 Increases with distance and independent of frequency
 Spherical (deep water) and cylindrical (shallow water <100m)
types
Wireless Sensor Networks
Akyildiz/Vuran
21
Transmission Loss
TL  20  Log (d )   ( f )  d  A
Attenuation due to
geometric spreading
Attenuation due to
absorption
Attenuation due to
multipath
TL increases with increasing freq. & distance
Wireless Sensor Networks
Akyildiz/Vuran
22
Factors Influencing Acoustic
Communications
 Noise
 Man Made Noise:
 Machinery noise (pumps, reduction gears, power plant) and

shipping activity
 Ambient Noise:
 Hydrodynamics (currents, storms, wind, rain), seismic, and
biological phenomena
Multi-Path Propagation
 Generates Inter-Symbol Interference (ISI), especially in shallow
waters
Wireless Sensor Networks
Akyildiz/Vuran
23
Factors Influencing Acoustic
Communications
 High Delay and Delay Variance
 The propagation speed in the UW-A channel is five orders
of magnitude lower than in the radio channel (0.67 s/km)
 Doppler Spread  Can be significant in UW-A and causes
ISI
Wireless Sensor Networks
Akyildiz/Vuran
24
Factors that Affect Sound Velocity
 The sound velocity increases with the increase of
temperature, pressure and salinity
Temperature
Wireless Sensor Networks
Akyildiz/Vuran
Depth
Pressure
Depth
Depth
Salinity
25
Typical Ocean Sound Velocity Profile
Speed of Sound (meters/sec)
Depth of Water (meters)
1480
1500
1520
Surface Layer
Seasonal Thermocline
Permanent Thermocline
1000
2000
Deep Isothermal Layer
3000
Wireless Sensor Networks
Akyildiz/Vuran
26
Sound Bends Down When Water Grows
Cooler With Depth
Warm Water
Depth
Shadow Zone
Cool
Water
Wireless Sensor Networks
Akyildiz/Vuran
27
Sound Bends Sharply Up
Cool Water
Depth
Shadow Zone
Warm Water
Wireless Sensor Networks
Akyildiz/Vuran
28
Sound Beam Splits When Temperature is
Uniform at Surface and Cool at Bottom
Depth
Isothermal
Shadow
Zone
Temperature
Cool
Wireless Sensor Networks
Akyildiz/Vuran
29
FOLLOWING TCP/IP PROTOCOL STACK
Data Link Layer
Physical Layer
Wireless Sensor Networks
Akyildiz/Vuran
Task Management Plane
Network Layer
Mobility Management Plane
Transport Layer
Power Management Plane
Application Layer
30
PHYSICAL LAYER


Until 1990  Non-coherent FSK modulation schemes (high power but low BW
efficiency)
In recent years  Fully coherent schemes such as PSK, and QAM (long range and
high throughput)
Also  OFDM (high throughput, robust, high spectral efficiency)






YEAR RATE[kbps] Band[kHz]
Range[km]
FSK 1984
1.2
5
3 (d)
PSK 1991
1.25
10
2 (d)
PSK 1994
0.02
20
0.9 (sh)
DPSK 1997
20
10
1 (d)
QAM 2001
40
10
0.3 (sh)

Wireless Sensor Networks
Akyildiz/Vuran
31
PHYSICAL LAYER
 RESEARCH CHALLENGES:
 Need to develop inexpensive transmitter/receiver
modems
 Low complex sub-optimal filters for real-time
communication with minimum energy expenditure
 Overcome the stability problem in coupling between PLL
and DCE (decision feedback equalizer)
 Channel estimation techniques
Wireless Sensor Networks
Akyildiz/Vuran
32
ERROR CONTROL
 ARQs  not useful!!
 FECs 
the way to go but which one????
 Hybrid ARQ
Wireless Sensor Networks
Akyildiz/Vuran
33
OPTIMAL PACKET SIZE
M. C. Vuran and I. F. Akyildiz, ``Cross-layer Packet Size Optimization for Wireless Terrestrial,
Underwater, and Underground Sensor Networks,’’ to appear in IEEE INFOCOM'08, Phoeniz, AZ,
April 13-18, 2008.
Deep water
Shallow water
Wireless Sensor Networks
Akyildiz/Vuran
34
MAC LAYER
 FDMA MACs
 Not suitable for UW-ASN
Narrow bandwidth in UW channels
Limited spectrum bands are vulnerable to fading and
multi-path
Wireless Sensor Networks
Akyildiz/Vuran
35
MAC LAYER
 TDMA MACs
 Not easy to achieve precise synchronization due to
variable delays
 High delay and delay variance  large guard times 
limited efficiency
Wireless Sensor Networks
Akyildiz/Vuran
36
MAC LAYER
 CSMA MACs
 Impractical due to “Idle Listenings” and “Collisions”
Wireless Sensor Networks
Akyildiz/Vuran
37
CSMA MAC with Sleep Schedules
V. Rodoplu and M. K. Park, “An Energy-Efficient MAC Protocol for Underwater Wireless
Acoustic Networks,” in Proc. IEEE/MTS OCEANS 2005, September 2005
 The objective is to save energy based on sleep periods with
low duty cycles
 However,
 Solution tied to the assumption that nodes follow sleep
periods (synchronization?)
 It mainly aims at efficiently organizing the sleep
schedules
Wireless Sensor Networks
Akyildiz/Vuran
38
SLOTTED FAMA
M. Molins and M. Stojanovic, “Slotted FAMA: A MAC Protocol for UW Acoustic
Networks”, Proc. of IEEE OCEANS, Sept 2006.
 Combines Carrier Sensing and a handshake between
sender and receiver
 Time slotting eliminates long control packets  energy
saved
 PROBLEM:
 Handshaking  low system throughput in very high
propagation delay UW networks
 CS may sense the channel idle while a transmission may
still be going on
Wireless Sensor Networks
Akyildiz/Vuran
39
Why CDMA?
 Robust to frequency-selective fading
 Compensates for the effect of multipath (Rake filters)
 Receivers can distinguish signals simultaneously
transmitted by multiple devices
 Increases channel reuse
 Reduces packet retransmissions
 Decreases energy consumption
 Increases network throughput
Wireless Sensor Networks
Akyildiz/Vuran
40
CDMA MAC Protocol
L. Freitag, M. Stojanovic, S. Singh, and M. Johnson, “Analysis of Channel Effects on Direct Sequence and Frequency-hopped
Spread-Spectrum Acoustic Communication,” IEEE Journal of Oceanic Engineering, Oct. 2001.
 Two SS-CDMA access techniques in shallow water are
compared
Direct Sequence Spread Spectrum (DSSS)
Frequency Hopping Spread Spectrum (FHSS)
 FHSS is shown to lead to a higher BER than DSSS
Wireless Sensor Networks
Akyildiz/Vuran
41
UW-MAC
D. Pompili, T. Melodia, I.F. Akyildiz, ``A distributed CDMA MAC Protocol for Underwater
Acoustic Sensor Networks,’’ IFIP Med Hoc Network Conf, Corfu, Greece, June 2007.
 Challenges:
 Harsh characteristics of the underwater medium, e.g.,
multipath, fading, limited bandwidth, shadow zones….
 A Distributed CDMA MAC Protocol
 Maximizes the network throughput
 Minimizes the access delays
 Minimizes the required energy to successfully transmit
data packets
Wireless Sensor Networks
Akyildiz/Vuran
42
UW-MAC: Main Features
 Unique solution for different architecture scenarios (2D and 3D)
 Distributed closed-loop algorithm to set the optimal transmit power
and code length
 Distributed (no centralized entity to select codes and transmit power)
 Intrinsically secure (use of chaotic codes)
 Supports multicast transmissions (codes decided at the transmitter)
 Robust against inaccurate node position and interference information
Wireless Sensor Networks
Akyildiz/Vuran
43
Network Layer
 Existing routing solutions
are unsuitable for UW Sensor
Networks !!!
 Proactive (DSDV, OLSR)
 Reactive (AODV; DSR)
 Geographical
Wireless Sensor Networks
Akyildiz/Vuran
44
Proactive Approaches
 Large signaling overhead to establish routes for the first
time and each time the network topology is modified
because of mobility or node failures
 Updated topology information has to be propagated to all
nodes in the network
Wireless Sensor Networks
Akyildiz/Vuran
45
Reactive Approaches
 Latency for source-initiated flooding of control packets to
establish paths amplified underwater by the slow
propagation of acoustic sound
 The topology in UW-ASNs is unlikely to vary dynamically on
a short-time scale
Wireless Sensor Networks
Akyildiz/Vuran
46
Geographical Routing
 Very promising for their scalability feature and their
limited
required signaling
 However, GPS receivers do not work in the UW properly
 Need to develop localization algorithms for underwater to
be able to associate each information measured by sensor
with the position of the sensing node !
Wireless Sensor Networks
Akyildiz/Vuran
47
Existing Solutions

G. Xie and J. H. Gibson. “A network layer protocol for UANs to address propagation
delay induced performance limitations,” Proc. of IEEE OCEANS’01, Honolulu, Nov.
2001.

E. M. Sozer, M. Stojanovic, and J. G. Proakis, ``UW Acoustic Networks,’’ IEEE
Journal of Oceanic Eng, Jan. 2000.

P. Xie, J.-H. Cui, and L. Lao, “VBF: Vector-Based Forwarding Protocol for
Underwater Sensor Networks,” in Proc. Networking 2006, May 2006

D. Pompili and I. F. Akyildiz, "Overview of Networking Protocols for Underwater
Wireless Communications," to appear in IEEE Communications Magazine, January
2009.
Wireless Sensor Networks
Akyildiz/Vuran
48
NETWORK LAYER CHALLENGES
 Develop routing algorithms which can cope with
disconnections due to shadow zones, failures, mobility or
battery replacement
 Develop algorithms to deal with possible intermittent
connectivity (SHADOW ZONES)
Wireless Sensor Networks
Akyildiz/Vuran
49
NETWORK LAYER CHALLENGES
 Develop algorithms to provide strict or loose latency
bounds for time-critical applications
 NOTE: Delay depends on the distance while delay jitter also
depends on the nature of the link, e.g., (horizontal links >
vertical links)
Wireless Sensor Networks
Akyildiz/Vuran
50
TRANSPORT LAYER
 It is totally unexplored so far except:
 Peng Xie and Jun-Hong Cui, ``SDRT: A Reliable Data
Transport Protocol for Underwater Sensor Networks,’’
to be published in Elsevier Ad Hoc Networks, Special Issue
on Underwater Networks, July 2007
Wireless Sensor Networks
Akyildiz/Vuran
51
TRANSPORT LAYER
 Classical end-to-end reliability notion is not applicable!
 Very high and very variable RTT
 Window Based or Rate Based TCPs will not work
 Need to distinguish between congestions and errors
 A New End-to-End Reliability notion is needed
Wireless Sensor Networks
Akyildiz/Vuran
52
APPLICATION LAYER
It is totally unexplored so far.
Wireless Sensor Networks
Akyildiz/Vuran
53
Further Research
MULTIMEDIA UNDERWATER SENSOR NETWORKS
 Surveillance using sensor/AUV hybrid networks
 Multimedia (Audio, Video, Data, Still Image) Traffic
 Reconstruction of underwater 3D images and videos
 Computer Vision, Stereovision
 Channel Allocation and Scheduling (Multimedia Traffic
Management)
Wireless Sensor Networks
Akyildiz/Vuran
54
Further Research
MULTIMEDIA UNDERWATER SENSOR NETWORKS
 Protocols, Algorithms and Architectures to maximize the network
lifetime while providing QoS required by the applications
 Distributed Data Mapping
 Convert “raw” readings into metrics of direct user interest
 In-network processing (network & node levels) using classification
methods
 Synchronization (intra-media, inter-media)
Wireless Sensor Networks
Akyildiz/Vuran
55