Transcript Underground Sensor Networks: Research Challenges

NEXT GENERATION SENSOR
NETWORKS
I. F. AKYILDIZ
Broadband Wireless Networking Laboratory
School of Electrical and Computer Engineering
Georgia Institute of Technology
Tel: 404-894-5141; Fax: 404-894-7883
Email: [email protected]
Web: http://www.ece.gatech.edu/research/labs/bwn
NEXT GENERATION SENSOR NETWORKS OVERVIEW
* DATA SENSOR NETWORKS
* MULTIMEDIA SENSOR NETWORKS
* WIRELESS UNDERWATER SENSOR NETWORKS
* WIRELESS UNDERGROUND SENSOR NETWORKS
IFA’2007
PUC
2
RESEARCH ON COMMUNICATION
PROTOCOLS SO FAR !!!

TWEAKING KNOWN CLASSICAL
COMMUNICATION PROTOCOLS FOR
MAC, ROUTING, TRANSPORT LAYERS

TWEAKING KNOWN TOPOLOGY and POWER
MANAGEMENT SCHEMES
 A “PAPER WRITING RACE !!!!”
IFA’2007
PUC
3
DATA SENSOR NETWORKS
SEVERAL GRAND CHALLENGES ARE STILL EXISTING!!!
IFA’2007
PUC
4
GRAND CHALLENGE:
Traditional layered approach is not suitable for WSNs !
Physical Layer
Traditional Approach
IFA’2007
PUC
Our View
Power Management Plane
MAC Layer
Cross-Layer
Melting
Application Layer
Cross-Layer Management Plane
Network Layer
Power Management Plane
Transport Layer
Task Management Plane
Application Layer
5
XLM: Cross-Layer Module
I.F. Akyildiz, M. C. Vuran, O. B. Akan, “XLM: Cross Layer Module for
Efficient Communication in Wireless Sensor Networks,"
Proc. of Int. Conf on Info Science and Systems, Princeton, March 2006.
Application Layer
Transport
 Initiative Concept
– Communication incentive is passed to the
receiver
 Receiver Contention
– Potential receivers contend for packets and
become next-hop
Network
 Local XL congestion control
– Highly congested nodes do not participate in
communication
MAC
 Angle-based routing
PHY
 Channel adaptive operation
– Adaptive to local minima in case of ‘voids’ in the
network
– Receivers adapt communication parameters based
on channel conditions
 Duty cycle operation
IFA’2007
– Energy consumption centric operation via duty
cycle
PUC
6
Initiative Concept
Core of XLM
A node participates in communication based on its
initiatives
When a node has a packet to send, it broadcasts
an RTS packet
A neighbor node contends for routing of the packet
based on its initiatives
IFA’2007
PUC
7
Initiative Concept
 Node Initiative

  RTS  Th 




Th
 relay  relay 

1, if 


max
I 
 



min 

 Erem  Erem 



0, otherwise

IFA’2007
PUC
8
Initiative Concept
 Node Initiative depends on

Received RTS packet’s signal
to noise ratio (SNR) –
channel quality

  RTS  Th 




Th
 relay  relay  Input packet rate for Cong.

1, if 
 Control

max
I 
 

 Buffer level for Cong. C.

min 

 Erem  Erem 

 Remaining energy

0
,
otherwise


 If all the inequalities are satisfied, node participates
in communication
9
IFA’2007
PUC
Transmission Initiation
 When a node has packet to send
– Listens to channel
– If channel is busy, it performs backoff with CWRTS
– If channel is idle, broadcasts an RTS packet
 Nodes receiving an RTS packet
– Check their location relative to source and destination
– Measure RTS packet’s SNR, RTS
IFA’2007
PUC
10
Transmission Initiation
Infeasible nodes
Feasible nodes
Sink
IFA’2007
PUC
11
Receiver Contention

A3
A2
Nodes closer to the sink are
highly likely to win the
contention
A1
Sink
RTS
A1
A2
A3
CTS
IFA’2007
PUC
CW1
CW2
CW3
CW4
12
Receiver Contention
A3
A2
A1
Sink
RTS
CTS
IFA’2007
PUC
CW1
CW2
13
Receiver Contention
A3
A2
A1
Sink
DATA
RTS
CTS
IFA’2007
PUC
14
Receiver Contention
A3
A2
A1
Sink
DATA
RTS
CTS
IFA’2007
PUC
ACK
15
Receiver Contention

A3

A2

A1
I = 1 may not hold for any node (high
congestion)
After CW4 if no CTS is heard,
neighbors send KEEP ALIVE packet
Sender determines congestion and
decreases transmission rate
Sink
A1
A2
A3
KEEP ALIVE
IFA’2007
CW1
PUC
CW2
CW3
CW4
16
XLM: Implementation
IFA’2007
PUC
17
XLM: Implementation
 Simple code
 Various functionalities performed through only
initiative concept
–
–
–
–
–
Error control
Medium access control
Routing
Congestion control
Energy conservation
 No neighbor tables
 No state information
 Efficient, adaptive, cross-layer operation
IFA’2007
PUC
18
XLM: Comparison with 5 Different
Layered Protocol Stacks
Energy consumption  avg. 65% reduction
Throughput  avg. 32% increment
Goodput  avg. 39% increment
Implementation  avg. 25% reduction in code
space
IFA’2007
PUC
19
GRAND CHALLENGE:
HOW TO REALIZE THE MAPPING??
Architecture
and Topology
User Requirements/
Applications
Communication
Protocols
IFA’2007
PUC
20
FURTHER GRAND CHALLENGES
 Cost Reduction to CENTS ??
 Deployment (Architecture) Decisions
(optimum # of sensors, optimum # of sinks,
optimal locations, fast deployment, reusability,
terrain considerations)
IFA’2007
PUC
21
FURTHER GRAND CHALLENGES
 TERABYTE of sensed information
 Network Monitoring and Management
 How to integrate WSNs into NGWI ?
IFA’2007
PUC
22
SEMI-GRAND CHALLENGE:
ERROR CONTROL & OPTIMAL PACKET SIZE
M. C. Vuran and I. F. Akyildiz, “Cross-layer Analysis of Error
Control in WSNs, IEEE SECON ‘06, September 2006.
 First work that jointly considers
 Broadcast wireless channel
 Multi-hop communication
 Realistic channel model
 Realistic hardware models (Mica2 and MicaZ)

Optimization Metrics
 Packet Throughput
 Energy per useful bit
 Resource utilization
IFA’2007
PUC
23
SEMI-GRAND CHALLENGE:
ERROR CONTROL & OPTIMAL PACKET SIZE
Application Requirement
Error
Control
Type
Optimal
Packet
Size
(byte)
Energy
Throughput
Delay
Reliability
-
High
-
-
ARQ
152
Low
-
-
-
FEC
250
Low
-
Low
-
ARQ
25
Low
-
Low
High
ARQ
162
IFA’2007
PUC
24
WIRELESS MULTIMEDIA SENSOR NETWORKS
I.F. Akyildiz, T. Melodia, K. Chowdhury, “A Survey on Wireless Multimedia
Sensor Networks”, Computer Networks (Elsevier), March 2007.
 Sensors with video and audio streams, still
images, and scalar sensor data.
 Also able to store process in real-time, correlate
and fuse multimedia data originated from
heterogeneous sources.
IFA’2007
PUC
25
Wireless Multimedia Sensor Networks
Sink
Interne
t
LEGEND
Multimedia
processing hub
Video sensor
Gateway
Audio sensor
High end video
sensor
Scalar sensor
Wireless gateway
Storage hub
(a)
(b)
Single-tier flat, homogeneous
Single-tier clustered, heterogeneous
sensors, distributed processing, sensors, centralized processing,
centralized storage
centralized storage
IFA’2007
PUC
(c)
Multi-tier, heterogeneous sensors,
distributed processing, distributed
storage
26
Video Sensors
IFA’2007
PUC
27
Stargate + Garcia = Multimedia Mobile
Sensor
IFA’2007
PUC
28
APPLICATIONS
 Tracking
 Home Automation
 Environmental monitoring
 Multimedia Surveillance Sensor Networks
(against crime and terrorist attacks, law
enforcement agencies to monitor areas, public
events, private properties and borders).
IFA’2007
PUC
29
APPLICATIONS
 Advanced Health Care Delivery
(Patients will carry medical sensors to monitor parameters such as
body temperature, blood pressure, pulse oximetry, breathing activity)
 Automated Assistance for the Elderly and Family Monitors
 Environmental Monitoring (acoustic and video feeds)
 Person Locator Services (locate missing persons)
 Industrial Process Control
IFA’2007
PUC
30
GENERAL WMSNs CHALLENGES
Resource Constraints
Variable Channel Capacity
Application-Specific QoS Requirements
High Bandwidth Demand
IFA’2007
PUC
31
GENERAL WMSNs CHALLENGES
 Power Consumption
 Integration with Internet (IP) Architecture and
other Wireless Technologies
 Protocols, Algorithms and Architectures to
maximize the network lifetime while providing QoS
required by the applications
IFA’2007
PUC
32
COMMUNICATION PROTOCOLS
Network Layer
Data Link Layer
Physical Layer
IFA’2007
PUC
Task Management Plane
Transport Layer
Power Management Plane
Application Layer
33
OPEN RESEARCH ISSUES IN APPLICATION LAYER
• Perform source coding according to application requirements and
hardware constraints, by leveraging advanced multimedia encoding
techniques (Distributed Source Coding Techniques)
• Traffic management and admission control functionalities
• Provide flexible and efficient system software
• Provide primitives for applications to leverage collaborative,
advanced in-network multimedia processing techniques
IFA’2007
PUC
34
OPEN RESEARCH ISSUES IN APPLICATION LAYER
 Differentiation between traffic types
Integrated Traffic: (AUDIO, VIDEO, DATA, STILL IMAGE)
– Delay in/sensitive, Jitter in/sensitive, Loss
in/sensitive, Different data rates
 How to guarantee delay bounds and jitter bounds?
 Data aggregation?
 Explore the tradeoffs between quality and energy
consumption!!
IFA’2007
PUC
35
OPEN RESEARCH ISSUES IN TRANSPORT LAYER







TCP/UDP Co-existence (Unfairness/TCP Starvation)
Distinction between Congestions and Errors
ACK Elimination?
Buffering Problem
Tradeoff between Reliability and Congestion Control
Real-Time Communication Support
Relation between Multimedia Coding Rate and
Reliability
IFA’2007
PUC
36
Routing Algorithms Overview
Network Condition
Based
Traffic Classes
Based
Metrics -
Metrics -
•
Position wrt sink
• Radio characteristics
• Error rate
• Residual energy
• Backlogged packets
IFA’2007
•
QoS profiles/Traffic
classes
• Dropping rate
• Latency tolerance
• Desired bandwidth
PUC
Real Time Streaming
Based
Metrics • Spatio-temporal
character
• Probabilistic delay
guarantees
37
QoS-based Routing Protocols
Energy-Aware QoS Routing
K. Akkaya, M. Younis, “An Energy-Aware QoS Routing
Protocol for
Wireless Sensor Networks,”
IEEE Int. Conf. on Distributed Computing Systems, 2003.
 Reliable Information Forwarding using Multiple Paths
B. Deb, S. Bhatnagar, B. Nath, “ReInForm: Reliable
Information
Forwarding Using Multiple Paths in Sensor Networks,”
IEEE Int. LCN Conf. 2003.
IFA’2007
PUC
38
38
QoS-based Routing Protocols
 SPEED
T. He, J.A. Stankovic, C. Lu, T. Abdelzaher, “SPEED: A
Stateless
Protocol for Real-Time Communication in Sensor
Networks,”
IEEE Int. Conf. on Distributed Computing Systems, 2003.
 MMSPEED
A. Felemban, C.G. Lee, E. Ekici, “MMSPEED: Multi-Path
MultiSPEED Protocol for QoS Guarantee of Reliability and
Timeliness in
IFA’2007
PUC
39
39
Open Research Issues in NETWORK LAYER
 Still most of the work are on “best effort” services.
 Strict delay guarantees is a difficult problem !!
 MMSPEED takes a probabilistic approach… more
research needed !!!
IFA’2007
PUC
40
Open Research Issues in NETWORK LAYER
 What are the optimal routing metrics?
 Metrics, like energy, delay etc., form a cost function which
is then minimized in previous work.
 Choice of the weights for these metrics are done heuristically;
 However, they are subject to dynamic network conditions,
thus, more research is needed!!
IFA’2007
PUC
41
Open Research Issues in NETWORK LAYER
 End-to-end QOS guarantees are not easy to achieve!
 When sensed data from the field is sent via the Internet, a single
routing metric is unsuitable for the entire path between source and
end user.
 Decoupling of reliability and routing parameters at such
network boundaries and a seamless integration of schemes better
suited to wired or wireless domains, respectively, need to be explored.
IFA’2007
PUC
42
QoS-based MAC Protocols
Y.Liu,
I. Elhanany,
H. MAC
Qi, “An Energy-Efficient QoSQoS
Aware
Aware Media Access Control Protocol for Wireless Sensor
Networks,” IEEE Int. MASS Conf., 2005.
 Coloring-based Real Time
H.Li,Communication
P. Shenoy, K. Ramamritham, “Scheduling Communication in
Real-Time Sensor Applications,” IEEE Int. RTAS Conf., 2004
Scheduling
IFA’2007
PUC
43
43
Open Issues in MAC LAYER
* Efficient Channel Access Policies
* Scheduling and Buffer Management
IFA’2007
PUC
44
FURTHER LINK LAYER OPEN RESEARCH PROBLEMS
ERROR CONTROL
FEC  Which one?
ARQ  Depends !!
Hybrid ARQ ??
Multimedia Packet Size Optimization !!
IFA’2007
PUC
45
SENSOR NETWORKS IN
CHALLENGED ENVIRONMENTS
 UNDERGROUND APPLICATIONS
 UNDERWATER APPLICATIONS
IFA’2007
PUC
46
Wireless Underground Sensor Networks
I.F. Akyildiz and Erich Stuntebeck, “Underground Sensor Networks:
Research Challenges”, Ad Hoc Networks (Elsevier) Journal, Nov. 2006.
Sink
Soil Condition
Sensor
-Water
-Salinity
-Temperature
IFA’2007
PUC
47
FURTHER APPLICATIONS
 Sports field monitoring
– Golf courses
– Soccer fields
– Baseball fields
– Grass tennis courts
IFA’2007
PUC
48
FURTHER APPLICATIONS

Infrastructure monitoring
–
pipes
–
electrical wiring
–
liquid storage tanks
–
underground fuel tanks
–
septic tanks
 Monitoring the structural health of any underground
components of a building, bridge, or dam
 Border Patrol and Security
IFA’2007
PUC
49
Major Undetected Pipe Leak in 2006
[New York Times, March 15, 2006 ]
 The largest oil spill occurred on the tundra of Alaska's North
Slope
– 270 K gallons of thick crude oil spilled over two acres
 Oil escaped through a pinprick-size hole in a corroded 34inch pipe
 Most of the oil seeped beneath the snow without attracting
the attention of workers monitoring alarm systems
 The spill went undetected for as long as five days
IFA’2007
PUC
50
Underground Pipeline Monitoring
Sink
Flow Direction
IFA’2007
Sensor (powered by fluid flow)
PUC
51
Wireless Underground SNs:
Research Challenges
Parameter
 Extreme Path Losses
– Models predict up to 150
dB/m loss, dependent on
the factors at right
Effect on Path
Loss
Frequency
Water content
– Soil water content has the
most effect on path loss of
any soil parameter
Temperature
% Sand particles in soil
% Clay particles in soil
IFA’2007
PUC
52
Underground Channel Challenges
Power Constraints
– Difficult/impossible to change the batteries for
underground devices
– High radio power necessary due to extreme path losses
Low data rate
– Channel conditions are best at low carrier frequencies
– Less bandwidth is available at lower frequencies
IFA’2007
PUC
53
Underground Channel Challenges
 Antenna Design
 Extremely Lossy Environment
– Strong FEC needed to help overcome weak signals,
but must not use excessive energy in processing
 A comprehensive channel model for the
underground did not exist*
IFA’2007
PUC
54
* Underground Channel Modeling & Analysis
L. Li, M. C. Vuran, I. F. Akyildiz, “Characteristics of Underground
Channel for Wireless Underground Sensor Networks”, IFIP Med-Hoc
(Mediterranean Ad Hoc Netw.) Conference, Corfu, Greece, June 2007.
Modeling Approach
1. Path Loss
2. Propagation Characteristics
3. Multi-path Fading
4. Bit Error Rate Performance
IFA’2007
PUC
55
Wireless Underground SNs:
Current Research
Modeling and Analysis of UNDERGROUND TUNNELS
IFA’2007
PUC
56
Wireless Underground SNs:
Research Challenges
New Communication Protocols needed for both
underground and tunnel applications!!
IFA’2007
PUC
57
WHY UNDERWATER SENSOR NETWORKS?
I.F. Akyildiz, D. Pompili, T. Melodia, “Underwater
Acoustic Sensor Networks: Research Challenges”,
Ad Hoc Networks (Elsevier) Journal, March 2005
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 and
INTERACTIONS WITH ON-SHORE HUMAN OPERATORS!!!
IFA’2007
PUC
58
Applications
• Ocean Sampling Networks
• Pollution Monitoring and other environmental monitoring
(chemical, biological)
• Disaster Prevention
• Assisted Navigation
• Distributed Tactical Surveillance
• Mine Reconnaissance
IFA’2007
PUC
59
UNDERWATER SENSOR NETWORKS ARCHITECTURE
Drifters, Gliders
IFA’2007
PUC
60
Ocean Sampling Sensors
Spread Spectrum Modem
http://www.dspcomm.com/
IFA’2007
Precision Marine Geodetic
Systems
Acoustic Transponders
http://www.link-quest.com
http://www.link-quest.com
PUC
61
Ocean Sampling Sensors
Point measurements in
upper water column 10 &
25 mi off Moss Landing
http://www.mbari.org/aosn/
IFA’2007
Drift buoy: Path followed
by surface currents
Surface station
http://www.mbari.org/aosn/
http://www.link-quest.com
PUC
62
Autonomuos Underwater Vehicles (AUVs)
CARIBOU
IFA’2007
by
Bluefin Robotics Corporation
PUC
63
Autonomuos Underwater Vehicles (AUVs)
Solar recharged AUV
Phantom HD2 ROV
http://www.mbari.org/aosn
http://www.link-quest.com
IFA’2007
PUC
64
Challenges for Underwater Sensor Networks
• Available bandwidth is severely limited
• UW channel is severely impaired (in particular due
to multi-path and fading)
• Very long and extremely variable propagation delays
. Very high bit error rates and temporary losses
of connectivity (SHADOW ZONES)
IFA’2007
PUC
65
Challenges for Underwater Sensor Networks
. Battery power is limited and usually batteries
cannot be recharged; no solar energy!!
• Very prone to failures because of fouling,
corrosion, etc.
• New communication protocols needed!!
IFA’2007
PUC
66
Our Contributions for UW-ASNs
D. Pompili, T. Melodia, and I.F. Akyildiz, “Deployment Analysis in
Underwater Acoustic Sensor Networks,” in Proc. of ACM WUWNet,
Los Angeles, CA, September 2006
 Deployment strategies for 2D and 3D UW-ASNs
– Objective: Developed target sensing and communication
coverage algorithms
 Study of the characteristics of the acoustic
propagation
– Multipath, fading, propagation delay, connectivity,
shadow zones
IFA’2007
PUC
67
ROUTING 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
 Develop algorithms to provide strict or loose latency bounds
for time-critical applications
IFA’2007
PUC
68
Distributed Routing Algorithms
D. Pompili, T. Melodia, and I. F. Akyildiz, "Routing Algorithms for
Delay-insensitive and Delay-sensitive Applications in Underwater
Sensor Networks," in Proc. of ACM MobiCom, LA, CA, Sept. 2006
 Delay-Insensitive Applications:
– Minimize energy by capturing the interactions between
the routing functions and the characteristics of the UW
channel
 Delay-Sensitive Applications:
– Minimize energy while statistically bounding the end-toend packet delay and the packet error rate required by
the application
IFA’2007
PUC
69
MAC PROTOCOLS
 TDMA (difficult synchronization, large guard bands due high delay and variance)
 FDMA (Narrow BW in UW channels; vulnerable to fading and multi-path)
 CSMA (Impractical due to “Idle Listenings” and “Collisions”)
 NOT SUITABLE FOR UNDERWATER APPLICATIONS!
IFA’2007
PUC
70
Our Contributions for UW-ASNs
D. Pompili, T. Melodia, and I. F. Akyildiz, “A Distributed
CDMA Medium Access Control for Underwater Acoustic Sensor
Networks,“Med Hoc Conference, Corfu, Greece, June 2007
A Distributed CDMA MAC Protocol
-
Maximizes the network throughput
– Minimizes the access delays
– Minimizes the required energy to successfully
transmit data packets
IFA’2007
PUC
71
UW- CDMA MAC Main Features
 Unique solution for different architecture scenarios (2D, 3D, AUVs)
 Distributed Closed Loop Algorithm (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
IFA’2007
PUC
72
Power and Code Self Assignment Problem
 Deep Water Channels
 Shallow Water Channels
 Optimization Problem (low computational complexity)
IFA’2007
PUC
73
Our Contributions for UW-ASNs:
J-Sim based UW-ASN Simulator
SensorApp
Developed @BWN-Lab:
-AcousticProp Model
-AcousticChannel
-AcousticPhy
-CDMA-based MAC
-SensorRouting
WirelessAgent
SensorRouting
LL
Queue
CDMA-based MAC
AcousticPhy
AcousticProp Model
IFA’2007
PUC
AcousticChannel
74
Ongoing Research
D. Pompili and I.F. Akyildiz, “Cross-layer Protocol Suite
for Underwater Acoustic Sensor Networks,”
in preparation.
PHY Layer
Our View
Traditional Approach
IFA’2007
Application Layer
Energy Management Plane
MAC Layer
Cross-Layer
Melting
Cross-Layer Management Plane
Network Layer
Energy Management Plane
Transport Layer
Cross-Layer Management Plane
Application Layer
PUC
75
OPEN RESEARCH ISSUES
 Transport Layer (No TCP solutions)
 Error Control Schemes
 Application Layer
IFA’2007
PUC
76