+ All Categories
Home > Technology > Data fusion

Data fusion

Date post: 11-Apr-2017
Category:
Upload: yousef-emami
View: 281 times
Download: 1 times
Share this document with a friend
28
A New Approach to Data Fusion in UnderWater Wireless Sensor Networks YOUSEF EMAMI [email protected] Department of Computer Engineering and Information Technology-Shiraz University of Technology 1
Transcript
Page 1: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

1

A New Approach to Data Fusion inUnderWater Wireless Sensor Networks

YOUSEF [email protected]

Page 2: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

2

Agenda AgendaRudiments

ApplicationsChallengesAlgorithms

Literature ReviewConclusionReference

Page 3: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

3

IEEE Remote Sensing Society Data Fusion Technical Committee defines data fusion as below:

The process of combining data provided by different instruments and source in order to improve the processing and interpretation of these data.

Hall defines data fusion as below: A combination of data from multiple sensor to accomplish improved accuracy and more specific inferences that could achieved by the use of single sensor alone

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 4: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

4

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

The relationship among Fusion Terms

Page 5: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

5

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Data Fusion in the Information-Processing Cycle

Page 6: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

6

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Data Mining in the Information-Processing Cycle

Page 7: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

7

Rudiments Data Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Classification Based on Relationship among Resources

Page 8: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

8

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review Reference

Conclusion

Page 9: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

9

The major challenges in the design of underwater sensor networks are as follows:

The available bandwidth is severely limited Propagation delay under water is five orders of magnitude higher than that in RF

terrestrial channels and extremely variable High bit error rates and temporary losses of connectivity Battery power is limited and usually batteries cannot be recharged; also, solar energy

cannot be exploited.

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 10: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

10

Terrestrial Sensor Networks vs. Underwater Sensor Networks

Size and cost Deployment Power Memory

RudimentsData Fusion

UWSNsApplications Challenges

AlgorithmsMethods

TechniquesLiterature

Review ReferenceConclusion

Page 11: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

11

Robotics Intelligent Transportation Systems Precision

Agriculture Security, Improving Intrusion Detection Data

Privacy Assessing and Monitoring Civil Infrastructures

Environmental Monitoring Fire Detection Financial analysis Fault diagnosis Medical Diagnoses

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 12: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

12

Improving Barrier Coverage in Wireless Sensor Networks Detect routing failures Collect link statistics for routing protocols

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 13: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

13

Data imperfectionData provided by sensors is always affected by some level of impreciseness as well as uncertainty in the measurements.

Data modalitySensor networks may collect the qualitatively similar (homogeneous) or different (heterogeneous) data, Both cases must be handled by a data fusion scheme.

Outliers and spurious dataData fusion algorithms should be able to exploit the redundant data to alleviate such effects

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 14: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

14

Processing frameworkData fusion processing can be performed in a centralized or decentralized manner. Data dimensionalityThe measurement data could be preprocessed, either locally at each of the sensor nodes or globally at the fusion center to be compressed into lower dimensional data, assuming a certain level of compression loss is allowed Data alignment/registrationSensor data must be transformed from each sensor’s local frame into a common frame before fusion occurs. Data registration is of critical importance to the successful deployment of fusion systems in practice.

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 15: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

15

Inference Bayesian Dempster-Shafer Fuzzy Logic Neural Network

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 16: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

16

Estimation Least squares Maximum likelihood Moving average filter Kalman filter Particle filter

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 17: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

17

Rough set based fusion

A mathematic method which analyses and treats vagueness and uncertainty, and offers an effective method to data fusion system.

It can analyze incomplete or inaccurate data to extract useful information.

Rough set theory is now mainly applied in decision making of distributed data fusion system..

Introduction Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 18: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

18

Data fusion in intelligent transportation systems: Progress and challenges – A survey by El Faouzi(2011)

A variety of functions are assigned to ITS to address the traffic congestion and safety problems:

Automatic incident detection (AID)

Advanced driver assistance (ADAS)

Network control, crash analysis and prevention

Traffic demand estimation

Traffic forecast and monitoring

Accurate position estimation

Each of these sub-systems can make use of different information sources. DF techniques can then be used to combine them to yield better results.

Rudimentd Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 19: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

19

Energy efficient wireless sensor network communications based on computational intelligent data fusion for environmental monitoring by D.F.Larios(2012)

Goal: Optimizing energy consumption in an environmental monitoring process

Method: The data fusion is based on a local Self-Organized Map(SOM)

Results: Enhancing network lifetime

RudimentsApplications ChallengesAlgorithmsMethods

Techniques

Literature Review Reference

Conclusion

Page 20: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

20

An approach to implement data fusion techniques in wireless sensor networks using genetic machine learning algorithms by Pinto(2014)

Goal: Optimizing the communication efficiency in dense wireless sensor network

Method: Genetic Algorithm

Result: The proposed approach is able adjust the sending rate of a WSN

RudimentsApplications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 21: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

21

A Power-Aware Framework for Distributed Data Fusion Application in Wireless Sensor Networks by Zongqing Lu(2011)

Goal: Mapping distributed data fusion application into networks.

Method: Fusion function placement and fusion function transfer maintenance.

Result: PAFusion has better performances than DFuse and PAFusion has much less overhead and causes only slightly more transmission cost than optimal solution.

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 22: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

22

Multisensor data fusion for underwater navigation by Majumder(2001)

Goal: Integrating information from a number of sensors to navigate a sub-sea vehicle

Method: Data Fusion

Result: Successful Navigation

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 23: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

23

Multi-sensor Data Fusion for Underwater Target Recognition Under Uncertainty by Tang Zheng(2010)Goal: Under Water Target Recognition

Method: A dynamic information fusion framework which is based on discrete dynamic bayesian network (DDBN) for representing, integrating and inferring various target characteristics from dynamic sensory information of different modalities

Result: The experimental results demonstrate the utility of the proposed method in efficiently modeling and inferring dynamic events.

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 24: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

24

An Energy-Efficient Data Fusion Protocol for Wireless Sensor Network by Bin Zeng (2007)

Goal: Proposing a data fusion protocol for gathering event data in sensor networks

Results: LEECF not only optimizes the transmission and fusion energy costs, but also increases the fusion speed for sensor nodes. LEECF is completely distributed, requiring no control information from the sink after node level is set up at first, and the nodes do not require knowledge of the global network in order for LEECF to operate.

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 25: Data fusion

Other applications:

Data fusion for underwater target tracking by S. Koteswara Rao(2009)Data Fusion Method for Underwater Object Localization by Yun Lu (2013)Application of the Multi-sensor Fusion Method for Underwater Landscape Modeling By Noel(2014)

Department of Computer Engineering and Information Technology-Shiraz University of Technology 25

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 26: Data fusion

To minimize energy consumption of Underwater wireless sensor networks ,Data fusion techniques are good candidate and can prolong lifetime of UWSNs.

Department of Computer Engineering and Information Technology-Shiraz University of Technology 26

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review ReferenceConclusio

n

Page 27: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

27

THANK YOU FOR YOUR KIND ATTENTION

RudimentsApplications

Challenges AlgorithmsMethods

Techniques

Literature Review

ReferenceConclusion

ProblemStatemen

t

Page 28: Data fusion

Department of Computer Engineering and Information Technology-Shiraz University of Technology

28

1. Lawrence A. Klein, “Sensor and Data Fusion” ,SPIE Press,2010

2. Ahmed Abdelgawad,Magdy Baypumi ,” Resource-Aware Data Fusion Algorithms for Wireless Sensor Networks”, Springer Publication, 2012

3. Martin E. Liggins ,David L. Hall,James Llinas,”Handbook of Multisensor Data Fusion”,CRC Press,2009

4. E. F. Nakamura, et al., "Information fusion for wireless sensor networks: Methods, models, and classifications," ACM Comput. Surv., vol. 39, p. 9, 2007.

5. Ian F. Akyildiz, Mehmet Can Vuran, “Wireless Sensor Networks”, John Wiley & Sons,2010

6. B. Khaleghi, et al., "Multisensor data fusion: A review of the state-of-the-art," Information Fusion, vol. 14, pp. 28-44, 2013.

7. N.-E. E. Faouzi, et al., "Data fusion in intelligent transportation systems: Progress and challenges – A survey," Information Fusion, vol. 12, pp. 4-10, 2011.

8. B. J. Larios D.F, Rodriguez G, "Energy efficient wireless sensor network communications based on computational intelligent data fusion for environmental monitoring," Communication, vol. 6, pp. 2189-2197, 2012.

9 A.R.Pinto,et al.,” An approach to implement data fusion techniques in wireless sensor networks using genetic machine learning algorithms”, Information Fusion,vol.15,pp.90-101,2014

10. S. Koteswara Rao, K.S. Linga Murthy, K. Raja Rajeswari.” Data fusion for underwater target tracking “,vol.4,pp.576-585,2010

Rudiments Applications ChallengesAlgorithmsMethods

Techniques

Literature Review

Reference

Conclusion


Recommended