Information Fusion, Decision and Control ofensor Network Based Intelligent Systems
Date Issued
2009
Date
2009
Author(s)
Huang, Chu-Hsiang
Abstract
Sensor networks to collect various information from environments enable deployment andpplication of many intelligent devices and systems, such as robots, intelligent vehicles, and eveniomedical instruments. Observing traditional approach separately executing information fusionrom sensor networks, decision, and later control functions, we propose a novel intelligent decisionramework to allow thorough system modeling of such devices, and thus further enhancementeyond traditional approach. Intelligent decision framework improves traditional estimation theoryy separating the mapping from event to observation into two mappings, the mapping frombserved physical quantity to sensor observation and the mapping from target event to physicaluantity. The mathematical formulation is constructed and applied in the firefighting robotavigation scenario to illustrate its effectiveness. We further shows that the intelligent decisionramework can be degenerated to traditional decision schemes under special conditions. Moremportantly, we can extend the framework to fuse observations from multiple kinds of physicaluantities and derive the optimal decision, beyond traditional statistical decision mechanisms. Forhe decision with limited knowledge of the correlations among physical quantities, we proposebservation Selection and derive the equality condition with optimal decision. While fuzzy logic ofess strict-sense mathematic structure is commonly employed to resolve this application scenario,e can demonstrate that Observation Selection derived from well-defined decision theory can be degenerated to fuzzy logic of multiple kinds of observations. Finally, simulation results show thathe proposed intelligent decision framework indeed improves the accuracy of the decision andnhances system performance. In addition to sensor network, this framework can also be applied inarious intelligent system or cognitive systems. We propose a novel cognitive radio spectrumensing scheme, Dual-way Time-Division Spectrum Sensing, derived under intelligent decisionramework to demonstrate the application of this general framework other than sensor network.his scheme mitigates the hidden terminal problem by only one node taking multiple observationsrom independent sensing channel, while cooperative spectrum sensing needs multiple nodes toerform multiple observation. Moreover, this scheme takes the path-loss due to geographicaleparation into consideration to improve the sensing performance. Analytical and simulation resulthows that the proposed spectrum sensing scheme significantly improves the performance ofraditional spectrum sensing.eywords: Sensor network, information
Subjects
Sensor network
information fusion
intelligent decision
data
fusion
multiple observation
intelligent system
robot
navigation
decision theory
cognitive radio
spectrum sensing
receiver sensing
DTD spectrum sensing
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-98-R96942046-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):d486cf0776db3b5bf23a568e3d38f081