Sensor and Data Fusion: A Tool for Information Assessment and Decision MakingAnnotation This book describes the benefits of sensor fusion as illustrated by considering the characteristics of infrared, microwave, and millimeter-wave sensors, including the influence of the atmosphere on their performance, sensor system application scenarios that may limit sensor size but still require high resolution data, and the attributes of data fusion architectures and algorithms. The data fusion algorithms discussed in detail include classical inference, Bayesian inference, Dempster-Shafer evidential theory, artificial neural networks, voting logic as derived from Boolean algebra expressions, fuzzy logic, and detection and tracking of objects using only passively acquired data. A summary is presented of the information required to implement each of the data fusion algorithms discussed. Weather forecasting, Earth resource surveys that use remote sensing, vehicular traffic management, target classification and tracking, military and homeland defense, and battlefield assessment are some of the applications that will benefit from the discussions of signature-generation phenomena, sensor fusion architectures, and data fusion algorithms provided in this text. |
From inside the book
Results 1-3 of 74
Page 198
... output is adapted first by inverting its binary output . If the number of output errors produced by the training patterns is reduced by the trial adaptation , the weights of the selected elements are changed by the a - LMS error ...
... output is adapted first by inverting its binary output . If the number of output errors produced by the training patterns is reduced by the trial adaptation , the weights of the selected elements are changed by the a - LMS error ...
Page 204
... output y is correct and the linear output s falls outside the dead zone , the weights are not updated . In this case Wk + 1 = Wk if & k = 0 and sky . ( 7-23 ) If the quantizer output is incorrect or if the linear output falls within the ...
... output y is correct and the linear output s falls outside the dead zone , the weights are not updated . In this case Wk + 1 = Wk if & k = 0 and sky . ( 7-23 ) If the quantizer output is incorrect or if the linear output falls within the ...
Page 259
... output . Fuzzy systems contain membership functions and production rules or fuzzy associative memory . Membership functions define the boundaries of the fuzzy sets consisting of the input and output variables . The production rules ...
... output . Fuzzy systems contain membership functions and production rules or fuzzy associative memory . Membership functions define the boundaries of the fuzzy sets consisting of the input and output variables . The production rules ...
Other editions - View all
Sensor and Data Fusion: A Tool for Information Assessment and Decision Making Lawrence A. Klein No preview available - 2012 |
Sensor and Data Fusion: A Tool for Information Assessment and Decision Making Lawrence A. Klein No preview available - 2004 |
Common terms and phrases
a₁ absorption Adaline algorithm angle tracks application artificial neural networks atmospheric attenuation B₁ Bayesian inference calculated classical inference classification coefficient coherent processor combined computed conditional probability confidence interval confidence levels correlation data association data fusion decision Dempster-Shafer Dempster's rule detection modes direction angle measurements distribution elements emission emitters error estimation example false alarm probability frequency fusion algorithms fusion architecture fusion process fuzzy logic fuzzy sets H₁ hypothesis identification infrared Kalman filter layer likelihood function linear membership functions MODTRAN multibeam antenna multiple sensor multisensor neuron node object operating optimal output P-value parameters passive percent perceptron probability mass Proc produced production rules proposition radiometer rain range ratio resolution scans sensor data Sensor Fusion sensor system signal processing signatures standard deviation statistical surveillance radar Table techniques temperature threshold track from Sensor unsupervised learning update vector voting logic wavelength weights