Subscribe
Issue No.08 - August (2001 vol.23)
pp: 904-909
ABSTRACT
<p><b>Abstract</b>—In a multiple sensor system, sensor <tmath>$S_i$</tmath>, <tmath>$i=1, 2 \ldots , N$</tmath>, outputs <tmath>$Y^{(i)}\in [0,1]$</tmath>, according to an <it>unknown</it> probability distribution <tmath>$P_{Y^{(i)} | X }$</tmath>, in response to input <tmath>$X \in [0,1]$</tmath>. We choose a fuser—that combines the outputs of sensors—from a function class <tmath>${\cal{F}} = \{ f : [0,1]^N \mapsto [0,1] \}$</tmath> by minimizing empirical error based on an iid sample. If <tmath>$\cal{F}$</tmath> satisfies the isolation property, we show that the fuser performs at least as well as the best sensor in a probably approximately correct sense. Several well-known fusers, such as linear combinations, special potential functions, and certain feedforward networks, satisfy the isolation property.</p>
INDEX TERMS
Sensor fusion, multiple sensor system, information fusion, fusion rule estimation.
CITATION
Nageswara S.V. Rao, "On Fusers that Perform Better than Best Sensor", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.23, no. 8, pp. 904-909, August 2001, doi:10.1109/34.946993
333 ms
(Ver 2.0)

Marketing Automation Platform