Summary: IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS--PART B: CYBERNETICS, VOL. 34, NO. 2, APRIL 2004 1007
Localization-Based Sensor Validation Using the
Parham Aarabi, Member, IEEE
Abstract--A sensor validation criteria based on the sensor's
object localization accuracy is proposed. Assuming that the true
probability distribution of an object or event in space (x) is
known and a spatial likelihood function (SLF) (x) for the
same object or event in space is obtained from a sensor, then
the expected value of the SLF [ (x)] is proposed as a suitable
validity metric for the sensor, where the expectation is performed
over the distribution (x). It is shown that for the class of
increasing linear log likelihood SLFs, the proposed validity metric
is equivalent to the KullbackLeibler distance between (x) and
the unknown sensor-based distribution (x) where the SLF (x)
is an observable increasing function of the unobservable (x).
The proposed technique is illustrated through several simulated
and experimental examples.
Index Terms--KullbackLeibler divergence, object localization,
sensor validation, sound localization.