Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!agate!eos!shelby!lindy!news
From: GA.CJJ@forsythe.stanford.edu (Clifford Johnson)
Newsgroups: comp.ai
Subject: Re: DATA FUSION
Message-ID: <4749@lindy.Stanford.EDU>
Date: 15 Sep 89 16:45:04 GMT
Sender: news@lindy.Stanford.EDU (News Service)
Distribution: usa
Lines: 32

In article <1882@csm9a.UUCP>,
fhadsell@csm9a.UUCP ( GP) writes:
>I have received a flyer concerning a course by Llinas & Hall on Data
>Fusion.  As I read it the flyer doesn't say what Data Fusion is.  Would
>you please enlighten me?

See Defense Electronics' 1986 handbook of Command, Control,
Communications, and Intelligence (C3-I), at 217,223:

    As C3-I systems have increased in complexity and scope,
    manual methods of merging data are no longer adequate,
    resulting in the need for fully automated methods, variously
    referred to as data fusion, multisource correlation or
    multisensor integration...  The price of the many benefits
    of automated data fusion is, of course, an increase in
    system complexity...  The simplest approach combines
    hard-decision data -- declarations -- by logical rules,
    voting, or weighted summations.  For small numbers of
    sensors and hypotheses, this may be implemented as a lookup
    table of decisions, one for each possible combination of
    sensor reports.  When the number or sensor hypotheses is
    large, logical -- e.g. Boolean -- equations, arithmetic
    equations or semantic networks may be used to more
    efficiently perform the combination and decision-making
    process...  While hard-decision sensors generally include
    internal quantitative signal processing, they also include a
    decision rule -- a preset threshold to achieve defined
    probabilities of correct identification and false alarm --
    which passes only firm decisions for data combination.
    Consider a target approaching two sensors...

Etc.  Anyone got better references on this subject?