ECA
Rule-based
RLID
Data
Management
Jie
Wu,
Dong
Wang,
Huanye
Sheng
Department
of
Computer
Science
&
Engineering,
Shanghai
Jiaotong
University
jwugsjtu.edu.cn
Abstract-
RFID
reader
networks
generate
an
abundance
of
event
data
every
day.
Those
data
are
raw
streams
of
RFID
data,
and
do
not
bear
any
business
meaning.
A
challenge
issue
in
RFID
middleware
is
the
design
of
an
appropriate
mechanism
for
dealing
with
the
mass
data
so
that
logical
conclusions
and
state
of
the
business
can
be
derived.
The
paper
investigates
the
issues
related
to
effective
RFID
data
management
and
introduces
a
great
way
to
collect
complex
decision-making
logic
and
handle
the
mass
data
generated
by
RFID
reader
networks
using
rule
engine.
Event-Condition-Action
(ECA)
rules
are
adopted.
In
our
approach,
RFID
middleware's
reactive
functionality
is
specified
and
managed
within
a
rule
base
rather
than
being
encoded
in
programs.
The
flexibility
and
efficiency
of
RFID
data
processing
is
achieved.
And
the
modularity,
maintainability
and
extensibility
of
the
RFID
middleware
are
enhanced.
I.
INTRODUCTION
RFID
Reader
networks
generate
an
abundance
of
event
data
every
day.
For
RFID
tags,
the
more
the
data
storage,
the
higher
the
cost.
Consequently,
in
most
cases,
an
RFID
tag
placed
on
a
case
or
a
pallet
emits
only
a
unique
product
code.
By
using
an
RF
protocol,
an
RFID
reader
observes
RFID
tags
and
gets
product
codes
from
them.
But
the
codes
contained
in
each
observation
alone
do
not
provide
further
information.
Those
data
are
raw
streams
of
RFID
data,
and
do
not
bear
any
business
meaning.
However,
enterprises
need
to
make
instant
decision
on
high-volume,
high-velocity
data
streams.
And
the
advantage
of
RFID
is
real
time
knowledge.
So
the
captured
RFID
events
need
to
be
transformed
into
high
level
semantic
events
automatically,
that
is
to
turn
simple
events
into
meaningful
events
so
that
logical
conclusions
and
state
of
the
business
can
be
derived.
After
every
observation,
the
observation
time
and
the
location
information
of
the
tag
read
are
added
to
it
by
RFID
reader
and
RFID
Middleware.
During
the
process
of
event
management,
filtering
in-memory
stored
event
data
sets,
aggregating
data,
changing
the
states
of
objects,
and
directing
data
to
appropriate
network
or
enterprise
systems
are
carried
out.
We
are
faced
with
many
challenges
in
the
process:
(1)
Data
Aggregation.
The
associations
between
tagged
objects
are
changing
while
they
move
through
the
supply
chain.
How
to
make
it
flexible
to
allow
"one
to
one",
"one
to
many",
or
"many
to
many"
data
aggregations
and
associations?
(2)
State
Transformation.
Under
special
conditions,
e.g.
the
observations
made
by
special
RFID
readers
suggest
the
state
transformation
of
the
tagged
objects,
from
a
truck
to
a
warehouse,
from
on
sale
to
being
sold
out.
Based
on
those
recorded
state
changes,
users
can
have
improved
visibility
into
moving
objects'
statuses.
A
problem
is
rising,
"How
to
efficiently
get
such
meaningful
state
changes
from
raw
observations?"
(3)
Business
process.
Actions
should
be
taken
immediately
after
the
detection
of
misdirected
materials,
the
identification
of
counterfeit
goods
and
so
on.
How
to
organize
the
knowledge
of
what
happened
leading
up
to
this
event
and
what
actions
should
be
taken
under
this
condition
so
that
it
can
work
in
a
correct
and
efficient
way?
Due
to
these
challenges,
we
argue
that
rule
engine
can
be
efficiently
employed.
Rule
engine
brings
many
advantages:
First,
a
rule
engine
can
make
decisions
based
on
hundreds
of
thousands
of
facts,
quickly,
reliably
and
repeatedly.
Secondly,
rule-based
specifications
are
flexible,
therefore
easy
to
adapt,
alter,
and
maintain
as
requirements
change.
Thirdly,
rules
are
well-suited
for
processing
and
analyzing
by
machines.
Finally,
rules
can
be
managed
in
a
single
rule
base
as
well
as
in
several
rule
bases
possibly
distributed
over
the
Web
[1].
We
propose
a
way
using
ECA
rules
to
process
RFID
data.
Using
ECA
rules
to
define
RFID
event
management
does
not
restrict
the
functionality
of
RFID
event
management.
Arbitrary
functions
can
be
contained
in
the
action.
It
is
just
a
framework
through
which
the
logic
of
RFID
event
management,
or
at
least
some
part
of
it,
is
defined
in
a
declarative
data-oriented
way.
This
paper
puts
forward
a
solution
using
ECA
rules
to
process
RFID
data
efficiently
and
flexibly.
The
remainder
of
this
paper
is
organized
as
follows:
The
second
section
is
the
related
work
section.
The
third
section
provides
an
overview
about
ECA
rules
and
how
does
it
work
in
RFID
processing.
The
fourth
section
is
the
research
method
section.
This
section
explains
the
design,
implementation
and
performance
issues
of
using
ECA
rules
in
RFID
data
processing.
In
the
fifth
section,
we
present
our
conclusions.
II.
RELATED
WORK
The
concept
of
the
architecture
of
Savant[12]
(ALE's
predecessor)
the
RFID
middleware
component
in
the
EPC
Network
was
initially
proposed
by
the
Auto-ID
Center,
an
industry-sponsored
research
program
to
foster
RFID
adoption.
While
it
features
functionality
for
adapting
different
kinds
of
RFID
readers,
event
management,
task
management
and
real
time
in-memory
event
database,
it
was
programmed
in
java
language,
and
does
not
use
rules
to
process
RFID
data.
The
ALE
specification
is
the
application-level
interface
standard
developed
by
EPCglobal
to
allow
clients
to
obtain
filtered
and
consolidated
EPC
observations
from
a
variety
of
Authorized licensed use limited to: HUNAN UNIVERSITY. Downloaded on July 1, 2009 at 21:41 from IEEE Xplore. Restrictions apply.
