This invention relates to multipoint communications systems which
employ handshaking, random access and collision detection techniques, and more
particularly to such systems where information must be received from a plurality
of responding terminals. Specifically, the invention can be applied to a supermarket
checkout system designed to reduce checkout time at the point of sale where, typically,
many items are purchased by a customer.
Other possible applications of the invention include, for example,
retail sales in general; delivery control; inventory control; security check of
objects or beings occupying, or entering or leaving a defined area; automatic
toll collection or monitoring of moving vehicles; telemetry; and network data
communications in general.
Several approaches have been taken to implement communications systems
which receive information from multiple terminals. In general, these can be referred
to as transponder implementations, and communications protocols, including the
Aloha technique, carrier sense multiple access with collision detection, and time
division multiple access.
An advantage to using transponders is their relative simplicity.
Transponders receive electromagnetic energy at a specific frequency from an interrogator
and transmit back a signal which may be a continuous tone or an information-containing
message such as an ID code. Typically, transponders do not permit sophisticated
two-way communications, using a protocol, to enable the interrogator to coordinate
or separate the responses from multiple transponders. Because of this, in applications
where two or more transponders are identical and are located closely together,
it may not be possible for the interrogator to determine the number of communicating
transponders or to separate out communications from multiple transponders.
The Aloha technique provides one way in which multiple stations can
share a communication channel over which only one transmission can be received
at a time. Using the Aloha technique, when a station has information to transmit,
it simply transmits it, without first being signaled that the communication channel
is available. Because it is possible for more than one station to transmit at the
same time, several transmissions may overlap or "collide," as it is called in
the art. When a collision occurs, the network controller can either signal the
station to retransmit its data or the network controller can simply drop the transmission
and wait for the station to resend it. The principal disadvantage of the Aloha
technique is that it can only be employed effectively when the likelihood of collision
is very low, such as when the communication channel is idle most of the time.
One specific example of a system using the Aloha technique is found
in U.S. Patent No. 4,352,183 to Davis et al. In an effort to overcome
problems in the most basic form of the Aloha technique, the system described in
U.S. Patent No. 4,352,183 employs a protocol for determining when a particular
transmitter is allowed to transmit. A controlling transmitter first broadcasts
an initial command signal to all the transmitters to indicate that a communication
channel has become available. Each transmitter having data to send transmits a
request for information transmission message to the controlling transmitter in
a randomly chosen one of a predetermined number of timeslots following receipt
of the initial command signal. If no collisions are detected, the controlling
transmitter responds with an acknowledgment signal addressed to the particular
transmitter that sent a request for information transmission message, thereby
permitting that transmitter to send a data-containing message to the controlling
transmitter. A disadvantage of this system is that the controlling transmitter
must periodically issue command signals to indicate when a communication channel
becomes available. Another disadvantage is that the controlling transmitter must
issue a command signal addressed specifically to the requesting transmitter before
that transmitter is permitted to send a data-containing signal.
The carrier sense multiple access with collision detection protocol
provides a way of reducing collisions between transmissions in more heavily utilized
networks. This protocol is well-known in the art and has evolved into an industry
standard: IEEE Standard 8802.3-1992, Information Technology -- Local and Metropolitan
Area Networks --Part 3: Carrier Sense Multiple Access with Collision Detection
(CSMA/CD) Access Method and Physical Layer. Specifications, Institute for Electrical
and Electronics Engineers, Piscataway, New Jersey 08855-1331.
Under this protocol, multiple stations can respond to the same interrogator,
hence the name multiple access. However, before attempting to transmit, each station
monitors the channel to determine if another station is communicating (i.e. the
station senses the presence of a carrier frequency), hence the name carrier sense,
and if so, waits for the other station to finish transmission before attempting
to transmit information. The interrogator responds according to whether or not
two or more stations attempt to transmit simultaneously (i.e. collision detection).
A disadvantage of this protocol when applied to systems where a single device
interrogates a large number of stations simultaneously is that collisions are frequent
and throughput low because several stations attempt to transmit whenever the channel
is found to be available.
Another protocol, time division multiple access, has been used in
multipoint communication systems. Under this approach, each station is assigned,
based on its own unique identifying characteristics, one of a predetermined number
of timeslots in which to transmit. To complete longer transmissions, each station
communicates during its assigned timeslot over multiple cycles. A problem with
this approach is that the interrogator must be able to distinguish between the
responding stations prior to requesting data from them in order to assign an individual
timeslot to each.
European Patent Application EP 0 409 016 A2 describes a system for
locating predetermined labeled objects. An interrogator has a narrow beamwidth
antenna for transmitting an energizing signal at a predetermined first frequency.
A transponder incorporated in the label on each of the labeled objects receives
the energizing signal and transmits a return signal at a predetermined second frequency.
The energizing signal incorporates a predetermined transponder identification
code. If the predetermined transponder identification code matches the identification
code stored in the transponder, the transponder transmits a return signal.
The system of EP 0 409 016 A2 may also be used to locate objects
within a predetermined category of a plurality of objects. In this mode, the interrogator
transmits a predetermined category code to all of the transponders, and waits
for receipt of signals containing an identification code from each of the transponders
which have a stored category code which matches the predetermined category code
that was transmitted. To assist in avoiding collisions between their signals when
responding, each transponder waits after energization for a predetermined delay
period before responding, the length of the delay period being determined from
a code stored in the transponder's memory.
A disadvantage of the system of EP 0 409 016 A2, where used with
a large number of objects to be identified, is the need for the interrogator to
be cognizant of the transponder identification codes a priori; that is,
in advance of any communications made by the transponder. Another disadvantage
of the system, where used with a large number of objects, is the need to store
a predetermined time delay code in each transponder which is distinct from all
the other time delay codes of transponders belonging to the same category. In
general, this characteristic can be expected to increase the total time required
to locate a subset of the objects in a category because many of the possible time
delay codes of the category will be unused.
European Patent Application EP 0 494 114 A2 describes a supermarket
checkout system having an interrogator and a plurality of transponders, each transponder
being attached to an individual object to be identified. In that system, an interrogation
signal is first transmitted by a central interrogator to all of the transponders.
On receipt of the interrogation signal, each transponder transmits a response
identifying the particular transponder. Without waiting for further communication
from the interrogator, and at intervals which are determined randomly or pseudo-randomly
by circuitry within each transponder, each transponder repeats its identifying
response two more times in succession to increase the probability of successful
reception of its response by the interrogator.
That system further describes use of an interrogation signal which
can be modulated intermittently with the identification code of a particular transponder,
or with a code identifying a category of transponders, so as to cause to respond
only the particular transponder or category of transponders which have the same
identification code or category code stored in the particular transponder's memory.
Identification codes are transmitted and received digitally with error correction
and detection bits to improve the chances that the interrogator determines when
it has received a response correctly from a transponder.
Under the system of EP 0 494 114 A2, if the interrogator receives
a response from a transponder correctly, the interrogator signals the same to the
transponder by momentarily interrupting the interrogation signal. The transponder
is adapted to sense the interruption of the interrogation signal and stops transmitting
its response under such conditions.
A problem of the system of EP 0 494 114 A2, where used with multiple
objects to be identified, is that the response signals can be expected to always
collide during the first transmission because they are timed to be sent by all
transponders immediately on receipt of the interrogation signal. Thus, under this
system the response signals during the first transmission can generally be expected
to be received in error. Unnecessary delay and possibly undetected errors may result
from a system the operation of which is conditioned on detecting errors in the
responses of multiple transponders which are all intended to be transmitted simultaneously.
Another problem of this system, where used with large numbers of objects to be
identified, lies in the lack of an affirmatively granted transmission time for
each transponder to respond. Although each transponder is adapted to repeat its
response signal, the system is not designed to ensure that each transponder has
a time to transmit which is distinct from the times that all other transponders
are transmitting, as might be the case if the interrogator were to affirmatively
signal the individual transponder that a transmission time had been granted.
US-A-5 124699 discloses an electromagnetic detection and identification
system comprising an interrogator and a plurality of transponders each having a
unique code. In use each transponder is power-on by an interrogation field having
a frequency of 120kHz and thereafter the transponder is reset/initialised. The
interrogator then starts a selection procedure during which one of the transponders
will be read, by transmitting an interrogation field which varies in frequency
e.g. from 120kHz to 119kHz. When the transponders detect this variation within
a period of 256 ms, a random number is generated by a random number generator and
a counter begins to count from this random number. The transponder whose counter
first reaches the transition from value 127 to value 0 transmits a starting block.
When the interrogator detects the starting block, the interrogator field frequency
119kHz is changed to 120kHz. This causes transponders whose counters have not
yet reached the transition from value 127 to value 0 to switch into a passive mode
i.e. their counters are temporarily suspended. Thus, the 120kHz signal normally
results in all but one of the active transponders being inhibited and passes control
to the remaining selected transponder which is enabled to transmit its unique code
to the interrogator. When the interrogator has correctly received the unique code
from the selected transponder, the interrogation field frequency is again changed
to 119kHz. In response to this 119kHz frequency the selected transponder is switched
to a permanent passive mode and is excluded from further participation in any subsequent
selection procedure. The remaining transponders which were previously inhibited
are reactivated in response to the change in field frequency to 119kHz so that
they may participate in the next selection procedure. The selection procedure may
continue until the codes of all the transponders have been read.
SUMMARY OF THE INVENTION
Accordingly, it is an object of the present invention to provide
a method by which the identifying information from a large group of data transponders
can be received intelligibly and efficiently over a single communications medium
without having to address each data transponder individually or distinguish between
them prior to receiving the identifying information.
An object of an embodiment of the invention is to provide a method
by which a network controller can indicate to the transmitting data transponder
the- positive success of receiving communication.
A further object of the invention is to provide a method by which
the network controller can signal a specific requesting, but as yet unidentified,
data transponder that it is ready to receive the identifying information from the
particular data transponder.
It is another object of embodiments of the invention to provide an
RFID tag which can be attached conformably to articles of varying shapes and sizes
and which can be used in a supermarket checkout system to permit rapid checkout
of items contained in the supermarket checkout cart.
Still further, it is an object of embodiments of the invention to
provide an RFID tag which can be used in applications such as retail sales in general,
delivery control, inventory control, surveillance, telemetry, automatic toll collection
and monitoring of moving vehicles, and tracking, where the use of such tags can
be effectively employed.
These and other objects of the invention are obtained through a method
of communicating between a plurality of data transponders and a network controller
comprising the steps of: (a) initializing each data transponder in response to
receipt of a signal transmitted from the network controller to all the data transponders;
(b) generating a random number within each data transponder following initialisation;
(c) incrementing a counter in each data transponder in response to the passage
of successive time periods; (d) transmitting a "request to transmit" signal from
the data transponder to the network controller; and transmitting a data-containing
signal to the network controller if a first acknowledegment signal is received
from the network controller, which first acknowledgement signal does not specifically
identify the requesting transponder characterised by the step of comparing the
random number with the count maintained by the counter during successive time periods;
and in that the "request to transmit" signal is transmitted when the random number
in the data transponder equals the count maintained by the counter ; and in that
the data-containing signal is transmitted only if the first acknowledgement signal
is received from the network controller within a predetermined time period after
the data transponder transmits the "request to transmit" signal.
In the preferred embodiment, the network controller transmits a second
acknowledgement signal to all the data transponders if it determines that it has
received a data-containing signal successfully. If a particular data transponder
does not receive a second acknowledgement signal within a predetermined time period
after transmitting a data-containing signal, the particular data transponder transmits
again a data-containing signal. Alternatively, if the second acknowledgement signal
is received within the predetermined time period, communication is terminated.
Communication is also terminated if, after initialization, the data transponder
neither receives a first acknowledgement signal nor a second acknowledgement signal
from the network controller within a predetermined time period. Once the data-containing
signals are received successfully from all the data transponders, the network
controller broadcasts a third acknowledgement signal to all the data transponders.
Each data transponder is powered off and disabled from further communication if
it receives a third acknowledgement signal from the network controller after it
receives a second acknowledgement signal within the predetermined time period after
transmitting a data-containing signal as described above. Preferably, each data
transponder is also capable of being powered on by a signal from the network controller
which, in some applications, may contain the energy necessary for the data transponder's
operation. In other applications, the data transponder may be powered by a battery
or other power source.
In the preferred embodiment, if a data transponder does not receive
a first acknowledgement signal within a predetermined time period after transmitting
a request to transmit signal, that data transponder generates a new random number
and begins a new attempt to communicate with the network controller, that data
transponder interpreting the absence of the first acknowledgement signal as the
failure to obtain permission to transmit at that time from the network controller.
On the other hand, if a data transponder receives the first acknowledgement signal
prior to transmitting the request to transmit signal, incrementing of that
data transponder's counter is suspended temporarily. In this way, incrementing
of the counters in all of the data transponders is suspended except in the one
data transponder which has timely received a first acknowledgement signal from
the network controller in response to a request to transmit. Suspending the other
data transponder in this way assists the data transponder which has been given
permission to transmit its data-containing signal at a time when other data transponders
are not attempting to transmit. Afterwards, when the network controller receives
the data-containing signal and transmits the second acknowledgement signal, incrementing
of the counter in all the data transponders where it was suspended temporarily
is resumed again.
In another preferred embodiment, a type code is incorporated in a
command signal broadcast from the network controller prior to initiation of the
random delay in the data transponders. Each data transponder compares the type
code which is received from the network controller to a type code stored in the
data transponder's memory. If the two match in a particular data transponder, that
data transponder begins the communication sequence described above. In this way,
stored information can be retrieved from data transponders belonging only to a
certain group at a time, a function which may be advantageous in avoiding transmissions
of the data transponders from colliding. However, if the received type code and
the stored type code do not match in a particular data transponder, that data transponder
is powered off until again powered on by a subsequent power-on signal from the
Preferably, the data transponder is embodied in an RFID tag which:
- is attached conformably to articles;
- is adapted to communicate with the network controller through serial wireless
radio frequency (RF) transmission;
- is adapted to receive the energy necessary for its operation from a battery
or, alternatively, from a wireless RF transmission from the network controller;
- is adapted to transmit a data-containing signal which includes an article identification
code identifying the article to which it is attached, the article identification
code being read from memory located within the RFID tag.
The thus-adapted RFID tag embodiment of the data transponder is suitable
for use in a supermarket "Checkout system," such as that disclosed in U.S. Patent
No. 5,239,167. For such use, RFID tags are attached conformably to articles held
for sale in a supermarket. The purchaser loads up a shopping cart with the tagged
articles to be purchased, and moves the cart into an enclosure at the point of
sale which is appropriately shielded from the entrance or escape of radiofrequency
For the supermarket checkout system where RFID tags are not battery
powered, inside the shielded enclosure a network controller transmits a power on
signal having a high energy content to all the RFID tags attached to articles situated
within the enclosure. The RFID tags receive and store the energy in an energy
storage capacitor provided in each RFID tag. Once the accumulated voltage across
the capacitor exceeds a predetermined threshold, the individual RFID tag is ready
to receive communications from the network controller.
The network controller then transmits a begin signal to initiate
communication with all the RFID tags attached to articles situated within the enclosure.
Following the communication methods described above, an article identification
(ID) code is received by the network controller for each RFID tagged article located
within the shielded enclosure.
Preferably, a central supermarket computer attached to the network
controller is programmed to retrieve product information such as name, brand, size,
weight, etc. associated with the article ID code and to output the same to a display
located outside the shielded enclosure. The same product information can also be
output to a printer outside the enclosure for generation of a checkout invoice.
The enclosure may preferably be provided with a scale for registering the combined
weight of articles in the shopping cart, such as is described in U.S. Patent No.
5,239,167. The supermarket's computer can then determine if all the article ID
codes have been registered successfully by comparing the combined weight of the
articles in the cart to the computed weight of the articles which have been successfully
Once ID codes have been successfully received for all tagged articles,
the network controller transmits a final (third) acknowledgment signal to the
RFID tags, signaling the successful receipt of the article ID codes and permanently
disabling the RFID tags from attempting to communicate further. The computer then
permits the door of the shielded enclosure to be opened, and the cart containing
the articles is permitted to be removed by the customer. Near the exit of the
supermarket, a security check is made, which can be accomplished through interrogation
by another network controller. The security check need only determine if any RFID
tags in the customer's control are in the communication-enabled active state.
If such is the case, an alarm is sounded by the interrogating device to alert store
The thus-adapted RFID tag embodiment can also be used in conjunction
with a network controller for delivery control or inventory control purposes. In
such cases, the network controller preferably will selectively interrogate a group
of RFID tags attached to articles by incorporating a type code in a command signal
broadcast to all RFID tags within the range of the network controller transmitter.
A computer connected to the network controller can then be used, in the delivery
control application, for example, to automatically increment, or decrement the
present inventory tally, and update shipping records. Similarly, in the inventory
control application, the network controller can be used with a computer to generate
and update inventory records in real time, if desired.
An embodiment of the present invention will now be described, by
way of example, with reference to the accompanying drawings in which:
- FIG.1 is a perspective view of a supermarket checkout system employing an RFID
tag embodiment of the data transponder of the present invention;
- FIG.2 is a physical component diagram of the RFID tag;
- FIG.3 (3A) shows a flowchart of operations for the RFID tag embodiment; and
- FIG.4 (4A) is a block diagram of the RFID tag embodiment.
Referring to FIG. 1, the data transponder of the present invention
as embodied in RFID tags 10 is shown in use as part of a supermarket checkout system.
As shown, cart 22 holding randomly disposed articles 24 to which RFID tags 10 have
been conformably attached is moved into an RF shielded enclosure 16 via a conveyor
18 mounted below the enclosure. An external network controller 12 of the type generally
called a Point of Sale Terminal (POST), described below, is located inside the
enclosure. A display and printer 14 displays the item and/or price and prints
a record for the customer, as is well known in the art. In general, the POST transmits
a high energy content transmission signal to power up the RFID tags. The POST
then broadcasts a 'BEGIN' signal to all the RFID tags to stimulate the tags to
initiate a communication sequence which will ultimately result in the transmission
of article-identifying information (an article ID card) from each of the RFID
tags to the POST. As will become apparent from the description of the data terminal
operation to follow, a communication protocol determines the priority of and scheduling
of signals back and forth between the data terminal and the POST 12 such that
the POST 12 correctly receives the article-identifying information from each of
the data terminals.
Referring to FIGS. 2 and 3, the RFID tag remains in a normally powered
off condition 50 until powered on 52 and initialized 54 in response to receipt
of a high energy content transmission broadcast by the network controller to all
the RFID tags. When the RFID tag 10 receives the high energy content transmission
from the network controller on receiver antenna 40, a voltage is accumulated across
the terminals of energy storage capacitor 36. When the voltage exceeds a threshold
value, a RESET signal is generated within the RFID tag to initialize 54 the digital
circuitry of the integrated circuit (IC) 30 to predetermined states. The methods
and structures by which a high energy content transmission can be received by
an RFID tag will be appreciated by persons skilled in the art. After initialization,
the RFID tag remains idle until a BEGIN signal containing a command code is received
56 from the external network controller 12.
The command code is comprised of several portions which convey various
information to the RFID tag. A portion of the command code indicates to the RFID
tag that the command originated from a POST. Another portion of the command code
indicates whether article-identifying information is to be read from (READ mode)
or written to the RFID tag (WRITE mode). Still another portion of the command code
contains a type-identifier which is used by the individual RFID tag to determine
whether it belongs to the group of tags which are addressed by the particular command.
When READ mode is indicated 58 by the command code, the portion of
the command code which indicates the origin of the command (i.e. POST) is examined,
62, by circuitry within the particular RFID tag. If the command code is from a
POST a disable flag is raised, 64, which sets the particular RFID tag to be permanently
disabled at the conclusion of the communication with the POST. If the command is
not from a POST, the RFID tag determines if the type-identifier contained in the
command code matches, 60, the type-identifier stored in memory within the RFID
tag. If the command is from a POST, or if the received type-identifier matches,
60, the stored type-identifier in a particular RFID tag, the communication sequence
is allowed to proceed for that tag. If the command is not from a POST and the two
type-identifiers do not match, the RFID tag is powered off again, 50,and remains
in a powered off state until powered on again by receipt of a high energy content
transmission as described above.
With reference to FIG. 4, each RFID tag has a random number generator
266 and a digital counter 264 responsive to a local clock signal 8KCLK, 228, within
the RFID tag. Counter 264 is reset at the time of initialization. Those RFID tags
whose type-identifiers match the type-identifier in the command code received
from the network controller or which are in communication with a POST, proceed
with the following communication sequence. Referring back to FIG. 3, the random
number generator 66 delivers a random number to be held for comparison, 70, with
the output of the digital counter. When the random number has been generated,
the counter begins counting up, 68, towards the random number with the passage
of successive time periods. When the output of the counter and the random number
are EQUAL, 70, in a particular RFID tag, that RFID tag transmits, 72, a request
to transmit signal (RTT) to the external network controller 12 to request permission
at that time to begin transmitting its data. It will be appreciated that the request
to transmit signal, RTT, is a generic signal which can be the same for all requesting
RFID tags, as the RTT signal does not identify the particular requesting RFID tag.
The requesting RFID tag then waits, 74, for a first acknowledgment
signal from the external network controller. A delay count is initiated from the
time the request to transmit signal is transmitted by the RFID tag. If the first
acknowledgment signal is received by the requesting RFID tag before the delay
count reaches a value indicating that a predetermined time period has been exceeded,
76, i.e., before the delay counter "times out", then the requesting RFID tag determines
next whether a collision has occurred, 78. If the first acknowledgment signal is
not received by the requesting RFID tag within a predetermined time period after
transmission of the request to transmit signal, the delay counter times out, 76,
which indicates to the requesting RFID tag that the external network controller
did not receive the RFID tag's request to transmit signal correctly. Receipt of
the first acknowledgment signal by the RFID tag too early after transmission of
the request to transmit signal indicates to the RFID tag that a collision has
occurred, 78, in that the first acknowledgment signal had been an intended response
to some other RFID tag which had sent a request to transmit signal. In either
case, receipt of the first acknowledgment signal too early or too late with respect
to when the request to transmit signal is sent causes the requesting RFID tag to
generate, 66, a new random number, to reset and restart its counter 68, and to
reinitiate the above-described communication sequence with the network controller.
Receipt of the first acknowledgment signal by the RFID tag within
the predetermined time period permits the RFID tag to perform the next step of
the communication sequence, that of transmitting a data-containing signal: for
example, transmitting, 80, the article ID code to the external network controller.
After a particular RFID tag transmits, 80, its article ID code to the network controller,
the RFID tag then waits for receipt, 82, of a second acknowledgment signal from
the network controller which signifies that the ID code was received. A delay
count is initiated in the particular RFID tag from the time that the article ID
code is transmitted. If the second acknowledgment signal is received, 82, from
the network controller before the delay count reaches a value indicating the passage
of a predetermined time period and times out, 100, then the particular RFID tag
concludes that its transmission of the ID code has been successful.
The RFID tag then checks again to determine whether it is communicating
with a POST type network controller, as is the case if the disable flag has been
raised, 84, in the particular RFID tag. If the disable flag has not been raised,
the RFID tag powers itself off, 50, to await the next command from the network
controller. If the disable flag has been raised, the particular RFID tag then waits
for receipt, 86, of a third acknowledgment signal from the POST which signifies
that all RFID tags have been read successfully. Once the third acknowledgment is
received from the POST the RFID tag disables itself permanently from further communication,
Since the communication protocol relies on broadcast transmissions
between the RFID tags and the external network controller which do not identify
the particular RFID tag, a particular RFID tag may receive communications from
the network controller that were not intended for it. Should the first acknowledgment
signal be received by a particular RFID tag before that tag has transmitted, 98,
a request to transmit signal, but after the counter in the particular RFID tag
has begun incrementing, then that RFID tag concludes that the first acknowledgment
was intended for some other RFID tag which has been given permission to transmit.
The counter of the particular RFID tag is therefore halted temporarily, 96, to
permit the RFID tag having permission to transmit to complete transmission of
its data-containing signal to the external network controller. Only after the
second acknowledgment is sent by the external network controller and received,
94, by the halted RFID tag does the halted RFID tag resume incrementing, 68, its
counter again. It will be appreciated that the first acknowledgment signal will
generally cause all of the RFID tags to be temporarily halted except for the one
which currently has been given permission to transmit.
Occasionally, the delay count following the transmission of the data-containing
signal by a particular RFID tag will time out, 100, indicating to the particular
RFID tag that the network controller has not received the data-containing signal
correctly from the particular RFID tag. When this happens, the particular RFID
tag resends, 80, the data- containing signal to the network controller and again
waits for receipt of a second acknowledgment signal.
Provision is made for the possibility that a problem between the
network controller on the particular RFID tag will prevent the RFID tag from receiving
acknowledgments from the network controller after the particular tag has been powered
up. In tags which are powered by reception of external RF transmissions, the tags
will revert to a powered off condition automatically within a predetermined time
period after termination of the external RF transmissions. In such tags, the predetermined
time period is determined by the time for the voltage across the terminals of
energy storage capacitor 36 to decay below the threshold required to keep the tag
operating. In applications where the RFID tags are powered by batteries instead
of external RF transmissions, the RFID tag is provided with an additional "watchdog
timer" to power off the tag in case of communication malfunction. The watchdog
timer located within the RFID tag (not shown) terminates the communication sequence
for that tag and powers off the tag if no acknowledgments are received within a
longer predetermined time period than the predetermined delay timeout periods
which follow the first and second acknowledgments.
A special type of network controller called an inventory terminal
(INVT) can be used to receive and record article-identifying information for the
number and type of articles which are displayed or stored. All aspects of the
communication sequence remain the same with the INVT as for the POST, except that
network controller commands originating from the INVT, e.g. the BEGIN signal and
acknowledgment signals, do not contain a bit which identifies the command as having
come from a POST, such that a disable flag will be raised in the RFID tag.
An additional type of network controller called a programmer terminal
(PROGRAMMER) can be used to send and store article-identifying information to
the RFID tags after the time of their manufacture. After the RFID tags are powered
up, 52, and initialized, 54, by the high energy content transmission, the initial
command sent by the PROGRAMMER contains a bit which indicates to the RFID tag
that WRITE mode is selected, 58. The initial command also contains article-identifying
information, i.e. a type-identifier and a specific article ID code. Once the RFID
tag recognizes that WRITE mode is selected, 58, the article-identifying information
from the command is loaded, 90, into the RFID tag's shift register, and then transferred
and stored, 92, in memory located within the RFID tag.
The structure which implements the function of the RFID tag is described
with reference to FIG. 4. Incoming command and acknowledgment signals, 234, which
have been amplitude modulated with a binary signal code and transmitted as a wireless
radio frequency (RF) signal by the network controller are received at the receiver
270. The receiver demodulates the incoming signal and passes a raw binary signal
RxD, 272, to the decoder 274. The decoder recovers the received clock RxC, 220,
associated with the signal and, as will be further described, regenerates the
binary signal to prevent noise in the raw signal RxD, 272, from being passed to
other elements within the RFID tag. The recovered clock signal RxC, 220, is used
to transfer the regenerated binary signal SI, 210, into the shift register 244.
Shift register 244 can be used to shift a serial binary signal, such as a request
to transmit signal, RTT, or an article ID code, into encoder 236 serially to be
amplitude modulated and transmitted as a wireless RF signal by the transmitter
238 to the external network controller 12. In another mode, the shift register
244 can be used to output command code data 250 from the incoming signal on a parallel
output interface (not pictured as a separate element) into a type comparator 282
which compares the type-identifier in the command code 254 with a type-identifier
256 retrieved from a programmable read only memory ("PROM") 252 within the RFID
tag. The parallel output interface of the shift register is also used to output
individual signal lines or pairs HOLD, 246, and ACK [2,1], 248, to the control
The control element 200 controls the sequence of operations and data
flow within the RFID tag in response to the presence of internal signals within
the RFID tag at a given point in time. These internal signals correspond to the
RFID tag's logical "state." It will be appreciated that the control element 200
can be constructed by a read only memory having as inputs the internal signals
within the RFID tag and which address, in their various combinations, locations
within the memory which contain outputs of the control element which correspond
to the next "state" of the RFID tag.
Random number RN, 268, is delivered by a random number generator
266 to a comparator 286 for comparison with the output of a digital counter 264.
When the two match, the RNMATCH output 310 of the comparator 286 is raised to
signal the control element 200 to send the request to transmit signal. A timeout
detector 289 uses the six lowest order bits [5..0] of the counter output 312 to
determine when any of three particular delay counts: TIMEOUT 1, 304, TIMEOUT 2,
306, and CDET, 308,have been met or exceeded.
The RFID tag also includes a multiplexer MUX, 260, used to channel
a received clock signal RxC, 220, or a transmitted clock signal TxC, 218, to the
shift register 244 as SCLK, 212, under the control of clock select signal CLKSEL,
222, for use in inputting or outputting a binary signal to the shift register from
the decoder or encoder, respectively. The RFID tag runs on a CLOCK, 294, provided
by a 1 Mhz local oscillator 292, the output of which is divided 16 times to form
the clock inputs 3TC (204) and 16RC (276) to encoder 240 and decoder 274 respectively.
The 3TC clock input is further divided eight times to form a clock signal running
at 7.8125 kHz ("8KCLK"), 228, which is the clock signal used to increment the
digital counter 264.
The decoder regenerates the raw binary signal RxD, 272, in the following
manner. By use of clock input 16RC (276) the incoming raw binary signal RxD, 272,
is sampled at the rate of 16 times per data bit. The samples are recorded within
the decoder as low or high voltage samples. An edge detector (not shown) within
the decoder is used to recover the clock signal RxC, 278, from the raw binary signal
RxD, 272. Clock signal RxC, 278, is further used to determine when recording of
the voltage samples for the set of 16 samples for each data bit should begin. If
the number of high voltage samples in each set of 16 exceeds the number of low
voltage samples,then the incoming data bit is regenerated as a binary one. Otherwise,
if the number of low voltage samples exceeds the number of high voltage samples,
the incoming data bit is regenerated as a binary zero. The stream of regenerated
binary ones and zeros are output from the decoder as SI, 210, which is the serial
input to the shift register 244. Decoder input RMOD, 232, is a predetermined number
which indicates the length of the incoming signal to be decoded. The value of RMOD
may be programmed to permit decoding of signals of other lengths, but the programming
must be done prior to the time of signal reception. When the decoder determines
that it has finished decoding the signal of the appropriate length, it signals
RxRDY, 312, to the control element 200 to enable further processing to proceed.
A typical communication begins as follows. The external network controller
12 such as a POST sends a high energy content transmission to the RFID tag 10.
As described above, when the voltage across the energy storage capacitor 36 (FIG.
2) exceeds a minimum threshold value, the RFID tag is powered on and initialized.
Exceeding the threshold raises the RESET, 290, input to the control element 200,
thereby causing the control element 200 to reset elements under its control, e.g.
resetting the counter 264 by means of the CRST line, permitting the local oscillator
292 to begin running, and resetting all internal signal lines within the RFID tag.
The RFID tag now waits for receipt of a 'BEGIN' command from the network controller.
When the BEGIN command is received at the receiver antenna 40 (FIG 2.), the receiver
demodulates the incoming amplitude-modulated RF signal, and outputs a raw binary
signal RxD, 272, to the decoder. As described above, the decoder 274 outputs a
regenerated binary signal containing a command code as serial input SI, 210, to
the shift register 244. When the decoder 274 has finished regenerating the binary
signal, it raises signal line RxRDY, 312, to the control element 200. Control
element 200 then activates the shift register 244 to output the received command
code over its parallel interface 250.
A portion of the received command code is held as the received type-identifier
input 254 to the type comparator 282. The received type-identifier is compared
to a type-identifier stored in the RFID tags which has been retrieved into LATCH
258 for comparison by the following procedure. Upon receipt of the RESET signal
290 by the control element 200 during initialization of the RFID tag, the control
element has raised PROM chip select line /CS, 214, and retrieve line /RECALL,
215, to retrieve the type-identifier from its stored location in the PROM, 252.
The retrieved type-identifier is stored into and held in a LATCH 258 by the signal
TLATCH, 217, from the control element 200 until it is compared to the received
type-identifier by the type comparator 282. When the two type-identifiers match,
the TMATCH line 300 is raised. The control element 200 is programmed to react
to an active TMATCH line 300 by raising the CRST line 224 to reset and start the
counter 264 and by raising the SET RN line 230 to cause the random number to be
The random number RN, 268, is delivered from the random number generator
266 to a latch (not shown) located within the comparator circuit 286. With each
clock cycle, the comparator compares the 13 highest order bits [14..2] of the
count 313, to determine if the count and the latched random number match. When
the two match, the RNMATCH line 310 is raised to signal the control element 200
that the request to transmit (RTT) signal can be sent. In response thereto, the
control element 200 signals the encoder 240 through TMOD 202 to output a generic,
"hard-coded" RTT stored within the encoder to the transmitter 238 through the
line TxD 237. This TxD signal,237, is then amplitude modulated onto an RF carrier
by the transmitter 238 and transmitted on the transmitter antenna 38 (FIG. 2)
as the output signal 280.
When the encoder has completed signal transmission, the TDONE line
288 is raised to notify the control element 200 of the same. If the RTT signal
that was transmitted results in the network controller granting permission to transmit
to the requesting RFID tag, the next command to be received by the RFID tag will
be a first acknowledgment. The first acknowledgment is received as a serial binary
signal as described above for the BEGIN command. When the first acknowledgment
signal has been shifted completely into the shift register 244, the ACK1 line,
248, of the shift register's parallel output interface becomes active. The control
element 200 acts upon the ACK1, 248, input in accordance with the state of other
inputs, RNMATCH, 310, and CDET, 308, to determine the appropriate action to take
next. For example, if ACK1, 248, and RNMATCH, 310, are active but CDET, 308, is
not, the control element 200 concludes that it has received the first acknowledgment
signal at the correct time. The control element then raises /RECALL, 216, with
the chip select lines /CS 214 to retrieve and output the article ID code from the
PROM, 252, to the shift register 244 to be encoded and transmitted externally.
However, if ACK1, 248, is received while RNMATCH, 310, is inactive, the start-suspend
counter line SS 226 is raised to stop the counter 264 from incrementing. Then,
when ACK2, 248, is raised on receipt of the next second acknowledgment sent by
the network controller, start-suspend counter line SS 226 is lowered, thereby
permitting the counter to resume incrementing.
If ACK1, 248, RNMATCH, 310 and CDET, 308, are all active at the same
time, but TIMEOUT 1 is not, then the first acknowledgment has been received too
early after the sending of the request to transmit, which indicates that a collision
has occurred. It will be appreciated that CDET, 308,is held active from the time
that the request to transmit or the article ID code is transmitted (as is indicated
to the control element when the signal TDONE, 288, is raised) until sufficient
time has elapsed to permit a response to the transmission, e.g. a first acknowledgment,
to have been received.
The delay count following the sending of the request to transmit
signal is maintained and checked in the following manner. With every clock cycle,
the timeout detector 284 tests the six lowest order bits [5..0] of the counter
output 313 to determine the elapsed time. When the delay count following the sending
of the request to transmit signal is exceeded, TIMEOUT 1, 304, is raised. If ACK1,
248, is not active by that time to indicate that the first acknowledgment has
been received, the control element concludes that it has not been granted permission
to transmit at that time. In response to either CDET, 308, or TIMEOUT 1, 304,
being raised, the control element 200 raises SET RN, 230, and raises CRST, 224,
momentarily to signal the random number generator 266 to generate a new random
number and to restart the counter 264, respectively, in order to begin the communication
sequence again. Similarly, after the RFID tag transmits the data-containing signal,
the control element 200 resets the counter 264 by raising CRST, 224.
If the article-identifying information is received from the data-containing
signal correctly by the network controller, the next command to be received by
the RFID tag will be a second acknowledgment signal from the network controller.
Receipt of the second acknowledgment will be indicated as ACK2, 248, to the control
element 200. In response to ACK2, 248, the control element 200 determines if the
disable flag has been raised 84 (Fig. 3). This disable flag is a bit stored in
a bit-storing device, e.g. a latch or a "flip-flop" (not shown) located in the
control element 200. The status of the disable flag is dependent on whether the
command code sent by the network controller at the beginning of the communication
session contains a bit that indicates that the network controller is a POST. If
the disable flag has not been raised, the communication is terminated upon receipt
of the second acknowledgment, and the RFID tag waits for the network controller
to power it down, i.e. to discontinue sending the high energy content RF signal.
If the disable flag has not been raised, thus indicating that the network controller
is a POST, the RFID tag remains active until receipt of a third acknowledgment
After sending the data-containing signal, the RFID tag again maintains
and checks a delay count while waiting for receipt of a second acknowledgment
signal. If the second acknowledgment signal has not been received before the delay
count is exceeded, TIMEOUT 2, 306, is raised. The control element 200 then raises
the RECALL, 216, and /CS, 214, inputs to the PROM 252 to begin transmitting the
data-containing signal again to the network controller.
Once the article-identifying information has been received correctly
by the network controller for all the RFID tags, the network controller sends a
third acknowledgment which causes line ACK3 302 in the RFID tag to become active.
The ACK3 signal, 302, causes the RFID tag to become disabled from further communication
with the network controller. For tags used in a supermarket checkout system, permanent
disablement may be accomplished through a switch (not shown) located in the PROM
252 which is operable by the control element 200 upon receipt of ACK3, 302, to
short circuit the terminals of energy storage capacitor 36 (Fig. 2).
For tags used in applications where reusability is desired, such
as on clothing sold in retail department stores, an effective disabling mechanism
can be provided which is reversible at a later time by a PROGRAMMER type network
controller. For such applications, the RFID tag can be provided with a switch
(not shown), again located in the PROM, 252, which is operable by the control element
200 upon receipt of ACK3, 302, to disable transmissions from the RFID tag, but
which does not affect reception by the tag of incoming commands. Such a switch
may be provided by a latch in the PROM 252 which, when in the 'disable' state,
causes the control element 200 or to hold start-suspend counter line SS, 228, in
the suspend state, or, hold encoder control inputs TMOD, 202, in a transmission-disabled
condition. The PROGRAMMER network controller can then be used at a later time
to "flip" the switch of the disabled tag, i.e. to reset the latch in the PROM 252
to the fully enabled state, to permit the RFID tag to transmit responses to the
network controller again.
It will be appreciated that the data terminal of the present invention
can be put to many uses besides that of an RFID tag attached to articles for use
in a supermarket checkout system. By example and without intending to be limiting,
among the uses to which the data terminal is adapted are: airwave or network data
communications in general; telemetry; retail sales in general; delivery control;
inventory control; automatic toll collection, speed checking or identification
of vehicles on a highway; identification of and the locating of persons, animals,
or component parts or stock items and the like which occupy, enter, or leave a
defined area. Furthermore, these and other modifications of the invention will
be readily apparent to those skilled in the art in light of the description of
the preferred embodiments set forth herein without departing from the scope of
the invention which is set forth in the appended claims.