GB2317483A - Transaction system - Google Patents

Abstract

A transaction system comprises a plurality of contactless smart cards and a reader. In conducting a transaction with one of the smart cards the reader sends a signal to the cards and each card sends a return signal after a relatively short or a relatively long period of time, as determined by the first digit of an identifier of the card. Those cards responding after a relatively long period of time are excluded from further communication with the reader. Thus certain cards are excluded until only the card having the highest or lowest identifier remains in communication with the reader. A transaction between the reader and that card can then occur.

Description

TRANSACTION SYSTEM This invention relates to a transaction system in which a portable token, for example a card, is used in conjunction with a device, often termed a reader or terminal, to perform a transaction of some kind. The invention is particularly, but not exclusively, related to contactless smart cards.

Contactless smart cards work on, or close to, a reader which provides power. This power is supplied via a RF (radio frequency) induction field and is akin to the reader being a primary coil of a transformer and the card a secondary coil. Both the reader and card typically have a single coil aerial. As technology improves the power consumption of smart cards is being reduced. This means that cards can work further away from the reader, that is the volume or field of operation of the reader is larger. If the reader maintains a fixed power output, it means more cards can be powered in the field of operation at the same time. Therefore the probability of there being more than one card in the field of operation of the reader is greater.

There can be difficulties in the reader communicating with a specific card in that signals from the reader may be picked up by more than one card. These difficulties could be alleviated or solved by reducing the power output of the reader which would allow only one card to work within its field of operation. However, this would also reduce the field of operation of the reader which is a disadvantage from an operational point of view.

A system could be configured such that cards could enter the field of operation of the reader one at a time. This would allow the reader to interrogate a single card for its identity. At any stage the reader could use a list of stored identification numbers to enquire whether a particular card was still in the field of operation. This would be inefficient because it would reduce the amount of time available for the transaction to take place. The reader could then ask if there was a new card in the field of operation after asking all known cards to be quiet. This would also be inefficient because the reader would have to keep repeating this task quickly enough so as not to allow two cards to enter the field of operation at the same time. Furthermore it is undesirable to request that the user of a card remove his card from the field of operation and then reenter the card simply because the reader is confused about the cards present in its field of operation. This is because the user of the card may not carry out the request either by accident or intentionally. For example, a card could be left intentionally in the field of operation to defraud the system or to prevent the system from working.

According to a first aspect the invention provides a method of selecting one token from a plurality of tokens, the tokens having individual identifiers in which a reader sends an outgoing signal to the tokens and at least one of the tokens sends a return signal after a time period which depends on the value of part of the identifier.

Preferably the reader has a field of operation and the tokens are present in the field of operation.

Preferably the identifiers are identity numbers of the tokens. Conveniently, the identifier is a string of digits, for example, a string of "O"s and "l"s.

Preferably a return signal is sent at the same time by all tokens having "1" as a particular part of their identifier. Preferably a return signal is sent at the same time by all tokens having "0" as a particular part of their identifier. The return signals for a "1" digit and a "0"digit may occur after different time periods following sending of the outgoing signal.

Preferably the tokens send a first return signal corresponding to a first digit in their identifiers in response to the reader sending an initiation signal.

Preferably once the reader has received at least one return signal it sends an interrupt signal to the tokens to prevent one or more tokens which have not sent a return signal from sending return signals.

Preferably the reader sends a plurality of outgoing signals to the tokens to determine for at least one token a first digit of its identifier, through all intermediate digits of its identifier to a last digit of its identifier. Alternatively the reader determines a selection of the digits of the identifier. It may not need to determine them all.

Conveniently the method is a method of determining the complete identifier of at least one token. Preferably the method is a method of determining the complete identifier of more than one token. It may also sort the determined identifiers into ascending or descending order.

Preferably the tokens operate in a contactless manner. They may be contactless smart cards.

Preferably the tokens contain hardware logic capable of implementing the method.

Alternatively the tokens contain software capable of implementing the method.

The tokens may be powered by an electrical power source such as an RF field or they may be powered by an internal power source such as a battery.

According to a second aspect the invention provides a transaction system adapted to carry out the method according to the first aspect of the invention.

An embodiment of the invention will now be described, by way of example only, with reference to the accompanying Figure which shows communication signals between a reader and a number of cards.

In a transaction system operating with a plurality of smart cards, it is possible that two or more cards will enter the field of operation of a reader at the same time. This can occur if a person is carrying two contactless cards of the same type in his wallet and waves his wallet near the reader. In this event either of the cards could power-up first.

Therefore if there are two or more cards in the field of operation of the reader at the same time the reader needs to identify the cards in order to transact only with one card at one time. It should be noted that once a card has been identified then it is a simple matter to address commands only to that card, for example by appending the identification number of the card to the command in question.

A method according to the invention is set out in the following and described with reference to the Figure. The method relates to three contactless smart cards A, B and C each having a unique identification number. In the Figure communication signals issued by the reader and each of the cards are shown to the right. Signals are issued during time slots 10, 12, 14 and 16. Time slot 10 occurs first and time slot 16 occurs last.

In time slot 10, the reader first sends a command (a) to all the cards in the field of operation to instruct them to go into an identify procedure. The command (a) will be a certain sequence of bits. Alternatively, it may be a single bit. As all of the cards are in the field of operation of the reader, they will all receive this command at substantially the same time.

In time slot 12, the cards each inspect the first bit of their identification number. Those cards having a "1" as a first bit send a signal (b) to the reader after a time interval tl.

Those cards having a "0" as a first bit send a signal (d) to the reader after a time interval tO. The difference in the time intervals tl and to is sufficient to allow the reader to send a signal (c) between signals (b) and (d) being sent. The time intervals are directly calculated from a clock generated on each card, which is dependent upon the frequency of the RF field. Each card has the same clock frequency and so can respond after substantially similar time intervals tl and tO. Alternatively the card may have an accurate internal time source such as one which is crystal based.

If a card receives a signal (c) from the reader before it has sent its pulse (b) or (d), it stops transmitting and does not communicate, remaining silent unless it is spoken to specifically with the correct card identification number or unless it receives the command to start the identify procedure again. In the example above, cards A and B have "1" as their first bit. Card C has "0" as its first bit so its signal (d) does not actually occur and therefore its occurrence is only shown as a broken line. Thus, card C takes no further part in this particular example of the method because it detected the reader signal (c) before it was due to transmit its own signal (d). The reader signal (c) was in response to signals (b) from cards A and B.

Cards A and B have both replied and so the reader knows there are still one or more cards ready to have further bits identified. In this case both the second bits are "0" and so in time slot 14 the reader receives signals only after tO. It then sends another signal (e) which leads to identification of the third digit of each card in time slot 16. Since there is a difference between the respective third bits of cards A and B, card A does not send its response and remains silent during the rest of the identify procedure. Card B is the only card left responding and a complete identification of card B is straightforward.

The method is terminated if no response is received after a time interval to because this means that the reader has reached the end of the bit sequence or the card or cards which continue to be involved in the identify procedure have been removed from the field of operation. In an embodiment in which the reader knows how many digits are present in an identification number it can terminate the identify procedure once signals representing all of the digits have been received from a card.

The method works with any number of cards as long as there is sufficient power from the reader to energise them all.

In the example shown the logic is based upon the transmission of "0" bits, that is a pulse which is a power reduction below a datum power level. Inverse logic may be used in which "1" bits are used instead of "0" bits.

An advantage of the method is that not only is one card left activated at the end of the procedure but that the reader can determine the identification number of that card. This can be done by recording the time intervals between the signals which are sent by the card in response to signals sent by the reader.

In the example the card with the highest binary number (card B) completes the identify procedure. If the procedure were to be repeated with card B having been deactivated (which can be done because its identification number can be determined and the appropriate command sent to it), then the card with the next highest binary number (card A) would complete the procedure. Therefore the method can be used to sort a number of cards into either descending order, if tl < tO, or ascending order, if tl > tO.

In the method each card in the field of operation may be identified within a fixed time.

An upper limit of this time is related to the number of bits in the identification number multiplied by the maximum time between reader pulses. An example based on bit times follows. If a signal is equal to one bit time, tl is also equal to one bit time and tO is equal to three bit times, then the maximum time to identify one bit would be five bit times. If there are 32 bits in the card identification number (equivalent to 4000 000 000 unique combinations or card identities) then the overall identify sequence per card would take 160 bit times or 17mS at 9600 baud.

It should be noted that the signals sent by the reader and the card may not be single pulses (although this is simplest) but may be a sequence of pulses so as to reduce the possibility of noise interfering with operation of the method.

This invention uses the fact that most modern smart cards have a unique identification number stored in their memory. This is for security reasons so that all cards can be audited and traced whilst in the manufacturing phase of their life and when they are issued to the end customer. The identification number can be used to blacklist a stolen card or to allow the card to be mutually authenticated with a reader. The size of the identification number is large enough to enable all cards to have a unique number. The number normally cannot be changed because the memory where it is stored is made to be written on once (typically EPROM or EEPROM with security logic).

All the functionality described could be implemented either as hardware logic in an ASIC or as software in a smart card microcontroller. Both these devices can contain memory to store the unique identification number.

Claims (15)

1. A method of selecting one token from a plurality of tokens, the tokens having individual identifiers in which a reader sends an outgoing signal to the tokens and at least one of the tokens sends a return signal after a time period which depends on the value of part of the identifier.

2. A method according to claim 1 in which the reader has a field of operation and the tokens are present in the field of operation.

3. A method according to claim 1 or claim 2 in which the identifiers are identity numbers of the tokens.

4. A method according to any preceding claim in which the identifiers are a string of"O"s and "l"s.

5. A method according to claim 4 in which a return signal is sent at the same time by all tokens having "1" as a particular part of their identifier.

6. A method according to claim 4 or claim 5 in which a return signal is sent at the same time by all tokens having "0" as a particular part of their identifier.

7. A method according to claim 5 or claim 6 in which the return signals for a "1" digit and a "O"digit occur after different time periods following sending of the outgoing signal.

8. A method according to any preceding claim in which the tokens send a first return signal corresponding to a first digit in their identifiers in response to the reader sending an initiation signal.

9. A method according to any preceding claim in which once the reader has received at least one return signal it sends an interrupt signal to the tokens to prevent one or more tokens which have not sent a return signal from sending return signals.

10. A method according to any preceding claim which determines the complete identifier of at least one token.

11. A method according to any preceding claim in which the tokens communicate with the reader in a contactless manner.

12. A method according to claim 11 in which the tokens are contactless smart cards.

13. A method substantially as described herein with reference to the Figure.

14. A transaction system comprising a plurality of tokens and a reader adapted to carry the method according to any preceding claim.

15. A transaction system comprising a plurality of tokens and a reader substantially as described herein with reference to the Figure.