MXPA99008387A - Improved method for collision management in a non-contact data exchange system - Google Patents

Improved method for collision management in a non-contact data exchange system

Info

Publication number
MXPA99008387A
MXPA99008387A MXPA/A/1999/008387A MX9908387A MXPA99008387A MX PA99008387 A MXPA99008387 A MX PA99008387A MX 9908387 A MX9908387 A MX 9908387A MX PA99008387 A MXPA99008387 A MX PA99008387A
Authority
MX
Mexico
Prior art keywords
response
terminal
message
messages
portable
Prior art date
Application number
MXPA/A/1999/008387A
Other languages
Spanish (es)
Inventor
Grieu Francois
Original Assignee
Innovatron Electronique
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Innovatron Electronique filed Critical Innovatron Electronique
Publication of MXPA99008387A publication Critical patent/MXPA99008387A/en

Links

Abstract

The invention relates to a method for managing collisions in a data exchange system between portable objects and at least one send-receive terminal capable of cooperating with a plurality of said portable objects which are simultaneously present in the communications field of the terminal. The method consists of the following steps:a) transmission of call messages by the terminal to the portable objects;b) conditional transmission from the portable object to the terminal of a response message to the call message with a probability of possibly less than 100%, where the response message contains an identification code specific to the portable object;c) when the terminal receives a response from an identified portable object and if said response does not collide with the response sent by another portable object, a specific data communication link is established between the terminal and identified portable object, data are exchanged and the process returns to step a) for a further iteration;and d) if there is no reception without collision, return to step a) for a further iteration. According to the invention, terminal transmissions comprise two types of call messages:query messages, which initialise the probability of response by the portable objects, and marker messages sent out in succession after query messages. The marker messages are able to instruct the portable objects to adjust the probability value on reception of a marker message in the absence of a previous response to a call message.

Description

PROCEDURE FOR COLLISION MANAGEMENT IN A CONTACT-FREE DATA EXCHANGE SYSTEM FIELD OF THE INVENTION The invention relates to non-contact communication between a portable object and a terminal. BACKGROUND OF THE INVENTION The contactless data exchange is well known; Among the applications of this technique, we find, among others, access control and electronic toll collection, for example for access and tolling of collective means of transport. In this last example, each user is provided with a portable object of the type "contactless card" or "contactless plate", which is an object capable of exchanging information with a terminal or fixed station when the portable object is brought closer to the latter. , so as to allow non-galvanic mutual coupling ("terminal" means the transmitting / receiving terminal capable of cooperating with portable objects). SUMMARY OF THE INVENTION The invention points to the particular situation in which multiple portable objects can be simultaneously present in the field of action of the terminal, and where there is no physical medium, for example the insertion in a slot, to isolate the arrival of a single object within this field. The object is to exchange data with a chosen portable object. For this, a unique or quasi-unique identification is affected to each portable object. If a portable object is within the range of the terminal, it transmits its identification to the terminal. This transmission can be disrupted if several portable objects respond at the same time. On the contrary, if the terminal receives an identification correctly, the terminal returns a message that includes the identification, which will only be processed by the cards that receive it if the identification is identical to yours, triggering in this case the exchange phase of the data between this specific card and the terminal. To the extent that the signals emitted by the different portable objects can overlap each other over time, the terminal must be provided with an "anti-collision" mechanism that allows detecting these situations and managing them, causing a new emission of the signals by the parties. portable objects. There are currently several types of anti-collision procedures, among which the following can be identified: - "Slotted Alona", in which each of the portable objects responds to a general call issued by the terminal in one of several previously defined time intervals (the "slots") randomly chosen by the cards, and the terminal selects a single sequence for dialogue between them (see FR-A-2,666,187). The probability of response to the call by the card is fixed and equal to 100%. The selection is achieved if the response of a card is alone in a slot; - "Probabilistic", in which the probability of response of each card to a general call after a collision of responses received in the terminal, is less than 100% and is reduced until a single response is obtained (comp. French patent 97 02501 of March 3, 1997 for "Proceedings of management des collisions dans un systéme d 'échange de données sans contact"). The general call may include a simple collision warning or not at the time of the previous general call, in which case the card defines the increase or decrease in the probability of its response, as in EP-A-0,702,324, but preferably, as in the French application 97 02501, the general call includes a parameter defined by the terminal as a function of the collision history and it is the value of this parameter that defines the probability of response of the card. The present invention starts from the verification that some aspects of the two systems can be combined in a useful way to obtain an improved compromise between the different factors that influence the performances of the system., such as the maximum number of portable objects simultaneously present that can be processed, allowing the duration of the general call cycle, the response of all portable objects eventually present and the duration of the general call cycle in the presence of a single portable object ( according to the most frequent case). Furthermore, in the presence of the coexistence of populations of portable objects of various types, there is an interest in a "hybrid" system in which the same terminal can operate interchangeably with portable objects of different types, and even with portable objects of type mixed; preferably a hybrid portable object would also work with terminals designed by only one of the types of anti-collision procedure. More precisely, the method of the invention is characterized by comprising the following successive steps: a) emission by the terminal, towards the portable objects, of call messages; b) conditional emission by the portable object, towards the terminal, of a response message to the general call with a probability that may be less than 100%, this response containing an identification of the portable object itself; c) with the reception by the terminal of a response of an identified portable object, and in the absence of a collision of this response with a response emitted by another portable object, establishment of a specific data communication connection between the terminal and the identified portable object and continuation of a data exchange, and return to stage a) for a new iteration; and d) in the absence of collision-free reception, return to stage a) for a new iteration. According to the invention, the emissions by the terminal comprise two types of call messages: interrogation messages, which initialize the value of the probability of response by portable objects, and marker messages issued successively after the messages of interrogation, the appropriate marker messages being to order the portable objects to eventually modulate in the absence of a previous response to a call message, this probability value against receiving a marker message. According to the various advantageous embodiments: - the portable objects respond selectively in the timing intervals following the reception of the marker messages, the choice between these timing intervals being determined in a probabilistic manner by means of a random or pseudorandom draw made by the user. portable object against receiving an interrogation message, the terminal can also reissue an interrogation message against the detection of a collision; the modulation by the portable object, in the absence of a previous response to a call message, of the probability value against receiving a marker message may be a function of the value of a modified index against the reception of a marker message, especially of an increased index on each reception of a new marker message and reinitialized against receipt of a subsequent interrogation message; - the terminal reduces the time interval separating two successive call messages in case of absence of response of a portable object identified in a predetermined period, - the issuance of a response by the portable object to the call messages is susceptible to be inhibited by the reception of a specific message issued by the terminal and containing an identification corresponding to that of the portable object, the inhibition being lifted with the initialization of the portable object and / or with the receipt of a message to that effect.
DETAILED DESCRIPTION OF THE INVENTION Next, a detailed description of the method of the invention will be consigned, with different variants and forms of embodiment. In principle, the probabilistic system of the French application 97 02501 will be resumed and the notation used will also be indicated. On the other hand, being portable objects frequently in the form of cards, they will be called that way for convenience, without it having a limiting character. - Repeatedly the reader issues the general call, an interrogation message, (hereinafter called "Query") that contains a parameter P, then puts itself in a position to receive the eventual response of a card. Each card that detects Query chooses to answer or not with a probability that is a function of P and a random number X, extracted from the card. The response of the card (or "Answer to Query" ATQ) comprises at least one unique identification on each card, and optionally other identification elements. - At a later stage the reader issues at least one application message containing this identification, to which only the card carrying this identification will respond (ie the cards comprise, for that or those subsequent messages, a stage of comparison with the identification they contain). - Optionally, an application message is accompanied by a shorter identification chosen by the reader. In the application messages following the acceptance of the card for that first application message, the short identification will be used to replace the initial unique identification (in order to decrease the volume of data exchanged). The unique identification can be a predetermined serial number, or a random number of four octets, for example. This second possibility has the advantage that it does not allow the card to be recognized from one use to the other, preventing its use by the unique identification mechanism. For a random identification, it will be chosen for each card in the initialization or the reception of Query and can be used after the choice of answers, in a procedure of the type of the variant of the "Slotted Aloha" mentioned above. The present invention proposes to determine the probability of response of the card to an interrogation message of the reader by means of a calculation of the card that makes this probability evolve, according to whether, at the time of the previous interrogations, the card has already responded or not. This assumes the existence of a state variable, on the card that reflects your past decisions (designated I below). The Query interrogation message preferably contains a probability modulation parameter P, such as the system of the French application 97 02501, although this is optional. In a first embodiment example, particularly simple: a) the reader periodically issues completely identical interrogations without parameter; b) at the time of tensioning or entering the field, the card initializes an internal variable 1 = 1; c) upon receipt of any call from the reader, the card randomly chooses an integer in the interval [1 ... 1]; if the chosen number is 1, the card responds to the interrogation of the reader by issuing its identification, and reinitializes I = 8; in the contrary case the card decreases I of 1 and does not emit its identification. It is observed that the probability of response of the card to the call of the reader is 1 at the end of stage b), all the cards respond to the first call, then the probability is 1/8 at the end of stage c ). In subsequent calls, the probability is returned to 1/8 if the card has responded; in the opposite case, as long as the card remains unanswered, the probability of response successively evolves 1/7, 1/6, 1/5, 1/4, 1/3, 1/2, 1. There are a maximum of eight interrogations between two responses of the card, which guarantees that in eight interrogations at most, the reader will know the identification of the card, from the moment in that the card is close enough so that there is no communication error and no collision occurs. A variant (Example B), gives the same behavior of the card but simplifies the random draws and reduces its quantity: c) with the receipt of a call from the reader, the card examines I; if I = 1, the card responds to the interrogation of the reader by issuing its identification and reinitializes I in an integer randomly chosen in the interval [1 .... 8]; otherwise, the card decreases I by 1 and does not emit its identification. An improvement, applicable more generally, is that the reader can define the parameters of the probability of response of the card; for example, it joins each interrogation a parameter P used in step c), replacing the constant 8. The reader can thus provide an optimal value, for example higher because the size of the useful communication area of the reader and its destination , make the simultaneous presence of several cards more likely. An additional improvement is that the reader increases this parameter if it detects collisions and decreases it if it does not detect them. An improvement applicable independently of the preceding one, consists in foreseeing in the call of the reader, another parameter A, compared with a predetermined value inscribed in the memory of the card, inhibiting any answer to a call according to the result of this comparison. The inhibition, for example, can take place if parameter A, when identifying the type of application (between electronic purse, debit / bank credit, telephone, pool access control, etc.) intended for the reader does not correspond to the inscribed value on the card and that represents your destination. Or even, in the absence of the element in the intersection of the set of applications processed by the reader and described by A, with the set of applications processed by the card and described in the memory of the card. By means of this improvement, the number of cards susceptible to respond is limited to those that are relevant, reducing the probability of collision. An improvement, independently applicable, is that the reader modifies the rhythm of their interrogations according to whether or not a message is detected on at least one card: the next interrogation is closer to the previous one, if, at the end of the latter, the reader has not detected any response, if it has detected the response of a card. The effect is to increase the number of interrogations, allowing, however, the complete reception of the response of a card if such a response is received. The waiting period between the end of one interrogation and the beginning of the next in the absence of answer detection of the cards is chosen as the sum of the maximum period between the end of the interrogation and the start of the card response, plus at least, the activation period of the device of the reader that detects the card response (usually, a device that detects the subcarrier issued by the card). An improvement, applicable independently, consists in providing two types of interrogations of the reader, designated "Query" and "Marker". Query reinitializes the response probability (Query can contain a P parameter). Marker responds to the card with a probability that evolves according to whether it has previously responded or not. For example, this refinement can modify Example A in the following manner: b) by subjecting it to voltage or entering it in the field, the card reinitializes the internal variable 1 = 0. c) upon receipt of a call from the reader, the card determines if it is a Query type and, if so, reinitializes I in an integer chosen randomly in the interval [1 ... 8], then acts as in the case of a Marker. In both cases, if I = 1, the card responds to the interrogation of the reader by issuing its identification, then in all cases it decreases I in 1 except if it is already null. It is observed that the probability of response of the card after a Query is 1/8, then if it continues without answering, the probability after a Marker evolves successively of 1/7, 1/6, 1/5, 1/4 , 1/3, 1/2, 1. According to a response from the card or at the end of stage b), the probability of response to a marker, if any, is 0 until the next Query. An advantage is that, if the parameter P and / or A is used, it is only necessary in Query. When a card has received an application message for which the comparison with its identification, short or long, has given a positive result, the card memorizes this event and no longer responds to the Query or Marker of the reader until a particular event, such as the reinitialization of the card and / or the reception of a message from the reader that presents a particular characteristic. This allows the cards already processed to be excluded from the enumeration mechanism, which reduces the probability of collision and increases the temporal density of the Query and Marker.
A variant is that the Markers are numbered by the reader and that this incremental number is taken into account in the probability of answering the card; for example, this number can be compared to. I (the response of the card is inhibited in case of disagreement), or more generally combined with the state variable contained in the card. Another variant consists in providing for an inhibition and / or modification of the state variable in case of incoherence of certain characteristics (such as the checksum) of the information received by the card. In a more general embodiment of the present invention, the probability of response to Query of the cards is a function of the two integers M and N. The decision to answer or not is taken on the card and comprises the following stages: • determination of the numbers M and N, of which at least one is a function of P, • uniformly distributed random drawing of an integer X, with 0 < . X < N, • answer if X is less than M, that is, with the probability M / N. To the existing systems would correspond two particular cases of such procedures: - in an example of the probabilistic system, N = 64 and M = P + 1, that is to say probability of response (P + D / 64; in a system of the type Slotted Alona, it would be more exactly: N = 2P and M = 1, that is, response probability 1 / (2P), however, in both cases, only one of the parameters M and N is variable, the other being fixed. The two parameters M and N could be numbers contained in the cycle of the general call message, but at least one of the parameters of the two is variable, giving the advantage of flexibility in the choice of characteristics of the anti-collision function. In the following examples, it is assumed that the Query message issued by the terminal contains an M number that defines P and may vary, for example, in the manner described in the French application 97 02501, as in a preferred embodiment of the invention. The dynamic evolution of the parameter P as a function of the detected collisions, is still applicable. The different variable parameters described in the following examples can be used alone with the use of a variable M parameter, or they can be associated with several of them, or they can even be used in several without using said M parameter.
Example No. 1: Query comprises an application selection field A, which conditionally inhibits the response of the cards; for example, the card responds if the preceding conditions on P and X are met, and if A equals a predetermined value contained on the card. This allows to reduce the probability of collision when several cards are in the field of the reader, but it is known that only those carriers of a certain application designated A are susceptible to be processed. The condition on A, of course, can be more complex, for example A can be a list of applications, the card responding if one of these applications is present in an application list memorized on the card. Example No. 2: the reader can issue, after Query and in a way that can be distinguished from Query, one or several items called "Marker". These Marker messages can serve to temporize the ATQ responses of the cards, which allows a gain of time in the transmission and management of the responses. There are several possible variants. Case 2.1: Marker messages contain, or allow to derive from the cards, an index I (Optionally they could also include the parameter P and eventually A). I is initialized (for example, 0) during each Query and incremented in each Marker. This numbering mechanism can be implemented in the reader and included in the Marker and / or in the card (which initializes I to 0 in the issuance and / or reception of Query and increases it in the issuance and / or reception of Marker). This numbering mechanism is implemented at least in the reader or the cards, so that upon receipt of Marker the value I is known by the cards. The card responds an ATQ to the Marker in function of I, P, of a random draw X (P and X can be those of a previous Query or Marker and are not necessarily recalculated in each Marker). The probability of the card's response is higher if the card has not yet responded from the previous Query, if it has already responded. In addition, in case of a collision after a Marker, the procedure can return to the Query message and repeat the call cycle with reinitialization of I, so that the overall probability of response of the cards within the range of the terminal increases when there is no collision and decreases when there is. This method example is advantageous when a message at the end of the data exchanges inhibits any response from a card that has already responded, at least until a subsequent general call. Usually A, P and the random draw X are defined by the Query message and the response condition is that I * M <; X < I * M + M. In this way, after the messages N / M (one Query, and N / M-l Marker, after which the cycle recommences) all the cards will have answered at least once. Thanks to this improvement, an advantageous feature of the "Slotted Aloha" is obtained: for any value of the P parameter, for a single card in the field of the reader, and in the absence of communication error, there is the certainty of capturing the identification of the card in a limited number of 2 * N / M messages from the entry of the card in the field, assuming that the card only responds from the first received Marker. As can be seen, a result comparable to the "Slotted Alona" with N / M slots is obtained, with the difference that the beginning of the slots is explicitly indicated by the reader when issuing Marker, instead of being implicitly by timed by the cards In relation to "Slotted Alona" this avoids exact timing, both through the cards and through the reader. Another advantage in relation to the probabilistic system is that if P is absent from the Markers (and A, see below) the duration of the Marker is reduced in relation to Query, and that the card does not necessarily have to draw another random number from each Marker. , further reducing the duration of the cycle. Even, the card can systematically answer the nth Marker, where n is a characteristic number of the card, for example, a part of its serial number. It will be noted that if the terminal only emits Query, without Marker, the terminal is equivalent to the probabilistic system, and the cards in this example are compatible with the terminals of the two systems. Case 2.2: It can be foreseen that the reader will include the index I in the subsequent application message (s) (especially the first one) and that this field of the applicative message will be compared by the card in index I, for the same reason as the card performs the comparison with its identification of an application message field; in this way there is a reinforcement of the selectivity of the selection mechanism without elongation of the identification field in the ATQ. Case 2.3: It can be foreseen that the numbering is both explicit in the Markers and implicit in the card, and that the card only answers if there is agreement; this case above all is advantageous as a complement to the previous improvement. Example No. 3: As a variant or complement to the preceding examples, the terminal may vary the interval between two successive Markers. In particular, if no ATQ response is received at the terminal, it is no longer necessary to wait for the usual time of data exchange between a card and the terminal. In this example, if the card responds to Query or Marker, it does so by issuing a signal (beginning of the ATQ) within a maximum period T0 following the end of Query or Marker. The reader looks for this signal, and if it is not present in a suitable period (Tx> T0 + reader detection period) following the end of Query, it then issues a new Query or Marker type message. In the presence of the signal, the Query and Marker emission is then inhibited, until the end of the emission by the cards (at least to the extent that a predetermined period T2 corresponding to the maximum length of the expected response did not pass, and / or to the extent that a collision error has not been detected by the reader in the message received from the card (s), the advantage is to allow a faster scan of the cards, since the number of Query (and / or Marker) ) per unit of time is clearly increased In the hypothesis that the duration of the ATQ is large in relation to the Query and Marker, and especially in a system that uses the example No. 1, the response opportunities for the cards are much more numerous.

Claims (8)

  1. NOVELTY OF THE INVENTION Having described the present invention is considered as a novelty and therefore the content of the following claims is claimed as property. A collision management procedure in a data exchange system between portable objects and at least one data sending / receiving terminal capable of cooperating with multiple portable objects simultaneously present in the communication field of the terminal, comprises the following successive stages: a) emission by the terminal, towards portable objects, of call messages; b) conditional emission by the portable object, towards the terminal, of a response message to the call with a probability, which may be less than 100%, this response containing an identification of the portable object itself; c) against the receipt by the terminal of a response of an identified portable object, and in the absence of collision of this response with a response emitted by another portable object, establishment of a specific data communication connection between the terminal and the portable object identified and continuation of the data exchange, and return to stage a) for a new iteration; and d) in the absence of collision-free reception, return to stage a) for a new iteration; characterized in that the emissions by the terminal comprise two types of call messages: interrogation messages, which initialize the value of the probability of response by the portable objects and marker messages issued successively after the interrogation messages, being the marker messages, suitable for ordering the portable objects that eventually modulate, in the absence of a previous response to a call message, this probability value against receipt of a marker message. The method according to claim 1, characterized in that the portable objects respond selectively in the timing intervals following the reception of the marker messages. 3 - The method according to claim 2, characterized in that the choice between the timing intervals is determined probabilistically by a random or pseudorandom draw made by the portable object against receipt of an interrogation message. 4. The method according to claims 1 to 3, characterized in that the terminal re-emits an interrogation message against detection of a collision. The method according to claims 1 to 4, characterized in that the modulation by the portable object, in the absence of a previous response to a call message, of the probability value against reception of a marker message is a function of the value of an index modified against receiving a message marker. The method according to claim 5, characterized in that said index is increased at each reception by a new marker message and reinitialized against the reception of a subsequent interrogation message. The method according to one of claims 1 to 6, characterized in that the terminal reduces the time interval separating two successive call messages in the absence of response of a portable object identified in a predetermined time. The method according to one of claims 1 to 7, characterized in that the sending of a response by the portable object to the call messages is susceptible to being inhibited by the reception of a specific message issued by the terminal containing an identification. corresponding to the portable object, inhibiting the initialization of the portable object and / or the reception of a message to that effect. SUMMARY This procedure manages collisions in a data exchange system between portable objects and at least one data sending / receiving terminal capable of cooperating with portable objects. It comprises the following stages: a) issuance by the terminal, towards portable objects, of call messages; b) conditional emission by the portable object, towards the terminal, of a response message to the call with a probability, which may be less than 100%, this response containing an identification of the portable object itself; c) against the receipt by the terminal of a response of an identified portable object, and in the absence of collision of this response with a response emitted by another portable object, establishment of a specific data communication connection between the terminal and the portable object identified and continuation of the data exchange, and return to stage a) for a new iteration; and d) in the absence of collision-free reception, return to stage a) for a new iteration. According to the invention, the emissions by the terminal comprise two types of call messages: interrogation messages, which initialize the value of the probability of response by the portable objects and marker messages issued successively after the messages of interrogation, being the marker messages, suitable for ordering the portable objects that eventually modulate, in the absence of a previous response to a call message, this probability value against receipt of a marker message.
MXPA/A/1999/008387A 1998-01-15 1999-09-13 Improved method for collision management in a non-contact data exchange system MXPA99008387A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
FR98/00383 1998-01-15

Publications (1)

Publication Number Publication Date
MXPA99008387A true MXPA99008387A (en) 2000-07-01

Family

ID=

Similar Documents

Publication Publication Date Title
EP1072128B1 (en) Method and system of identifying wireless units
US6456191B1 (en) Tag system with anti-collision features
US7274284B2 (en) Method for RF card detection in a contactless system
USRE44411E1 (en) Method of addressing messages, method of establishing wireless communications and communications system
AU729742B2 (en) A method of handling collisions in a system for interchanging data without contact
US8054162B2 (en) Communication method in RFID or remote sensor systems
AU755365B2 (en) Improved method for collision management in a non-contact data exchange system
MXPA99008387A (en) Improved method for collision management in a non-contact data exchange system
CA2266337C (en) Tag system with anti-collision features
EP1596544B1 (en) Method and devices for establishing wireless communication