EP1721285A1

EP1721285A1 - Method for identifying an electronic label, electronic label, and base station for implementing said method

Info

Publication number: EP1721285A1
Application number: EP04816358A
Authority: EP
Inventors: Pierre Raimbault
Original assignee: Individual
Current assignee: Individual
Priority date: 2003-11-13
Filing date: 2004-11-10
Publication date: 2006-11-15
Also published as: FR2862406B1; FR2862406A1; US20090160623A1; WO2005050538A1

Abstract

The invention relates to a method for identifying electronic labels provided with an identification code consisting of a plurality of rows, by means of a base station emitting interrogation commands (104, 108, 124, 128, 140) specifying at least one row of said code, each label emitting, in response to the base station, the value of the code thereof for each said row (110, 126, 130, 142). The inventive method consists of the following steps: the electronic labels are interrogated (104, 108), in an interrogation step, for at least one row of the code, and, if at least one electronic label emits a response (110), a signalisation step is performed by the base station when it has received two responses, the labels that do not have a pre-determined value for the interrogated row are subjected to a temporary inhibition step (114, 134) when the base station signals that it has received two responses, and a storage step (114, 134) is performed for the storage of each row in the labels that are temporarily inhibited. At least one subsequent interrogation step (124, 128, 140) is performed for each stored inhibition row.

Description

METHOD OF IDENTIFYING ELECTRONIC LABELS, ELECTRONIC LABELS AND BASE-STATION USING THE SAME

The present invention relates to a method for identifying electronic tags, electronic tags and a base station implementing it. It applies, in particular to the identification of objects provided with labels communicating by radio frequency waves. RFID tag identification methods are known as presented in document EP 96402555.5. These methods may be too slow to identify a large number of moving labels, for example on a forklift or in another vehicle. The present invention aims to remedy these drawbacks. To this end, the present invention relates, according to a first aspect, to a method of identifying electronic labels equipped with a memory of an identification code specific to each label, this code comprising a certain number of rows, identified by a base station adapted to transmit interrogation commands, explicit or implicit, specifying at least one rank of said code, each label being within range of the base station being adapted to transmit, in response to an interrogation command , a signal representative of the value of its code for each said rank, characterized in that it comprises

a step of interrogating the electronic tags and, on reception, by the base station, of the responses of the electronic tags, for at least one rank of code, a step of passing to the next rank during which the base station n 'transmits a signal to the electronic tags only in the case where the responses of the electronic tags represent at least two different code values for the rank interrogated;

- when the base station has reported having received replies representative of at least two different code values for the rank interrogated:. a step of temporarily inhibiting the labels which have a predetermined code value for said interrogated rank, inhibition step from which said label no longer responds to interrogations before receiving a de-inhibition instruction, and

"SB,",. "- =] ïOT n," -3M? 1 * "<&% && . a step of memorizing said rank in the labels performing the inhibition step,

- for each stored inhibition rank, at least one subsequent re-interrogation step relating to said memorized inhibition rank, and, in response to said re-interrogation, for each label which has memorized said inhibition rank:. a de-inhibition step and. a response step representative of the value of its code in this row or in another row, the base station associating with each said label a part of code common to all the labels performing said response step. Thanks to these provisions, the electronic tags are divided into sub-groups whenever they do not have the same code value in a row and, for each sub-group, part of the code or identifying labels of this sub -group is already known when these electronic tags are re-interrogated. Recognition of the labels of this subgroup is therefore faster than if all the rows of their code had to be interrogated again. In addition, since the base station only transmits a signal in one of the two cases, the method is faster than the methods known in the prior art. According to particular characteristics, the signal transmitted by the base station upon receipt of the responses from the electronic tags is a silence signal during which the base station stops transmitting a magnetic field. Since the signal is limited to silence, it is faster than if a frame had to be transmitted, for example to indicate the value of the code for the rows already identified. According to particular characteristics, the magnetic field that the base station stops emitting during the signal transmitted by the base station upon receipt of responses from the electronic tags is a magnetic field supplying the electronic tags. Thanks to these provisions, the downward channel between the base station and the labels is limited to the magnetic field for supplying electronic labels, which simplifies the implementation of the method which is the subject of the present invention. According to particular characteristics, the base station performs the subsequent re-interrogation steps, starting with the last row of inhibition and then carrying out a succession of interrogation steps until at least one label is fully identified. Thanks to these provisions, a large part of the code is already known for the labels responding to re-interrogations. The identification speed of these labels is therefore high. According to particular characteristics, when the code of a label is fully identified, said label performs a final inhibition step, starting from which said tag no longer responds to interrogations before receiving a de-inhibition instruction. Thanks to these provisions, the number of active labels is gradually reduced until all the labels are identified. According to particular characteristics, at each interrogation step the labels determine the instant of emission of their response as a function of the value of their code for the rank concerned by the response. According to particular characteristics, at the start of a re-interrogation, the labels emit a response which concerns the rank which follows that for which the inhibition has taken place. Thanks to these provisions, the labels do not emit a response already known to the base station, in the case where the codes are binary and the interrogations relate to a single rank of the codes. According to a second aspect, the present invention relates to an electronic label equipped with a memory of an identification code specific to each label, this code comprising a certain number of rows, identified by a base station suitable for transmitting commands d interrogation, explicit or implicit, specifying at least one rank of said code, each label being within range of the base station being adapted to transmit, in response to an interrogation command, a signal representative of the value of its code for each said row, characterized in that it comprises: - means for responding to a query of the electronic labels by the base station, for at least one row of code, and, for each query:

a control means suitable for detecting a signal transmitted by the base station only in the case where the base station has received responses representative of at least two different code values for the rank interrogated, and, in the event of detection of this signal, to determine if said label has a predetermined code value for said rank and, in this case, to inhibit the electronic label, that is to say no longer allow it to respond to interrogations until that she receives a de-inhibition instruction,

an inhibition rank memory in which the control means stores the rank of the code to which the label has possibly been inhibited. According to a third aspect, the present invention relates to a base station for identifying electronic labels equipped with a memory of an identification code specific to each label, this code comprising a certain number of rows, said base station being suitable to issue interrogation commands, explicit or implicit, specifying at least one rank of said code, each label being within range of the base station being adapted to emit, in response to an interrogation command, a signal representative of the value of its code for each said rank, characterized in that it comprises: a means of memorizing each row called "divergence" for which at least two labels have sent two different responses to said interrogation,

a suitable means of communication, upon reception by the base station of the responses of the electronic labels, for at least one row of code, to transmit a signal from the base station to the electronic labels only in the case where the responses from electronic tags represent at least two different code values for the rank queried;

- control means adapted to re-interrogate the labels for each stored row of divergence and to associate with their response a part of code common to all the labels which respond to said re-interrogation. The advantages, purposes and specific characteristics of this label and of this base station being similar to those of the process as succinctly set out above, they are not repeated here. Other advantages, aims and characteristics of the present invention will emerge from the description which follows, given for explanatory purposes and in no way limitative with regard to the appended drawings in which: - Figure 1 shows an example of implementation of the object method of the present invention, by a succession of steps carried out by a base station and labels in accordance with the present invention; - Figure 2 shows an operating algorithm of a base station according to an embodiment of the present invention; - Figure 3 shows an algorithm for operating a label according to an embodiment of the present invention; FIG. 4 represents a factor of gain in speed of identification of the method which is the subject of the present invention compared to the fastest method known at present; - Figure 5 shows a base station and a label of a particular embodiment of the present invention and - Figure 6 shows a timing diagram of signals exchanged between the base station and electronic labels. Throughout the description, rank is called the address of a bit of the identification code of a label and a slot or time slot is called a predefined time interval in which a label can emit a signal for its identification. . The principle of the process which is the subject of the present invention is as follows:

- the base station issues a "Start"("KD") instruction which initializes the interrogation process and which makes it possible to identify the electronic label with the lowest code or identifier ("ID"). "Return"("KR") instructions are used to perform the interrogations to identify the other labels, at the rate of a single "KR" instruction per identified label;

- the instructions "KD" and "KR" are followed by a scan of the identifiers (ID), bit by bit (rank by rank), going, for example, from the most significant bit ("MSB"), rank "0" and ending on the least significant bit ("LSB"), at the highest rank (equal to the number of bits to be identified - 1);

- for each row tested, a pair of time slots (sO and si) is provided; the active tags respond with a "BEEP" in the slot sO if the value of their code at the rank considered is "0". If the value of their code at the rank considered is "1", they respond in the slot if; - the number of the reading rank is incremented after each pair of slots, implicitly, that is to say without an explicit instruction being issued by the base station;

- when there are only responses in the slot sO, the base station stores the value "0" at the reading rank considered;

- when there are only responses in the slot if, the base station stores the value "1" at the reading rank considered;

- when several labels are active, since they cannot have the same identifier "ID", it necessarily occurs, in a given row, the case where at least one label responds in sO while at least one label responds in the slot if, this case is called a "divergence"; - this discrepancy is signaled to the labels by the production of a base station signal at the end of the slot s This signal consists, for example, in carrying out a brief carrier interruption called "silence", a brief carrier interruption which does not require no resynchronization of label clocks When there is no divergence, the base station does not transmit any signal to electronic labels. Silence and the absence of silences constitute implicit interrogation commands, specifying to electronic labels to pass to the next rank;

- the appearance of the divergence causes the following actions:. after detection of silence, the active tags (not inhibited) which emitted a "BEEP" in the slot if go into the inhibited state, that is to say that they no longer respond to interrogations until 'they receive a new interrogation ("re-interrogation") for this rank (see below) and memorize the rank at which inhibition occurred and. in the base station, incrementation of a list pointer and memorization of the rank at which the divergence has been noted (without erasing of the other divergence ranks already memorized) and memorization of the value "0" at the reading rank considered; - the interrogation continues only for the labels not inhibited (in the case of a divergence, those which responded in the sO slot at the time of the divergence). If other divergences arise, proceed as above;

- when the final rank has been processed, the last active label is completely identified. It goes into permanent inhibition and its code is stored by the base station in the list of identified codes;

- it may happen that a divergence occurs in the last row, in the case where two labels have identical "IDs" except for the least significant bit, LSB, present in the last row. In this case, there is a double identification and the two labels are put into inhibition by memorizing the last rank as inhibition rank. This discrepancy is not entered in the base station list. There is no generation of silence for the last row;

- after identification of a label, the base station returns to the last row of divergence registered in its list (if the list is empty there is no more label to identify). This row number is sent to all the inhibited tags. All those whose inhibition rank memory is equal to the divergence rank sent by the base station are reactivated or "de-inhibited". The base station clears the memory of the last divergence rank by decrementing its pointer of the divergence rank list and takes, as part of the code already identified, the part of the code of the previous identified label which extends from the first rank to the rank preceding the divergence rank by adding, for the divergence rank, the value "1" (for example, if the previous identified label had the code "01001011" ... and the last divergence rank is the sixth rank, the part of code already identified for the next label being identified is, for the first six rows, "010011") and - reading of the labels resumes from the next row (we know that all the labels thus re-activated have an "ID" in state "1" at the rank of divergence; it is therefore useless to scan this rank again). As before, the reading continues until the last row (LSB) with possibly new divergences, treated as above. The instructions or commands used are as follows: - "KD": "Start" instruction. It is followed by NBID data which indicates the number of rows or bits of the "ID" identifiers (for example 32 bits), which allows the labels to identify the end of a scan (rank = NBID -1). All the labels contained in the Base-station field become active (Inhib = "0"). The row counters (Nrang), and the tag inhibition row memories (MRI), are set to zero. In the Base station, the rank counter (Nrang), and the list pointer of the Base station (Ptr), are set to zero. - "KR": "Return" instruction. It is followed by a MRDV data which indicates the rank of the last discrepancy noted. This row number is the value indicated by the Ptr pointer read in the base station, then the Ptr pointer is decremented. All tags whose MRI inhibition rank memory is equal to this MRDV divergence rank become active again. Their rank counter (Nrang) is incremented and the interrogation resumes from MRDV + 1 until the last rank. It is observed that the same interrogation process is carried out following the "KD" and "KR" instructions. After a fixed dead time including the turnaround time of the base station (which starts listening for "BIPs"), counting the clock makes it possible to define the time position of the pairs of slots. Only if a divergence is detected on a row Nrang> NBID -1:

- the base station produces a silence (SIL) which causes the inhibition of all the labels which have responded in the slot if at the time of the divergence. These labels store the current rank in their inhibition rank memory (MRI = Nrang); - the base station increments its pointer Ptr and writes the number of the divergence rank in its list: MRDV (Ptr) = Nrang;

- the rank is incremented;

- the scanning continues, the temporal position of the following slots being shifted as a function of the silence transmission time and of the associated time delays. If no divergence is detected, the base station produces no signal or silence and the electronic tags go to the next row. When the last rank (LSB) is reached (Nrang = NBID - 1):

- one label is completely identified (or two if a discrepancy has occurred in the rank of the LSB); - if the list pointer Ptr is at "0", all the labels have been identified (end of the process)

- if the list pointer Ptr is not at "0", a "Return" "KR" instruction is issued. An example of exchanges carried out between a base station in accordance with the present invention and a label in accordance with the present invention is given below to implement the method which is the subject of the present invention. The example given below, with reference to FIG. 1, is limited to identifiers "ID" with a length of six bits (Nrang = 0 to 5) and to a case where four labels identified A to D are identified. :

During a step 102, the base station issues an "KD" instruction: "Start" instruction. During an initialization step 104, all the labels which receive the "KD" instruction are activated, their rank counter and their inhibition rank memory are set to "0". During a step 106, the list pointer and the rank counter of the base station are set to "0". During a step 108, the number of bits of the identifier (NBID) is transmitted to the labels and the scanning begins at the first rank, the rank "0". During a step 110, because the labels A and B respond in the slot s0 and the labels C and D respond in the slot if, a divergence is detected in Nrang = 0. During a step 112, because there has been a divergence, the base-station generates a silence and increments its pointer Ptr and writes in its memory the rank number of the first divergence MRDV (1) = Nrang = 0. Since there has been a silence from the base station, the labels that responded in the slot if, that is to say the labels C and D (whose ID bit is "1"), inhibit and memorize MRI = Nrang = 0 during a step 114. Similarly, during a step 116, a second divergence is detected in Nrang = 3, the label A responding in the slot sO and the label B in the slot if. During a step 118, the base station registers in its list MRDV (2) = 3 and generates silence. During a step 120, the label B, which responded in the slot if inhibits and stores MRI = 3. For the following rows, there is no longer any divergence and therefore no more silence from the leaves the base station until the last row is reached and the label A is identified by the base station, during a step 122 and is permanently inhibited until the next start instruction " KD ". (in the following table "X" represents a "BEEP" emitted by the label and "Inh" does so, for it to inhibit itself)

During a step 124, the base station sends an instruction "KR", instruction "Return", followed by the last number of divergence rank registered in the list: MRDV (2) = 3. The pointer is decremented and returns to Ptr = 1. During a step 126, the label B for which MRI = MRDV = 3 is reactivated. The interrogation resumes at rank: Nrang = MRDV + 1 = 4 and continues until the last rank, without divergence. Label B is identified by the base station and is permanently disabled until the next start instruction "KD".

During a step 128, the base station issues an instruction "KR", instruction "Return", followed by the number of the divergence rank entered in the list: MRDV (2) = 0. The pointer is decremented and returns to Ptr = 0. During a step 130, the labels C and D for which MRI = MRDV = 0 are reactivated. The interrogation resumes at the rank: Nrang = MRDV + 1 = 1. During a step 132, a divergence is detected in Nrang = 1 and the base station generates a silence. During a step 134, the label D which responded in the slot if inhibits and stores MRI = 1. During a step 136, the Base-station increments its pointer and writes MRDV (1) = 1. During a step 138, the interrogation continues until the last rank without further divergence. Label C is identified and is permanently inhibited until the next start instruction "KD".

During a step 140, the base station sends an instruction "KR", instruction "Return", followed by the last number of divergence rank registered in the list: MRDV (2) = 1. The pointer is decremented and returns to Ptr = 0. During a step 142, the label D for which MRI = MRDV = 1 is reactivated. The interrogation resumes at the rank: Nrang = MRDV + 1 = 2. The interrogation continues, without divergence, until the last row. During a step 144, the label D is identified and is definitively inhibited until the next start instruction "KD". The pointer Ptr being at Ptr = 0, the process is finished, all the labels have been identified.

FIG. 2 represents an operating algorithm of the base station (the labels are designated by the word "tags"). During a step 200, the base station is initialized. During a step 202, the tags synchronize and get in phase (see document EP 0 777 192). During a step 204, the base station transmits a start instruction "KD" indicating the number of bits of the identification codes NBID. During a step 206, the base station sets the variable Nrang and the list pointer Ptr to "0". During a step 208, the base station performs a waiting step (equal to the turnaround time). During a step 210, the base station waits for the end of the slot s0. During a step 212, the base station determines whether it has received a “BEEP” signal during the slot s0. If yes, during a step 214, the variable s0 is set to "1". Otherwise, during a step 216, the variable s0 is set to "0". During a step 218, the base station waits for the end of the slot s During a step 220, the base station determines whether it has received a "BEEP" signal during the slot s If yes, during 'a step 222, the variable si is set to "1". Otherwise, during a step 224, the variable si is set to "0". During a step 226, the base station determines whether the variables s0 and si are both at the value "0". If so, during a test 228, the base station determines whether the variable "Nrang" is equal to "0". If so, during a step 230, the base station concludes that there is no active tag in its vicinity. Otherwise, during a step 232, the base station determines that there has been an anomaly and stops its operation, for example to return to step 200. If the result of test 226 is negative, during d In a step 234, the base station determines if the variable s0 is worth "1" and the variable if is worth "0". If so, during a step 236, ID (Nrang) = 0 is stored. This table represents the values of the bits of the ID being identified. Otherwise, during a step 238, the base station determines if the variable s0 is worth "0" and the variable if is worth "1". If yes, during a step 240, ID (Nrang) = 1 is stored. In these two cases, the labels which have responded are not inhibited. On the other hand, if during a step 242, the base-station has received responses during each of the slots s0 and if, it determines that there has been a "divergence", that is to say that the labels who carried out a response are found on two branches of the tree of identification codes which separate at rank Nrang considered. In this case, during a step 244, the base station determines whether the rank Nrang is equal to NBID - 1 (number of bits of the identifier minus one). If yes, during step 246, two labels are identified, they have identical identifiers except for their last bit. If not, during a step 254, the value of the bit for the rank considered and for the labels which will continue to respond (that is to say those which will not inhibit) is equal to "0 ": ID (Nrang) = 0. Then, during a step 256, the row number considered is memorized and the variable MRDV (Ptr) takes the value Nrang and the pointer Ptr is incremented by 1. During in a step 258, the base station generates a silence "SIL" and, during a step 260, the variable Nrang is incremented. Following one of steps 236 or 240, during a step 248, the base station determines whether the variable Nrang = NBID - 1. Otherwise, step 260 is carried out. If so, during a step 250, a label is identified, that is to say that its code is fully known. Following one of the steps 246 or 250, during a test 252, the base station determines whether the pointer Ptr has a value greater than "0". Otherwise, the label recognition and identification process is completed and the base station can return to step 200. If yes, during a step 262, the base station issues a return instruction "KR" by providing as data the value of the variable MRDV (Ptr), ie the rank number of the last divergence in the list of divergence ranks. Then, during a step 264, the variable Nrang is set to the value MRDV (Ptr), the ID (Nrang) = 1 is stored, the variable Nrang is incremented and the variable Ptr is decremented. Then the base station returns to step 208. FIG. 3 represents an algorithm for the operation of each label. During a step 300, the label receives a signal from a base station which allows it to receive the energy necessary for its operation and initializes. During a step 302, the label determines whether its synchronization is carried out (see document EP 0 777 192). Otherwise, it returns to step 302. If yes, during a step 304, the label determines whether it has received an instruction. Otherwise, it returns to step 302. If yes, during a step 306, it determines whether the instruction received is a start instruction "KD". Otherwise, it returns to step 302. If yes, during a step 308, the label initializes the value of a variable NBID indicating the number of bits of the identifier supplied by the base station. Then, during a step 310, the label initializes: the variable "Inhib" to the value "0" signifying that the label is activated, - the variable "MRI" to the value "-1", MRI representing an inhibition rank number and "- 1" a value signifying that no inhibition has yet taken place, - the variable CPtr to the value " 0 ", CPtr being the slot counter and - the row number being interrogated Nrang is set to 0. During a step 312, the label waits for the end of the turnaround time. During a step 314, the label determines whether the slot s0 has started. Otherwise, step 314 is repeated. If yes, during a step 316, the label determines whether it is in the inhibited state (Inhib = 1). If so, the label proceeds to step 328 described later. Otherwise, during a step 318, the label determines whether, at the rank number indicated by Nrang, the bit read on its identifier is equal to "0". If yes, during a step 324, the label emits a “BEEP” signal during the slot s0. Otherwise, during a step 320, the label determines whether the slot has started. Otherwise, step 320 is repeated. If yes, during a step 322, the label emits a “BEEP” signal during the slot if. Following one of the steps 322 or 324, during a step 326, the label determines whether the slot if is finished. Otherwise, step 326 is repeated. If so, during a step 328, the label determines whether it detects a "SIL" silence. If so, during a step 330, the label is warned that a divergence has been detected by the base station. It stops counting the slots for the duration of the silence and the turnaround time. Then, during a step 332, the label determines whether the variable "Inhib" is at "1". If so, it proceeds to step 338 described later. Otherwise, during a step 334, the label determines whether the value of its identification code for the rank considered is "1". Otherwise, it goes to step 338. If yes, during a step 336, the label sets its variable "Inhib" to "1" and its variable "MRI" to the value of the variable "Nrang" then goes to step 338. During a step 338, the label determines whether the silence is ended, as well as the turnaround time. Otherwise, step 338 is repeated. If so, during a step 340, the label resumes the counting of the slots and, during a step 342, the value of the variable "Nrang" is incremented and the label returns to step 314. When the result of test 328 is negative (no silence is detected), during step 344, the label determines whether the variable "Nrang" is equal to NBID - 1. Otherwise, the label performs the step 342. If yes, during a step 346, the label determines whether its variable "Inhib" is equal to "1". If so, it performs step 352 described below. Otherwise, (that is to say if the label is active), during a step 348, the label knows that it has been fully identified and, during a step 350, the variable " Inhib "is set to" 1 "and the variable" MRI "is set to the value NBID - 1. In step 352, the label determines whether an instruction has been received. Otherwise, step 352 is repeated. If so, during a step 354, the label determines whether the instruction received is a "KR" return instruction. Otherwise, step 306 is repeated. If yes, during from a step 356, the variable “MRDV” takes the value provided by the instruction KR, that is to say the rank number of the last divergence from the list of divergences stored by the base station. Then, during a step 358, the label determines whether the variable "MRI" has for value the value of the variable "MRDV". If yes, during a step 360, the value of the variable "Inhib" is set to "0", that is to say that the label is reactivated. Following step 360, or if the result of step 358 is negative, during a step 362, the variable "Nrang" is set to the value MRDV + 1 and step 312 is repeated. The inventor compared the identification speed of the currently known fastest algorithm on the market, with that of the algorithm exposed with regard to FIGS. 1 to 3, for three types of identifier ID and for a number of labels ranging from two to one hundred. This comparison shows a time saving factor illustrated in FIG. 4. With the nominal durations and 100 labels, the exposed algorithm improves the speed by a factor of at least 1.41 compared to the conventional algorithm with a random identifier. This improvement reaches 1.83 when the identifier has a constant prefix and a random 9-bit suffix. This factor increases to 2.92 when the identifier is sequential. FIG. 5 shows a base station 500 comprising a control and command module 505, an antenna 510, a database 515 and operating instructions 520 in a non-volatile memory 525 and a variable memory 530. We also observe in FIG. 5, a label 550 comprising an antenna 555, a processor 560, a code memory 565, a variable memory 570 and a program memory 575. The operating instructions 520 implement one of the embodiments of the method which is the subject of the present invention, for example the embodiment illustrated in FIG. 1 or the embodiment illustrated in FIG. 2. The program memory 575 keeps an operating program of the label 550, implementing one of the modes for carrying out the process which is the subject of the present invention, for example the embodiment illustrated in FIG. 1 or the embodiment illustrated in FIG. 3. Alternatively, the code can take more than two different values in a row or more than one row of the code is interrogated during an interrogation step. In this case, the number of time slots which follow a interrogation step is increased to correspond to the number of possible responses and following the de-inhibition step, the de-inhibited labels transmit their response for the rank inhibition. It is observed that, in this case, the silence which indicates the occurrence of a divergence is located after the first slot for which a response is received by the base station so that the labels can determine whether it should be inhibited, or not. According to a sub-variant, following a response from the labels, the base station transmits a signal representative of the most low for said row so that the labels can determine whether it should inhibit or not. In FIG. 6, there is a chronogram of signals exchanged between the base station and the electronic tags which, in the example given here, have the same identifiers as above:

In Figure 6, the scales are not respected. To give an idea, in the prototype produced by the inventor, which operates with a magnetic field of frequency 125 KHz, the duration of a response is 1 ms, the duration of a silence is 0.05 ms and the turnaround time is 1, 2 ms. The upper part of FIG. 6 relates to the signals transmitted by the base station and the lower part of FIG. 6 relates to the signals transmitted by the electronic tags considered jointly as a group. In the upper part relating to the base station, the highest signal represents the emission of a magnetic field supplying electronic tags and the momentary silences or interruptions of this field to signal the presence of a divergence. In the upper part concerning the base station, the lower level signal represents the time intervals for sending explicit requests. In the lower part concerning electronic labels, a signal represents the time intervals during which at least one electronic label emits a signal to indicate to the base station the value of its code in a row. The transmitter labels are indicated by the letters "A", "B", "C" and "D" in the transmission time intervals. Below the lower part, the rank concerned by each response has been indicated and, in line below, the meaning of the responses given. During a first phase 602, starting arbitrarily at the origin of the times, during which the base station transmits a request for initialization of the electronic tags. Then, during a phase 604, the base station transmits, to the initialized labels, a request for interrogation of their first rank of code. During a phase 606, the electronic tags reserve a turnaround time which allows the base station to listen to their responses. During a step 608, the labels "A" and "B" emit a response indicating that they have, in the first row of their identifier, the value "0". During a step 610, the labels "C" and "D" emit a response indicating that they have, in the first row of their identifier, the value "1". During a step 612, since there has been a divergence, the base station emits a silence which indicates that it has received responses representing different code values and which causes the temporary inhibition of the labels "C " and D". During a phase 614, the electronic tags reserve a turnaround time which allows the base station to listen to their responses. During a step 616, no label emits because the labels "C" and "D" are inhibited and the labels "A" and "B" do not have, in the second row of their identifier, the value "0 ". During a step 618, the labels "A" and "B" emit a response indicating that they have, in the second row of their identifier, the value "1". Since there has been no divergence, the base station does not emit silence, which indicates that it has received responses representing the same code value and which does not cause any new temporary label inhibition . During a step 620, the labels "A" and "B" detect that there has been no emission of silence by the base station and, without reserving a turnaround time, emit a response indicating that they have, in the third row of their identifier, the value "0". During a step 622, no label emits because the labels "C" and "D" are inhibited and the labels "A" and "B" do not have, in the third row of their identifier, the value "1 ". Since there has been no divergence, the base station does not emit silence, which indicates that it has received responses representing the same code value and which does not cause any new temporary label inhibition . During a step 624, the label "A" detects that there has been no emission of silence by the base station and, without reserving a turnaround time, emits a response indicating that it has, in the fourth row of its identifier, the value "0". During a step 626, the label "B" emits a response indicating that it has, in the fourth row of its identifier, the value "1". During a step 628, since there has been a divergence, the base station emits a silence which indicates that it has received responses representing different code values and which causes the temporary inhibition of the labels "B". During a phase 630, the label "A" reserves a turnaround time which allows the base station to listen to its response. The rest of the exchanges between the base station and the labels are not described here. It is observed, with reference to FIG. 6, that the characteristic of the present invention according to which upon reception, by the base station, of the responses from the electronic tags, during a step of passing to the next row, the base station only emits a signal to electronic labels if the responses from electronic labels represent at least two different code values for the rank questioned This saves a lot of time for the identification of electronic labels, in addition to the time saved thanks to the memorization of the inhibition ranks which makes it possible to know part of the code of the electronic labels, here the label "B" of which three code values are already known for at least the first three rows. Indeed, the duration necessary to interrogate a rank where there is divergence is the sum of: - twice the duration of a response by the electronic tags, - the duration of a silence, - the duration of reversal. On the other hand, the duration necessary to interrogate a rank where there is no divergence is only equal to the sum of: - twice the duration of a response by electronic tags and - the duration of detection of absence of silence (duration less than or equal to the duration of a silence). We also observe that the total number of discrepancies encountered during the identification of a number n of electronic tags is equal to n-1. The time saved by the characteristic of the invention indicated above is therefore all the more important as the number of rows to be interrogated is. Although, in the description, it is the labels which have a "1" in a row where a divergence is detected which are inhibited but the present invention is not limited to this case and extends to the case where it is the labels which have a "0" in a row where a divergence is detected which are inhibited. Similarly, the meaning of the successive time intervals can be reversed and the direction of interrogation of the codes in ascending order can be reversed without departing from the scope of the present invention.

Claims

1 - Method for identifying electronic labels (550) equipped with a memory (565) of an identification code specific to each label, this code comprising a certain number of rows, identified by a base station (500 ) adapted to issue interrogation commands (104, 108, 124, 128, 140), explicit or implicit, specifying at least one rank of said code, each label being within range of the base station being adapted to transmit, in response to an interrogation command, a signal representative of the value of its code for each said rank (110, 126, 130, 142), characterized in that it comprises: - an interrogation step (104, 108) electronic tags and, upon reception, by the base station, of the responses from the electronic tags, for at least one code row, a step of passing to the next row during which the base station does not transmit a signal at destination electronic tags only in case the responses from electronic tags represent at least two different code values for the rank queried;

- when the base station has reported having received replies representative of at least two different code values for the rank interrogated:. a temporary inhibition step (114, 134) of the labels which have a predetermined code value for said interrogated rank, inhibition step from which said label no longer responds to interrogations before receiving a de-inhibition instruction, and. a storage step (114, 134) of said rank in the labels performing the inhibition step,

- for each stored inhibition rank, at least one subsequent re-interrogation step (124, 128, 140) relating to said memorized inhibition rank, and, in response to said re-interrogation, for each label which has memorized said rank inhibition:. a de-inhibition step (126, 130, 142) and. a response step (126, 130, 142) representative of the value of its code in this row or in another row, the base station associating with each said label a part of code common to all the labels performing said response step (126, 130).

2 - Method according to claim 1, characterized in that the signal transmitted by the base station on receipt of responses from the electronic tags is a silence signal during which the base station stops transmitting a magnetic field.

3 - Method according to claim 2, characterized in that the magnetic field that the base station stops transmitting during the signal transmitted by the base station upon receipt of responses from electronic tags is a magnetic field for supplying electronic tags . 4 - Method according to any one of claims 1 to 3, characterized in that the base station (500) performs the subsequent re-interrogation steps (124, 128, 140) starting with the last row of inhibition then by performing a succession of interrogation steps until at least one label (550) is fully identified. 5 - Method according to any one of claims 1 to 4, characterized in that, when the code of a label (550) is fully identified, said label performs a final inhibition step (122, 126, 138, 144 ), from which said label no longer responds to interrogations before receiving a de-inhibition instruction.

6 - Method according to any one of claims 1 to 5, characterized in that, at each interrogation step (104, 108, 124, 128, 140), the labels (550) determine the time of emission of their answer according to the value of their code for the rank concerned by the answer.

7 - Method according to any one of claims 1 to 6, characterized in that, at the start of a re-interrogation (124, 128, 140), the labels (550) emit a response concerning the rank following that for which inhibition has taken place.

8 - Electronic label (550) equipped with a memory (565) of an identification code specific to each label, this code comprising a certain number of rows, identified by a base station (500) adapted to issue commands interrogation (104, 108, 124, 128, 140), explicit or implicit, specifying at least one rank of said code, each label being within range of the base station being adapted to transmit, in response to a command from interrogation, a signal representative of the value of its code for each said rank (110, 126, 130, 142), characterized in that it comprises:

- a response means (560, 555, 575) to an interrogation of the electronic labels by the base station, for at least one rank of code, and, for each interrogation: - a control means (560, 575) adapted to detecting a signal transmitted by the base station only in the case where the base station has received responses representative of at least two different code values for the rank interrogated, and, in the event of detection of this signal, in determining whether said label has a predetermined code value for said rank and, in this case, to inhibit the electronic label, that is to say no longer allow it to respond to interrogations until it receives an instruction to de-inhibition,

- an inhibition rank memory (570) in which the control means stores the rank of the code at which the label has possibly been inhibited.

9 - Electronic label according to claim 8, characterized in that the control means is adapted to detect as a signal transmitted by the base station to signal that the base station has received responses representative of at least two different code values is a silence signal during which the base station stops transmitting a magnetic field. 10 - Electronic label according to claim 9, characterized in that the control means is adapted to detect said signal on the magnetic field of supply of electronic labels by the base station.

11 - Label (550) according to any one of claims 8 to 10, characterized in that the control means (560, 575) is adapted, in response to a new interrogation relating to the inhibition rank (124, 128 , 140), to de-inhibit said label (126, 130, 142).

12 - Label (550) according to claim 11, characterized in that, when the label is de-inhibited (126, 130, 142), it transmits a response representative of the value of its code to the rank following the rank d 'inhibition. 13 - Base station (500) for identifying electronic labels (550) equipped with a memory (565) of an identification code specific to each label, this code comprising a certain number of rows, said base station being suitable for transmitting interrogation commands (104, 108, 124, 128, 140), explicit or implicit, specifying at least one row of said code, each label being within range of the base station being adapted to transmit, in response to an interrogation command, a signal representative of the value of its code for each said rank (110, 126, 130, 142), characterized in that it comprises:

a storage means (505, 530) for each row called "divergence" for which at least two labels have issued two different responses to said interrogation,

- a means of communication (505, 510) adapted, upon reception by the base station of the responses of the electronic labels, for at least one rank of code, to transmit a signal from the base station intended for the electronic labels only in the case where the responses of the electronic tags represent at least two different code values for the rank questioned;

- control means (505, 520, 525) adapted to re-interrogate the labels for each stored divergence rank (124, 128, 140) and to associate with their response a part of code common to all the labels which respond to said re-interrogation. 14 - Method according to claim 13, characterized in that the signal sent by the base station on receipt of responses from electronic tags is a silence signal during which the base station stops transmitting a magnetic field. 15 - Method according to claim 14, characterized in that the magnetic field that the base station stops transmitting during the signal transmitted by the base station upon receipt of responses from electronic tags is a magnetic field for supplying electronic tags . 16 - Base station (500) according to any one of claims 13 to 15, characterized in that the control means (505, 520, 525) are adapted to perform re-interrogations starting with the last row of inhibition then by performing a succession of interrogation steps until at least one label (550) is fully identified.