WO2012148363A1

WO2012148363A1 - Rfid sigulation process

Info

Publication number: WO2012148363A1
Application number: PCT/TH2012/000019
Authority: WO
Inventors: Tanee DEEMECHAI; Siwaruk Siwamogsatham
Original assignee: National Science And Technology Development Agency
Priority date: 2011-04-29
Filing date: 2012-04-27
Publication date: 2012-11-01

Abstract

The present invention discloses a method to incorporate the CDMA technique into a slotted Aloha-based RFID singulation process without incurring bandwidth expansion.

Description

Title of the Invention

RFID SIGULATION PROCESS

Technical Field

The present invention relates generally to RFID technology, more particularly, to the

RFID singulation process

Background Art

RFID singulation process is a process that enables a reader to communicate with several tags in its read zone on one-by-one basis. In an RFID system, a tag that has just been energized is said to be in unsingulated state, while a tag that has communicated successfully with the reader is said to be in singulated state.

The state-of-the-art singulation process is based on the ISO/IEC 18000-6 Type C (ISO-C) standard. The process employs a multiple access technique called framed slotted Aloha, such that the RFID system performs the singulation process in a number of rounds where in each round the reader and tags cooperate as follows. The reader starts a round by issuing a query command with a parameter Q indicating that the number of available slots within the round is 2^Q. In this regard, the first slot (slot 0) begins just after the query command and a slot ends when the reader issues a QueryRep command. Hence, slot n+1 begins just after the QueryRep command that closes slot n. Note that the number of available slots within a round is a critical system parameter.

After processing the query command, each unsingulated tag selects out of the available slots one slot on which the tag shall transmit a reservation signal called RN16 signal, and after the transmission of the RN16 signal the tag shall wait for an Ack signal from the reader. If the reader can detect an RN16 signal after the beginning of a slot (that is possible when only one tag selects the slot), it transmits an Ack signal. When a tag receives a valid Ack signal, it can then enter a communication session with the reader by first transmitting its identification code. A communication session shall be closed by the QueryRep command of the reader and then the communicated tag shall change its state to be singulated. Note that the RN16 signal of a tag is a packet containing a 16-bit number randomly chosen by the tag itself, and the correspondingly valid Ack signal is simply an echo packet containing the same 16-bit number.

After transmitting a signal indicating the beginning of a slot (query or QueryRep), the reader waits for an RN16 signal for a suitable moment. If it cannot detect any RN16 signal, it shall issue a QueryRep command to end the slot. Otherwise, it shall transmit the Ack signal and then wait for receiving a tag identification code, which starts a communication session. When a communication session is to be finished, the reader shall issue a QueryRep command.

Fig. 1 illustrates the ISO-C singulation process for the case that there are three tags, where tags 0, 1, and 2 select respectively slots 0, 0, and 2. It may be noted that only slot 2 is used for a communication session. It may be noted also that the number of available slots in a round should be suitably set according to the number of unsingulated tags in the round. It is well known that the number of slots in a round with a large number of unsingulated tags should equal the number of the unsingulated tags. This optimum setting would result in about one third of the slots being used for the communication sessions, and about two third of the slots being unusable because each of them is not selected by a tag or is selected by multiple tags.

The ISO-C singulation method is based on the framed slotted Aloha technique, as it lets each tag randomly selects a time slot for a possible communication session and hopes to take the opportunity that a slot is uniquely selected so that the slot can be used for a successful communication session. However, the code-division multiple-access (CDMA) technique may be used also in this context. In the CDMA technique, the spread-spectrum modulation is employed so that tags with different spread-spectrum codes can communicate with the reader at the same time. However, the CDMA technique generally requires multiply larger transmission bandwidth, where the. multiplicity is proportional to the number of tags we could allow them to communicate with the reader simultaneously.

Prior arts about using CDMA in RFID system may be mentioned as follows. (1) The US patent application US 2008/0036573 Al "CDMA-RFID" has described an Aloha technique wherein the tag signal consists of a non-data part and a data part, and the data part is based on spread-spectrum modulation (see the paragraph [1067] and Fig. 5 of the application). (2) D.-F. Tseng and Z.-C. Lin, "Anti-collision algorithm with the aid of interference cancellation and tag set partitioning in radio-frequency identification systems," IET Communications, 3 (1), 2009, pp. 143-150, has described a modification of the ISO-C singulation process wherein spread- spectrum modulation is applied on the R 16 signal in order to increase the reader's ability to detect two simultaneous tags' signals that are much different on the power level. However, the drawback of the above two prior arts is that the tag signal requires multiply larger bandwidth as a result of the spread-spectrum modulation. Summary of the Invention

The present invention employs a CDMA technique in the framework of the framed slotted Aloha-based RFID singulation process so that significant speed improvement is obtained without requiring bandwidth expansion. Various objectives, specific characteristics, and other related issues of the present invention should be obvious after considerations of the forthcoming enclosed drawings and detailed descriptions of the invention and preferred embodiment.

Brief Description of the Drawings

Fig. 1 shows an example of events in the ISO-C singulation process.

Fig. 2 shows the content of the RN16 signal.

Fig. 3 shows an example of signal encoding for the 16-bit data part of an RN16 signal and for the SS signal.

Fig. 4 shows an example of events in the singulation process of an exemplary mode of the present invention.

Fig. 5 shows the simulation results on the performance of an exemplary mode of the present invention, compared with the performance of the ISO-C singulation process.

Detailed Description of the Invention

Embodiments are described more fully below with reference to the accompanying figures, which form a part hereof and show, by way of illustration, specific exemplary embodiments. These embodiments are disclosed in sufficient detail to enable those skilled in the art to practice the invention. However, embodiments may be implemented in many different forms and should not be construed as being limited to the embodiments set forth herein. The following detailed description is, therefore, not to be taken in a limiting sense in that the scope of the present invention is defined only by the appended claims.

In the ISO-C singulation process, the RN16 signal may be considered as a temporary identification signal generated by a tag for the reservation of an exclusive communication session with the reader. One embodiment of the present invention is similar to the ISO-C singulation process, i.e., each tag also transmits a temporary identification signal for the reservation of an exclusive communication session with the reader in a framed slotted Aloha framework. However, the temporary identification signal of the present invention is clearly different from the RN16 signal, because it is derived from different principle and results in different performance, which may be described as follows.

First of all, the characteristics of the RN16 signal should be described as follows. The RN16 signal consists of fixed preamble and random 16-bit data as shown in Fig. 2, where the fixed preamble is the same for all tags, while the random 16-bit data represents the 16-bit number chosen randomly by the underlining tag. Duties of the fixed preamble comprise helping the reader detect the selection status of each slot based on the following principles. (1) If the reader cannot detect significant variation of the receive signal, it would deem that the underlining slot has not been selected by any tags. (2) If the reader can detect significant variation of the receive signal and also the fixed preamble, it would deem that the underlining slot has been selected by only one tag. (3) If the reader can detect significant variation of the receive signal but

cannot detect the fixed preamble, it would deem that the underlining slot has been selected by multiple tags.

However, for the case the reader deems that the underlining slot has been selected by only one tag, the slot may have been actually selected by multiple tags if the conditions related to the signal transmission of the tags are very well matched. The effect of this rare undesirable situation can be mitigated by the use of the Ack signal that will specify only a single tag for the possible communication session.

The temporary identification signal of the present invention is not structured similarly to that of the prior arts, where the signal generally consists of preamble and data. It consists of only one part that represents a signature sequence that has been selected randomly by the underlining tag for the underlining round (and hence we will call it an SS signal), where the selection is from a set designed such that (1) the reader can detect different SS signals (i.e., representing different signature sequences) that have been transmitted on overlapping time durations, and (2) the reader can detect multiple SS signals of the same signature sequence that have been transmitted on different but overlapping time durations. It is well known that deriving such a set is actually an art of the CDMA principle.

According to one embodiment, the set of signature sequences is a 17-member subset of a Gold code (R. Gold, "Optimal binary sequences for spread spectrum multiplexing (Corresp.)," IEEE Transactions on Information Theory, 13 (4), 1967, pp. 619-621) of length 31 chips:

· signature sequence 1 = 1001 0000 1010 1110 1100 0111 1100 110;

signature sequence 2 = 0100 0101 1111 0110 0111 0000 1101 010;

signature sequence 3 = 1101 0101 0101 1000 1011 0111 0001 100;

signature sequence 4 = 1000 0111 0111 0111 0000 0100 1001 111;

signature sequence 5 = 1011 1111 0001 1 101 0100 0001 01 10 100;

signature sequence 6 = 1100 1111 1 100 1001 1100 1010 1000 010; signature sequence 7 = 0010 1110 0110 0000 1101 1101 0101 110;

signature sequence 8 = 0110 101 1 1001 01 10 1010 1 101 1000 100;

signature sequence 9 = 0110 0110 1101 11 10 0001 0011 0100 011;

signature sequence 10 = 0100 1001 01 10 1 101 1001 0101 11 10 001 ;

signature sequence 11 = 0101 11 10 1011 0100 0101 01 10 10.1 1 000;

signature sequence 12 = 0000 1100 1001 101 1 11 10 0101 001 1 Oi l;

signature sequence 13 = 1001 1101 11 10 0110 01 11 1001 0000 001 ;

signature sequence 14 = 1000 1010 0011 11 11 1011 1010 0101 000;

signature sequence 15 = 1010 0101 1000 1 100 0011 1 100 11 11 010;

signature sequence 16 = 001 1 1001 1011 1001 0001 1110 0000 1 11 ;

signature sequence 17 = 0000 0001 1101 0011 0101 101 1 1 1 11 100.

It should be noted that the above Signature Sequences are by way of example, therefor some other orthogonal sequences can also be used, such as Kasami codes, Walsh Codes, and the like.

In addition, the SS signal is obtained from the selected signature sequence by nonreturn-to-zero (NRZ) encoding at chip rate equal to twice the bit rate of the RN16 signal. Note that the RN16 signal is obtained based on FM0 encoding. Therefore, the SS signal would require similar bandwidth to that of the RN16 signal. Fig. 3 shows example of encoding for an RN16 signal (for the 16-bit data part, assumed to be 1011 0001 1111 0010) and for an SS signal (for the signature sequence 1, i.e., 1001 0000 1010 11 10 1 100 0111 1 100 110) on the same time axis.

For a better clarification on the different between the SS signal and prevalent information-bearing signal of the prior arts, we would like to add more explanation as follows. Typically an information-bearing signal in a digital communication system is structured as a packet consisting of a preamble and a data part as shown in Fig. 2. An important duty of the preamble is to help the receiver estimate the transmission signal parameters (e.g., signal timing) necessary for detecting the data part. Due to such duty, the preamble is typically designed to represent a sequence known to the receiver (while the data part is a signal representing a sequence unknown to the receiver). Accordingly, detecting of the information from a packet is typically done by first detecting the preamble (representing a known sequence) while deriving the transmission signal parameters that are best matched to the signal, and then detecting the content of the data part based on the derived transmission signal parameters. However, if we want each packet to carry just a little amount of information, we may design the packet structure to consist of only an information-bearing preamble (the SS signal may be considered as such a preamble) as follows. Instead of using a preamble obtained from a known sequence, we use a preamble obtained from a sequence randomly selected from a known small set, where the selection depends on the information to be carried. For example, if we want each packet to carry just 4 bits, we should design the set to have 2⁴ members and use the 4-bit information of a packet to determine which member of the set is to be used for obtaining the preamble of the packet. According to one embodiment in which the SS signal is obtained from a set of 17 members, we could say that an SS signal is an information-bearing preamble that carries log2(17) = 4.0875 bits.

In one embodiment, the SS signal transmission is subjected to a random delay. In each round of the singulation, each tag also selects a number (denoted as D) randomly from 0 to 13 (for example), and when the selected slot begins the tag shall wait for D chip durations before it transmits the SS signal. This is to decrease the probability that multiple tags selecting the same signature sequence transmit the SS signals simultaneously, which usually causes the reader to misunderstand that there is only one tag selecting the sequence.

Another embodiment implements a random delay in a slightly different way as the SS signal is obtained from the selected signature sequence by the following steps: (1) concatenating the selected sequence with a cyclic prefix of random length or with a cyclic prefix of random length and also a cyclic postfix; and (2) obtaining the SS signal from the concatenated sequence instead of the selected sequence alone. More explanation about the mentioned concatenation is as follows. Concatenating a signature sequence of length 31 chips (for example) with a cyclic prefix of random length may be done by the following steps: (1) selecting a number (denoted as D) randomly from 0 to 13 (for example); (2) obtaining the concatenated sequence as [s(31-D:30) s(0:30)] where s(n:m) is a part of the signature sequence consisting of the elements from the n-th element to the m-th element. It may be noted that the cyclic prefix here is s(31-D:30), and the corresponding length is D chips. Concatenating a signature sequence of length 31 chips (for example) with a cyclic prefix of random length and also a cyclic postfix may be done by the following steps: (1) selecting a number (denoted as D) randomly from 0 to 13 (for example); (2) obtaining the concatenated sequence as [s(31-D:30) s(0:30) s(0:12-D)]. It may be noted that the cyclic prefix and the cyclic postfix here are respectively s(31-D:30) and s(0:12-D), and their corresponding lengths are respectively D and 13-D chips.

The singulation process according to one embodiment of the present invention may be obtained by a modification from that of the ISO-C singulation process as follows. According to the embodiment, the reader issues a query command that contains a parameter indicating the number of available slots in the query round. The first slot begins just after the reader's query command, while other slots begin respectively just after the reader's QueryRep commands as in the ISO-C process. In each round, a tag among others participating in the round selects one slot out of the available slots and transmits its SS signal on the beginning of the selected slot. The reader attempts to detect the SS signals during an expected time interval, in which several tags may transmit, their SS signals. For each SS signal being detected as representing a signature sequence that has been selected for the underlining slot by only one tag, the reader begins a communication session with that possible tag by issuing an Ack command indicating the detected signature sequence. The underlining tag responds to the Ack command by first transmitting its identification code and then responding to the reader's other commands (if any) until the reader issues another Ack command indicating another signature sequence or the reader issues a QueryRep command ending the slot. When the reader finds no more tags to communicate in a slot, it issues a QueryRep command. One bit of the Ack and QueryRep commands shall indicate the status of the just- finished communication session within the slot in order to inform the underlining tag about whether it should change its state of singulation.

Fig. 4 illustrates a singulation process that uses the temporary identification signals of the present invention for the case that tags 0 and 1 similarly select slot 0, but they differently select signature sequences 3 and 4 respectively.

Although the present embodiment creates an opportunity that multiple temporary identification signals may be detected simultaneously, it also increases the chance that multiple tags selecting the same slot also select the same temporary identification signal (because there are only 17 choices for the SS signal). Therefore, it may not be obvious that the present invention will create advantage by a greater extent than disadvantage. Then, in order to clearly see the potential advantage of the present invention, the inventors have simulated a conventional detection algorithm (with the same principle as that cited as 'Conventional Detector' in S. Moshavi, "Multi-user detection for DS-CDMA communications," IEEE Communications Magazine, October 1996, pp. 124-136) for detecting the SS signals in a slot under the same noise condition as when the ISO-C-based system can reliably perform. As a result, by setting the number of slots in a round to be one third of the number of tags, it is found that for each slot (1) the average number of correctly detected SS signals is 2.1, (2) the average number of false alarms is 0.12, (3) the average number of incorrect detections, in the case that the detector detects a multiply selected signature sequence as a uniquely selected one, is 0.09, and (4) the probability that no SS signal is detected is 0.07. However, for an ISO- C-based system, by assuming perfect signal detection, i.e., (1) if multiple tags transmit the RN16 signals, the reader will know that the received signal is not usable, (2) if a single tag transmits the RN16 signal, the reader will understand the signal content, and (3) if no tag transmits the RN16 signal, the reader will know that the slot is empty, it is found that for each slot

(1) the average number of correctly detected RN16 signals is 0.37,

(2) the average number of false alarms is 0,

(3) the average number of incorrect detections, in the case that the detector detects a multiply selected slot as a uniquely selected one, is 0, and

(4) the probability that no RN16 signal is detected as a usable one is 0.63.

From the numerical results mentioned above, the inventors have computed the read speeds of the two systems based on a common operating condition (as assumed by You- Chang Ko, Sumit Roy, Joshua R. Smith, Hyong-Woo Lee, and Choong-Ho Cho, "RFID MAC Performance Evaluation Based on ISO/IEC 18000-6 Type C," IEEE Communications Letters, 12 (6), 2008, pp. 426-428). It is then found that, at the uplink raw data rate of 160 kbps and the downlink symbol duration (Tari) of 6.25 us, the present invention-based system is 64% faster than the ISO-C-based system.

The simulation result in the previous paragraph is presented only for a single operating condition. Actually the inventors have performed the simulation for various operating conditions from which the results are shown in Fig. 5. The figure shows the values of read efficiency (the ratio of tag's data reading rate over the uplink raw data rate) at various operating conditions (as determined by uplink raw data rate, Tari, and the ratio of the number of slots over the number of tags in a query round). From the figure, we should note that the present invention-based process will work best when the number of slots is less than the number of tags in each query round, while the ISO-C-based process will work best when the number of slots is equal to the number of tags in each query round. We should also conclude from Fig. 5 that the present invention-based process is faster the ISO-C-based process in every operating condition if we have properly set the ratio of the number of slots over the number of tags in a query round.

The embodiment we have described is only an exemplary one, which is based on the ISO-C framework of framed slotted Aloha. However, we may apply the present invention in a different framework, e.g., You-Chang Ko, Sumit Roy, Joshua R. Smith, Hyong-Woo Lee, and Choong-Ho Cho, "RFID MAC Performance Evaluation Based on ISO/IEC 18000-6 Type C," IEEE Communications Letters, 12 (6), 2008, pp. 426-428. By the description presented above, a preferred embodiment of the present invention uses at least (1) a preferred set of signature sequences that may be obtained from simulation experiments, (2) a preferred method of signal encoding that maps from a signature sequence to an SS signal, and (3) a preferred detection algorithm that may have the same principle as that of the Linear Multiuser Detector or the Subtractive Interference Cancellation Detector mentioned in S. Moshavi, "Multi-user detection for DS-CDMA communications," IEEE Communications Magazine, October 1996, pp. 124-136.

Although the invention has been described in connection with an embodiment, it should be understood that various modifications, additions and alterations may be made to the invention by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Preferred Embodiment

A preferred embodiment of the present invention is as described in the Detailed Description of the Invention.

Claims

1. An RFID singulation process between a single reader and a plurality of tags in an Frame- slotted ALOHA RFID system, comprising the steps of: (a) in each tag, upon receiving a query command from the reader, transmitting a temporary identification signal over a randomly selected time slot, and waiting for an acknowledgment signal from the reader indicating that the reader has detected the temporary identification signal from a particular tag as a selected tag; and (b) in the selected tag, establishing a communication between the selected tag and the reader and transmitting data comprising at least an identification data of the selected tag to the reader,

the process being characterized in that

the temporary identification signal comprising only a signature sequence that is randomly selected from a plurality of predetermined sequences by the tag, and the signature sequence is a orthogonal data sequence.

2. A method according to claim 1, wherein the detection process further has a duty of (a) detecting multiple temporary identification signals that represent the same signature sequence even though the signals have been transmitted on different and overlapping intervals, or (b) detecting multiple temporary identification signals that represent the same signature sequence but have been generated from mutually different sequences that have been resulted differently from a sequence-concatenation process comprising concatenating the signature sequence with a cyclic prefix of a selected length.

3. A method according to claim 1 , wherein the temporary identification signal is transmitted over a time slot within a slot that has been randomly selected by the tag.

4. A method according to claim 2, wherein the temporary identification signal is transmitted over an interval within a slot that has been randomly selected by the tag.

5. A method according to claim 4, wherein (a) the transmission of the temporary identification signal starts at a time that has been selected randomly from within the slot by the tag, or (b) the temporary identification signal has been generated from a sequence resulted from concatenating the selected signature sequence with a cyclic prefix of random length, or (c) the temporary identification signal has been generated from a sequence resulted from concatenating the selected signature sequence with a cyclic prefix of random length and also a cyclic postfix.