CN103942514A - RFID collision avoiding method based on related IDs - Google Patents

Abstract

The invention provides an RFID collision avoiding method based on related IDs. The RFID technique is one of key techniques required to be researched in the internet-of-things perception layer, is an identity card of an object in real life, and is a core foundation link achieving the internet of things. RFID labels send data to a reader-writer in a modulation binary pulse mode; in a conventional RFID label identification method, the reader-writer sends a recognition command to actively search for the labels, each label will occupy N average recognition times, and the N is greater than 1. According to the RFID collision avoiding method based on the related IDs, the incidence relation among the labels is enhanced, so that the labels can actively send the IDs of themselves under a certain trigger condition. The RFID collision avoiding method is applied to the deterministic collision avoiding circumstance based on a multi-branch search tree, the labels can be recognized in single communication, recognizing efficiency is greatly improved, and collision of the labels is avoided.

Description

RFID collision avoidance method based on associated ID

Technical field

The invention belongs to the technical field that computer utility is combined with Internet of Things.

Background technology

Internet of Things will be realized thing and thing, what between thing and people, engage in the dialogue is mutual, thing just equally needs a legal identity with people, realize legalizing of physical world thing identity, RFID(is radio-frequency (RF) identification) technology is crucial, its radio-frequency (RF) tag has unique electric product coding, can be used as " I.D. " of object in actual life object is carried out to Division identification.

RFID (Radio Frequency Identification), it is radio-frequency (RF) identification, it is a kind of contactless automatic identification technology rising the nineties in 20th century, identifying is without manual intervention, utilize radiofrequency signal to realize the non contact transmission of information by Space Coupling, and reach the object of identification by institute's transmission of information.A complete rfid system mainly comprises following three parts: label (Tag), and read write line (Reader) and application controls system, application controls system is generally computer network system.Read write line sends the radiofrequency signal of certain frequency, in the time that electronic tag enters the magnetic field of read write line perform region, activated with the form of radiofrequency signal self entrained coded message is sent out by read write line, read write line receives information and sends computer network system to, computer system judges the legitimacy of this information etc. according to certain operation rule, make corresponding judgement and control according to entrained information, thereby complete the once overall process of identification.

RFID electronic tag (Tag) is made up of small label chip and antenna conventionally, there is the function of intelligent read and coded communication, Tag is attached to the mark as object on object conventionally, because Tag has unique electronic code (Electronic Product Code), this coding can be used as No. ID, unique I.D. of this label of identification.Compare with bar code in the past, RFID electronic tag has the dust of not being afraid of, humidity, and break flour, the character such as high temperature, not affected by environment, the life-span is longer.

Rfid interrogator (Reader), is the equipment that reads (being again to write) label information, and a typical read write line is conventionally by control system, high-frequency model (transmitter, receiver), and antenna, the compositions such as interface, can be designed to hand-held or fixed.Read write line reads the data of RFID label or write, and can accomplish good encryption.

When rfid system work, read write line is by the radiofrequency signal of antenna transmission certain frequency, in the time that entering read write line perform region, passive (or passive) label (Passive Tag) is activated owing to producing induction current, the energy that relies on induction current to obtain sends the product information being stored in label by the built-in antenna of label, and active (or initiatively) label (Active Tag), all the time in activating, initiatively duty, initiatively send out the radiofrequency signal of a certain frequency, read write line received RF signal, after reading information decoding, information is delivered to computer system, computer system judges the legitimacy of this radio-frequency (RF) tag according to certain logical operation, make corresponding processing and control for different settings.

Rfid system has following three kinds of fundamental modes: full duplex (Full Duplex) and half-duplex (Half Duplex) system and sequential (SEQ) system.In full duplex rfid system, electronic tag and read write line can transmit information mutually at synchronization, half-duplex RF ID system electronic label and read write line also can two-way transmission information, but synchronization can only a direction pass, these two kinds of working methods are all read write line emitting electromagnetic waves, and electronic tag is made certain response.Sequential working method is electromagnetic field short time, the periodically disconnection that read write line gives off, and this spacer segment of electronic label identification, utilizes these intervals to realize the data transmission between electronic tag and read write line, is a kind of typical Radar operation modes.

The anticollision problem of RFID is mainly divided into two classes, and a class is the collision of read write line, and a class is the collision of label.The collision of so-called label; be that read write line is in work; synchronization, has in the read-write reach that large-scale label is present in read write line, in the time having multiple Unidentified label in read write line reach conventionally; each label can respond the querying command of read write line; transmission information is to read write line, and the same time of read write line can only be carried out real-time Communication for Power with a label, in the time that multiple labels send order to read write line simultaneously; any one label of read write line None-identified, produces collision.

The collision of read write line mainly contains three types:

(1) interference of frequency.When read write line work, can be centered by read write line, outwards radiated electromagnetic wave, in the time that two or more read write line distances frequency of operation too near or two read write lines is close, read write line is to can mutually producing and disturb between the electromagnetic signal of label transmitting, cause label successfully to identify, this phenomenon is called the interference of read write line frequency.

(2) interference of label.In the time that label is positioned at the reach of two or more read write lines simultaneously, different read write lines all can send read-write to label, in the time that label receives, can there is the overlapping of signal in the signal that at this moment different read write lines sends, cause label can not correctly receive the read write command of one of them read write line, can not send out correct answer signal, at this moment this label will be by skip, the different read write line interference problem to label that Here it is.

(3) interference of concealed terminal.In actual rfid system, due to the limitation of communication distance between read write line and label, conventionally in limited scope, need the large-scale read write line of disposing, this has all strengthened the collision problem of read write line, thereby in the time disposing RFID application system, the collision problem of read write line is also the factor that must consider.

What between rfid interrogator and label, carry out is short-distance wireless communication, the collision problem solving between its label mainly contains based on FDMA, CDMA, the anti-collision algorithm of SDMA and TDMA, due to the restriction of RFID label especially passive label power and function, adopt FDMA, CDMA, the anti-collision algorithm of SDMA thought design can make the hardware design complexity of read write line, increase cost, thereby common employing is the anti-collision algorithm based on the design of TDMA thought, at present, label anti-collision algorithm based on TDMA is broadly divided into two classes: a kind of is random algorithm based on ALOHA agreement, another kind is the definite algorithm based on search tree.Random algorithm based on ALOHA agreement mainly contains three major types: ALOHA algorithm, Frame Slotted Aloha algorithm, dynamically Frame Slotted Aloha algorithm; Definite algorithm based on search tree mainly contains radix-2 algorithm, inquiry tree algorithm, random tree generating algorithm etc.

Algorithm based on ALOHA class can produce the phenomenon that label " is died of hunger ", the label having in read write line identification range is not successfully identified by read write line, occur skip phenomenon, therefore the algorithm based on ALOHA class is mainly applicable to the less application scenario of number of tags among a small circle.And algorithm based on search tree class is deterministic algorithm, can ensure that the whole labels in read write line reach can both all be identified, but having relatively high expectations to read write line.

Summary of the invention

The present invention seeks to reduce the collision problem of label, improve the recognition efficiency of system.The present invention on the basis of having understood RFID gordian technique primary study the anti-collision algorithm of RFID label, new method is avoided in the RFID collision proposing based on associated ID, can be applied in the RFID technology of different frequency range, as logistic storage, taking care of books, the various fields such as military management, can effectively avoid tag-collision and improve system recognition rate.

RFID collision avoidance method based on associated ID provided by the invention, mainly comprises following committed step:

1st, increase the incidence relation between label, that is: set up certain father between label, sub-mapping relations, in the time that father's label is searched, subtab will be triggered to send the ID of oneself;

1.1st, read write line sends Article 1 instruction: " GET_IDUNIT ", all labels are received after instruction will be by compare with self ID, if identical, will corresponding be converted to binary pulse and send to read write line;

1.2nd, read write line receives the pulse that label sends, and value corresponding pulse is stacked respectively, and from large to small, same value is only once stacked for stacked order;

1.3rd, read write line is popped stack top location, and adds in order parameter gParam, and parameter l evel is updated to the nlevel value of current new stack top location, and bconflict=0 identifies part label successively;

2nd, all labels of having identified detect whether there is father's label, if having, on its corresponding father's tag pulses position, put 1, and send this binary pulse, claim that this pulse is subtab identification pulse;

2.1st, read write line receives after this subtab identification pulse, just can judge which label and have subtab, and read while write device this subtab identification pulse is sent to label,

2.2nd, all subtabs judge the subtab which epicycle has and can read, and according to all subtabs the position in pulse train, corresponding time delay be set send successively oneself ID;

2.3rd, read write line receives subtab ID successively, until parameter stack is empty, all labels complete identification.

Suppose existing ID_0 to ID_8 totally 9 labels,, in introduced algorithm, the identifying of ID as shown in Figure 1.

In Fig. 1, dotted line has represented the identification path of algorithm.From figure, be not difficult to find out, in multiway tree algorithm, all labels are all comprised on identification path, suppose that the number of communications of successfully having identified these 9 labeling requirements is .

In multiway tree algorithm, introduce the relationship maps relation between label, for the reliability that ensures that algorithm design is analyzed, still adopt 9 ID used in Fig. 1, the mapping relations of all labels adopt random fashion to generate.Suppose the mapping relations of ID wherein and identification path as shown in Figure 2.

In figure, Fat has indicated father's label position.As shown above, ID_4, ID_5 taking ID_1, ID_2 as father's label, are also identified respectively when ID_1, ID_2 are identified.Therefore identify number of times and reduced 1 time, identification path has also shortened.

In Fig. 2, the mapping relations of father's label and subtab are Random Maps, and as can be seen from Figure 2, father's label ID_7 is identified after subtab ID_3, and therefore, the mapping relations between two labels do not play a role.Discuss for the mapping relations between further optimization label below.

Not hard to imagine, if all mapping relations will be used as much as possible, before will ensureing that so father's label will appear at subtab.Simulation process below uses restraint to mapping relations, that is: require the mapping relations of all random generations will ensure that father's label ID of all labels except first label is less than the ID of label itself.The label ID choosing is identical with Fig. 2, again plans mapping relations by above-mentioned constraint, and the identifying of label as shown in Figure 3.

ID_3, ID_4, ID_5 can both be recognized in advance, therefore in read write line access the 2. when number branch, all labels can not be replied, therefore with respect to the original multiway tree algorithm shown in Fig. 1, number of communications has reduced 4 times altogether, improvement multiway tree algorithm shown in comparison diagram 2, identification number of times has reduced 3 times, and efficiency will inevitably get a greater increase.

advantage of the present invention and good effect

Collision based on associated ID provided by the invention avoids new method can effectively solve tag-collision problem, improves system identification efficiency.

The method is applied in the determinacy anticollision situation based on multi-fork search tree, and its benefit is that single communication can be identified multiple labels simultaneously, has improved greatly recognition efficiency, has avoided tag-collision; The method is applied in probabilistic ALOHA anticollision situation, and read write line can go out according to the position judgment of pulse the position of empty slot, thereby in the time reading time slot, can avoid the tag-collision problem causing owing to reading empty slot.Theoretical and experiment shows, the method is not subject to ID length restriction, and has higher recognition efficiency, under the situation of or many labels less in number of labels, recognition efficiency is all more than 85%, in the time that number of tags is more, efficiency can exceed 90%, has effectively avoided the collision problem of label.

This collision avoids new method invention to can be applicable to multiple RFID technical applications, for example, logistic storage, retailing, industry is manufactured, apparel industry, medical control, traffic administration, food security, animal identification, taking care of books, aviation management, military management, mine safety etc.

[brief description of the drawings]

Fig. 1 is the recognition sequence of ID in multiway tree algorithm.

Fig. 2 is multiway tree algorithm identified path and the mapping relations figure introducing after mapping relations.

Fig. 3 doses identification path and the mapping relations figure after constraint to mapping relations.

Fig. 4 is multiway tree collision-proof method read write line workflow diagram.

Fig. 5 is multiway tree collision-proof method label workflow diagram.

Fig. 6 is label ID arrangement plan.

[embodiment]

In method for designing of the present invention, as shown in Figure 4, label workflow as shown in Figure 5 for read write line workflow.

All labels will detect in the label recognizing, and whether comprise father's label, if having, in pulse corresponding to father's label, put 1, and send this binary pulse, claim that this pulse is subtab identification pulse.Read write line receives after this subtab identification pulse, just can judge which label and have subtab, read while write device and also this subtab identification pulse will be sent to label, all subtabs also can be judged the subtab which epicycle has to read, and the position according to all subtabs in pulse train, arranges corresponding time delay and sends successively oneself ID.

The concrete identifying of the present invention:

We are with the identifying that is identified as example and describes in detail algorithm for design of 10 labels.Suppose label length is 16bit, and every 4bit forms one , have 4 , be divided into 4 , therefore in process flow diagram value is 4, is provided with label ID configuration as shown in Figure 6.

(1) read write line sends Article 1 instruction: " GET_IDUNIT ", wherein order parameter: =0, =NULL; All labels are received after instruction will be by with self compare, wherein be equivalent to prefix, if identical, will to just be converted to binary pulse and send to read write line, in this instruction, because of value is 0, i.e. requirement send , there is no prefix, therefore before for sky, all all return .

(2) it is all stacked that the label that read write line is received is replied result.

(3) read write line is popped the stack top location content of result stack, join in order parameter, there is level=0+1=1, gParam is that current nIDUNITValue value is 1, due to level<M-1=4-1=3, therefore next instruction is still: " GET_IDUNIT ", and wherein order parameter is: =1, =1; All RFID labels will with self after contrast, only have ID_3 have with identical, therefore will =1 is specified =9 send out.

(4) after read write line receives, stack top location is popped, the nIDUNITValue of this unit is joined in order parameter gParam, there is level=1+1=2, gParam is 1_9, due to level<M-1=4-1=3, therefore next instruction is still: " GET_IDUNIT ", and wherein order parameter is: =2, =1_9; All RFID labels will with self after contrast, only have ID_3 have with identical, therefore will =2 is specified =12 send out.

(5) after read write line receives, stack top location is popped, the nIDUNITValue of this unit is joined in order parameter gParam, have level=2+1=3, gParam is 1_9_12, current + 1 rear value is 3, equals M, gets in RFID last , that is, and after obtaining LVL3, can basis with send after this instruction the data of replying draw complete label .Instruction type is: " GET_ID ", and order parameter is: =3, =1_9_12; All do contrast after ID_3 prefix with identical, therefore ID_3 returns to LVL3, has completed once identification.

After ID_3 successfully identifies, because the father's label shining upon in ID_5 is ID_3, therefore, send after self ID at ID_3, after time delay T finishes, ID_5 is triggered, and also sends himself ID, has completed " sending function automatically " of label.ID_5 successfully identifies.

Identifying and the said process of residue label are basic identical, until complete the identification of all labels.

Claims (1)

1. the RFID collision avoidance method based on associated ID, is characterized in that the method mainly comprises following committed step:

1st, increase the incidence relation between label, that is: set up certain father between label, sub-mapping relations, in the time that father's label is searched, subtab will be triggered to send the ID of oneself;

1.1st, read write line sends instruction: " GET_IDUNIT ", all labels are received after instruction will be by compare with self ID, if identical, will corresponding be converted to binary pulse and send to read write line;

1.2nd, read write line receives the pulse that label sends, and value corresponding pulse is stacked respectively, and from large to small, same value is only once stacked for stacked order;

1.3rd, read write line is popped stack top location, and adds in order parameter gParam, and parameter l evel is updated to the nlevel value of current new stack top location, and bconflict=0 identifies part label successively;

2nd, all labels of having identified detect whether there is father's label, if having, on its corresponding father's tag pulses position, put 1, and send this binary pulse, claim that this pulse is subtab identification pulse;

2.1st, read write line receives after this subtab identification pulse, just can judge which label and have subtab, and read while write device this subtab identification pulse is sent to label,

2.1st, all subtabs judge the subtab which epicycle has and can read, and according to all subtabs the position in pulse train, corresponding time delay be set send successively oneself ID;

2.3rd, read write line receives subtab ID successively, until parameter stack is empty, all labels complete identification.