CN102521627A - Redundancy eliminating algorithm of RFID (radio frequency identification) system reader based on middleware - Google Patents

Abstract

The invention discloses a redundancy eliminating algorithm of an RFID (radio frequency identification) system reader based on a middleware. The method relies on an EPC (electronic product code) network architecture to utilizes label information of an RFID middleware to judge a redundant reader. The algorithm does not require system topology or require a reader to write information into a label, and has good flexibility and expandability. Compared with the typical algorithm LEO+RRE (Layered Elimination Optimization+ Recurrence Risk Estimator), the algorithm improves the detection rate of the redundant reader for 6.27-20.80%, and reduces data quantity of the labels to be obtained and processed by the system for 4.50-35.73%, so as to further optimize rationality of system deployment.

Description

RFID system reader redundancy removing algorithm based on middleware

Technical Field

The invention relates to the technical field of Radio Frequency Identification (RFID) networks, in particular to a middleware-based redundancy removing algorithm for an RFID system reader.

Background

The rfid (radio Frequency identification) is a system for automatically identifying and wirelessly acquiring data. With the rapid development of the electronic industry, the RFID technology has been widely applied to a plurality of fields such as supply chain management, medical monitoring, positioning navigation, smart home, etc., and is regarded as one of the core technologies for implementing the Internet of Things (Internet of Things) in the industry.

Redundant-reader-Problem is one of the fundamental problems affecting the performance of RFID systems. In a large-scale, especially densely deployed RFID system, the energy consumption of the RFID system can be reduced by judging and closing the redundant readers, the system deployment is optimized, and the quantity of the tag data which needs to be acquired and processed by the system is reduced. Therefore, the aim of the research on the redundancy removal of the RFID reader is to optimize the overall performance of the system by minimizing the number of working readers in a coverage area, which is beneficial to solving the problem of reader-level read-write collision and can also be used for guiding the planning and design of the RFID system. Two classic reader redundancy removal algorithms:

(1) RRE (redundant Reader Elimination) algorithm: the algorithm is an RRE algorithm based on a greedy strategy, the number of tags in a reader RA is taken as a weight, and the reader with the largest weight is ensured to preferentially lock all tags in the RA. The "locking" process refers to the multiple read-write communication process of each reader with its RA internal tag. Each reader determines whether to write locking information into the label or not by comparing the weight of the reader with the weight of the information returned by the label, the operation is iterated until all labels are locked by the reader, and the reader which does not lock any label is a redundant reader. But the RRE algorithm may fail under certain system topologies. Meanwhile, the operating characteristics of the reader and the tag for writing information one to one may cause the occurrence of a plurality of readers with the same weight, thereby causing the RRE algorithm to miss the redundant reader.

(2) LEO (layered animation optimization) algorithm: the algorithm is based on a first-come-first-obtain strategy, all readers read tags in RA sequentially in a layering mode, and each tag is ensured to be written with locking information by the reader which reads for the first time, and the reader which does not write information into any tag is a redundant reader. Compared with the RRE algorithm, the read-write times of the reader to the label in the LEO algorithm are greatly reduced, but the read sequence of the reader is random, so the reliability of the algorithm is poor. On the basis, an LEO author further provides an LEO + RRE algorithm, and the algorithm integrates the advantages of the two algorithms and has better redundancy removal performance and reliability. However, the communication mode of exchanging information between readers by reading and writing tags may cause the LEO (LEO + RRE) algorithm to miss the redundant reader or fail to achieve the optimal performance of the algorithm.

Meanwhile, many algorithms, including the RRE algorithm and the LEO algorithm, require the reader to write "lock" information to the tag, and this information can also be read by other readers. Thus, these algorithms all contain an implicit premise: the maximum distance a tag can be read by a reader (read distance) is equal to the maximum distance it can be written to (write distance). However, the chip nature of the RFID tag dictates that more power is consumed by the reader to write information to the tag, and thus the write distance of the reader is generally less than its read distance. The operating characteristics of unequal read-write distances of the reader to the tags can cause the situation that the redundancy removing algorithm fails to judge and misjudges the redundant reader, so that the system cannot be ensured to read all tag information. Therefore, various reader redundancy elimination algorithms, such as RRE, LEO, etc., which require the reader to write "lock" information to the tag, have unreliability in practical application environments. In addition, in a large-scale, especially wireless communication, RFID system, both the reader and the tag are generally resource-limited, and multiple times of reading and writing of the tag by the reader and the redundant discrimination function of the reader will increase the system load.

Therefore, in the implementation of the RFID system reader redundancy elimination algorithm, the operating characteristics of the RFID system have a large influence on the performance and reliability of the algorithm, and overcoming these adverse effects is a problem to be solved in the existing algorithm.

Disclosure of Invention

In order to solve the defects of the prior art that the redundant reader is not judged, the resources of the reader and the tag are generally limited in a large-scale RFID system, particularly in a wireless communication RFID system, the system load is increased by the multiple reading and writing of the tag by the reader and the redundancy judgment function of the reader, and the like, the invention provides the middleware-based RFID system reader redundancy removing algorithm which has higher detection rate of the redundant reader and can reduce the tag amount to be processed by the system.

The technical scheme of the invention is as follows: an RFID system reader redundancy removing algorithm based on middleware comprises the following steps:

s1 defines the MRT matrix with reader number R_iLine of action, tag number T_jWhen the reader R is column_iReadable label T_jLet the matrix element MRT (R)_i，T_j) If not, the value of (A) is 0;

s2, calculating the sum of element data of each row and the sum of elements of each column of the matrix MRT to respectively obtain a one-dimensional array MR and an MT;

s3 calculating element maximum value in array MR, if only one element corresponds to maximum value, recording the element reader number as R_minOtherwise, recording the element number set corresponding to the maximum value of the element in the array MR as RCS;

s4 records the number of the same tags in the reader and its peripheral readers in the RCS as RRF, calculates the RRF value of each reader in the RCS, and calculates the reader R with the minimum RRF value_minSelecting the reader as a working reader;

s5 reaction of R_minUpdating the corresponding element value in the matrix MRT to be 0;

s6 updating the array MR data, repeating the steps S3 to S5 until the MRT matrix is empty, and the reader which is not designated as the working state at this time is the redundant reader.

In the above scheme, the matrix element MRT (R)_i，T_j) Is defined asThe following:

<math> <mrow> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Subset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&NotSubset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mi>RS</mi> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mi>TS</mi> <mo>)</mo> </mrow> </mrow> </math>

wherein, RS and TS are reader set and label set stored in the middleware in a certain time period respectively,indicating that the reader can read the corresponding tag, otherwise, indicating that the tag is not within the reader Radius (RA).

Further, the one-dimensional arrays MR and MT are respectively defined as follows:

<math> <mrow> <mi>MR</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>j</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS

<math> <mrow> <mi>MT</mi> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>i</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS

the element number in the array MR is the number of the reader, and the element value is the number of the tags in the corresponding reader RA; the element numbers in the array MT represent tag numbers, and the element values are the number of readers capable of reading corresponding tags.

Further, the definition of the element (reader) number set RCS corresponding to the maximum value of the element in the array MR is as follows:

RCS＝{R_i∈RS|MR(R_i)＝max(MR)}

further, the RRF calculation method of the readers in the RCS set is as follows:

<math> <mrow> <mi>RRF</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>m</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>j</mi> </munder> <mo>[</mo> <mi>MT</mi> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>-</mo> <mi>CoT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>m</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> <mo>]</mo> <mo>&CenterDot;</mo> <mi>CoT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>m</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </mrow> </math> R_m∈RCS

wherein R is_mFor readers in the RCS set, CoT is a comparison matrix of MRTs, i.e., CoT ═ MRT.

Said R_minThe definition is as follows:

$R_{\min }=\arg \underset{m}{\min }\mathrm{RRF}\left(R_m\right)$

wherein R is_minThe representation is that in all reader sets RCS containing the same maximum number of tags, the reader has the same minimum number of tags with all the readers around the reader, namely the reader contains the maximum number of tags alone。

Said R_minThe update formula of the corresponding element in the matrix MRT is as follows:

R_i∈RS，T_j∈TS

in the scheme, the redundancy removing algorithm of the RFID system reader based on the middleware does not need the reader to read the tag for multiple times in the redundancy removing process, and does not need the reader to write any information into the tag.

Compared with the prior art, the invention has the beneficial effects that: by means of the EPC network architecture, the algorithm obtains the tag information and judges the redundant reader by using the RFID middleware in the EPC network architecture. Different from the existing redundancy removing algorithm, the algorithm does not need a reader to read and write information for a plurality of times on the label, so that the influence of the working characteristics of the RFID system is avoided. In addition, the algorithm can also be applied to an RFID system with tags without storage capacity. Compared with the RRE algorithm and the LEO + RRE algorithm, the MRRE algorithm has higher detection rate of redundant readers, effectively reduces the label amount to be processed by the system and further optimizes the rationality of system deployment. Compared with the classical algorithm LEO + RRE, the algorithm improves the detection rate of the redundant reader by 6.27-20.80%, reduces the quantity of label data required to be acquired and processed by the system by 4.50-35.73%, and further optimizes the rationality of system deployment.

Drawings

FIG. 1 is a schematic diagram of an EPC network architecture upon which the present invention is based;

FIG. 2 is a schematic diagram of the algorithm flow of the present invention.

Detailed Description

The invention is further described below with reference to the accompanying drawings.

The invention relates to a redundancy removing algorithm of an RFID system reader based on middleware, which utilizes the RFID middleware in an EPC network architecture to obtain label information and judge a redundant reader. The EPC network is one of typical applications of the internet of things based on the RFID technology, and integrates an RFID system and an internet application, and a network architecture thereof is shown in fig. 1, and realizes real-time remote tracking of a tag by acquiring and retrieving a unique EPC Code (Electronic Product Code) of the tag. The RFID Middleware (Middleware) IS an important component of an EPC network, IS a novel software system, mainly obtains tag data through a reader, performs operations such as aggregation, filtering and storage on the data, and provides high-quality data flow for EPC application service and EPC information service (EPC IS).

[0038] The MRRE algorithm provided by the invention does not need to know the system topology, and the requirements of the algorithm on the RFID system are as follows:

the reader can read all the tag information in the RA, and the reader is not required to write information into the tags;

the RFID middleware may receive and store tag information transmitted by each reader.

The storage format of the RFID middleware tag information is set as < R, T, timestamp >, that is, a reader with the serial number of R reads the tag information with the serial number of T at a certain time (timestamp). The MRRE algorithm requires middleware to build a "reader-tag" matrix MRT, defined as follows:

the reader set and the label set stored in the middleware in a certain time period are respectively marked as RS and TS, and the reader number R is used_iLine of action, tag number T_jFor a column, a matrix MRT is constructed and matrix elements MRT (R) are defined_i，T_j) Is composed of

<math> <mrow> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Subset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&NotSubset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mi>RS</mi> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mi>TS</mi> <mo>)</mo> </mrow> </mrow> </math>

Wherein,

indicating that the reader can read the corresponding tag, otherwise, indicating that the tag is not in the reader RA, and defining a comparison matrix CoT as MRT.

The MRRE algorithm redundancy removal step is shown in fig. 2:

(step 201) calculating the sum of the element data of each row of the matrix MRT to obtain a one-dimensional array MR:

<math> <mrow> <mi>MR</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>j</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS

calculating the sum of the element data of each row of the matrix MRT to obtain a one-dimensional array MT:

<math> <mrow> <mi>MT</mi> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>i</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS

the element number in the array MR is the number of the reader, and the element value is the number of the tags in the corresponding reader RA; the element numbers in the array MT represent tag numbers, and the element values are the number of readers capable of reading corresponding tags.

(step 202) calculating the maximum value of the element in the array MR, wherein a plurality of elements possibly exist in the array corresponding to the maximum value. If only one element corresponds to the maximum value, the element (reader) is marked as R_minOtherwise, the element (reader) number set corresponding to the maximum value of the element in the array MR is recorded as RCS:

RCS＝{R_i∈RS|MR(R_i)＝max(MR)}

(step 203) recording the reader R in the set RCS_mThe number of tags having the same number as its surrounding readers is RRF:

R_m∈RCS

to obtain

$R_{\min }=\arg \underset{m}{\min }\mathrm{RRF}\left(R_m\right)$

Wherein R is_minThe representation shows that in all reader sets RCS containing the same maximum number of tags, the reader has the same number of tags with all the readers around the reader, namely the reader has the maximum number of tags contained independently. Therefore, it is least likely to be a redundant reader and should be selected as the working reader first.

(step 204) the data in the update matrix MRT is:

R_i∈RS，T_j∈TS

(step 205) updating the array MR data, and repeating (step 202) to (step 204) until the matrix MRT is a null matrix, wherein the reader which is not designated as the working state at the moment is a redundant reader.

Compared with the classical algorithm LEO + RRE, the algorithm improves the detection rate of the redundant reader by 6.27-20.80%, reduces the quantity of label data required to be acquired and processed by the system by 4.50-35.73%, and further optimizes the rationality of system deployment.

In summary, those skilled in the art may make various changes and modifications to a middleware based RFID system reader redundancy elimination algorithm of the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (8)

1. An RFID system reader redundancy removing algorithm based on middleware is characterized by comprising the following steps:

s1 defines the MRT matrix with reader number R_iLine of action, tag number T_jWhen the reader R is column_iReadable label T_jLet the matrix element MRT (R)_i，T_j) If not, the value of (A) is 0;

s2, calculating the sum of element data of each row and the sum of elements of each column of the matrix MRT to respectively obtain a one-dimensional array MR and an MT;

s3 calculating element maximum value in array MR, if only one element corresponds to maximum value, recording the element reader number as R_minOtherwise, recording the element number set corresponding to the maximum value of the element in the array MR as RCS;

s4 records the number of the same tags in the reader and its peripheral readers in the RCS as RRF, calculates the RRF value of each reader in the RCS, and calculates the reader R with the minimum RRF value_minSelecting the reader as a working reader;

s5 reaction of R_minUpdating the corresponding element value in the matrix MRT to be 0;

s6 updating the array MR data, repeating the steps S3 to S5 until the MRT matrix is empty, and the reader which is not designated as the working state at this time is the redundant reader.

2. The middleware-based RFID system reader redundancy elimination algorithm of claim 1, wherein: the matrix element MRT (R)_i，T_j) The definition is as follows:

<math> <mrow> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>1</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Subset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&NotSubset;</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mi>RS</mi> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mi>TS</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math>

wherein, RS and TS are reader set and label set stored in the middleware in a certain time period respectively,

indicating that the reader can read the corresponding tag, otherwise, indicating that the tag is not within the reader Radius (RA).

3. The middleware-based RFID system reader redundancy elimination algorithm of claim 2, wherein: the one-dimensional arrays MR and MT are respectively defined as follows:

<math> <mrow> <mi>MR</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>j</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS，

<math> <mrow> <mi>MT</mi> <mrow> <mo>(</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&Sigma;</mi> <mi>i</mi> </munder> <mi>MRT</mi> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>T</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> R_i∈RS，T_j∈TS，

the element number in the array MR is the number of the reader, and the element value is the number of the tags in the corresponding reader RA; the element numbers in the array MT represent tag numbers, and the element values are the number of readers capable of reading corresponding tags.

4. The middleware-based RFID system reader redundancy elimination algorithm of claim 3, wherein: the definition of the element (reader) number set RCS corresponding to the maximum value of the element in the array MR is as follows:

RCS＝{R_i∈RS|MR(R_i)＝max(MR)}。

5. the middleware-based RFID system reader redundancy elimination algorithm of claim 4, wherein: the RRF calculation method of the readers in the RCS set is as follows:

wherein R is_mFor readers in the RCS set, CoT is a comparison matrix of MRTs, i.e., CoT ═ MRT.

6. The middleware-based RFID system reader redundancy elimination algorithm of claim 5, wherein: said R_minThe definition is as follows:

$R_{\min }=\arg \underset{m}{\min }\mathrm{RRF}\left(R_m\right),$

wherein R is_minThe representation shows that in all reader sets RCS containing the same maximum number of tags, the reader has the same number of tags with all the readers around the reader, namely the reader has the maximum number of tags contained independently.

7. The middleware-based RFID system reader redundancy elimination algorithm of claim 6, wherein: said R_minThe update formula of the corresponding element in the matrix MRT is as follows:

R_i∈RS，T_j∈TS。

8. the middleware-based RFID system reader redundancy elimination algorithm of any one of claims 1 to 7, wherein: the RFID system reader redundancy removing algorithm based on the middleware does not need a reader to read the tag for multiple times in the redundancy removing process, and does not need the reader to write any information into the tag.

Images