CN103927633B - A kind of Item Information traceability system perceived with relevance and method - Google Patents

Abstract

The invention discloses an article information traceability system and method with correlation perception. The information server accepts the raw RFID data processed by the middleware through the capture interface; combines logic processing to form RFID identification data and stores it in the information database; publishes the RFID mapping data to the discovery server through the registration interface; opens the RFID identification data to the application through the query interface Shared query. The discovery server accepts the release of RFID mapping data from the information server through the registration interface; stores the RFID mapping data in the mapping database after correlation analysis; provides sharing of RFID mapping data between different discovery servers through the communication interface; opens to applications through the query interface Query for RFID mapping data. The present invention ensures the integrity of the retrospective query results on the basis of fully considering the dynamic association of RFID data caused by the change of inclusive relationship such as combination or split among items.

Description

An item information traceability system and method with correlation perception

technical field

The invention belongs to the technical field of computer networks, and in particular relates to an article information traceability system and method with correlation perception.

Background technique

The International Organization for Standardization GS1 has given a clear definition of traceability, that is, whether the history, location and other application conditions of an item can be queried and verified through identification records. It can be seen that the core issues that traceability services are most concerned about include two points, that is, how to identify an item and how to find the corresponding identification record data.

Radio-frequency identification technology (Radio-frequency identification, RFID) is such an automatic identification and data collection technology without direct contact, which can be used to track and locate target objects. Its main features are:

1) For the unique identification problem: using RFID technology as the means of item identification, compared with traditional one-dimensional barcodes and two-dimensional barcodes, the biggest advantage is that it can automatically identify target objects and obtain relevant data through radio frequency signals. The entire identification process Virtually no human intervention is required.

2) For the identification record search problem: you can choose to store the identification record directly in the RFID tag based on the "data-on-tag" mode; you can also choose to store the identification record in the network based on the "data-on-network" mode , and then use the mapping relationship between the RFID identifier and the network storage address as the routing prompt for the search.

Therefore, RFID technology has received high attention from academia and industry in recent years, and has been widely used in many application fields such as logistics management, food traceability, anti-counterfeiting verification, product recall, etc. A ubiquitous network for tracking and location.

At present, systems that provide traceability services based on RFID technology mainly include the EPCglobal network proposed by the EPCglobal organization, the DIALOG system, and the ePedigree system. However, the existing schemes generally lack the consideration of the correlation between RFID data, which leads to the impact on the traceability service quality (integrity). The disadvantage of the existing schemes is that they do not pay enough attention to the inclusive correlation between RFID data, and only use the inclusive correlation as a basis for RFID data compression, but do not consider the analysis of inclusive correlation in the process of traceability. . As a result, if the query function involves the correlation between data, the existing scheme cannot guarantee the integrity of the query results.

As shown in Figure 1, the drug supervision application based on RFID technology is taken as an example. The complete life cycle of a drug includes four stages of "production-transportation-sales-use". Assume that every participant in the supply chain, from manufacturers, large distributors, small distributors, retailers to buyers, maintains its own information data independently. The RFID tag adopts EPC code, and the item itself, the case and the pallet carrying the item will be affixed with different RFID tags. A single item (identified as SGTIN1) is first packed into a box (identified as SSCC1) at <L1, T2>, and then the box is placed on a container pallet (identified as SSCC2) at <L2, T4>. After the item is transferred from a large distributor to a small distributor, it undergoes packaging in another container pallet (identified as SSCC3). The final item is sold to ordinary buyers. In this way, RFID data may be distributed to multiple different servers for storage, and there will be dynamic containment associations between tags. For ordinary users, they do not know the combination and disassembly process of items in the supply chain.

For example, the drug regulatory agency received reports from the masses and found that although the drug with the single product label SGTIN1 has not expired, it has deteriorated and lost its efficacy. After analysis, the most likely reason for the deterioration of the drug is that it was not guaranteed to be stored in a low temperature environment of 2-10°C during transportation. In order to locate the specific location of the deterioration and the relevant responsible person, so as to avoid similar problems in future transportation, it is necessary to trace the temperature data of each logistics warehouse during the transportation of medicines. If the medicines are transported uniformly in large containers, each warehouse keeps the historical records of the containers. Then only querying the item information data associated with the identifier SGTIN1 cannot solve this actual demand.

Contents of the invention

Aiming at the above problems, the present invention provides an item information traceability system with correlation perception and a corresponding traceability query method, which ensures traceability on the basis of fully considering the dynamic association of RFID data caused by changes in the inclusive relationship between items such as combination or split. Integrity of query results. In the case where the user only inputs the item ID, not only the related item information of the ID is returned, but also all container information with temporal characteristics that have occurred with the item.

The technical scheme that the present invention adopts is as follows:

An item information traceability system with relevance perception, which includes at least one discovery server and at least one information service system;

The discovery server includes:

The registration interface is connected with the information server in the information service system to receive the RFID mapping data release;

Query interface, open to applications for querying RFID mapping data;

A query processing module, configured to process a query request for RFID identification;

The correlation analysis module is used to analyze the containment relationship that may exist between the mapping data of the item RFID identification, and perform correlation analysis;

The mapping database is used to store the mapping relationship between the RFID identification of the item obtained through correlation analysis and the address identification of the information server storing the item information;

The information service system includes:

The reader is used to obtain the item identification carried in the RFID tag through radio frequency communication;

The middleware is used to preprocess the raw RFID data acquired by the reader and form RFID events with certain semantics;

The information server includes: a registration interface, which is connected with the registration interface of the discovery server to register and issue the RFID mapping data; a capture interface, which is connected with the middleware to receive the RFID identification uploaded by it; a query interface, which is opened to the application program to perform Shared query of RFID identification data; information database for storing RFID identification data processed by middleware.

Further, the discovery server also includes a communication interface, which is open to other discovery servers for sharing mapping data between discovery servers.

Further, the discovery server further includes a filtering module, configured to use time attribute information to filter the association analysis to avoid information redundancy.

Further, the preprocessing performed by the reader on the original RFID data includes filtering processing and deduplication processing.

A method for tracing item information with correlation perception using the above-mentioned system, the steps of which include:

1) The information owner registers the RFID event records owned by the discovery server, and announces the corresponding security policy at the same time;

2) The query requester initiates a query request for the RFID identification to the discovery server, and submits the identity capability certificate at the same time;

3) The discovery server retrieves the mapping record with the RFID identification as the key in step 2), and performs correlation analysis to find the mapping record of the container identification associated with the RFID identification;

4) The discovery server checks whether the inquirer has legal access rights to the information data corresponding to the RFID identification; if it is an authorized legal user, it returns an accessible list of relevant addresses to the query requester, otherwise it refuses to provide services or directly Discard the query request;

5) According to the address list obtained in step 4), inquire about all information servers corresponding to the address of the requester and establish a communication connection, and simultaneously submit identity certificates to all information servers respectively;

6) After all the information servers pass the authentication, the information data of the items associated with the RFID tags will be returned to the query requester.

The present invention proposes a new traceability system architecture, including RFID tags, readers, middleware, information services, discovery services, and applications; the data structure of traceability information is established, including the original records of the readers, the information stored in the information server Identification records and mapping records required for query, etc.; an improved discovery service model (directory service model) is proposed, and a correlation analysis module of RFID data and a redundant data filtering module are added; and the corresponding retrospective query algorithm: According to The item ID entered by the user will return the related item information of the ID, as well as the information of all container items with temporal characteristics that have occurred with the item. Compared with the prior art, the present invention not only returns the related item information of the item when the user only inputs the item ID, but also returns all container information with temporal characteristics that have occurred with the item; while ensuring that the query results While ensuring completeness, it also ensures its accuracy, and filters through time attributes to avoid redundancy.

Description of drawings

Figure 1 is a schematic diagram of the application of drug supervision based on RFID technology;

Figure 2 is a traceability system architecture diagram with correlation perception;

Figure 3 is an example diagram of the data structure of traceability information;

Fig. 4 is a schematic diagram of a traceability method with correlation perception;

Fig. 5 is a flow chart of retrospective query with correlation analysis added.

Detailed ways

The present invention will be further described below through specific embodiments and accompanying drawings.

1. Architecture of traceability system

Figure 2 shows the architecture diagram of the item information traceability system with correlation perception, including the discovery server (or discovery service module), two independent information service systems (corresponding to company A and company B respectively), and the discovery server and Two independent information service systems are connected through the registration interface. Any independent information service system includes information server, middleware, reader and label. The reader obtains the item identification carried by the label through radio frequency communication. The reader is connected with the middleware, and the middleware is connected with the information server and ERP system. connected.

Among them, the core of the traceability system is the Information Service and the Discovery Service. On this basis, it assists other applications to realize functions such as traceability query, location query, and anti-counterfeiting authentication.

As shown in Figure 2, the discovery server specifically includes a registration interface, a query interface, a communication interface, a query processing module, a correlation analysis module, a filtering module and a mapping database. The query processing module is used for processing the query request for the RFID identification initiated by the query requester. The mapping database is used to store the mapping relationship between the item ID and the address ID of the information server storing the detailed item information. The association analysis module is used to analyze the containment relationship that may exist between the identity mapping data. The filtering module is used to filter the above association analysis with the help of time attribute information to avoid information redundancy. The information service system includes readers, middleware and information servers. The reader is used to obtain the item identification carried in the RFID tag through radio frequency communication. The middleware is used to perform preprocessing such as filtering and deduplication on the original RFID data acquired by the reader, and form RFID events with certain semantics. Information server includes: registration interface, capture interface, query interface and information database.

Through the capture interface, the information server accepts the upload of the original RFID data processed by the middleware; after combining logical processing, the RFID identification data is formed and stored in the information database of the information server; through the registration interface, the RFID mapping data is released to the discovery server; through Query interface, which opens shared query of RFID identification data to applications. The discovery server accepts the release of the RFID mapping data of the information server through the registration interface; after correlation analysis, stores the RFID mapping data in the mapping database of the discovery server; provides sharing of RFID mapping data between different discovery servers through the communication interface; Through the query interface, the query of the RFID mapping data is opened to the application.

Specifically, when an item enters the reading range of the reader, the reader can automatically obtain the item identification carried in the electronic tag through radio frequency communication, thereby generating the original RFID identification data. In order to remove data redundancy, it is necessary to go through pre-processing operations such as deduplication and filtering of middleware, and combine other item-related information in the enterprise information system to form RFID events with certain semantics, and finally store them in the information server. However, unlike the search service on the Internet that can use the hyperlink relationship between web pages to grab web page information, the data source in the RFID network is an information server containing corporate secrets, and retrospective inquiries are only open to authorized partners. Therefore, in order to achieve flexible traceability of RFID data information on a global scale, the basic premise is that the information owner needs to publish certain index information, that is, the mapping relationship between RFID data and the data source, to the index server for other nodes to query . In order to ensure the privacy of information owners, it is also necessary to issue corresponding security policies. From the perspective of implementation, the infrastructure of information service and discovery service can be maintained by large enterprises or companies, and discovery service, as a linking Unicom information service, can also be hosted by a third party.

2. Data structure of traceability information

The data involved in the traceability can be divided into three categories, namely the original record of the reader reading the electronic tag, the RFID event record stored in the information server, and the index information record stored in the discovery server. According to the query request, the discovery server uses the returned data to generate the location information or running track of the commodity with the RFID tag. Specifically, the various data involved are shown in Figure 3, and are described as follows:

1) Original record (Initial_record): The unprocessed original observation data includes three-dimensional attributes, namely the observation target oid, the observation event occurrence time eventTime and the observation location readPoint.

2) Item information (Thing_information): In addition to the check-in information reflecting whether the RFID tag appears at a certain location at a certain time, IoT applications based on RFID networks usually need to pay attention to other attribute information of the object to which the tag is attached, such as item type , place of origin, etc., thereby providing more complex query responses. These application data can be uniformly abstracted into the form of "key-value" pair <propertyID, propertyValue>.

3) Event record (event_record): Through the middleware processing program, a meaningful RFID event record can be formed after integrating the original record and item information. According to whether the containment relationship changes, it can be divided into ordinary observation events (Observation_event) and aggregation events (Aggregation_event). The observation objects of the latter include the item identifier cid and the container identifier pid, and the concept of event behavior is added. The action option can take the value of "add" (combination) or "delete" (decomposition). The original information of the RFID tag read by the reader contains a special field to indicate whether it is an observation event or an aggregation event, so as to identify whether a containment relationship transition occurs.

4) Mapping record (map_record): The mapping record, namely index information, is stored in the mapping database of the discovery server, and the index information is just a simple summary of the detailed item information in the information server. The core of this mapping relationship is the object identifier oid as a key value and the information source address map that stores related event records. In order to increase the analysis of the inclusive correlation between data, the possible behavior attribute action and eventTime when the event occurs are also saved. If the value of action is empty, it means that it is just a common observation event, and the container identifier pid is also empty. The registrant registers with the discovery server, and the address of the registrant is the source address here. After the discovery server receives the registration information, it can generate the corresponding mapping record.

5) Publisher information (Publisher_profile): In order to ensure the confidentiality of the publisher's index information, corresponding security policies can be set on the discovery service. This often requires the publisher to publish certain identity information, such as publisher ID, name, address, email address, affiliation (unit or organization), and identity public key and other authentication auxiliary information that can be used for encryption and decryption.

3. Relational-aware traceability process

Figure 4 is a schematic diagram of the interaction process between the information publisher and the information inquirer during a complete traceability process, and the specific description is as follows:

1) Information owners (information servers A, B) register their own RFID event records with the discovery server (discovery server), and publish corresponding security policies at the same time.

2) The query requester initiates a query request for a specific RFID identification (or identification range) to the discovery server, and submits an identity capability certificate at the same time.

3) The discovery server first retrieves the mapping record with the RFID identifier requested by the user as the key, and then performs correlation analysis through the correlation analysis module to find the mapping record of the container identifier associated with the RFID identifier.

The mapping record reflects the mapping relationship between the identification and the address, that is, in which address the information of the identified item is stored in the server. In addition, registration time, registrant, affiliation and other auxiliary information will also be included.

4) The discovery server checks whether the inquirer has legal access to the information data corresponding to the above-mentioned RFID identification. If it is an authorized legal user, it can return a list of relevant addresses that can be accessed to the query requester, that is, according to the identification of the query, go to A database in a directory server such as DNS is searched for the relevant ID mapping record (which contains address information). If it is not an authorized legal user, refuse to provide services or directly discard the query request.

5) According to the address list obtained in step 4), the query requester establishes communication connections with all the information servers corresponding to the addresses, and submits identity and capability certificates to all information servers at the same time.

6) If the information server passes the authentication, it will return the detailed item information data associated with the RFID tag.

After adding association analysis, the main idea of discovery service for mapping record retrieval is: first, search for mapping records based on the RFID identifier requested by the user as a keyword; then analyze whether there is a mapping record with aggregation behavior; The container ID is added to the queue to be checked, and then the query is performed recursively. The execution flow of the algorithm adding correlation analysis is shown in Figure 5, and the description of the recursive query is as follows:

a) Set the upper limit and lower limit of time, and query the mapping records with id as the key;

b) If there is a mapping record with the query ID as the key, extract the non-aggregated records and add them to the DS response queue, and enter step c); otherwise, return an empty DS response queue;

c) Judging whether there is an aggregation record, if not, return to the DS response queue; if there is, enter step d);

d) group the mapping records corresponding to the aggregation behavior according to the container identifier pid;

e) For each group, determine whether there is a combination event. A combination event refers to an aggregation event that occurs between multiple items due to the coexistence of locations. Usually, there is an outer container item; if there is a combination event, identify the pid The lower limit time is modified to the earliest combination event occurrence time; if not, then judge whether there is a split event, the split event refers to the release of the state of the above combination relationship; if there is a split event, modify the upper limit time identified by the pid to the latest The later split event occurrence time, if there is no split event, end the recursive query.

Considering that there are mainly two basic forms of retrospective query, namely location query (track) and tracking query (trace), the former only focuses on the current latest position of the item, while the query target of the latter is the complete movement track of the item, so when discovering Before starting the service, the user query type should be judged first. If it is a positioning query, it is necessary to sort the mapping records according to the time attribute after each query, and always keep the latest mapping record.

In addition, there is another very important factor to consider, namely the reusability of the container. Since RFID electronic tags usually coexist with the marked items in their entire life cycle, and containers such as boxes and container pallets are often used for a long time in the process of logistics and transportation, this makes the inclusive relationship between single items and containers only temporarily. If only the RFID identification of the container is used for direct retrieval, it is likely to obtain a large amount of irrelevant information. On the one hand, redundant information reduces query performance and increases network communication load. On the other hand, a logistics company may cooperate with multiple companies with competing businesses. Inquirers usually only have access to container information within a certain period of time based on business logic. If the access rights are arbitrarily enlarged without control, the privacy of the information owner may be leaked. Therefore, in the present invention, the time attribute is used for filtering to avoid obtaining too many redundant results, that is, filtering is realized by adjusting the upper and lower limits of the time and narrowing the query interval.

The above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Those of ordinary skill in the art can modify or equivalently replace the technical solution of the present invention without departing from the spirit and scope of the present invention. The scope of protection should be determined by the claims.

Claims (4)

1. a kind of Item Information retroactive method perceived with relevance, this method are applied to a kind of Item Information traceability system, The Item Information traceability system includes at least one discovery server and at least one information service system；

The discovery server includes：

Registration interface, it is connected with the information server in described information service system to receive RFID mapping data publications；

Query interface, opened to application program to carry out the inquiry of RFID mapping data；

Query processing module, for handling the inquiry request for RFID marks；

Correlation analysis module, for analyze article RFID mark mapping data between containment relationship that may be present, carry out Correlation analysis, the single product inputted according to user identify, and return to the relative article information of the mark, and all sent out with the article Gave birth to the information of the container article with temporal feature；

Mapping database, for storing the article RFID identity obtained through correlation analysis and the information of storage article information Mapping relations between server address mark；

Described information service system includes：

Reader, for obtaining the article identity carried in RFID label tag mark by twireless radio-frequency communication mode；

Middleware, for being pre-processed to the raw RFID data that reader obtains, and formed with certain semantic RFID Event；

Information server, including：Registration interface, it is connected with the registration interface of the discovery server to carry out RFID mapping data Registration issue；Interface is captured, is connected with middleware and is identified with receiving the RFID of its upload；Query interface, opened to application program Put to carry out the shared inquiry of RFID mark datas；Information database, number is identified for storing the RFID after middleware processes According to；

The Item Information retroactive method perceived with relevance comprises the following steps：

1) information owner is to the RFID event record for finding that server registration is possessed, while announces corresponding security strategy；

2) inquiry request person to it is described discovery server initiate for RFID mark inquiry request, while submit identity powers and functions with Card；

3) it is described to find that server retrieval is identified as the map record of keyword with RFID step 2) described, and carry out relevance point Analysis, search the map record of the container identification of RFID mark associations；

4) whether the discovery server check inquiry has legal access to information data corresponding to RFID marks Authority；If the validated user authorized then returns to addressable relative address list to inquiry request person, otherwise refusal provides Service directly abandons inquiry request；

5) address list obtained according to step 4), inquiry request person's all information servers corresponding with address are established logical Letter connection, while identity powers and functions voucher is submitted respectively to all information servers；

6) all information server certifications are by rear, by RFID identify associated by the information data of article return to inquiry request Person.

2. the method as described in claim 1, it is characterised in that：Step 2) the inquiry request for RFID marks is to be directed to The inquiry request of particular RFID mark, or the inquiry request for RFID marks section；Before discovery service starts first Judge user's query type：Locating query and tracking are inquired about.

3. the method as described in claim 1, it is characterised in that：Step 3) the correlation analysis, is asked according to user first RFID be identified as keyword search map record；Then analyse whether to there are the map record of polymerization behavior；If so, then Will likely container identification add queue to be checked, then recursively inquired about.

4. the method as described in claim 1, it is characterised in that：Time attribute is used to filter inquiry request to avoid obtaining Take excessive redundant results.