RFID intelligent analysis, tracking, and statistics system and operation method thereof

Abstract

The application relates to the technical field of electronic tags, and discloses an RFID intelligent analysis, tracking and statistics system and an operating method thereof, which comprises a computer module, a reader module, an antenna module and an RFID tag module; the computer module is connected with the reader module through data communication; the reader module is connected with the RFID tag module through the antenna module; the RFID tag carries EPC data; the computer module carries a modeling unit, a tracking unit and a statistics unit; the tracking unit collects data of the RFID tag module through the reader module and the antenna module; the collected data is transmitted to the modeling unit and the statistics unit respectively; the modeling unit analyzes the data and constructs an information model of the RFID tag; and the statistics unit analyzes the data and stores information of the RFID tag; the system realizes reading and collecting electronic tag information, effectively tracking information and statistics, etc.

Description

Technical Field

[0001] This invention relates to the field of electronic tag technology, specifically to an RFID intelligent analysis, tracking, and statistical system and its operation method. Background Technology

[0002] With the rapid development of material products, a wide variety of products are quickly applied to people's lives and production. For many material products, traditional labeling methods for classification are cumbersome, complex, time-consuming, and labor-intensive. Later, to address the drawbacks of traditional labeling, designers began to design RFID electronic tags. These tags, through the integration of chips and transceiver antennas, overcome the shortcomings of traditional labels. RFID electronic tags can be used in many fields, facilitating the encoding and classification of information for various items. With the development of RFID technology, many types of RFID electronic tags have emerged. However, these tags still have many shortcomings, such as difficulties in reading and collecting information, difficulties in information statistics, and difficulties in tracking electronic tags, which limit their use.

[0003] Therefore, existing electronic tag technology needs further improvement to address the above issues. Summary of the Invention

[0004] The purpose of this invention is to address the difficulties in reading and collecting information from electronic tags, the difficulties in information statistics, and the difficulties in tracking electronic tags in existing technologies, which greatly limit the use of electronic tags. Through the rational design of electronic tags, using computer modules, reader modules, antenna modules, and RFID tag modules, it is possible to effectively read and collect electronic tag information, and effectively track and statistically analyze information.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] An RFID intelligent analysis, tracking, and statistics system is provided, comprising: a computer module, a reader module, an antenna module, and an RFID tag module. The computer module is interconnected with the reader module via data communication, and the reader module is interconnected with the RFID tag module via the antenna module. The RFID tag carries EPC (Electronic Product Code) data.

[0007] The computer module carries a modeling unit, a tracking unit, and a statistics unit. The tracking unit collects data from the RFID tag module through the reader module and the antenna module, and transmits the data collected by the tracking unit to the modeling unit and the statistics unit respectively. The modeling unit performs data analysis and constructs an information model of the RFID tag, and the statistics unit performs data analysis and statistically stores the information of the RFID tag.

[0008] The modeling unit is used to construct an information model of RFID tags by combining pre-collected tag information and / or tag information being collected. The information model of RFID tags constructed using pre-collected tag information is a regression model. The regression model generates fingerprints and forms a fingerprint database, which is part of the database. The tag information being collected refers to the continuous collection of RFID tags and their storage in the database.

[0009] Furthermore, the RFID tag module selects several passive RFID tags.

[0010] Furthermore, the reader module includes a radio frequency transceiver circuit and a main control circuit, wherein the main control circuit is electrically connected to the radio frequency transceiver circuit.

[0011] Furthermore, the main control circuit is selected as either an MCU main control circuit or a CPU main control circuit.

[0012] Furthermore, the MCU main control circuit uses the STM32F103RET6 chip.

[0013] Furthermore, the modeling unit includes a pre-label information element, a regression modeling element, a fingerprint generator, and a fingerprint database element. The pre-label information element transmits the pre-collected label information to the regression modeling element. After modeling, the regression modeling element transmits the modeling information to the fingerprint generator. The fingerprint generator generates fingerprints and transmits them to the fingerprint database element.

[0014] Furthermore, the tracking unit includes a tracking tag information element, a position sorting element, a transformation matrix generator, a placement matrix generator, and a tracking result element. The tracking tag information element transmits the tracked and collected RFID tag information to the position sorting element. The position sorting element sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator and the placement matrix generator, respectively. The transformation matrix generator and the placement matrix generator generate the tracking result and transmit it to the tracking result element for display and storage.

[0015] Furthermore, the radio frequency transceiver circuit includes a radio frequency transceiver chip and a power amplifier circuit, and the radio frequency transceiver chip and the power amplifier circuit are electrically connected to each other.

[0016] Furthermore, the radio frequency transceiver chip is selected as the MagicRF M100 chip.

[0017] Furthermore, the reader module is selected from the ThingMagic M7e-Pico reader.

[0018] Furthermore, the antenna module is selected as the Times-7 A5010 antenna.

[0019] Furthermore, the RFID intelligent analysis, tracking, and statistics system also includes a cloud computing module, a back-end server module, a main reader module, and a WLM server module. The reader module is interconnected with the RFID tag module, the reader module is interconnected with the main reader module through the WLM server module, the main reader module is interconnected with the back-end server module, the back-end server module is interconnected with the cloud computing module, and the cloud computing module is interconnected with the computer module.

[0020] Furthermore, a traceability method for an RFID intelligent analysis, tracking, and statistical system includes the following specific steps:

[0021] Step 1: Hardware Setup

[0022] A computer module, a reader module, an antenna module, and an RFID tag module are selected. The computer module is interconnected with the reader module via data communication. The reader module is interconnected with the RFID tag module via the antenna module. The RFID tag carries EPC data. The hardware Internet of Things is formed through signal interconnection. The computer module carries a modeling unit, a tracking unit, and a statistical unit to complete the hardware construction.

[0023] The modeling unit includes a pre-label information element, a regression modeling element, a fingerprint generator, and a fingerprint database element. The pre-label information element transmits the pre-collected label information to the regression modeling element. After modeling, the regression modeling element transmits the modeling information to the fingerprint generator. The fingerprint generator generates fingerprints and transmits them to the fingerprint database element.

[0024] The tracking unit includes a tracking tag information element, a position sorting element, a transformation matrix generator, a placement matrix generator, and a tracking result element. The tracking tag information element transmits the RFID tag information collected during tracking to the position sorting element. The position sorting element sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator and the placement matrix generator, respectively. The transformation matrix generator and the placement matrix generator generate the tracking result and transmit it to the tracking result element for display and storage.

[0025] Step 2: Reader module assembly:

[0026] Based on step 1, a power supply, an MCU chip, an RF transceiver, a power amplifier, a directional coupler, and an antenna are selected. The power supply provides power to the reader module. The power output terminal of the power supply is electrically connected to the power input terminals of the MCU chip, the RF transceiver, the power amplifier, the directional coupler, and the antenna, respectively. The output signal terminal of the MCU chip is electrically connected to the input signal terminal of the power amplifier. The output signal terminal of the power amplifier is electrically connected to the input signal terminal of the RF transceiver. The output signal terminal of the RF transceiver is electrically connected to the input signal terminal of the directional coupler. The output signal terminal of the directional coupler is electrically connected to the signal terminal of the antenna.

[0027] Step 3: Pre-reading of RFID tags:

[0028] Based on step 2, select a number of RFID tags and place them within the range that the reader module can read. The computer module transmits the set program instructions to the reader module, and the reader module sends multiple frame lengths and transmits them to the RFID tags through the antenna module. The RFID tags will generate non-interfering random numbers from the multiple frame lengths. The corresponding time slot is determined based on the random numbers of each RFID tag and fed back to the reader module for pre-reading of the RFID tags.

[0029] Step 4: The reader module reads the RFID tag information.

[0030] Based on the pre-reading of the RFID tag in step 3, the reader is powered on and enters the initialization state. It sends a continuous carrier signal and an access command to the RFID tag. After the RFID tag receives the continuous carrier signal and the access command, the internal circuit of the RFID tag is powered on and starts to work, and establishes a data transmission channel. Data transmission also begins. Then the reader selects the RFID tag through the query command, and the RFID tag will feed back the EPC data information to the reader for reading.

[0031] Step 5: Reader module filters and selects GS1 standard EPCs:

[0032] Based on the reading of RFID tag information in step 4, since there may be RFID tags without coded data among many RFID tags, it is necessary to remove data tags that do not meet the standard according to the GS1 standard EPC tag rules.

[0033] Step 6: Parse the read data:

[0034] Based on the standard-compliant data tags from step 5, the original data is parsed out through reverse engineering by the computer module in the system. This data includes information such as serial number, UPC, and prefix.

[0035] Step 7: Internal program calculation and storage:

[0036] Select the data tags that meet the standards from step 6. After the reader filters out the data identified by the sensor, these data need to be stored in a list and compared after each data reading to ensure that the data is not duplicated. If the data read already exists, it is filtered out.

[0037] Step 8: Modeling the modeling unit:

[0038] Select the standard-compliant data tags from step 7. Since there is pre-existing tag information, the computer module creates a database that stores the standard-compliant data tags and assigns a unique location number to each tag.

[0039] Step 9: Predict the label position:

[0040] Based on step 8, the computer module sends instructions to the reader module, which then finds the next possible location of the current location. This location is used as the predicted label location. Since there are different label locations, these different label locations are used to generate a regression model in order to locate the location of the target label, form fingerprint data, and then generate a fingerprint database.

[0041] Step 10: Track tag information in real time:

[0042] Based on step 9, the tracking unit, and the fingerprint database, the possible location of each tag is predicted through dynamic M nearest neighbor. Dynamic M uses the nearest neighbor (DKNN) algorithm. This model searches for the next possible location, which is sorted according to the location probability score provided by its DKNN algorithm, and converts RSSI into location coordinates. The transformation matrix approximates the distance from the current estimated location to the next set of possible locations. This framework iteratively updates the regression model in the modeling step as the environment changes, thereby tracking tag information in real time.

[0043] Step 11: Front-end data display:

[0044] If the data on the front end of the computer software changes due to changes in the data on the back end, it will be updated in real time and displayed to the user. The data can also be exported, and the parsed data will be automatically saved to the handheld device.

[0045] The modeling unit, tracking unit, and statistics unit work together to continuously track the location of the target tag. The tracking process first collects data through an ultra-high frequency radio frequency identification (UHF RFID) system (i.e., RFID reader, AP or antenna, RFID tag and data processor, etc.). The UHF RFID system collects the backscattered signals of the reference tag and the target tag. The modeling unit uses the information of the reference tag to build a regression model, and then uses the model to build a fingerprint database. Over time, it adapts to environmental changes. All modeling parameters and fingerprint information are stored in the database.

[0046] The tracking device continuously collects RSSI information from the target tags; this information is sampled in half-second intervals, forming a hidden Markov chain. The data from each sampling period is then used to locate possible positions using the DKNN algorithm. After position ranking and estimation, a transition matrix and an emission matrix are generated based on the sum of squared errors (SSE). Finally, a Viterbi search algorithm is used to solve the Hidden Markov Model (HMM).

[0047] Furthermore, the UHF RFID system here is equipped with a computing unit, an RFID reader, an access point, and tags (i.e., target tags and reference tags). The received tag backscatter signal is then transmitted to the reader through the access point. The reader transmits the data to the data processor for further processing. After processing the tag information, the data processor sends the information to the database, which is then used by other units within the framework.

[0048] Furthermore, the modeling unit utilizes a regression model to generate a fingerprint database to derive the relationship between distance and RSSI. This unit comprises two independent steps: 1) establishing a regression model and 2) generating fingerprints. In short, the regression modeling step provides an effective relationship between distance and RSSI by utilizing preprocessed signal information collected from reference marker locations, and the fingerprint generation step provides an estimated RSSI database of virtual reference points. These virtual reference points facilitate localization using the DKNN algorithm.

[0049] Furthermore, the tracking unit estimates the position of the target tag and identifies the most likely movement path. Generally, the motion of an object can be defined as a state transition process, where the object's next position depends entirely on its current position. Since we predict these position sequences based on observed RSSI, this process can be represented as an Hidden Markov Model (HMM).

[0050] Furthermore, in a passive radio frequency identification (RFID) system, the reader processes the signal and acts as a bridge between the access point (i.e., antenna) and the data processor. Typically, the reader transmits the signal and analyzes the signal received from the tag via the wireless access point; the tag is placed on the object to be identified or tracked. The area covered by this signal is called the interrogation zone, and the passive tag must be located within the interrogation zone to backscatter the signal. Each passive tag consists of an antenna and an integrated circuit (chip); directly connected to the antenna terminal. After receiving the reader's transmitted signal, the antenna terminal powers the tag's chip, which then sends a signal to the reader; the tag's unique ID is sent to the reader by modulating the backscattered signal; the backscattered signal detected by the access point is then sent to the data processor (such as a computer) for further processing by the reader.

[0051] The computer module sends instruction codes to the reader module to perform the reading action. The reader module sends radio signals to the RFID tag module through the antenna module to read the information of the RFID tag module. After receiving the radio signal, the RFID tag module charges and backscatters the modulated signal. It receives the RFID tag signal through the antenna module and sends the data to the reader module. The reader module then sends the data to the computer module for processing.

[0052] The information obtained by the RFID intelligent analysis, tracking, and statistics system is stored in the database. Pre-label information is sent to the modeling unit, which generates a fingerprint database and then transmits it to the database. Tracking tag information is sent to the tracking unit, which generates tracking results and then transmits them to the database.

[0053] Beneficial effects

[0054] This invention improves positioning and tracking accuracy through a rational design of the electronic tag system. Multiple experiments were conducted in various scenarios to test the proposed system. Results show that the system can track objects, deploy reference tags in the environment, and store their location information in a database. A Gaussian filter is used to filter out abnormal RSSIs, thereby improving positioning accuracy. This RFID system can resist tracking attacks, as well as reader and tag forgery attacks. It offers better security and performance and can be applied to distributed environments, such as cloud database environments. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the structure of an RFID intelligent analysis, tracking, and statistics system according to the present invention.

[0056] Figure 2This is a schematic diagram of the structure of the RFID intelligent analysis, tracking, and statistical system of the present invention.

[0057] Figure 3 This is a schematic diagram of the modeling unit of an RFID intelligent analysis, tracking, and statistics system according to the present invention.

[0058] Figure 4 This is a schematic diagram of the structure of the tracking unit of an RFID intelligent analysis, tracking, and statistics system according to the present invention.

[0059] Figure 5 This is a schematic diagram of the process structure of an RFID intelligent analysis, tracking, and statistics system according to the present invention.

[0060] Figure 6 This is another schematic diagram of the process structure of an RFID intelligent analysis, tracking, and statistics system according to the present invention.

[0061] Figure reference numerals: 01, Computer module; 02, Reader module; 03, Antenna module; 04, RFID tag module; 1, RFID intelligent analysis and traceability system; 2, Database; 3, Modeling unit; 4, Tracking unit; 31, Pre-tag information element; 32, Regression modeling element; 33, Fingerprint generator element; 34, Fingerprint database element; 41, Tracking tag information element; 42, Position sorting element; 43, Transformation matrix generator element; 44, Placement matrix generator element; 45, Tracking result element. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] See Figure 1-6 As shown, the present invention provides an RFID intelligent analysis, tracking, and statistics system, which includes: a computer module 01, a reader module 02, an antenna module 03, and an RFID tag module 04. The computer module 01 is interconnected with the reader module 02 via data communication, and the reader module 02 is interconnected with the RFID tag module 04 via the antenna module 03. The RFID tag carries EPC (Electronic Product Code) data.

[0064] The computer module 01 carries an RFID intelligent analysis and traceability system 1, and is equipped with a modeling unit 3, a tracking unit 4, and a statistical unit. The tracking unit 4 collects data from the RFID tag module 04 through the reader module 02 and the antenna module 03. The data collected by the tracking unit 4 is transmitted to the modeling unit 3 and the statistical unit respectively. The modeling unit 3 performs data analysis and constructs an information model of the RFID tag. The statistical unit performs data analysis and statistically stores the information of the RFID tag.

[0065] The modeling unit 3 is used to construct an information model of RFID tags by combining pre-collected tag information and / or tag information being collected. The information model of RFID tags constructed using pre-collected tag information is a regression model. The regression model generates fingerprints and forms a fingerprint database, which is part of database 2. The tag information being collected is the continuous collection of RFID tags and their storage in database 2.

[0066] The RFID tag module 04 selects several passive RFID tags.

[0067] The reader module 02 includes a radio frequency transceiver circuit and a main control circuit, wherein the main control circuit is electrically connected to the radio frequency transceiver circuit.

[0068] The main control circuit is selected from either an MCU main control circuit or a CPU main control circuit.

[0069] The MCU main control circuit uses the STM32F103RET6 chip.

[0070] The modeling unit 3 includes a pre-label information element 31, a regression modeling element 32, a fingerprint generator 33, and a fingerprint database element 34. The pre-label information element 31 transmits the pre-collected label information to the regression modeling element 32. After modeling, the regression modeling element 32 transmits the modeling information to the fingerprint generator 33. The fingerprint generator 33 generates fingerprints and transmits them to the fingerprint database element 34.

[0071] The tracking unit 4 includes a tracking tag information element 41, a position sorting element 42, a transformation matrix generator 43, a placement matrix generator 44, and a tracking result element 45. The tracking tag information element 41 transmits the tracked RFID tag information to the position sorting element 42. The position sorting element 42 sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator 43 and the placement matrix generator 44 respectively. The transformation matrix generator 43 and the placement matrix generator 44 generate the tracking result and transmit it to the tracking result element 45 for display and storage.

[0072] The radio frequency transceiver circuit includes a radio frequency transceiver chip and a power amplifier circuit, and the radio frequency transceiver chip and the power amplifier circuit are electrically connected to each other.

[0073] The radio frequency transceiver chip selected is the MagicRF M100 chip.

[0074] The reader module 02 is selected from the ThingMagic M7e-Pico model reader.

[0075] The antenna module 03 is selected as the Times-7 A5010 antenna.

[0076] The RFID intelligent analysis, tracking, and statistics system further includes a cloud computing module, a back-end server module, a main reader module 02, and a WLM server module. The reader module 02 is interconnected with the RFID tag module 04. The reader module 02 is interconnected with the main reader module 02 through the WLM server module. The main reader module 02 is interconnected with the back-end server module. The back-end server module is interconnected with the cloud computing module. The cloud computing module is interconnected with the computer module 01.

[0077] Furthermore, a traceability method for an RFID intelligent analysis, tracking, and statistical system includes the following specific steps:

[0078] Step 1: Hardware Setup

[0079] A computer module, a reader module, an antenna module, and an RFID tag module are selected. The computer module is interconnected with the reader module via data communication. The reader module is interconnected with the RFID tag module via the antenna module. The RFID tag carries EPC data. The hardware Internet of Things is formed through signal interconnection. The computer module carries a modeling unit, a tracking unit, and a statistical unit to complete the hardware construction.

[0080] The modeling unit includes a pre-label information element, a regression modeling element, a fingerprint generator, and a fingerprint database element. The pre-label information element transmits the pre-collected label information to the regression modeling element. After modeling, the regression modeling element transmits the modeling information to the fingerprint generator. The fingerprint generator generates fingerprints and transmits them to the fingerprint database element.

[0081] The tracking unit includes a tracking tag information element, a position sorting element, a transformation matrix generator, a placement matrix generator, and a tracking result element. The tracking tag information element transmits the RFID tag information collected during tracking to the position sorting element. The position sorting element sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator and the placement matrix generator, respectively. The transformation matrix generator and the placement matrix generator generate the tracking result and transmit it to the tracking result element for display and storage.

[0082] Step 2: Reader module assembly:

[0083] Based on step 1, a power supply, an MCU chip, an RF transceiver, a power amplifier, a directional coupler, and an antenna are selected. The power supply provides power to the reader module. The power output terminal of the power supply is electrically connected to the power input terminals of the MCU chip, the RF transceiver, the power amplifier, the directional coupler, and the antenna, respectively. The output signal terminal of the MCU chip is electrically connected to the input signal terminal of the power amplifier. The output signal terminal of the power amplifier is electrically connected to the input signal terminal of the RF transceiver. The output signal terminal of the RF transceiver is electrically connected to the input signal terminal of the directional coupler. The output signal terminal of the directional coupler is electrically connected to the signal terminal of the antenna.

[0084] Step 3: Pre-reading of RFID tags:

[0085] Based on step 2, select a number of RFID tags and place them within the range that the reader module can read. The computer module transmits the set program instructions to the reader module, and the reader module sends multiple frame lengths and transmits them to the RFID tags through the antenna module. The RFID tags will generate non-interfering random numbers from the multiple frame lengths. The corresponding time slot is determined based on the random numbers of each RFID tag and fed back to the reader module for pre-reading of the RFID tags.

[0086] Step 4: The reader module reads the RFID tag information.

[0087] Based on the pre-reading of the RFID tag in step 3, the reader is powered on and enters the initialization state. It sends a continuous carrier signal and an access command to the RFID tag. After the RFID tag receives the continuous carrier signal and the access command, the internal circuit of the RFID tag is powered on and starts to work, and establishes a data transmission channel. Data transmission also begins. Then the reader selects the RFID tag through the query command, and the RFID tag will feed back the EPC data information to the reader for reading.

[0088] Step 5: Reader module filters and selects GS1 standard EPCs:

[0089] Based on the reading of RFID tag information in step 4, since there may be RFID tags without coded data among many RFID tags, it is necessary to remove data tags that do not meet the standard according to the GS1 standard EPC tag rules.

[0090] Step 6: Parse the read data:

[0091] Based on the standard-compliant data tags from step 5, the original data is parsed out through reverse engineering by the computer module in the system. This data includes information such as serial number, UPC, and prefix.

[0092] Step 7: Internal program calculation and storage:

[0093] Select the data tags that meet the standards from step 6. After the reader filters out the data identified by the sensor, these data need to be stored in a list and compared after each data reading to ensure that the data is not duplicated. If the data read already exists, it is filtered out.

[0094] Step 8: Modeling the modeling unit:

[0095] Select the standard-compliant data tags from step 7. Since there is pre-existing tag information, the computer module creates a database that stores the standard-compliant data tags and assigns a unique location number to each tag.

[0096] Step 9: Predict the label position:

[0097] Based on step 8, the computer module sends instructions to the reader module, which then finds the next possible location of the current location. This location is used as the predicted label location. Since there are different label locations, these different label locations are used to generate a regression model in order to locate the location of the target label, form fingerprint data, and then generate a fingerprint database.

[0098] Step 10: Track tag information in real time:

[0099] Based on step 9, the tracking unit, and the fingerprint database, the possible location of each tag is predicted through dynamic M nearest neighbor. Dynamic M uses the nearest neighbor (DKNN) algorithm. This model searches for the next possible location, which is sorted according to the location probability score provided by its DKNN algorithm, and converts RSSI into location coordinates. The transformation matrix approximates the distance from the current estimated location to the next set of possible locations. This framework iteratively updates the regression model in the modeling step as the environment changes, thereby tracking tag information in real time.

[0100] Step 11: Front-end data display:

[0101] If the data on the front end of the computer software changes due to changes in the data on the back end, it will be updated in real time and displayed to the user. The data can also be exported, and the parsed data will be automatically saved to the handheld device.

[0102] Among them, the modeling unit 3, the tracking unit, and the statistics unit work together to continuously track the position of the target tag. The tracking process first collects data through an ultra-high frequency radio frequency identification (UHF RFID) system (i.e., RFID reader, AP or antenna, RFID tag, and data processor). The UHF RFID system collects the backscattered signals of the reference tag and the target tag. The modeling unit 3 uses the information of the reference tag to build a regression model, and then uses the model to build a fingerprint database. Over time, it adapts to environmental changes. All modeling parameters and fingerprint information are stored in the database.

[0103] The tracking device continuously collects RSSI information from the target tags; this information is sampled in half-second intervals, forming a hidden Markov chain. The data from each sampling period is then used to locate possible positions using the DKNN algorithm. After position ranking and estimation, a transition matrix and an emission matrix are generated based on the sum of squared errors (SSE). Finally, a Viterbi search algorithm is used to solve the Hidden Markov Model (HMM).

[0104] The UHF RFID system here is equipped with a computing unit, an RFID reader, an access point, and tags (i.e., target tags and reference tags). The backscattered signal from the received tag is then transmitted to the reader through the access point. The reader transmits the data to the data processor for further processing. After processing the tag information, the data processor sends the information to the database, which is then used by other units within the framework.

[0105] Modeling unit 3 utilizes a regression model to generate a fingerprint database to derive the relationship between distance and RSSI. This unit comprises two independent steps: 1) establishing a regression model and 2) generating fingerprints. In short, the regression modeling step provides an effective relationship between distance and RSSI by utilizing preprocessed signal information collected from reference marker locations, and signal information collected from reference tags at known locations. The fingerprint generation step estimates an RSSI database of virtual reference points. These virtual reference points facilitate localization using the DKNN algorithm.

[0106] The tracking unit estimates the position of the target tag and finds the most likely movement path. Generally, the motion of an object can be defined as a state transition process, where the object's next position depends entirely on its current position. Since we predict these position sequences based on the observed RSSI, this process can be represented as an Hidden Markov Model (HMM).

[0107] In a passive radio frequency identification (RFID) system, the reader processes the signal and acts as a bridge between the access point (i.e., antenna) and the data processor. Typically, the reader broadcasts a signal and analyzes the signal received from the tag via the wireless access point. The tag is placed on the object to be identified or tracked. The area covered by the broadcast signal is called the interrogation zone; for the signal to backscatter, the passive tag must be located within the interrogation zone. Each passive tag consists of an antenna and an integrated circuit (chip); it is directly connected to the antenna terminal. After receiving the reader's transmitted signal, the antenna terminal powers the tag's chip, which then sends a signal to the reader. The tag's unique ID is sent to the reader by modulating the backscattered signal. The backscattered signal detected by the access point is then sent to the data processor (such as a computer) for further processing by the reader.

[0108] The computer module sends instruction codes to the reader module to perform the reading action. The reader module sends radio signals to the RFID tag module through the antenna module to read the information of the RFID tag module. After receiving the radio signal, the RFID tag module charges and backscatters the modulated signal. It receives the RFID tag signal through the antenna module and sends the data to the reader module. The reader module then sends the data to the computer module for processing.

[0109] The information obtained by the RFID intelligent analysis, tracking, and statistics system is stored in the database. Pre-label information is sent to modeling unit 3, which generates a fingerprint database and then transmits it to the database. Tracking tag information is sent to the tracking unit, which generates tracking results and then transmits them to the database.

[0110] In a passive radio frequency identification (RFID) system, the reader processes the signal and acts as a bridge between the access point (i.e., antenna) and the data processor. Typically, the reader broadcasts a signal and analyzes the signal received from the tag via the wireless access point. The tag is placed on the object to be identified or tracked. The area covered by the broadcast signal is called the interrogation zone; for the signal to backscatter, the passive tag must be located within the interrogation zone. Each passive tag consists of an antenna and an integrated circuit (chip); it is directly connected to the antenna terminal. After receiving the reader's transmitted signal, the antenna terminal powers the tag's chip, which then sends a signal to the reader. The tag's unique ID is sent to the reader by modulating the backscattered signal. The backscattered signal detected by the access point is then sent to the data processor (such as a computer) for further processing by the reader.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An RFID intelligent analysis, tracking, and statistical system, characterized in that, The RFID intelligent analysis, tracking, and statistics system includes: a computer module, a reader module, an antenna module, and an RFID tag module. The computer module is interconnected with the reader module via data communication. The reader module is interconnected with the RFID tag module via the antenna module. The RFID tag carries EPC data. The computer module carries a modeling unit, a tracking unit, and a statistics unit. The tracking unit collects data from the RFID tag module through the reader module and the antenna module, and transmits the data collected by the tracking unit to the modeling unit and the statistics unit respectively. The modeling unit performs data analysis and constructs an information model of the RFID tag, and the statistics unit performs data analysis and statistically stores the information of the RFID tag. The modeling unit is used to construct an information model of RFID tags by combining pre-collected tag information and / or tag information being collected. The information model of RFID tags constructed using pre-collected tag information is a regression model. The regression model generates fingerprints and forms a fingerprint database, which is part of the database. The tag information being collected refers to the continuous collection of RFID tags and their storage in the database.

2. The RFID intelligent analysis, tracking, and statistical system according to claim 1, characterized in that, The RFID tag module selects several passive RFID tags.

3. The RFID intelligent analysis, tracking, and statistical system according to claim 1, characterized in that, The reader module includes an RF transceiver circuit and a main control circuit, with the main control circuit electrically connected to the RF transceiver circuit.

4. The RFID intelligent analysis, tracking, and statistical system according to claim 3, characterized in that, The main control circuit is selected from either an MCU main control circuit or a CPU main control circuit.

5. The RFID intelligent analysis, tracking, and statistics system according to claim 4, characterized in that, The MCU main control circuit uses the STM32F103RET6 chip.

6. The RFID intelligent analysis, tracking, and statistical system according to claim 1, characterized in that, The modeling unit includes a pre-label information element, a regression modeling element, a fingerprint generator, and a fingerprint database element. The pre-label information element transmits the pre-collected label information to the regression modeling element. After modeling, the regression modeling element transmits the modeling information to the fingerprint generator. The fingerprint generator generates fingerprints and transmits them to the fingerprint database element.

7. The RFID intelligent analysis, tracking, and statistical system according to claim 1, characterized in that, The tracking unit includes a tracking tag information element, a position sorting element, a transformation matrix generator, a placement matrix generator, and a tracking result element. The tracking tag information element transmits the tracked RFID tag information to the position sorting element. The position sorting element sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator and the placement matrix generator, respectively. The transformation matrix generator and the placement matrix generator generate the tracking result and transmit it to the tracking result element for display and storage.

8. The RFID intelligent analysis, tracking, and statistical system according to claim 3, characterized in that, The radio frequency transceiver circuit includes a radio frequency transceiver chip and a power amplifier circuit, and the radio frequency transceiver chip and the power amplifier circuit are electrically connected to each other.

9. The RFID intelligent analysis, tracking, and statistical system according to claim 8, characterized in that, The radio frequency transceiver chip selected is the MagicRF M100 chip.

10. A traceability method for an RFID intelligent analysis, tracking, and statistical system, characterized in that, The specific steps are as follows: Step 1: Hardware Setup A computer module, a reader module, an antenna module, and an RFID tag module are selected. The computer module is interconnected with the reader module via data communication. The reader module is interconnected with the RFID tag module via the antenna module. The RFID tag carries EPC data. The hardware Internet of Things is formed through signal interconnection. The computer module carries a modeling unit, a tracking unit, and a statistical unit to complete the hardware construction. The modeling unit includes a pre-label information element, a regression modeling element, a fingerprint generator, and a fingerprint database element. The pre-label information element transmits the pre-collected label information to the regression modeling element. After modeling, the regression modeling element transmits the modeling information to the fingerprint generator. The fingerprint generator generates fingerprints and transmits them to the fingerprint database element. The tracking unit includes a tracking tag information element, a position sorting element, a transformation matrix generator, a placement matrix generator, and a tracking result element. The tracking tag information element transmits the RFID tag information collected during tracking to the position sorting element. The position sorting element sorts the positions of the RFID tags and then transmits the position sorting information to the transformation matrix generator and the placement matrix generator, respectively. The transformation matrix generator and the placement matrix generator generate the tracking result and transmit it to the tracking result element for display and storage. Step 2: Reader module assembly: Based on step 1, a power supply, an MCU chip, an RF transceiver, a power amplifier, a directional coupler, and an antenna are selected. The power supply provides power to the reader module. The power output terminal of the power supply is electrically connected to the power input terminals of the MCU chip, the RF transceiver, the power amplifier, the directional coupler, and the antenna, respectively. The output signal terminal of the MCU chip is electrically connected to the input signal terminal of the power amplifier. The output signal terminal of the power amplifier is electrically connected to the input signal terminal of the RF transceiver. The output signal terminal of the RF transceiver is electrically connected to the input signal terminal of the directional coupler. The output signal terminal of the directional coupler is electrically connected to the signal terminal of the antenna. Step 3: Pre-reading of RFID tags: Based on step 2, select a number of RFID tags and place them within the range that the reader module can read. The computer module transmits the set program instructions to the reader module, and the reader module sends multiple frame lengths and transmits them to the RFID tags through the antenna module. The RFID tags will generate non-interfering random numbers from the multiple frame lengths. The corresponding time slot is determined based on the random numbers of each RFID tag and fed back to the reader module for pre-reading of the RFID tags. Step 4: The reader module reads the RFID tag information. Based on the pre-reading of the RFID tag in step 3, the reader is powered on and enters the initialization state. It sends a continuous carrier signal and an access command to the RFID tag. After the RFID tag receives the continuous carrier signal and the access command, the internal circuit of the RFID tag is powered on and starts to work, and establishes a data transmission channel. Data transmission also begins. Then the reader selects the RFID tag through the query command, and the RFID tag will feed back the EPC data information to the reader for reading. Step 5: Reader module filters and selects GS1 standard EPCs: Based on the reading of RFID tag information in step 4, since there may be RFID tags without coded data among many RFID tags, it is necessary to remove data tags that do not meet the standard according to the GS1 standard EPC tag rules. Step 6: Parse the read data: Based on the standard-compliant data tags from step 5, the original data is parsed out through reverse engineering by the computer module in the system. This data includes information such as serial number, UPC, and prefix. Step 7: Internal program calculation and storage: Select the data tags that meet the standards from step 6. After the reader filters out the data identified by the sensor, these data need to be stored in a list and compared after each data reading to ensure that the data is not duplicated. If the data read already exists, it is filtered out. Step 8: Modeling the modeling unit: Select the standard-compliant data tags from step 7. Since there is pre-existing tag information, the computer module creates a database that stores the standard-compliant data tags and assigns a unique location number to each tag. Step 9: Predict the label position: Based on step 8, the computer module sends instructions to the reader module, which then finds the next possible location of the current location. This location is used as the predicted label location. Since there are different label locations, these different label locations are used to generate a regression model in order to locate the location of the target label, form fingerprint data, and then generate a fingerprint database. Step 10: Track tag information in real time: Based on step 9, the tracking unit, and the fingerprint database, the possible location of each tag is predicted through dynamic M nearest neighbor. Dynamic M uses the nearest neighbor (DKNN) algorithm. This model searches for the next possible location, which is sorted according to the location probability score provided by its DKNN algorithm, and converts RSSI into location coordinates. The transformation matrix approximates the distance from the current estimated location to the next set of possible locations. This framework iteratively updates the regression model in the modeling step as the environment changes, thereby tracking tag information in real time. Step 11: Front-end data display: If the data on the front end of the computer software changes due to changes in the data on the back end, it will be updated in real time and displayed to the user. The data can also be exported, and the parsed data will be automatically saved to the handheld device.

Images