CN117435902A - Method and device for determining RFID tag movement behavior based on machine learning - Google Patents
Method and device for determining RFID tag movement behavior based on machine learning Download PDFInfo
- Publication number
- CN117435902A CN117435902A CN202311761455.0A CN202311761455A CN117435902A CN 117435902 A CN117435902 A CN 117435902A CN 202311761455 A CN202311761455 A CN 202311761455A CN 117435902 A CN117435902 A CN 117435902A
- Authority
- CN
- China
- Prior art keywords
- data
- sequence data
- phase
- tag
- rfid tag
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000006399 behavior Effects 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000010801 machine learning Methods 0.000 title claims abstract description 63
- 238000000819 phase cycle Methods 0.000 claims abstract description 79
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 230000004927 fusion Effects 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 16
- 230000000386 athletic effect Effects 0.000 claims description 11
- 230000037147 athletic performance Effects 0.000 claims description 10
- 230000010354 integration Effects 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 9
- 238000007637 random forest analysis Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 229910000838 Al alloy Inorganic materials 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- RGCLLPNLLBQHPF-HJWRWDBZSA-N phosphamidon Chemical compound CCN(CC)C(=O)C(\Cl)=C(/C)OP(=O)(OC)OC RGCLLPNLLBQHPF-HJWRWDBZSA-N 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013068 supply chain management Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a method and a device for determining RFID tag movement behavior based on machine learning, belonging to the technical field of electronic information, wherein the method comprises the following steps: acquiring tag data of a target RFID tag read by a reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data; performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data; the phase unwrapping sequence data and/or RSSI value sequence data are used as input data and are input into a tag motion behavior recognition model, and the motion behavior of the target RFID tag is determined; the tag movement behavior recognition model is trained based on input sample data and movement behavior types corresponding to the input sample data. The invention utilizes the machine-learned tag movement behavior recognition model to automatically judge the movement behavior type of the RFID tag, thereby reducing the cost and improving the recognition precision of the in-out information.
Description
Technical Field
The invention relates to the technical field of electronic information, in particular to a method and a device for determining RFID tag movement behavior based on machine learning.
Background
Currently, the technology development of mobile internet, big data, internet of things and the like is very rapid, and intelligent logistics and intelligent warehousing have become focuses of attention of many industries and enterprises. Under the large background of the industry, links of purchasing, selling, warehousing, logistics and the like in the production and management of enterprises are gradually changed, the key of enterprise competition is not only concerned with the improvement of service and quality, and the modernization level of warehousing and logistics is also an important factor of enterprise competition. The importance of warehouse logistics in enterprises is increasingly prominent, and the warehouse logistics have evolved into an indispensable link in the modern supply chain management chain.
The RFID technology has the advantages of large data capacity, reusability and the like, so that the RFID technology can be applied to the field of warehouse management. The general principle is as follows: and the antenna emits signals, and when the RFID tag arranged on the goods recognizes the signals emitted by the antenna, the tag data of the RFID tag is fed back to the reader-writer so as to realize the warehousing judgment by using the tag data. The RFID technology is utilized, so that the workload of operators can be reduced, meanwhile, the occupation of storage fund flows can be reduced, and the circulation speed is accelerated and the reproduction efficiency is improved.
However, in the prior art, accurate judgment of the movement behavior of the RFID tag is difficult to realize, so that accurate judgment of the information of the goods in and out of the warehouse cannot be performed. In addition, the prior art scheme has at least the following defects in the process of judging the warehouse-in and warehouse-out information: on one hand, more hardware needs to be arranged, so that the cost is high, and the wide popularization is difficult; on the other hand, more human intervention is required, resulting in a lower degree of automation of the warehouse entry identification.
Disclosure of Invention
The invention provides a method and a device for determining the movement behavior of an RFID tag based on machine learning, which are used for solving at least one defect in the prior art.
In a first aspect, the present invention provides a method for determining RFID tag athletic activity based on machine learning, comprising: acquiring tag data of a target RFID tag read by a reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data; performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data; inputting the phase unwrapping sequence data and/or the RSSI value sequence data as input data to a pre-established tag movement behavior recognition model, and determining movement behaviors of the target RFID tag; the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
According to the method for determining the movement behavior of the RFID tag based on machine learning provided by the invention, under the condition that the antenna transmits electromagnetic signals of a plurality of carrier frequency bands, tag data of a target RFID tag read by a reader-writer is obtained, and the method comprises the following steps: reading a plurality of tag data of the target RFID tag corresponding to the carrier frequency bands by using a reader-writer; and carrying out average fusion on the phase sequence data in the plurality of tag data, and taking the fusion result as final phase sequence data.
According to the method for determining the motion behavior of the RFID tag based on machine learning, under the condition that the phase sequence data of the target RFID tag has a phase missing problem, the phase unwrapping is carried out on the phase sequence data, and the method comprises the following steps: determining a target group to which the target RFID tag belongs to determine all RFID tags in the target group; wherein the grouping of RFID tags is predetermined based on the similarity of the movement of the RFID tags; carrying out data fusion on the phase sequence data of all RFID tags to determine reference phase sequence data; and unwrapping the phase sequence data of the target RFID tag by using the reference phase sequence data to obtain phase unwrapped sequence data.
According to the method for determining the motion behavior of the RFID tag based on machine learning, which is provided by the invention, the phase sequence data of all RFID tags are subjected to data fusion, and the reference phase sequence data is determined, and the method comprises the following steps: determining an average rate of change of the phase sequence data for each RFID tag;
carrying out average fusion on all the average change rates; and carrying out time-based integration on the average fused data to determine reference phase sequence data.
Determining an average rate of change of the phase sequence data for each RFID tag; and carrying out average fusion on all the average change rates to obtain the reference phase sequence data.
According to the method for determining the motion behavior of the RFID tag based on machine learning provided by the invention, the phase sequence data of the target RFID tag is unwrapped by using the reference phase sequence data, and the phase unwrapped sequence data is obtained, and the method comprises the following steps: acquiring phase difference sequence data of the phase sequence data of the target RFID tag and the reference phase sequence data; disentangled the phase difference sequence data by utilizing a path integration method to obtain disentangled initial phase disentangled sequence data; and taking the sum of the initial phase unwrapping sequence data and the reference phase sequence data as final phase unwrapping sequence data.
According to the method for determining the RFID tag movement behavior based on the machine learning, the tag movement behavior recognition model comprises a first sub-model and a second sub-model; the first sub-model is a machine learning model obtained through training based on phase unwrapping sequence sample data and a motion behavior type corresponding to the phase unwrapping sequence sample data; the second sub-model is a machine learning model obtained through training based on RSSI value sequence sample data and a motion behavior type corresponding to the RSSI value sequence sample data; and comprehensively judging the motion behavior of the target RFID tag by using the output results of the first sub-model and the second sub-model.
According to the method for determining the RFID tag movement behavior based on the machine learning, the first sub-model and the second sub-model are random forest machine learning models.
In a second aspect, the present invention also provides an apparatus for determining RFID tag athletic activity based on machine learning, comprising:
the data acquisition module is used for acquiring tag data of the target RFID tag read by the reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data;
the phase unwrapping module is used for phase unwrapping the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data;
the motion behavior judging module is used for taking the phase unwrapping sequence data and/or the RSSI value sequence data as input data, inputting the input data into a pre-established tag motion behavior recognition model and determining the motion behavior of the target RFID tag;
the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
In a third aspect, the present invention also provides an RFID system comprising: a controller, a reader, at least one antenna, and at least one tag; the reader-writer sends electromagnetic signals with preset frequency to the tag through the antenna; after receiving the electromagnetic signal, the tag feeds back tag data to the reader; the reader transmits the tag data to a controller; the controller performs the method of determining RFID tag athletic performance based on machine learning as set forth in any one of the preceding claims.
In a fourth aspect, the present invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of determining RFID tag movement behaviour based on machine learning as described in any of the above when the program is executed.
In a fifth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a method of determining RFID tag movement behaviour based on machine learning as described in any of the above.
According to the method and the device for determining the RFID tag movement behavior based on machine learning, the tag movement behavior recognition model is trained based on the phase unwrapping sequence data and/or the RSSI value sequence data by utilizing a machine learning algorithm, so that the movement behaviors (static, warehouse-out, warehouse-in and the like) of the tag can be automatically judged, the labor and hardware cost is reduced, and the judgment precision of warehouse-in and warehouse-out information and the logistics management level are improved.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for determining RFID tag athletic activity based on machine learning provided by the present invention;
FIG. 2 is a schematic diagram of a phase unwrapping process according to the present invention;
FIG. 3 is a second schematic diagram of a phase unwrapping process according to the present invention;
FIG. 4 is a schematic diagram of a machine learning based device for determining RFID tag athletic activity in accordance with the present invention;
FIG. 5 is a schematic diagram of the RFID system provided by the present invention;
fig. 6 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that in the description of embodiments of the present invention, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. The orientation or positional relationship indicated by the terms "upper", "lower", etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of description and to simplify the description, and are not indicative or implying that the apparatus or elements in question must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and "coupled" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
The terms "first," "second," and the like in this application are used for distinguishing between similar objects and not for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type and not limited to the number of objects, e.g., the first object may be one or more. In addition, "and/or" indicates at least one of the connected objects, and the character "/", generally indicates that the associated object is an "or" relationship.
Methods and apparatus for determining RFID tag athletic activity based on machine learning provided by embodiments of the present invention are described below in connection with fig. 1-6.
FIG. 1 is a flow chart of a method for determining RFID tag athletic performance based on machine learning provided by the present invention, as shown in FIG. 1, including but not limited to the following steps:
step 101: and acquiring tag data of the target RFID tag read by the reader-writer.
The target RFID tag is a tag for judging the movement behavior; the tag data includes phase sequence data and initial RSSI value sequence data.
Step 102: performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; and filtering the initial RSSI value sequence data to obtain the RSSI value sequence data.
In the phase unwrapping process of the present invention, there may be pi phase jitter problems and 2pi period ambiguity problems. This part can be solved by means of existing technology, and the invention is only briefly described and not described in detail. In particular, both pi phase jitter and 2pi period ambiguity problems can be solved using existing path integration methods. When the path integration method is used, a precondition needs to be satisfied: the acquisition rate of the signals meets the Nyquist sampling theorem, namely, the acquisition frequency of the phase signals is 2 times higher than the highest frequency of the signals, and the correctness and the accuracy of the disentangled phase sequence obtained by the path integration method can be ensured only when the conditions are met.
In an actual signal sampling environment, the phase measurement value of the RFID may have a problem of phase loss due to insufficient data volume of the acquired phase signal. In the case of phase-missing problems, i.e. insufficient data, the phase-missing problem needs to be solved.
Of course, under the condition of sufficient data, other disentanglement steps are directly carried out, so that disentangled bit disentanglement sequence data can be obtained and used for a subsequent tag motion behavior recognition model.
Furthermore, in order to obtain a smoother RSSI value curve (which can be determined by RSSI value sequence data), the invention can also carry out filtering processing on the RSSI value sequence data. Specifically, firstly, performing outlier detection on initial RSSI value sequence data, and extracting statistical characteristics; then, filtering the initial RSSI value sequence data by using a Dixon test method to remove abnormal RSSI values with larger offsets; finally, the final RSSI value sequence data is obtained for the subsequent tag motion behavior recognition model. Step 103: and taking the phase unwrapping sequence data and/or the RSSI value sequence data as input data, inputting the input data into a pre-established tag movement behavior recognition model, and determining the movement behavior of the target RFID tag.
The label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
The specific athletic performance type may be set as desired. In general, the following four athletic performance types may be included: warehouse entry, ex-warehouse, resting, loitering at the warehouse door edge but not entering the warehouse door.
The invention can collect the tag sample data of the RFID tag under the four different motion types, and can acquire the training sample data (namely, input sample data) for training the machine learning model after processing the tag sample data in the step 102.
It will be appreciated that the input sample data in the present invention may include only phase unwrapped sequence sample data; the input sample data may also include only RSSI value sequence sample data; the input sample data may also include two types of data: phase unwrapped sequence sample data and RSSI value sequence sample data.
As the name implies, the phase unwrapped sequence sample data is sample data corresponding to the phase unwrapped sequence data, and the RSSI value sequence sample data is sample data corresponding to the RSSI value sequence data.
According to the method for determining the RFID tag movement behavior based on machine learning, provided by the invention, the tag movement behavior recognition model is trained based on the phase unwrapping sequence data and/or the RSSI value sequence data by utilizing a machine learning algorithm, so that the movement behaviors (static, ex-warehouse, warehouse-in and the like) of the tag can be automatically judged, the labor and hardware cost is reduced, and the judgment precision of the in-warehouse information and the logistics management level are improved.
Based on the foregoing embodiment, as an optional embodiment, the method for determining the motion behavior of an RFID tag based on machine learning provided by the present invention obtains tag data of a target RFID tag read by a reader-writer under the condition that an antenna transmits electromagnetic signals of a plurality of carrier frequency bands, including: reading a plurality of tag data of the target RFID tag corresponding to the carrier frequency bands by using a reader-writer; and carrying out average fusion on the phase sequence data in the plurality of tag data, and taking the fusion result as final phase sequence data.
Specifically, the antenna module of the invention can simultaneously transmit electromagnetic signals with a plurality of carrier frequencies, and correspondingly, the reader-writer can read a plurality of phase sequence data after the target RFID tag receives a plurality of electromagnetic signals.
Further, the invention performs average fusion on the plurality of obtained phase sequence data to obtain a group of average sequence phase data serving as the phase sequence data of the target RFID tag.
According to the method for determining the motion behavior of the RFID tag, when the phase data quantity is insufficient, the data of phase loss can be perfected, so that phase unwrapping can be ensured.
Under the RFID access door system use scene, when fork truck or dolly carry the goods motion, there is the probability and can meet the carrying operation machinery speed too fast, and sampling frequency is unable to keep up and leads to the condition that the data volume is insufficient for the unable unwrapping of phase place smoothly. For the phase-miss problem, the present invention provides a method for guided unwrapping using reference phase sequence data. Fig. 2 is one of the flow charts of phase unwrapping provided in the present invention, as shown in fig. 2, in the case that there is a phase missing problem in the phase sequence data of the target RFID tag, the phase unwrapping is performed on the phase sequence data, including:
step 201: determining a target group to which the target RFID tag belongs to determine all RFID tags in the target group; wherein the grouping of RFID tags is predetermined based on the similarity of the motion of the RFID tags.
It will be appreciated that the phase characteristics of the RFID tags of several items of cargo on the same carrier machine (which may be a forklift or a cart) will generally remain highly consistent, and based on this principle, the present invention may implement grouping of RFID tags. Specifically, the manner of determining the target packet to which the target RFID tag belongs includes:
the carrier machine to which the target RFID tag belongs is determined to determine all RFID tags loaded on the carrier machine (i.e., all RFID tags within the target group). The carrying work machine to which the target RFID tag belongs can be determined according to the tag data, namely the tag data also comprises information of the carrying work machine to which the target RFID tag belongs.
Alternatively, the present invention may pre-use the grouping information of the RFID tags as part of the tag data to facilitate determination of the specific grouping of the RFID tags.
Alternatively, in an RFID access door system, the density of tags is high, and the radial displacement of tags in the same area has very similar motion characteristics, so the invention also groups RFID tags according to different areas.
Step 202: and carrying out data fusion on the phase sequence data of all the RFID tags to determine the reference phase sequence data.
Specifically, first, an average rate of change of the phase sequence data of each RFID tag is determined; then, carrying out average fusion on all the average change rates; and finally, integrating the average fused data based on time to determine the reference phase sequence data.
The formula for the average fusion is as follows:
ntfor the number of tags in the set,for the phase sequence data of the RFID tag t, t represents the serial number of the RFID tag,/or->For average rate of change>Is the average fused data.
By combiningIntegrating the time to obtain reference phase sequence data +.>。
Step 203: and unwrapping the phase sequence data of the target RFID tag by using the reference phase sequence data to obtain phase unwrapped sequence data.
FIG. 3 is a second schematic diagram of a phase unwrapping process according to an alternative embodiment of the present invention; as shown in fig. 3, unwrapping the phase sequence data of the target RFID tag with the reference phase sequence data to obtain phase unwrapped sequence data includes:
step 301: acquiring phase difference sequence data of the phase sequence data of the target RFID tag and the reference phase sequence data;
step 302: disentangled the phase difference sequence data by utilizing a path integration method to obtain disentangled initial phase disentangled sequence data;
step 303: and taking the sum of the initial phase unwrapping sequence data and the reference phase sequence data as final phase unwrapping sequence data.
Specifically, assuming that the path integration method is U, the phase unwrapping process is as follows:
wherein,for the phase sequence data of the RFID tag with phase loss (target RFID tag), the +.>For reference phase sequence data, +.>Indicating that the phase unwrapping process is performed by means of path integration,/->Data representing the initial phase unwrapping sequence after phase unwrapping,/->The unwrapped generated phase unwrapped sequence data is guided for the reference phase sequence data.
Based on the foregoing embodiment, as an optional embodiment, the tag motion behavior recognition model includes a first sub-model and a second sub-model; the first sub-model is a machine learning model obtained through training based on phase unwrapping sequence sample data and a motion behavior type corresponding to the phase unwrapping sequence sample data; the second sub-model is a machine learning model obtained through training based on RSSI value sequence sample data and a motion behavior type corresponding to the RSSI value sequence sample data; and comprehensively judging the motion behavior of the target RFID tag by using the output results of the first sub-model and the second sub-model.
Optionally, the first sub-model and the second sub-model are both random forest machine learning models. Specifically, the invention employs a RF (random forest) random forest machine learning model that combines a plurality of decision trees together, randomly and with a set of data back selected each time, while taking randomly selected portions of the features as input, wherein the combiner selects a majority of the classification results as the final result in the classification problem.
In the implementation process, a first sub-model and a second sub-model are respectively constructed based on the input and output of different speeds and different paths, phase unwrapping sequence sample data and RSSI value sequence sample data which are close to a library gate but do not pass through the library gate.
And classifying the type of the movement behavior of the target RFID tag by using a random forest machine learning model in actual work, so as to judge the final movement behavior type of the target RFID tag.
The random forest machine learning model comprises a first sub-model and a second sub-model, and based on the output results of the first sub-model and the second sub-model, the motion behaviors of the target RFID tag are comprehensively judged, mutual demonstration is carried out, and the accuracy of the judging results is improved.
Fig. 4 is a schematic structural diagram of an apparatus for determining an RFID tag movement behavior based on machine learning according to the present invention, as shown in fig. 4, the apparatus includes:
the data acquisition module 401 is configured to acquire tag data of a target RFID tag read by the reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data;
a phase unwrapping module 402, configured to phase unwrap the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data;
the athletic performance discriminating module 403 is configured to input the phase unwrapping sequence data and/or the RSSI value sequence data as input data to a pre-established tag athletic performance identifying model, and determine an athletic performance of the target RFID tag;
the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
It should be noted that, when the device for determining the motion behavior of the RFID tag based on the machine learning provided by the embodiment of the present invention is specifically running, the method for determining the motion behavior of the RFID tag based on the machine learning described in any one of the above embodiments may be executed, which is not described in detail in this embodiment.
In yet another aspect, the present invention also provides an RFID system, the system comprising: a controller, a reader, at least one antenna, and at least one tag; the reader-writer sends electromagnetic signals with preset frequency to the tag through the antenna; after receiving the electromagnetic signal, the tag feeds back tag data to the reader; the reader transmits the tag data to a controller; the controller performs the method of determining RFID tag athletic performance based on machine learning as set forth in any one of the preceding claims.
Fig. 5 is a schematic structural diagram of the RFID system provided by the present invention, as shown in fig. 5, the reader-writer drives the antenna to transmit a radio frequency signal by supplying power to the antenna through the coaxial cable, the RFID tag is activated by the signal of the reader-writer, and converts the signal into a part of energy, so as to drive the RFID tag to feed back the tag data stored in the chip, thereby realizing the communication between the RFID tag and the reader-writer. After the RSSI value and the phase information of the RFID tag are obtained, the reader-writer transmits the RSSI value and the phase information to a computer (namely a controller) for data processing.
As an optimal embodiment, the invention also provides a design scheme of multiple antennas, and by adopting a method for arranging a pair of antennas at one side inside and outside a library, the data volume is ensured, the antenna signal range is enlarged, at least two and at most four antennas can receive the tag signals no matter in the library or outside the library, compared with the arrangement scheme of a single antenna at one side, the data volume is improved by two to four times, the tag phases can be smoothly unwound, meanwhile, denser data points in unit time can enable RSSI and unwinding phase curves to be smoother, and the observation of the motion track and machine learning modeling of the single tag are ensured.
The whole access door for entering and exiting is protected by an IP67 aluminum alloy shell and is arranged behind the door in the warehouse or at any required position.
Fig. 6 is a schematic structural diagram of an electronic device according to the present invention, and as shown in fig. 6, the electronic device may include: processor 610, communication interface (communications interface) 620, memory 630, and communication bus 640, wherein processor 610, communication interface 620, and memory 630 communicate with each other via communication bus 640. The processor 610 may invoke logic instructions in the memory 630 to perform a method of determining RFID tag athletic performance based on machine learning, the method comprising: acquiring tag data of a target RFID tag read by a reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data; performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data; inputting the phase unwrapping sequence data and/or the RSSI value sequence data as input data to a pre-established tag movement behavior recognition model, and determining movement behaviors of the target RFID tag; the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
Further, the logic instructions in the memory 630 may be implemented in the form of software functional units and stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the method for determining RFID tag movement behavior based on machine learning provided by the above embodiments, the method comprising: acquiring tag data of a target RFID tag read by a reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data; performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data; inputting the phase unwrapping sequence data and/or the RSSI value sequence data as input data to a pre-established tag movement behavior recognition model, and determining movement behaviors of the target RFID tag; the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of determining RFID tag athletic activity based on machine learning, comprising:
acquiring tag data of a target RFID tag read by a reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data;
performing phase unwrapping on the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data;
inputting the phase unwrapping sequence data and/or the RSSI value sequence data as input data to a pre-established tag movement behavior recognition model, and determining movement behaviors of the target RFID tag;
the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
2. The method for determining the movement behavior of an RFID tag based on machine learning according to claim 1, wherein acquiring tag data of a target RFID tag read by a reader/writer in a case where an antenna transmits electromagnetic signals of a plurality of carrier frequency bands, comprises:
reading a plurality of tag data of the target RFID tag corresponding to the carrier frequency bands by using a reader-writer;
and carrying out average fusion on the phase sequence data in the plurality of tag data, and taking the fusion result as final phase sequence data.
3. The method for determining RFID tag motion behavior based on machine learning of claim 1, wherein in the event of a phase-missing problem in the phase sequence data of the target RFID tag, phase unwrapping the phase sequence data comprises:
determining a target group to which the target RFID tag belongs to determine all RFID tags in the target group; wherein the grouping of RFID tags is predetermined based on the similarity of the movement of the RFID tags;
carrying out data fusion on the phase sequence data of all RFID tags to determine reference phase sequence data;
and unwrapping the phase sequence data of the target RFID tag by using the reference phase sequence data to obtain phase unwrapped sequence data.
4. A method for determining RFID tag athletic performance based on machine learning as claimed in claim 3, wherein data fusion of phase sequence data for all RFID tags to determine reference phase sequence data includes:
determining an average rate of change of the phase sequence data for each RFID tag;
carrying out average fusion on all the average change rates;
and carrying out time-based integration on the average fused data to determine reference phase sequence data.
5. A method of determining RFID tag movement behavior based on machine learning as claimed in claim 3, wherein unwrapping the phase sequence data of the target RFID tag with the reference phase sequence data to obtain phase unwrapped sequence data comprises:
acquiring phase difference sequence data of the phase sequence data of the target RFID tag and the reference phase sequence data;
disentangled the phase difference sequence data by utilizing a path integration method to obtain disentangled initial phase disentangled sequence data;
and taking the sum of the initial phase unwrapping sequence data and the reference phase sequence data as final phase unwrapping sequence data.
6. The method of determining RFID tag athletic activity based on machine learning of claim 1, wherein the tag athletic activity identification model includes a first sub-model and a second sub-model; the first sub-model is a machine learning model obtained through training based on phase unwrapping sequence sample data and a motion behavior type corresponding to the phase unwrapping sequence sample data; the second sub-model is a machine learning model obtained through training based on RSSI value sequence sample data and a motion behavior type corresponding to the RSSI value sequence sample data;
and comprehensively judging the motion behavior of the target RFID tag by using the output results of the first sub-model and the second sub-model.
7. The method of determining RFID tag athletic activity based on machine learning of claim 6, wherein the first sub-model and the second sub-model are both random forest machine learning models.
8. An apparatus for determining RFID tag athletic activity based on machine learning, comprising:
the data acquisition module is used for acquiring tag data of the target RFID tag read by the reader-writer; the tag data comprises phase sequence data and initial RSSI value sequence data;
the phase unwrapping module is used for phase unwrapping the phase sequence data to obtain phase unwrapped sequence data; filtering the initial RSSI value sequence data to obtain RSSI value sequence data; the motion behavior judging module is used for taking the phase unwrapping sequence data and/or the RSSI value sequence data as input data, inputting the input data into a pre-established tag motion behavior recognition model and determining the motion behavior of the target RFID tag;
the label motion behavior recognition model is a machine learning model which is obtained through training based on input sample data and motion behavior types corresponding to the input sample data.
9. An RFID system, comprising: a controller, a reader, at least one antenna, and at least one tag;
the reader-writer sends electromagnetic signals with preset frequency to the tag through the antenna;
after receiving the electromagnetic signal, the tag feeds back tag data to the reader;
the reader transmits the tag data to a controller;
the controller performs the method of determining RFID tag athletic activity based on machine learning of any one of claims 1-7.
10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the method of determining RFID tag movement behavior based on machine learning as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311761455.0A CN117435902B (en) | 2023-12-20 | 2023-12-20 | Method and device for determining RFID tag movement behavior based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311761455.0A CN117435902B (en) | 2023-12-20 | 2023-12-20 | Method and device for determining RFID tag movement behavior based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117435902A true CN117435902A (en) | 2024-01-23 |
| CN117435902B CN117435902B (en) | 2024-04-02 |
Family
ID=89552098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311761455.0A Active CN117435902B (en) | 2023-12-20 | 2023-12-20 | Method and device for determining RFID tag movement behavior based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117435902B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108414972A (en) * | 2018-04-26 | 2018-08-17 | 华中科技大学 | A kind of mobile robot RFID localization methods based on phase property |
| CN108734512A (en) * | 2018-05-03 | 2018-11-02 | 北京星网锐捷网络技术有限公司 | A kind of article attention rate processing method, equipment, system and storage medium |
| CN110711701A (en) * | 2019-09-16 | 2020-01-21 | 华中科技大学 | Grabbing type flexible sorting method based on RFID space positioning technology |
| US20200097690A1 (en) * | 2019-11-26 | 2020-03-26 | Intel Corporation | Behavior detection using rfid in environments with high rfid tag density |
| CN112698328A (en) * | 2020-11-30 | 2021-04-23 | 四川大学 | Phase unwrapping method and system for monitoring dam and landslide deformation GB-SAR |
| CN113361450A (en) * | 2021-06-24 | 2021-09-07 | 上海鼎算智能科技有限公司 | RFID-based activity sequence identification method, system, medium and terminal |
| WO2022214114A2 (en) * | 2021-08-10 | 2022-10-13 | 中咨数据有限公司 | Bridge deformation monitoring method fusing gnss data and insar technology |
| CN116486341A (en) * | 2023-04-27 | 2023-07-25 | 厦门大学 | Training and identifying method and device for human body behavior identification model based on RFID |
| CN116738144A (en) * | 2023-05-08 | 2023-09-12 | 厦门大学 | Human behavior detection method, device, medium and equipment based on RFID |
| CN117109581A (en) * | 2023-08-04 | 2023-11-24 | 华中科技大学 | Mobile robot self-positioning navigation method, system, electronic equipment and storage medium |
| CN117114030A (en) * | 2022-05-12 | 2023-11-24 | 中国移动通信有限公司研究院 | Warehouse-in and warehouse-out identification method, device, identification system and readable storage medium |
-
2023
- 2023-12-20 CN CN202311761455.0A patent/CN117435902B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108414972A (en) * | 2018-04-26 | 2018-08-17 | 华中科技大学 | A kind of mobile robot RFID localization methods based on phase property |
| CN108734512A (en) * | 2018-05-03 | 2018-11-02 | 北京星网锐捷网络技术有限公司 | A kind of article attention rate processing method, equipment, system and storage medium |
| CN110711701A (en) * | 2019-09-16 | 2020-01-21 | 华中科技大学 | Grabbing type flexible sorting method based on RFID space positioning technology |
| US20200097690A1 (en) * | 2019-11-26 | 2020-03-26 | Intel Corporation | Behavior detection using rfid in environments with high rfid tag density |
| CN112698328A (en) * | 2020-11-30 | 2021-04-23 | 四川大学 | Phase unwrapping method and system for monitoring dam and landslide deformation GB-SAR |
| CN113361450A (en) * | 2021-06-24 | 2021-09-07 | 上海鼎算智能科技有限公司 | RFID-based activity sequence identification method, system, medium and terminal |
| WO2022214114A2 (en) * | 2021-08-10 | 2022-10-13 | 中咨数据有限公司 | Bridge deformation monitoring method fusing gnss data and insar technology |
| CN117114030A (en) * | 2022-05-12 | 2023-11-24 | 中国移动通信有限公司研究院 | Warehouse-in and warehouse-out identification method, device, identification system and readable storage medium |
| CN116486341A (en) * | 2023-04-27 | 2023-07-25 | 厦门大学 | Training and identifying method and device for human body behavior identification model based on RFID |
| CN116738144A (en) * | 2023-05-08 | 2023-09-12 | 厦门大学 | Human behavior detection method, device, medium and equipment based on RFID |
| CN117109581A (en) * | 2023-08-04 | 2023-11-24 | 华中科技大学 | Mobile robot self-positioning navigation method, system, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 崔英花;: "快速运动物体的分布式多天线阅读器定位系统", 科技导报, no. 18, 28 September 2017 (2017-09-28), pages 60 - 65 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117435902B (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang | Intelligent and integrated RFID (II-RFID) system for improving traceability in manufacturing | |
| JP4963237B2 (en) | Tag reader | |
| CN103049838B (en) | Package shipment method for monitoring state based on finite state machine | |
| CN113859847B (en) | Storage checking method and storage system | |
| CN109492724A (en) | Warehouse, warehouse compartment state identification method, logistics execute system and electronic equipment | |
| CN112488615A (en) | Warehouse in and out and inventory management method | |
| CN102103716A (en) | Intelligent warehouse goods position distribution verification system and method | |
| CN109543782A (en) | A kind of automated warehousing freight detection method, apparatus and system | |
| CN112465085A (en) | Intelligent packing production management system of thing networking based on RFID technique | |
| Keller et al. | Decreasing false-positive RFID tag reads by improved portal antenna setups | |
| CN101739543A (en) | Ic tag reading necessity determination method and ic tag reading necessity determination apparatus | |
| CN117435902B (en) | Method and device for determining RFID tag movement behavior based on machine learning | |
| CN113283839B (en) | Error-preventing automatic opening system of cigarette sale ex-warehouse RFID code scanning equipment | |
| CN114291482B (en) | LiDAR-based monitoring in a materials handling environment | |
| US10885412B2 (en) | Systems and methods for tracking produce | |
| EP4302248A1 (en) | System and method of monitoring an industrial environment | |
| CN112581046A (en) | Warehouse logistics management system and method | |
| CN109509003A (en) | Supply chain system for tracing and managing based on RFID technique | |
| CN113570316A (en) | Access management method and access management system | |
| CN113159246A (en) | Steel mill cargo identification method and device based on two-dimensional code label and computer equipment | |
| Windmann et al. | A novel self-configuration method for RFID systems in industrial production environments | |
| CN111275386A (en) | Method and system for intelligent transportation, electronic device, and computer storage medium | |
| CN201698448U (en) | Trigger fault-tolerant addressing large-scale data warehouse collector based on wireless radio frequency identification | |
| CN101551659A (en) | Storage monitoring system integrating wireless radio-frequency identification | |
| US20230031656A1 (en) | Method and device for transferring information flow of customized furniture information management and control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |