A method for identifying electronic tags of metal cans in complex environments
By analyzing the difference in signal recognition quality and frequency domain energy distribution of metal tanks in the intelligent entry and exit control cabinet, combining the nearest neighbor interference degree and signal distortion degree, signal decomposition and denoising are solved, and signal interference problem in metal tank label identification environment is improved. Identification efficiency and accuracy.
Patent Information
- Application Number
- CN202411139493.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-08-20
AI Technical Summary
In the intelligent entry and exit control cabinet, the random access of metal cans leads to complex RFID tag identification environment, signal interference and dynamic changes, making it difficult to obtain accurate dynamic information, affecting the recognition efficiency and accuracy.
By obtaining the initial identification signal of each hole position and the before and after signals, analyzing the difference in signal identification quality and frequency domain energy distribution, combining the nearest neighbor interference degree and signal distortion degree, signal decomposition and denoising, and obtaining filtered identification signals to improve identification efficiency .
It improves the identification accuracy and signal quality of metal can label information, reduces the impact of dynamic interference, and ensures the clarity and accuracy of label information.
Smart Images

Figure CN118657160B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of radio frequency identification, and in particular to a method for identifying electronic tags of metal cans in a complex environment. Background Art
[0002] Automatic in-and-out identification of metal cans is an important part of intelligent manufacturing and warehouse management, mainly concentrated in the food, beverage and chemical industries. Metal cans may be used to store and transport various chemicals, liquid additives or gases, etc. The in-and-out management of metal cans is of great significance to the production cost and safety of enterprises. Usually, enterprises use intelligent in-and-out control cabinets to achieve automatic management and intelligent decision-making of materials. The intelligent in-and-out control cabinet is equipped with an RFID antenna board to track and identify the RFID tags on the metal cans.
[0003] Due to the random access of metal cans, the actual RFID tag recognition environment is very complex. In the smart access cabinet, multiple metal cans have RFID tags. Metal is a good conductor of electromagnetic waves, which can hinder the reader signal and make the metal surface behind the tag reflect the signal, reducing the recognition efficiency. The mutual interference between tags during collection will also lead to reading errors. At the same time, the tag information of the metal can is dynamically updated when it is continuously accessed. Once the relative position of the electronic tag and the card reader changes, the interference component will also change accordingly, which makes it difficult for the card reader to obtain the accurate dynamic information of the metal can tag, affecting the recognition and processing of the tag information. Summary of the invention
[0004] In order to solve the technical problem that it is difficult to obtain accurate dynamic information of metal can labels in the prior art, which affects the identification and processing of label information, the purpose of the present invention is to provide a method for identifying metal can electronic labels in a complex environment. The technical solution adopted is as follows:
[0005] The present invention provides a method for identifying electronic tags of metal cans in a complex environment, the method comprising:
[0006] Obtain the initial recognition signal of the electronic tag on each hole of the metal can in the intelligent access control cabinet, as well as the recognition signal of each metal can on the hole before taking and after returning;
[0007] According to the degree of confusion deviation between each well position and the local well position in the initial recognition signal, the signal recognition quality of each well position is obtained;
[0008] According to the difference in energy distribution of the identification signal of each metal can before and after it is returned in the frequency domain, and the difference in the signal recognition quality of the corresponding hole position before and after it is returned, the dynamic interference of each metal can is obtained;
[0009] According to the changes in the local distribution of the identification signals of each metal can before being taken and after being returned, as well as the dynamic interference of the metal can's taking, the identification signals of each metal can after being returned are decomposed and denoised to obtain the filtered identification signals of each metal can after being returned; the filtered identification signals of each metal can are stored.
[0010] Furthermore, the method for obtaining the signal recognition quality includes:
[0011] For any hole position, the neighbor interference degree of the hole position is obtained according to the distribution disorder of the phase difference between the hole position and the local hole position in the initial recognition signal;
[0012] Obtaining the signal distortion degree of the well position according to the degree of deviation of the initial recognition signal of the well position among the initial recognition signals of all well positions;
[0013] The signal recognition quality of the hole position is obtained according to the neighbor interference degree and signal distortion degree of the hole position; both the neighbor interference degree and the signal distortion degree are negatively correlated with the signal recognition quality.
[0014] Furthermore, the method for obtaining the neighbor interference degree includes:
[0015] Using other hole positions within a preset neighborhood of the hole position as local hole positions of the hole position;
[0016] Obtaining the phase spectrum of the initial identification signal of the hole position and each local hole position; converting the mean of the phase difference between the phase spectrum of the hole position and each local hole position at the same frequency into a data form as the phase deviation between the hole position and each local hole position;
[0017] The variance of the phase deviation between the hole position and all local hole positions is taken as the neighbor interference degree of the hole position.
[0018] Furthermore, the method for obtaining the signal distortion includes:
[0019] The average of the initial recognition signals of all well positions is taken as the average recognition signal;
[0020] The mean square error between the initial recognition signal of the hole position and the average recognition signal is taken as the signal distortion of the hole position.
[0021] Furthermore, the method for obtaining dynamic interference includes:
[0022] According to the frequency distribution difference of the identification signal of each metal can before and after it is taken out and returned in the frequency domain space, and the energy density proportion of the frequency in the frequency domain, the energy change degree of each metal can is obtained;
[0023] The difference between the signal recognition quality of the hole position of the metal can before taking and the signal recognition quality of the hole position after returning is used as the change degree of taking recognition of each metal can;
[0024] The difference between the access energy variation degree and the access identification variation degree of each metal can is taken as the access dynamic interference of each metal can.
[0025] Furthermore, the method for obtaining the energy variation degree includes:
[0026] For any metal can, when the frequency in the frequency domain space of the identification signal after the return is not in the frequency domain space of the identification signal before the use, the corresponding frequency is used as the newly added frequency;
[0027] In the frequency domain space of the identification signal corresponding to the returned metal can, the square modulus of the amplitude corresponding to each newly added frequency is taken as the energy density of each newly added frequency;
[0028] The sum of the energy densities of all newly added frequencies is normalized to obtain the energy variation of the metal can.
[0029] Furthermore, the method for obtaining the filtering identification signal includes:
[0030] Divide the identification signals of each metal can before it is taken out and after it is returned in the same way to obtain more than two signal analysis windows;
[0031] In each signal analysis window, according to the signal distribution difference and signal distribution correlation of the identification signal before taking and after returning, the local update amount of each signal analysis window is obtained;
[0032] According to the dynamic interference of each metal can's use and the local update amount of each signal analysis window, the value of the identification signal of the metal can after being returned in each signal analysis window is corrected to obtain the adjusted reference signal of each metal can in each signal analysis window;
[0033] The adjusted reference signal of each metal can in all signal analysis windows is subjected to signal decomposition and denoising to obtain a filtered identification signal of each metal can after it is returned.
[0034] Furthermore, the method for obtaining the local update amount includes:
[0035] For any signal analysis window of any metal can, the ratio of the variance of the identification signal before the metal can is taken out to the variance of the identification signal after the metal can is returned in the signal analysis window is taken as the signal-to-noise ratio of the signal analysis window;
[0036] In the signal analysis window, the amplitude distribution probability of the metal can on the identification signal before taking and after returning is obtained; the conditional entropy of the amplitude distribution probability of the metal can after returning under the amplitude distribution probability before taking is taken as the information consumption of the signal analysis window;
[0037] According to the signal-to-noise ratio and information consumption of the signal analysis window, the local update amount of the signal analysis window is obtained, and both the signal-to-noise ratio and the information consumption are positively correlated with the local update amount.
[0038] Furthermore, the method for obtaining the adjustment reference signal includes:
[0039] For any signal analysis window of any metal can, obtain the extreme value of the identification signal after the metal can is returned in the signal analysis window;
[0040] The ratio of the dynamic interference of the metal can to the local update amount of the signal analysis window is used as the adjustment coefficient;
[0041] The product of each extreme value of the identification signal after the metal can is returned in the signal analysis window and the adjustment coefficient is used as the adjustment extreme value of the signal analysis window;
[0042] An adjustment reference signal of the signal analysis window is obtained according to all adjustment extreme values in the signal analysis window.
[0043] Furthermore, the signal decomposition and denoising of the adjusted reference signal of each metal can in all signal analysis windows to obtain the filtered identification signal of each metal can after the return includes:
[0044] Performing ITD signal decomposition on each signal analysis window based on the adjusted reference signal of the signal analysis window to obtain a first component;
[0045] The first components in all signal analysis windows corresponding to the metal can are connected to obtain a first decomposition result of the metal can; the first decomposition result is removed from the identification signal of the metal can after it is returned to obtain a filtered identification signal of the metal can.
[0046] The present invention has the following beneficial effects:
[0047] The present invention first analyzes the signal recognition quality of the hole positions where the metal cans are placed in the intelligent access control cabinet, and obtains the signal recognition quality through the initial recognition signal confusion analysis of the hole positions and local hole positions, characterizing the static interference that the metal cans will be subjected to at different hole positions, so as to facilitate the subsequent analysis of the hole position changes when they are taken. Taking into account the random situation of the metal cans, the hole position changes of the metal cans when they are taken are comprehensively considered, and the dynamic interference of the use is obtained by combining the energy difference of the recognition signal of each metal can in the frequency domain before and after the use, reflecting the degree of dynamic noise brought by the metal cans after use and return, so as to adjust the filtering of the signal. Further, for the local changes of the recognition signal before and after the use, the degree of dynamic update of the signal information is locally estimated, combined with the dynamic interference of the use, the dynamic interference after the return and the degree of information update are decomposed and denoised, so as to filter the dynamic situation of the random use of the metal cans more accurately, and obtain a clearer and more accurate filtered recognition signal for storage. The present invention analyzes the static interference of the hole positions for placing metal cans, as well as the dynamic interference and dynamic information brought by the process of taking the metal cans, adaptively filters the identification information of each metal can, improves the signal quality and makes the identified label information clearer. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0049] Figure 1 A flow chart of a method for identifying electronic tags of metal cans in a complex environment provided by an embodiment of the present invention;
[0050] Figure 2 A schematic diagram of an RFID tag for a metal can provided by one embodiment of the present invention;
[0051] Figure 3 A schematic diagram of an RFID system provided by an embodiment of the present invention;
[0052] Figure 4 A flow chart of a method for obtaining signal recognition quality provided by an embodiment of the present invention;
[0053] Figure 5 A flow chart of a method for obtaining dynamic interference provided by an embodiment of the present invention;
[0054] Figure 6A flow chart of a method for obtaining a filtering identification signal provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0055] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of the metal can electronic tag identification method in a complex environment proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0056] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0057] The following is a detailed description of a method for identifying electronic tags of metal cans in a complex environment provided by the present invention with reference to the accompanying drawings.
[0058] See also Figure 1 , which shows a flow chart of a method for identifying a metal can electronic tag in a complex environment provided by an embodiment of the present invention, the method comprising the following steps:
[0059] S1: Obtain the initial recognition signal of the electronic tag at each hole of the metal can in the smart access cabinet, as well as the recognition signal of each metal can at the hole before taking and after returning.
[0060] In the embodiment of the present invention, the intelligent in-and-out storage control cabinet can accommodate 300 metal cans, and uses a plate-type antenna layer with holes. All holes on the antenna layer are arranged in a square matrix. Metal cans are placed in each hole, and each hole is evenly equipped with a small LED light to indicate the usability of the corresponding metal can. The control cabinet contains a database to record the storage information of the metal cans, etc., to achieve information management and processing of tag recognition.
[0061] Each metal can has an RFID anti-metal tag on the bottom. Figure 2, which shows a schematic diagram of an RFID tag for a metal can provided by an embodiment of the present invention. The RFID system consists of three parts: an electronic tag, a reader / writer, and a computer and its application. Among them, the electronic tag is mainly composed of a built-in antenna and a circuit chip, and has the function of completing communication with the radio frequency antenna. In addition, since the material stored in the metal can is a consumable, the electronic tag adopts an active tag, which is equipped with a built-in power supply and can actively update the status information of the metal can. The reader / writer is mainly composed of four parts: an antenna, a control unit, a wireless transceiver front end, and a communication interface. Its main function is to read or write the information of the electronic tag. The card reader transmits radio waves, which activate the active tag and provide it with sufficient energy for broadcasting. The activated tag uses the built-in antenna to send the information stored on the chip back to the card reader via radio waves. The computer and application software are connected to an external computer through the communication interface of the reader / writer to realize data exchange. The card reader can read the RFID tag information of each metal can and send the signal to the computer through the communication interface. Please refer to Figure 3 , which shows a schematic diagram of an RFID system provided by an embodiment of the present invention.
[0062] RFID is a wireless communication technology, also known as radio frequency identification technology, which can identify and track tagged objects through radio waves. Each RFID tag has a unique EPC serial number, which is used to represent a single item. The tag contains information related to the item, such as the serial number of the metal can, the storage material in the can, the storage content, the access record, and other real-time updated status data.
[0063] When the card reader identifies each metal can RFID tag, there may be multipath effects, signal reflections, signal attenuation, backscattering effects, and signal collisions and mutual interference when multiple metal cans are taken in and out at the same time. These problems are common problems in the multi-object in-and-out electronic tag identification scenario. Although they can be solved by corresponding protocols and strategies such as anti-collision protocols, antenna system design, or time division multiple access technology, interference problems will still exist. For example, the RFID tags of metal cans entering and leaving the warehouse cannot be identified, or the card reader cannot decode distorted wireless signals. When the wireless signals sent by the electronic tags are smoothed by the signal processing system in the card reader, the data after the RFID tags are dynamically updated will still be lost or deformed. Therefore, it is necessary to optimize the smoothing of dynamic information.
[0064] In an embodiment of the present invention, the metal cans in the intelligent access control cabinet are first filled, and the card reader is used to traverse the electronic tags of the metal cans to obtain wireless signals, and obtain the initial identification signal corresponding to each hole position. Since the state of the metal can changes when it is taken away and returned, the wireless signal of the metal can that has not left the hole position is used as the identification signal before each metal can is taken. Since the electronic tag has a unique EPC serial number, the hole position when the metal can is put back can be determined according to the serial number, and since the state may change, the re-scanned wireless signal is used as the identification signal after the metal can is returned. It should be noted that the specific signal acquisition process is a technical means well known to technicians in this field and will not be elaborated here.
[0065] S2: The signal recognition quality of each hole position is obtained according to the degree of confusion deviation between each hole position and the local hole position in the initial recognition signal.
[0066] Since the antenna layer is usually designed to radiate or receive signals in a specific direction, when the metal can electronic tags on different hole positions exchange data with the antenna layer, the interference and signal strength they receive are different due to the different position distances and receiving directions. The smaller the interference and the clearer the signal, the weaker the static interference in the signal recognition of the metal can.
[0067] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the signal recognition quality of each well position can be found in Figure 4 , which shows a flow chart of a method for obtaining signal recognition quality provided by an embodiment of the present invention, the method comprising the following steps:
[0068] S201: For any hole position, according to the distribution disorder of the hole position and the local hole position in the phase difference of the initial recognition signal, obtain the neighbor interference degree of the hole position.
[0069] The phase difference reflects the degree of signal interference between hole positions. Ideally, the initial state of the metal can is the same, and the initial identification signal obtained should also be the same. However, due to the close position and the strong electromagnetic conduction of the metal can, the signal reception strength and phase at different hole positions will differ.
[0070] In the embodiment of the present invention, first, other hole positions within the preset neighborhood range of the hole position are used as local hole positions of the hole position. The preset neighborhood range is the eight neighborhood range of the hole position, and the hole positions are distributed in a matrix in the intelligent in-and-out storage control cabinet. The implementer can adjust the preset range according to the specific implementation scenario, which is not limited here.
[0071] Furthermore, the phase spectrum of the initial identification signal of the hole position and each local hole position is obtained, and the signal can be converted into the frequency domain using Fourier transform to obtain the phase spectrum. The mean of the phase difference between the phase spectrum of the hole position and each local hole position at the same frequency is converted into data form as the phase deviation of the hole position and each local hole position, and the phase deviation reflects the interference component of the hole position and each local hole position. Since the phase may exist in a complex form, it can be converted into a digital form by a cosine function, which is convenient for subsequent data analysis and calculation. It should be noted that the Fourier transform to obtain the phase spectrum and the complex data form conversion are both technical means well known to technicians in this field, and will not be repeated here.
[0072] Finally, the variance of the phase deviation between the hole position and all local hole positions is taken as the neighbor interference degree of the hole position. The variance reflects the degree of distribution disorder of the phase deviation. The larger the neighbor interference degree, the greater the difference in interference between the hole position and the local hole position, and the more complex the interference component. On the contrary, the smaller the neighbor interference degree, the more similar the interference between the hole position and the local hole position, and the simpler the interference component.
[0073] S202: Obtain the signal distortion degree of the hole position according to the deviation degree of the initial recognition signal of the hole position in the initial recognition signals of all the hole positions.
[0074] The deviation of the initial recognition signal of the hole position from the overall initial recognition signal reflects the degree of signal distortion at the hole position. In the embodiment of the present invention, the mean of the initial recognition signals of all the hole positions is taken as the average recognition signal to characterize the ideal consistency of the recognition signals at the hole positions. The mean square error between the initial recognition signal of the hole position and the average recognition signal is taken as the signal distortion degree of the hole position to reflect the degree of distortion at the hole position. The greater the signal distortion degree of the hole position, the higher the deviation of the acquired signal is, and the more serious the interference effect on the hole position is.
[0075] S203: Obtaining the signal recognition quality of the hole position according to the neighbor interference degree and signal distortion degree of the hole position; both the neighbor interference degree and the signal distortion degree are negatively correlated with the signal recognition quality.
[0076] The signal recognition quality is obtained by combining the neighbor interference degree and signal distortion degree of the hole position. In the embodiment of the present invention, the product of the neighbor interference degree and the signal distortion degree of the hole position is negatively correlated and normalized to obtain the signal recognition quality. In a specific implementation method of the present invention, the expression of the signal recognition quality is:
[0077] β i =1-norm(A i ×B i );where β i It is expressed as the signal recognition quality of the ith well, Ai It is expressed as the neighbor interference degree of the i-th hole position, B i It is represented as the signal distortion of the ith hole position, norm() is represented as the normalization processing function. It should be noted that normalization is a technical means well known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.
[0078] The greater the signal recognition quality of the hole position, the more it means that the interference component at the hole position is single and the degree of distortion is small, and the recognition signal of the metal can at the corresponding hole position is clearer.
[0079] S3: The dynamic interference of each metal can is obtained according to the difference in energy distribution of the identification signal in the frequency domain before and after each metal can is taken and returned, and the difference in signal recognition quality of the corresponding hole position before and after taking and returning.
[0080] Since the use of metal cans is not disposable, that is, the objects in the metal cans may only be partially consumed each time they are taken out, and the metal cans will be returned when they are not completely consumed, and the circumstances in which the metal cans are taken out and the locations where they are returned are also random. Therefore, when each metal can is taken out and placed, part of the information will be automatically updated to the data stored in the metal can electronic tag through the RFID communication protocol, and the status information of the metal can will be updated once. The recorded information of each metal can's electronic tag before and after the update is different, and the wireless signal is also different.
[0081] The original RF signal received by the card reader from the electronic tag usually contains noise and possible interference. Since the antenna layer is usually designed to radiate or receive signals in a specific direction, when the hole position changes before and after placement, the strength of the interactive signal between the metal can electronic tag and the antenna layer is different, and the interference components attached to the signal will also change. Therefore, in order to ensure that the electronic tag signal can be accurately located and read when the metal can is randomly placed, it is necessary to compare and analyze the interference components in the wireless signal before and after placement, and comprehensively consider the static environmental interference of the hole position and the estimated dynamic interference level to improve the accuracy of subsequent analysis of the real signal.
[0082] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining dynamic interference is described in Figure 5 , which shows a flow chart of a method for obtaining dynamic interference provided by an embodiment of the present invention, the method comprising the following steps:
[0083] S301: Obtain the energy variation of each metal can according to the frequency distribution difference of the identification signal of each metal can before and after it is returned in the frequency domain space, and the energy density ratio of the frequency in the frequency domain.
[0084] Even if the electronic tag is updated, the frequency components of the wireless signal are fixed. When the frequency changes in the signal after placement, it may be caused by the change in the position of the hole or the dynamic noise introduced when it is returned. Therefore, a preliminary analysis of the energy density of the frequency change is performed to reflect the degree of the changed part.
[0085] In an embodiment of the present invention, for any metal can, when the frequency in the frequency domain space of the identification signal after return is not in the frequency domain space of the identification signal before use, the corresponding frequency is used as a newly added frequency to determine the new frequency change generated after return.
[0086] Furthermore, in the frequency domain space of the identification signal after the metal can is returned, the square modulus of the amplitude corresponding to each newly added frequency is used as the energy density of each newly added frequency, and the sum of the energy densities of all newly added frequencies is normalized to obtain the energy change degree of the metal can, reflecting the degree of interference after the signal is taken and placed.
[0087] S302: The difference between the signal recognition quality of the hole position of the metal can before taking and the signal recognition quality of the hole position after returning is used as the taking recognition change degree of each metal can.
[0088] Since the energy change includes the static interference caused by the hole position change, it is relatively convenient to eliminate the static interference part, for example, directly obtaining the static signal-to-noise ratio of the hole position based on the deviation between the recognition signals for denoising, etc., while the added dynamic noise will be directly processed, resulting in noise signal loss or deformation, etc., causing the information component to change. Therefore, it is necessary to reflect the significance of dynamic noise in combination with the hole position change. In an embodiment of the present invention, the absolute value of the difference in signal recognition quality between the hole position before taking and after returning is used as the recognition change degree to reflect the degree of influence caused by the hole position.
[0089] S303: The difference between the access energy variation degree and the access identification variation degree of each metal can is used as the access dynamic interference of each metal can.
[0090] Combining the changes in signal recognition quality between hole positions and the changes in the energy density ratio of the overall frequency, the correlation degree of the changes is analyzed. The more consistent the degree of change, the less significant the dynamic interference component is, and vice versa, the higher the impact of dynamic interference is.
[0091] In the embodiment of the present invention, the absolute value of the difference between the access energy variation degree and the access identification variation degree of the metal can is used as the access dynamic interference of the metal can. The expression of the access dynamic interference is:
[0092] ; In the formula, k z It is expressed as the dynamic disturbance of the access to the z-th metal can, It is expressed as the signal recognition quality of the hole position of the z-th metal can before taking it, It is expressed as the signal recognition quality of the hole position of the z-th metal can after it is returned, It is expressed as the energy density of the sth additional frequency of the zth metal can after it is returned, N z It is represented as the total number of new frequencies of the z-th metal can, || is represented as the absolute value extraction function, and norm() is represented as the normalization processing function.
[0093] in, It is expressed as the change degree of the access recognition of the z-th metal can, It is expressed as the energy variation of the z-th metal can. The closer the variation is, the more likely it is that the interference to the metal can is all static interference caused by the hole position. Conversely, the proportion of dynamic interference to the metal can is higher.
[0094] In other embodiments of the present invention, the difference between the ratio of the access energy variation degree to the access identification variation degree and the value 1 may also be used as the access dynamic interference, which will not be elaborated here.
[0095] At this point, the analysis of dynamic interference superposition during the process of picking and placing metal cans is completed.
[0096] S4: Based on the changes in the local distribution of the identification signals of each metal can before being taken and after being returned, as well as the dynamic interference of the metal can's use, the identification signals of each metal can after being returned are decomposed and denoised to obtain the filtered identification signals of each metal can after being returned; the filtered identification signals of each metal can are stored.
[0097] The addition of dynamic noise will affect the change of the real information components in the signal. Therefore, local analysis is performed on the recognition signal to estimate the degree of change of the dynamic information and analyze whether the dynamically changing information is real and effective information or caused by noise, so as to make the denoising process more reliable and reduce the degree of filtering of effective information.
[0098] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the filtering identification signal can be found in Figure 6 , which shows a flow chart of a method for obtaining a filtering identification signal provided by an embodiment of the present invention, the method comprising the following steps:
[0099] S401: Divide the identification signals of each metal can before use and after return in the same way to obtain more than two signal analysis windows.
[0100] Since the superposition distribution of effective dynamic information is relatively concentrated, while the superposition of dynamic interference is close to uniform distribution, it is necessary to perform local analysis on the recognition signal, so the recognition signal is divided into windows. In an embodiment of the present invention, the waveform characteristics of the signal can be analyzed by a window function, such as a Hanning window or a Hamming window, and the window function is slid along the recognition signal after the metal can is returned. The signal variance in the window is counted, and when it exceeds the preset division threshold, the part in the window is used as a signal analysis window, and continues to slide from the next position of the signal analysis window until all information analysis windows are obtained. The recognition signal before the metal can is taken is directly divided according to the signal analysis window of the recognition signal after the return. The window function can be used to divide according to the approximate distribution characteristics in the signal. It should be noted that the division of the window function is a technical means well known to technicians in this field, and will not be elaborated here.
[0101] In other embodiments of the present invention, a preset window size may be used to divide the identification signals before and after use to obtain multiple signal analysis windows. The implementers may adjust the specific settings according to the specific implementation situation to ensure that the division of the identification signals is the same. No limitation or elaboration is made here.
[0102] S402: In each signal analysis window, according to the signal distribution difference and signal distribution correlation of the identification signal before taking and after returning, a local update amount of each signal analysis window is obtained.
[0103] By analyzing the signal relationship of the identification signal in each window, a preliminary estimate of the information update status of the electronic tag is made to facilitate the subsequent analysis of the effectiveness and noise level of the information.
[0104] In an embodiment of the present invention, first, for any signal analysis window of any metal can, the ratio of the variance of the identification signal before the metal can is taken and the variance of the identification signal after the metal can is returned in the signal analysis window is used as the signal-to-noise ratio of the signal analysis window, reflecting the noise content in the window.
[0105] Secondly, in the signal analysis window, the amplitude distribution probability of the metal can on the identification signal before taking and after returning is obtained, and the conditional entropy of the amplitude distribution probability of the metal can after returning under the amplitude distribution probability before taking is used as the information consumption of the signal analysis window. The conditional entropy of the amplitude distribution reflects the degree of information consumption in the signal from before taking to after returning. The larger the conditional entropy, the higher the information consumption.
[0106] Finally, according to the signal-to-noise ratio and information consumption of the signal analysis window, the local update amount of the signal analysis window is obtained. The signal-to-noise ratio and the information consumption are both positively correlated with the local update amount. In an embodiment of the present invention, the product of the normalized signal-to-noise ratio and the information consumption can be used as the local update amount. In other embodiments of the present invention, other basic mathematical operations can be used to reflect that the signal-to-noise ratio and the information consumption are both positively correlated with the local update amount, such as addition or power operations, etc., which are not limited here. The larger the signal-to-noise ratio, the less noise content in the returned recognition signal. At the same time, the larger the information consumption, the larger the information consumption in the returned signal. At this time, the information consumption is more likely to be caused by dynamic valid information.
[0107] S403: According to the dynamic interference of each metal can's use and the local update amount of each signal analysis window, the value of the identification signal of the metal can after being returned in each signal analysis window is corrected to obtain the adjustment reference signal of each metal can in each signal analysis window.
[0108] When decomposing a signal, it will be decomposed based on the baseline of the signal. The baseline can obtain the upper and lower envelopes according to the maximum and minimum values in the signal, and then obtain the mean envelope as the baseline. For each signal analysis window, the extreme value data of the signal is adjusted to obtain an adjusted reference signal so that the subsequent decomposition adjustment based on the signal segment is more reliable.
[0109] In an embodiment of the present invention, for any signal analysis window of any metal can, the extreme value of the identification signal after the metal can is returned in the signal analysis window is obtained. It should be noted that the method for obtaining the extreme value is a technical means well known to technicians in this field and will not be elaborated here.
[0110] Furthermore, the ratio of the dynamic interference of the metal can to the local update amount of the signal analysis window is used as the adjustment coefficient, and the product of each extreme value of the identification signal after the metal can is returned in the signal analysis window and the adjustment coefficient is used as the adjustment extreme value of the signal analysis window. In a specific implementation of the embodiment of the present invention, the expression of the adjustment extreme value is:
[0111] ; In the formula, It is expressed as the adjusted extreme value of the wth extreme value in the cth signal analysis window of the zth metal can, It is expressed as the value of the wth extreme value in the cth signal analysis window of the zth metal tank, k z It is expressed as the dynamic disturbance of the access to the z-th metal can, It is expressed as the local update amount of the cth signal analysis window of the zth metal tank.
[0112] in, It is expressed as the adjustment coefficient of the cth signal analysis window of the zth metal can. When the dynamic interference is larger, it means that the dynamic noise in the signal is stronger. When the local update amount is larger, it means that there is more effective information. When the adjustment coefficient is greater than 1, it means that the noise intensity is high and the extreme points in the window need to be enhanced. When the adjustment coefficient is less than 1, it means that there is more effective information and the amplitude of the extreme points needs to be weakened.
[0113] Furthermore, an adjustment reference signal of the signal analysis window is obtained based on all adjustment extreme values in the signal analysis window, and a mean envelope is obtained based on the adjustment extreme values as the adjustment reference signal. It should be noted that the adjustment does not affect the maximum or minimum properties represented by the extreme value points themselves, and therefore the mean envelope can be directly obtained from the adjusted extreme values.
[0114] S404: performing signal decomposition and denoising on the adjusted reference signal of each metal can in all signal analysis windows to obtain a filtered identification signal of each metal can after it is returned.
[0115] When the adjustment extreme value is larger, more data information is separated, and the noise is more strongly weakened during decomposition. When the adjustment extreme value is smaller, more information components are retained, and valid information is protected. In an embodiment of the present invention, ITD signal decomposition is performed on each signal analysis window based on the adjustment reference signal of the signal analysis window to obtain the first component. It should be noted that ITD signal decomposition is a technical means well known to those skilled in the art. The first component is the first component decomposed in the decomposition, which will not be described in detail here. The first components in all signal analysis windows corresponding to the metal can are connected to obtain the first decomposition result of the metal can, and the first decomposition result is removed from the identification signal of the metal can after it is returned to obtain the filtered identification signal of the metal can.
[0116] The filtered identification signal of each metal can can be used as a pre-access identification signal when the metal can is taken and placed next time, and the filtered identification signal of each metal can is stored. In an embodiment of the present invention, the filtered identification signal of each metal can is input into a computer for decoding and reading, and the electronic tag of the metal can is accurately updated to obtain the real target signal, eliminating the dynamic interference caused by random artificial access, and retaining the updated information to the greatest extent. After the smart cabinet identifies the electronic tag information of the metal can entering the warehouse, it updates the status record data for each metal can, completes one entry and exit, and is recorded in the entry and exit ledger. Subsequently, the storage balance information of the metal can can be extracted and entered into the database for backup, and all metal cans are sorted from small to large according to the storage balance. Then, when the metal can is taken next time, the small LED light next to the hole where the metal can with the smallest storage balance is located will light up, prompting the metal can with less storage balance to be taken first. This recognition application can be adjusted according to the specific implementation scenario.
[0117] In summary, the present invention first analyzes the signal recognition quality of the hole positions where the metal cans are placed in the intelligent access control cabinet, and obtains the signal recognition quality through the initial recognition signal confusion analysis of the hole positions and local hole positions, characterizing the static interference that the metal cans will be subjected to at different hole positions, so as to facilitate the subsequent analysis of the hole position changes when they are taken. Taking into account the random situation of the metal cans, the hole position changes of the metal cans when they are taken are comprehensively considered, and the signal recognition quality differences of the hole positions before and after taking and returning are combined with the energy differences of the recognition signals of each metal can in the frequency domain before and after taking and returning to obtain the dynamic interference of taking, reflecting the degree of dynamic noise brought by the metal cans when they are used and returned, so as to adjust the filtering of the signal. Further, for the local changes of the recognition signals before and after taking and returning, the degree of dynamic update of the signal information is locally estimated, combined with the dynamic interference of taking, the dynamic interference after returning and the degree of information update are decomposed and denoised, so as to more accurately filter the dynamic situation of the random taking of the metal cans, and obtain a clearer and more accurate filtered recognition signal for storage. The present invention analyzes the static interference of the hole positions for placing metal cans, as well as the dynamic interference and dynamic information brought by the process of taking the metal cans, adaptively filters the identification information of each metal can, improves the signal quality and makes the identified label information clearer.
[0118] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0119] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
Claims
1. A method for identifying electronic tags of metal cans in complex environments. It is characterized in that The method comprises: Obtain the initial recognition signal of the electronic tag on each hole of the metal can in the intelligent access control cabinet, as well as the recognition signal of each metal can on the hole before taking and after returning; According to the degree of confusion deviation between each well position and the local well position in the initial recognition signal, the signal recognition quality of each well position is obtained; According to the difference in energy distribution of the identification signal of each metal can before and after it is returned in the frequency domain, and the difference in the signal recognition quality of the corresponding hole position before and after it is returned, the dynamic interference of each metal can is obtained; According to the change in the local distribution of the identification signal of each metal can before and after it is taken out and the dynamic interference of the metal can, the identification signal of each metal can after it is returned is decomposed and denoised to obtain the filtered identification signal of each metal can after it is returned; the filtered identification signal of each metal can is stored; The method for obtaining dynamic interference includes: According to the frequency distribution difference of the identification signal of each metal can before and after it is taken out and returned in the frequency domain space, and the energy density proportion of the frequency in the frequency domain, the energy change degree of each metal can is obtained; The difference between the signal recognition quality of the hole position of the metal can before taking and the signal recognition quality of the hole position after returning is used as the change degree of taking recognition of each metal can; The difference between the access energy variation degree and the access identification variation degree of each metal can is used as the access dynamic interference of each metal can; The method for obtaining the energy variation degree comprises: For any metal can, when the frequency in the frequency domain space of the identification signal after the return is not in the frequency domain space of the identification signal before the use, the corresponding frequency is used as the newly added frequency; In the frequency domain space of the identification signal corresponding to the returned metal can, the square modulus of the amplitude corresponding to each newly added frequency is taken as the energy density of each newly added frequency; The sum of the energy densities of all newly added frequencies is normalized to obtain the energy variation of the metal can.
2. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 1, It is characterized in that The method for acquiring the signal recognition quality includes: For any hole position, the neighbor interference degree of the hole position is obtained according to the distribution disorder of the phase difference between the hole position and the local hole position in the initial recognition signal; Obtaining the signal distortion degree of the well position according to the degree of deviation of the initial recognition signal of the well position among the initial recognition signals of all well positions; The signal recognition quality of the hole position is obtained according to the neighbor interference degree and signal distortion degree of the hole position; both the neighbor interference degree and the signal distortion degree are negatively correlated with the signal recognition quality.
3. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 2, It is characterized in that The method for obtaining the neighbor interference degree includes: Using other hole positions within a preset neighborhood of the hole position as local hole positions of the hole position; Obtaining the phase spectrum of the initial identification signal of the hole position and each local hole position; converting the mean of the phase difference between the phase spectrum of the hole position and each local hole position at the same frequency into a data form as the phase deviation between the hole position and each local hole position; The variance of the phase deviation between the hole position and all local hole positions is taken as the neighbor interference degree of the hole position.
4. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 2, It is characterized in that The method for obtaining the signal distortion comprises: The average of the initial recognition signals of all well positions is taken as the average recognition signal; The mean square error between the initial recognition signal of the hole position and the average recognition signal is taken as the signal distortion of the hole position.
5. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 1, It is characterized in that The method for obtaining the filtering identification signal includes: Divide the identification signals of each metal can before it is taken out and after it is returned in the same way to obtain more than two signal analysis windows; In each signal analysis window, according to the signal distribution difference and signal distribution correlation of the identification signal before taking and after returning, the local update amount of each signal analysis window is obtained; According to the dynamic interference of each metal can's use and the local update amount of each signal analysis window, the value of the identification signal of the metal can after being returned in each signal analysis window is corrected to obtain the adjusted reference signal of each metal can in each signal analysis window; The adjusted reference signal of each metal can in all signal analysis windows is subjected to signal decomposition and denoising to obtain a filtered identification signal of each metal can after it is returned.
6. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 5, It is characterized in that The method for obtaining the local update amount includes: For any signal analysis window of any metal can, the ratio of the variance of the identification signal before the metal can is taken out to the variance of the identification signal after the metal can is returned in the signal analysis window is taken as the signal-to-noise ratio of the signal analysis window; In the signal analysis window, the amplitude distribution probability of the metal can on the identification signal before taking and after returning is obtained; the conditional entropy of the amplitude distribution probability of the metal can after returning under the amplitude distribution probability before taking is taken as the information consumption of the signal analysis window; According to the signal-to-noise ratio and information consumption of the signal analysis window, the local update amount of the signal analysis window is obtained, and both the signal-to-noise ratio and the information consumption are positively correlated with the local update amount.
7. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 5, It is characterized in that The method for obtaining the adjustment reference signal includes: For any signal analysis window of any metal can, obtain the extreme value of the identification signal after the metal can is returned in the signal analysis window; The ratio of the dynamic interference of the metal can to the local update amount of the signal analysis window is used as the adjustment coefficient; The product of each extreme value of the identification signal after the metal can is returned in the signal analysis window and the adjustment coefficient is used as the adjustment extreme value of the signal analysis window; An adjustment reference signal of the signal analysis window is obtained according to all adjustment extreme values in the signal analysis window.
8. According to the method for identifying electronic tags of metal cans in complex environments as described in claim 5, It is characterized in that The step of performing signal decomposition and denoising on the adjusted reference signal of each metal can in all signal analysis windows to obtain a filtered identification signal of each metal can after it is returned includes: Performing ITD signal decomposition on each signal analysis window based on the adjusted reference signal of the signal analysis window to obtain a first component; The first components in all signal analysis windows corresponding to the metal can are connected to obtain a first decomposition result of the metal can; the first decomposition result is removed from the identification signal of the metal can after it is returned to obtain a filtered identification signal of the metal can.
