Full-chain traceability method and system for agricultural products based on traceability labels

Abstract

The invention relates to a full-chain tracing method and system for agricultural products based on a tracing label. When a traceability label reader is used to read a traceability label and a collision occurs, a signal received by each antenna of the reader is obtained to obtain the noise of the antenna received signal and the result after denoising; the average skewness of the noise of all antennas and the average skewness of the result after denoising are calculated, and the number of labels is estimated according to the average skewness; if the estimated number of labels is less than or equal to the number of antennas, a product code of the label is obtained according to the signal received by each antenna by using blind source separation; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, if the number of forks in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined according to the number of identified labels and the estimated number of labels, and when the timing is reached, the parameters of the request command sent by the reader to the label are adjusted; the product code of the label is obtained, and the tracing information corresponding to the product code is added to the full-chain tracing system.

Description

Technical Field

[0001] The present invention relates to the field of agricultural product traceability, and in particular to a full-chain traceability method and system for agricultural products based on traceability labels. Background Art

[0002] Among the three technologies of RFID, QR code and barcode, RFID (radio frequency identification) technology is increasingly widely used in agricultural product traceability systems due to its advantages of non-contact, long-distance and high efficiency. RFID technology can give each batch of agricultural products a unique identity, thereby realizing real-time collection and automatic identification of information, greatly improving the efficiency and accuracy of traceability. Information in various links such as production, processing, transportation, storage and sales of agricultural products can be recorded through RFID tags, and data can be read and transmitted through RFID readers, thereby realizing full monitoring and tracking of agricultural products. Through RFID technology, consumers can easily obtain data such as the source information, quality certification, production batches, etc. of agricultural products by scanning labels, thereby ensuring food safety and enhancing consumers' trust in product quality.

[0003] In the agricultural product industry operation brain platform, the whole chain information covers the processes of planting, crop management, picking, processing, warehousing, transportation, and sales. Among them, the traceability information of planting, crop management, picking, etc. is generally obtained through manual input or through the Internet of Things. During the transportation and warehousing process, there will be a lot of agricultural products, which requires batch identification of traceability labels and information entry. In this high-density tag environment, multiple tags may send signals to the reader at the same time, causing data collisions, resulting in information loss or reading errors, affecting the efficiency and accuracy of the system. Although traditional anti-collision algorithms such as ALOHA and tree structure algorithms can alleviate this problem to a certain extent, in highly concentrated batch tag recognition scenarios, there are still problems of high collision rate and low recognition efficiency, which limits the application effect of traceability labels in full-chain traceability. Summary of the invention

[0004] In the whole chain traceability of agricultural products during storage and transportation, due to the large number of agricultural products, it is easy to cause the collision of traceability labels, resulting in slow recognition speed and missed recognition. In a first aspect of the present invention, a whole chain traceability method of agricultural products based on traceability labels is provided, and the method comprises:

[0005] When a traceability tag reader is used to read a traceability tag and a collision occurs, the signal received by each antenna of the reader is obtained, and the signal received by each antenna is Gaussian filtered to obtain the noise of the antenna received signal and the result after denoising; the average skewness of the noise of all antennas and the average skewness of the denoising result are calculated, and the number of tags is estimated according to the average skewness;

[0006] If the estimated number of tags is less than or equal to the number of antennas, the product code of the tag is obtained by blind source separation based on the signals received by each antenna; otherwise, the search tree structure is determined based on the number of collision bits and the number of antennas. If the number of forks in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags. When the timing is reached, the parameters of the request command sent by the reader to the tag are adjusted;

[0007] Get the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system.

[0008] In an optional embodiment, estimating the number of labels according to the average skewness is specifically:

[0009] Correcting the average skewness of the denoised result using the average skewness of the noise;

[0010] Substitute the corrected average skewness into the pre-established functional relationship between the estimated number of labels and skewness to obtain the estimated number of labels.

[0011] In an optional embodiment, determining the search tree structure based on the number of collision bits and the number of antennas is specifically as follows:

[0012] The ratio of the maximum number of tags to the number of antennas under the collision bit number is calculated. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio.

[0013] In an optional embodiment, determining the timing of adjusting the execution of the search tree according to the number of identified tags and the estimated number of tags is specifically as follows:

[0014] If after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; or,

[0015] After searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached.

[0016] In an optional embodiment, the adjustment of the parameters of the request command sent by the reader to the tag is specifically:

[0017] Set all collision bits in the request parameter sent by the reader to the tag to 1; or,

[0018] The branches in the second half of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging results.

[0019] In a second aspect of the present invention, a full-chain traceability system for agricultural products based on traceability labels is provided, the system comprising:

[0020] The tag estimation module is used to obtain the signal received by each antenna of the reader when a collision occurs when the traceability tag is read by the traceability tag reader, perform Gaussian filtering on the signal received by each antenna, obtain the noise of the antenna received signal and the result after denoising; calculate the average skewness of the noise of all antennas and the average skewness of the denoising result, and estimate the number of tags according to the average skewness;

[0021] A tag identification module is used to obtain the product code of the tag by blind source separation based on the signals received by each antenna if the estimated number of tags is less than or equal to the number of antennas; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, a timing for adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, a parameter of the request command sent by the reader to the tag is adjusted;

[0022] The traceability content adding module is used to obtain the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system.

[0023] In an optional embodiment, estimating the number of labels according to the average skewness is specifically:

[0024] Correcting the average skewness of the denoised result using the average skewness of the noise;

[0025] Substitute the corrected average skewness into the pre-established functional relationship between the estimated number of labels and skewness to obtain the estimated number of labels.

[0026] In an optional embodiment, determining the search tree structure based on the number of collision bits and the number of antennas is specifically as follows:

[0027] The ratio of the maximum number of tags to the number of antennas under the collision bit number is calculated. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio.

[0028] In an optional embodiment, determining the timing of adjusting the execution of the search tree according to the number of identified tags and the estimated number of tags is specifically as follows:

[0029] If after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; or,

[0030] After searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached.

[0031] In an optional embodiment, the adjustment of the parameters of the request command sent by the reader to the tag is specifically:

[0032] Set all collision bits in the request parameter sent by the reader to the tag to 1; or,

[0033] The branches in the second half of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging results.

[0034] In a third aspect of the present invention, a computer program is provided, the computer program comprising the following modules:

[0035] The tag estimation module is used to obtain the signal received by each antenna of the reader when a collision occurs when the traceability tag is read by the traceability tag reader, perform Gaussian filtering on the signal received by each antenna, obtain the noise of the antenna received signal and the result after denoising; calculate the average skewness of the noise of all antennas and the average skewness of the denoising result, and estimate the number of tags according to the average skewness;

[0036] A tag identification module is used to obtain the product code of the tag by blind source separation based on the signals received by each antenna if the estimated number of tags is less than or equal to the number of antennas; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, a timing for adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, a parameter of the request command sent by the reader to the tag is adjusted;

[0037] The traceability content adding module is used to obtain the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system.

[0038] In yet another aspect, the present invention further provides a computer program storage medium, on which the computer program is stored.

[0039] In order to solve the problem that there are many agricultural products in the warehousing and transportation links and collisions are prone to occur during the traceability of the entire chain of agricultural products, resulting in slow or missed chain information entry, the present invention uses the noise of the antenna and the skewness of the denoised signal according to the central limit theorem to estimate the number of tags to be identified, and then dynamically adjusts the number of forks of the search tree according to the number of tags to be identified and the number of antennas of the reader, so as to find a balance between search efficiency and computational overhead, thereby optimizing the reading performance of RFID tags. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of Embodiment 1;

[0041] Figure 2 Search trees constructed for different cases of estimated number of tags;

[0042] Figure 3 A schematic diagram of the dynamic adjustment of the search tree;

[0043] Figure 4 This is a schematic diagram of the traceability functional unit in the industrial operation brain;

[0044] Figure 5 This is a structural diagram of the second embodiment. DETAILED DESCRIPTION

[0045] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Figure 1 The first embodiment of the present invention is shown. The first embodiment provides a full-chain traceability method for agricultural products based on a traceability label, including:

[0048] S100, when a traceability tag reader is used to read a traceability tag and a collision occurs, a signal received by each antenna of the reader is obtained, and a Gaussian filter is performed on the signal received by each antenna to obtain the noise of the antenna received signal and the result after denoising; an average skewness of the noise of all antennas and an average skewness of the result after denoising are calculated, and the number of tags is estimated according to the average skewness;

[0049] Specifically, during the storage or transportation process, the storage and transportation information needs to be added to the entire chain of agricultural products, such as storage location, temperature, time, transportation time, temperature, starting point, destination, etc. In the storage and transportation, agricultural products need to be identified in batches. When a large number of traceability tags are read by a traceability tag reader, collisions are likely to occur. The so-called collision means that when multiple traceability tags send information such as electronic product codes to the reader at the same time, these signals will interfere with each other, resulting in the reader being unable to correctly decode the information of any tag. Among them, the traceability tag is preferably an RFID tag or an NFC tag.

[0050] The reader is usually equipped with multiple antennas, each of which receives a superimposed signal mixed with multiple tag signals, and records the original mixed signal received by each antenna. Gaussian filtering is performed on the original mixed signal received by each antenna. After Gaussian filtering, the original signal is decomposed into two parts: one is the estimated noise signal, and the other is the signal after noise removal, that is, a relatively clean tag signal. The skewness of the noise signal received by each antenna and the skewness of the denoised signal are calculated separately, and then the noise skewness of all antennas is averaged to obtain the average noise skewness; similarly, the average of the skewness of the denoised signal of all antennas is calculated to obtain the average skewness after denoising. According to the central limit theorem, the distribution of the sum or average of a large number of independent and identically distributed random variables approaches the normal distribution, even if the original variable itself is not normally distributed.

[0051] When multiple tags send signals at the same time, the signal received by the reader is the superposition of multiple tag signals. Each tag signal will be affected by noise. If these noise sources are regarded as multiple independent and identically distributed random variables, then according to the central limit theorem, the distribution of the sum or average of these noises can be approximated as a normal distribution. Similarly, although the distribution of the tag signal itself may not be a normal distribution, due to the superposition of a large number of tag signals, some statistical characteristics of the signal finally received may be close to the characteristics of a normal distribution. And the more tags there are, the closer it is to a normal distribution. The skewness is used to determine the degree to which the signal conforms to the normal distribution. The closer the skewness is to 0, the more it conforms to the normal distribution. Since the noise is completely random, it conforms to the normal distribution more than the denoised signal. If it is affected by a larger external influence, the noise will also be greatly affected and deviate from the normal distribution. In order to eliminate or reduce the external influence, in one embodiment, the number of tags estimated according to the average skewness is specifically as follows:

[0052] The average skewness of the noise is used to correct the average skewness of the denoised result; in an optional embodiment, the average skewness of the denoised result minus the average skewness of the noise is used for correction. For example, the average skewness of the noise is calculated to be 0.05, and the average skewness of the denoised result is 1.5. The subtraction method is used for correction, and the corrected skewness is 1.45. This correction value is very close to the original skewness value after denoising because the noise skewness is very small.

[0053] Substitute the corrected average skewness into the pre-established function relationship between the estimated number of labels and skewness to obtain the estimated number of labels. Based on historical information, establish a function relationship between the number of labels and skewness, and substitute the corrected average skewness into the function relationship to obtain the estimated number of labels.

[0054] In another embodiment, the average skewness of the noise and the average skewness of the denoised signal are used as function independent variables, and the estimated number of tags is used as the function dependent variable. The functional relationship is established through a large amount of historical information.

[0055] S200, if the estimated number of tags is less than or equal to the number of antennas, the product code of the tag is obtained by blind source separation based on the signals received by each antenna; otherwise, the search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, the parameters of the request command sent by the reader to the tag are adjusted;

[0056] If the estimated number of tags is less than or equal to the number of antennas, the signal mixing degree is relatively low, and blind source separation technology is used. Blind Source Separation (BSS) can separate the original independent source signal from the mixed signal. Blind source separation is divided into underdetermined blind source separation and non-underdetermined blind source separation. The underdetermined blind source separation is when the number of tags is greater than the number of antennas; the non-underdetermined blind source separation is when the number of tags is not greater than the number of antennas. The independent component analysis (ICA) method can be used to obtain the signal sent by each tag. The number of antennas is preferably 2, 4 or 8.

[0057] When the estimated number of tags is greater than the number of antennas, the signal mixing degree is high, and the blind source separation technology may be difficult to effectively separate the signals of each tag. In this case, the search tree structure is determined according to the number of collision bits and the number of antennas, where the collision bit is the number of bits that collide during the collision detection process. For example, if the electronic product code (EPC code) of one tag is 1011 and the EPC of another tag is 1101, then the collision bit is 1xx1, where x is the collision bit. In an optional embodiment, the search tree structure is determined based on the number of collision bits and the number of antennas, specifically:

[0058] The ratio of the maximum number of tags to the number of antennas under the collision bit number is calculated. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio.

[0059] The collision bit number is the number of bits that collide. Taking the above data as an example, the collision bit number is 2. The maximum number of tags under the collision bit number is the maximum number of tags that may exist theoretically under the given collision bit number. For example, in the above 1xx1, since the first and last bits are both 1, the maximum number of tags is 1001, 1011, 1101, 1111, that is, the maximum number of tags is 4. A simpler calculation method is 2 to the power of the collision bit number. Assuming the collision bit number is 5, the maximum number of tags is 32. If the number of antennas is 4, the ratio is 8.

[0060] If the estimated number of tags is not less than or greater than half of the maximum number of tags, it means that the actual number of tags is close to the theoretical maximum value and the collision is serious. In order to narrow the search range more quickly, the number of forks in the search tree is set to the ratio calculated above. If the estimated number of tags is 10, which is greater than or equal to half of the maximum number of tags, that is, 8, the number of forks in the search tree is set to 8. Figure 2 shown.

[0061] If the estimated number of tags is less than or not greater than half of the maximum number of tags, it means that the actual number of tags is relatively small and the collision is relatively mild. To avoid the search tree being too complex, the number of forks in the search tree is set to half of the ratio. A smaller number of forks can reduce the complexity of the search and reduce unnecessary computational overhead. If the estimated number of tags is 5, which is less than half of the maximum number of tags, i.e. 8, the number of forks in the search tree is set to 4. Figure 2 shown.

[0062] The number of forks of the search tree mentioned above is the number of forks of the first fork of the search tree. In subsequent forks, each fork can be used as a new condition according to the above process. If the estimated number of tags is less than or equal to the number of antennas, the product code of the tag is obtained by blind source separation based on the signal received by each antenna; otherwise, the search tree structure is determined based on the number of collision bits and the number of antennas, and repeated continuously to obtain a complete search tree.

[0063] In one embodiment, if the fork in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined according to the number of identified tags and the estimated number of tags. When the timing is reached, the parameters of the request command sent by the reader to the tag are adjusted, specifically: if after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, then the timing is reached, and the collision bits in the request parameters sent by the reader to the tag are all set to 1. For example, there are 4 branches, if the estimated number of tags is 5, and 3 tags are identified after searching the first branch, the estimated number of tags minus the number of identified tags is 2, which is less than the number of antennas 4, then all the collision bits are set to 1. After identifying a tag, a deselect command is issued, and the tag will enter a dormant state and will no longer respond to any subsequent commands from the reader. After the tag leaves the scope of action of the reader, it will be reset when it enters the scope of action again, and all the collision bits will be set to 1. The tags that have been read will no longer respond, and no collision will occur. Since it is less than the number of antennas, the signal of the remaining unidentified tags, namely EPC, can be obtained through blind source separation.

[0064] Since the estimated tags have certain errors, in order to reduce the impact of the errors. In another optional embodiment, if the number of forks in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags. When the timing is reached, the parameters of the request command sent by the reader to the tag are adjusted. Specifically: after searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached, the second half of the branches of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging result. For example, there are 4 branches. If the estimated number of tags is 5, after searching the first half of the branches of the search tree, that is, after searching the second branch, a total of 3 tags are identified. The estimated number of tags minus the number of identified tags is 2, then the second half of the branches are merged. Assuming that the 4 branches are 00, 01, 10, and 11, respectively, the result of merging the second half of the branches is 1. Figure 3 As shown. Compared with the previous embodiment, although the number of searches increases, the impact of the error in the estimated number of labels is reduced. Preferably, in constructing the search tree, the branches are arranged in order of increasing bits, for example, the first branch is 00, the second branch is 01, the third branch is 10, and the fourth branch is 11. In one embodiment, the merging of the second half of the branches of the search tree specifically merges the second half of the branches into a branch whose first conflict or collision bit is 1. In the above example, the third branch and the fourth branch are 1 after merging.

[0065] If the number of forks in the search tree is equal to 2, then the two forks are searched separately. After the search tree is constructed, the process of searching the search tree for identification tags is to obtain a branch bit code in the search tree, such as Figure 2 The bit code of the first branch in is 00, and the aforementioned bit code bit of the collision bit is set as the bit code of the branch. Assuming that the EPC length is 8 and the collision bit is 1x01xx0x, then in the first branch, the first and second x are replaced with 00 to obtain 10010x0x; in the second branch, the first and second x are replaced with 0 and 1 respectively to obtain 10011x0x; in the third branch, the first and second x are replaced with 0 and 1 respectively to obtain 10011x0x; in the fourth branch, the first and second x are replaced with 1 to obtain 11011x0x.

[0066] When the parameter in the reader's Request is 10010, all tags starting with 10010 will respond. If multiple tags respond, a collision will occur. The method of S100 is used to estimate the number of colliding tags. If the estimated number of tags is not less than the number of antennas, which is 4, a search tree is constructed according to the process of S200. Of course, this search tree is a subtree of the previous search tree. If the estimated number of tags is less than the number of antennas, which is 4, blind source separation is used to obtain the product code of the tag based on the signals received by each antenna. It should be understood that the above subtree construction is a continuous iterative process. In order to avoid the tree being too deep, the present invention utilizes the estimated number of tags and the number of antennas when constructing the search tree. If the estimated number of tags is large, there will be more forks to avoid the search tree being too deep.

[0067] S300, obtaining the product code of the label, and adding the traceability information corresponding to the product code in the full-chain traceability system.

[0068] When the product code EPC is identified during storage or transportation, the storage or transportation related information is entered into the project corresponding to the ECP. When the full chain is traced later, the full chain information of the agricultural product can be obtained through the EPC of the label. Figure 4 It shows the entire chain information of agricultural products recorded in the agricultural products industry brain platform.

[0069] Figure 5 The second embodiment of the present invention is shown. The second embodiment provides a full-chain traceability system for agricultural products based on traceability labels, including:

[0070] The tag estimation module is used to obtain the signal received by each antenna of the reader when a collision occurs when the traceability tag is read by the traceability tag reader, perform Gaussian filtering on the signal received by each antenna, obtain the noise of the antenna received signal and the result after denoising; calculate the average skewness of the noise of all antennas and the average skewness of the denoising result, and estimate the number of tags according to the average skewness;

[0071] A tag identification module is used to obtain the product code of the tag by blind source separation based on the signals received by each antenna if the estimated number of tags is less than or equal to the number of antennas; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, a timing for adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, a parameter of the request command sent by the reader to the tag is adjusted;

[0072] The traceability content adding module is used to obtain the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system.

[0073] In an optional embodiment, estimating the number of labels according to the average skewness is specifically:

[0074] Correcting the average skewness of the denoised result using the average skewness of the noise;

[0075] Substitute the corrected average skewness into the pre-established functional relationship between the estimated number of labels and skewness to obtain the estimated number of labels.

[0076] In an optional embodiment, determining the search tree structure based on the number of collision bits and the number of antennas is specifically as follows:

[0077] The ratio of the maximum number of tags to the number of antennas under the collision bit number is calculated. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio.

[0078] In an optional embodiment, determining the timing of adjusting the execution of the search tree according to the number of identified tags and the estimated number of tags is specifically as follows:

[0079] If after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; or,

[0080] After searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached.

[0081] In an optional embodiment, the adjustment of the parameters of the request command sent by the reader to the tag is specifically:

[0082] Set all collision bits in the request parameter sent by the reader to the tag to 1; or,

[0083] The branches in the second half of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging results.

[0084] In a third aspect of the present invention, a computer program is provided, the computer program comprising the following modules:

[0085] The tag estimation module is used to obtain the signal received by each antenna of the reader when a collision occurs when the traceability tag is read by the traceability tag reader, perform Gaussian filtering on the signal received by each antenna, obtain the noise of the antenna received signal and the result after denoising; calculate the average skewness of the noise of all antennas and the average skewness of the denoising result, and estimate the number of tags according to the average skewness;

[0086] A tag identification module is used to obtain the product code of the tag by blind source separation based on the signals received by each antenna if the estimated number of tags is less than or equal to the number of antennas; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, a timing for adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, a parameter of the request command sent by the reader to the tag is adjusted;

[0087] The traceability content adding module is used to obtain the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system.

[0088] Through the description of the above implementation methods, technicians in this field can clearly understand that each implementation method can be implemented by adding a necessary general hardware platform, and of course can also be implemented by combining hardware and software. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a computer product, and the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it, and other embodiments may also be used. Although the present invention has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. The whole chain traceability method of agricultural products based on traceability labels is characterized by: The method comprises: When a traceability tag reader is used to read a traceability tag and a collision occurs, the signal received by each antenna of the reader is obtained, and the signal received by each antenna is Gaussian filtered to obtain the noise of the antenna received signal and the result after denoising; the average skewness of the noise of all antennas and the average skewness of the denoising result are calculated, and the number of tags is estimated according to the average skewness; If the estimated number of tags is less than or equal to the number of antennas, the product code of the tag is obtained by blind source separation based on the signals received by each antenna; otherwise, the search tree structure is determined based on the number of collision bits and the number of antennas. If the number of forks in the search tree is greater than 2, the timing of adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags. When the timing is reached, the parameters of the request command sent by the reader to the tag are adjusted; Get the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system; The determining of the search tree structure based on the number of collision bits and the number of antennas is specifically as follows: Calculate the ratio of the maximum number of tags to the number of antennas under the collision bit number. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio. The timing of adjusting the execution search tree according to the number of identified tags and the estimated number of tags is determined as follows: If after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; or, After searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; The adjustment of the parameters of the request command sent by the reader to the tag is specifically: Set all collision bits in the request parameter sent by the reader to the tag to 1; or, The branches in the second half of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging results.

2. The method according to claim 1, characterized in that The estimated number of labels according to the average skewness is specifically: Correcting the average skewness of the denoised result using the average skewness of the noise; Substitute the corrected average skewness into the pre-established functional relationship between the estimated number of labels and skewness to obtain the estimated number of labels.

3. The full-chain traceability system of agricultural products based on traceability labels is characterized by: The system comprises: The tag estimation module is used to obtain the signal received by each antenna of the reader when a collision occurs when the traceability tag is read by the traceability tag reader, perform Gaussian filtering on the signal received by each antenna, obtain the noise of the antenna received signal and the result after denoising; calculate the average skewness of the noise of all antennas and the average skewness of the denoising result, and estimate the number of tags according to the average skewness; A tag identification module is used to obtain the product code of the tag by blind source separation based on the signals received by each antenna if the estimated number of tags is less than or equal to the number of antennas; otherwise, a search tree structure is determined based on the number of collision bits and the number of antennas, and if the number of forks in the search tree is greater than 2, a timing for adjusting the execution of the search tree is determined based on the number of identified tags and the estimated number of tags, and when the timing is reached, a parameter of the request command sent by the reader to the tag is adjusted; The traceability content adding module is used to obtain the product code of the label and add the traceability information corresponding to the product code in the full-chain traceability system; The determining of the search tree structure based on the number of collision bits and the number of antennas is specifically as follows: Calculate the ratio of the maximum number of tags to the number of antennas under the collision bit number. If the estimated number of tags is greater than or not less than half of the maximum number of tags, the number of forks in the search tree is the ratio; otherwise, the number of forks in the search tree is half of the ratio. The timing of adjusting the execution search tree according to the number of identified tags and the estimated number of tags is determined as follows: If after searching a branch of the search tree, the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; or, After searching the first half of the branches of the search tree, if the estimated number of tags minus the number of identified tags is less than the number of antennas, the timing is reached; The adjustment of the parameters of the request command sent by the reader to the tag is specifically: Set all collision bits in the request parameter sent by the reader to the tag to 1; or, The branches in the second half of the search tree are merged, and the parameters of the request command sent by the reader to the tag are adjusted according to the search tree merging results.

4. The system according to claim 3, characterized in that The estimated number of labels according to the average skewness is specifically: Correcting the average skewness of the denoised result using the average skewness of the noise; Substitute the corrected average skewness into the pre-established functional relationship between the estimated number of labels and skewness to obtain the estimated number of labels.

Images