A cold chain logistics tracking method and system

Abstract

The application relates to a cold-chain logistics tracking method and system, which comprises the following steps: acquiring the unique identification code, space-time information and multi-dimensional preservation information of a plurality of commodities, wherein the multi-dimensional preservation information comprises temperature, humidity, pH value and marker gas content; based on maximum information entropy and a multivariate autoregressive model, the time domain characteristics and frequency domain characteristics of the multi-dimensional preservation and corresponding space-time information of each commodity are extracted; pseudo-noise sequences are used to spread spectrum modulate the time domain characteristics, frequency domain characteristics and unique identification code to generate a plurality of traceability codes; based on the traceability codes and preservation information, a self-encoder based on a convolutional neural network is trained; real-time traceability information of a commodity to be traced is input into the trained self-encoder to obtain one or more traceability codes. The unique identification code, preservation information and space-time information are spread spectrum modulated with a watermark, and the self-encoder is used for identification, so that the timeliness of cold-chain logistics tracking and the non-tamperability of traceability data are improved.

Description

Technical Field

[0001] This invention belongs to the field of logistics information processing technology, specifically relating to a cold chain logistics tracking method and system. Background Technology

[0002] Cold chain refers to a special supply chain system where goods are kept at the necessary low temperatures throughout their entire lifecycle, from acquisition at the place of origin or harvesting to processing, storage, transportation, distribution, and retail, until they reach the consumer. This ensures food quality and safety, reduces spoilage, and prevents contamination. The core purpose of logistics traceability and tracking is to define responsibility, and the core technological means is to establish and maintain the relationships between various traceability objects. Traceability objects mainly include responsible persons, main items, nodes (environment), and behaviors. By associating and recording the unique ID numbers of each traceability object during the production and circulation process, a full lifecycle traceability of the product is formed. Specifically, in cold chain food traceability, the key points include information collection of traceability objects, the supporting hardware and software management systems, the information carriers for labeling traceability objects, and an evaluation system for assessing the effectiveness of the traceability system. Traceability objects include the cold chain food itself, circulation nodes (warehouses, transport vehicles, etc.), behaviors (loading and unloading, distribution processing, packaging, transactions, etc.), and related personnel (handlers, managers, users, and related contacts).

[0003] The information carrier used to label traceability objects refers to the unique identification basis for these objects. Common forms include two-dimensional barcodes, one-dimensional barcodes, RFID, and optical or electronic tags such as strings. The evaluation system is a testing and evaluation method used to guide and assess the compliance of cold chain food traceability systems. Current technologies for cold chain traceability suffer from incompleteness, susceptibility to tampering, and delays, posing safety risks to cold chain products. Summary of the Invention

[0004] To improve the timeliness of cold chain logistics tracking and the immutability of traceability data, a cold chain logistics tracking method is provided in a first aspect of the present invention, comprising: acquiring unique identification codes, spatiotemporal information, and multidimensional preservation information of multiple commodities, wherein the multidimensional preservation information includes temperature, humidity, pH value, and content of labeled gases; determining the minimum dimension and maximum sampling time of the multidimensional preservation information of different commodities based on maximum information entropy and a multivariate autoregressive model, and extracting the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information of each commodity; spreading the temporal features, frequency domain features, and unique identification codes using a pseudo-noise sequence to generate multiple traceability codes; training an autoencoder based on a convolutional neural network by using one or more traceability codes of each commodity as tags and one or more preservation information as samples; and inputting the real-time traceability information of the commodity to be traced into the trained autoencoder to obtain one or more traceability codes.

[0005] In some embodiments of the present invention, determining the minimum dimension and maximum sampling time of multidimensional preservation information for different commodities based on maximum information entropy and a multivariate autoregressive model includes: determining the minimum dimension of multidimensional preservation information for each commodity based on maximum information entropy and a multivariate autoregressive model; and determining the maximum sampling time of multidimensional preservation information for each commodity based on maximum information entropy and a higher-order autoregressive model.

[0006] Furthermore, the determination of the minimum dimension of the multidimensional preservation information for each commodity based on the maximum information entropy and the multivariate autoregressive model includes:

[0007] The multivariate autoregressive model is expressed as follows:

[0008] y = β0 + β1X1 + β2X2 + ... + β n X n +e, where β0 is the initial freshness index, y represents the freshness index of the product, x1, x2, ..., x n Let represent the data for each dimension of the multidimensional archival information, and e represent the error vector of the multidimensional archival information.

[0009] In some embodiments of the present invention, the step of using pseudo-noise sequences to spread spectrum modulate time-domain features, frequency-domain features, and unique identification codes to generate multiple traceability codes includes: spreading the time-domain features and frequency-domain features using a direct-sequence spread spectrum method to obtain spread spectrum coding; using the maximum sampling time as the minimum period for watermark generation, with the position change rate within each minimum period as the watermark carrier, modulating the pseudo-noise sequence onto the spread spectrum coding to generate a watermark; and embedding the watermarks within multiple minimum periods into the unique identification codes to generate multiple traceability codes.

[0010] Furthermore, one or more traceability codes are arranged in chronological order and stored in a Merkle tree.

[0011] In the above embodiments, the convolutional neural network is a spatiotemporal graph convolutional neural network.

[0012] A second aspect of the present invention provides a cold chain logistics tracking system, comprising: an acquisition module for acquiring unique identification codes, spatiotemporal information, and multidimensional preservation information of multiple commodities, wherein the multidimensional preservation information includes temperature, humidity, pH value, and marker gas content; a generation module for determining the minimum dimension and maximum sampling time of the multidimensional preservation information of different commodities based on maximum information entropy and a multivariate autoregressive model, and extracting the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information of each commodity; spreading the temporal features, frequency domain features, and unique identification codes using a pseudo-noise sequence to generate multiple traceability codes; a training module for training an autoencoder based on a convolutional neural network, using one or more traceability codes of each commodity as labels and one or more corresponding preservation information as samples; and an input module for inputting the real-time traceability information of the commodities to be traced into the trained autoencoder to obtain one or more traceability codes.

[0013] A third aspect of the present invention provides an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the cold chain logistics tracking method provided in the first aspect of the present invention.

[0014] In a fourth aspect, the present invention provides a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the cold chain logistics tracking method provided in the first aspect of the present invention.

[0015] The beneficial effects of this invention are:

[0016] This disclosure provides a cold chain logistics tracking method, comprising: acquiring unique identification codes, spatiotemporal information, and multidimensional preservation information for multiple commodities, wherein the multidimensional preservation information includes temperature, humidity, pH value, and marker gas content; extracting the temporal and frequency domain features of the multidimensional preservation and its corresponding spatiotemporal information for each commodity based on a maximum information entropy and a multivariate autoregressive model; performing spread spectrum modulation on the temporal features, frequency domain features, and unique identification codes using a pseudo-noise sequence to generate multiple traceability codes; training an autoencoder based on a convolutional neural network based on the traceability codes and preservation information; and inputting the real-time traceability information of the commodity to be traced into the trained autoencoder to obtain one or more traceability codes. As can be seen, this invention reduces the dimensionality of multidimensional information by combining it with a multivariate autoregressive model, and establishes a correlation between its time-domain and frequency-domain features and spatiotemporal information. Then, a watermark is generated using a spread spectrum modulation method. Since the watermark is embedded in the location information and generates a traceability code, the real-time performance and tamper-proof nature of cold chain logistics tracking are improved. At the same time, the traceability code can be decoded into multidimensional information through a convolutional neural network, thereby improving the comprehensiveness of the traceability information. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the basic process of the cold chain logistics tracking method in some embodiments of the present invention;

[0018] Figure 2 This is a schematic diagram of the spread spectrum modulation process in some embodiments of the present invention;

[0019] Figure 3 This is a schematic diagram of the structure of a cold chain logistics tracking system in some embodiments of the present invention;

[0020] Figure 4 This is a schematic diagram of the structure of an electronic device in some embodiments of the present invention. Detailed Implementation

[0021] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0022] refer to Figure 1 In a first aspect, a cold chain logistics tracking method is provided, comprising: S100. acquiring unique identification codes, spatiotemporal information, and multidimensional preservation information of multiple commodities, wherein the multidimensional preservation information includes temperature, humidity, pH value, and labeled gas content; S200. determining the minimum dimension and maximum sampling time of the multidimensional preservation information of different commodities based on maximum information entropy and a multivariate autoregressive model, and extracting the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information of each commodity; using a pseudo-noise sequence to spread-spectrum modulate the temporal features, frequency domain features, and unique identification codes to generate multiple traceability codes; S300. using one or more traceability codes of each commodity as tags and one or more corresponding preservation information as samples to train an autoencoder based on a convolutional neural network; S400. inputting the real-time traceability information of the commodity to be traced into the trained autoencoder to obtain one or more traceability codes.

[0023] It should be noted that cold chain goods typically include food raw materials, processed food or semi-finished products, special biological products, and pharmaceuticals. A unique identifier (a string of numbers, text, or symbols) is generated by encoding one or more items from the product information and / or company information using a specific coding method, distinguishing each product from others. Product information includes the producer's country, name, company code, product name, product code, production date, shelf life, product category, and weight. Company information includes the company's location, company code, company type, company address, business license, permits, and contact information. Spatiotemporal information includes the station's location (latitude, longitude, and elevation), name, code, and address. Shipping information includes the tracking number, shipping company name, shipping company code, shipping time, and shipping date. The marker gases in the multidimensional preservation information include CO2, NH3, TVB-N, H2S, SO2, or other characteristic gases that can indicate the freshness or state of cold chain goods. The content of marker gases can also be determined by the nitrogen, oxygen, and sulfur content of the aforementioned mixed gases. For example, for meat products, freshness can be characterized using only labeled gases; while for eggs, freshness can be characterized using images and CO2 content. Optionally, an exponential decay model can be used to fit the freshness.

[0024] In S200 of some embodiments of the present invention, determining the minimum dimension and maximum sampling time of multidimensional preservation information of different commodities based on maximum information entropy and a multivariate autoregressive model includes: determining the minimum dimension of multidimensional preservation information of each commodity based on maximum information entropy and a multivariate autoregressive model; and determining the maximum sampling time of multidimensional preservation information of each commodity based on maximum information entropy and a higher-order autoregressive model.

[0025] Furthermore, the determination of the minimum dimension of the multidimensional preservation information for each commodity based on the maximum information entropy and the multivariate autoregressive model includes:

[0026] The multivariate autoregressive model is expressed as follows:

[0027] y = β0 + β1X1 + β2X2 + ... + β n X n +e, where β0 is the initial freshness index, y represents the freshness index of the product, x1, x2, ..., x n These represent the data for each dimension of the multidimensional preservation information, and e represents the error vector of the multidimensional preservation information. Dimensions include the product's temperature, humidity, pH value, and marker gas content, etc.

[0028] Furthermore, based on maximum information entropy and higher-order autoregressive models, the maximum sampling time for multidimensional preservation information of each commodity is determined, including: the final prediction error (FPE) criterion based on minimizing the mean squared error is a relatively effective criterion for determining the order of the AR model. Its definition is: given an observation length N, if prediction coefficients are estimated from one observation of a process, and then a system based on these prediction coefficients is used to process another observation, then the mean squared error of the prediction is obtained. Therefore, to ensure the maximum information entropy of the higher-order autoregressive model, this error must be minimized at the maximum sampling time.

[0029] It is understandable that the above-mentioned information on a product, such as product information, enterprise information, multi-dimensional preservation information, or spatiotemporal information, can all be used as a means of traceability or logistics tracking, that is, the whole or part of traceability information or traceability code.

[0030] refer to Figure 2 In S200 of some embodiments of the present invention, the step of using a pseudo-noise sequence to spread spectrum modulate time-domain features, frequency-domain features, and a unique identification code to generate multiple traceability codes includes: spreading the time-domain features and frequency-domain features using a direct-sequence spread spectrum method to obtain spread spectrum coding; using the maximum sampling time as the minimum period for watermark generation, with the position change rate within each minimum period as the watermark carrier, modulating the pseudo-noise sequence onto the spread spectrum coding to generate a watermark; and embedding the watermarks within multiple minimum periods into the unique identification code to generate multiple traceability codes. It can be understood that the position change rate usually represents the rate of change of position over time, and the position can be represented by latitude, longitude, elevation, and start and end points. The pseudo-noise sequence includes chaotic sequences, M-sequences, and other random sequences.

[0031] To improve the robustness of the watermark, this disclosure selects a set of time slots to jointly carry the watermark signal. Assume the watermark signal used during tracking has L bits, and the length of the PN code is N. Based on the definition of the time slot centroid, for all n time slots Io, J1…-1, they are randomly divided into L groups, each containing n / L time slots, corresponding to 1 bit of the watermark signal. The time slot group corresponding to the i-th bit of the watermark signal is denoted as G. Each bit of the watermark signal, after spread spectrum, corresponds to N chips. G' is randomly divided into N groups, each containing r = n / LN time slots, where r is called the redundancy. The time slot group corresponding to the i-th chip is denoted as G, and the time slots it contains are denoted as lj, (r = 0…r-1). The common centroid of o…l.1 is defined as:

[0032]

[0033] Furthermore, a large delay b is applied to all time slots. If the chip to be embedded is 1, then a delay greater than b is applied to each packet in the corresponding time slot group. This is equivalent to each packet being further delayed based on b, thereby increasing the centroid of the time slot group. Conversely, if the chip to be embedded is -1, then a delay less than b is applied to each packet in the corresponding time slot group. This is equivalent to each packet being advanced based on b, thereby decreasing the centroid of the time slot group.

[0034] In some embodiments, the traceability code generation process includes generating a check code based on spatiotemporal information, standard codes, and status information. In some embodiments, the traceability code generation process is further configured to generate a check code based on spatiotemporal information, standard codes, and status information using any feasible algorithm, such as using code distance, parity check, Hamming check, cyclic redundancy check, etc. In some embodiments, the system can also affix electronic tags to agricultural products or their packaging during production (harvesting) or storage. Users can scan these electronic tags using mobile terminals (e.g., mobile phones, tablets, laptops, etc.) to obtain the corresponding agricultural product traceability code. In some embodiments, the electronic tag can be RFID, NFC, or a QR code. One or more traceability codes are stored in a Merkle tree, maintained or updated via a consortium blockchain jointly constructed by the generator, transporter, consumer, and regulator, using Ripple or Algorand consensus. The Merkle tree may also contain a unique identifier.

[0035] Furthermore, one or more traceability codes are arranged in chronological order (timestamps) and stored in a Merkle tree.

[0036] In S300 of some embodiments of the present invention, one or more traceability codes for each product are used as tags, and one or more corresponding preservation information is used as samples to train an autoencoder based on a convolutional neural network. Specifically, a dataset is constructed using one or more traceability codes for each product as tags and one or more corresponding preservation information as samples, and a spatiotemporal graph convolutional neural network is trained.

[0037] Example 2

[0038] refer to Figure 3In a second aspect, the present invention provides a cold chain logistics tracking system 1, comprising: an acquisition module 11, configured to acquire unique identification codes, spatiotemporal information, and multidimensional preservation information of multiple commodities, wherein the multidimensional preservation information includes temperature, humidity, pH value, and content of marker gases; a generation module 12, configured to determine the minimum dimension and maximum sampling time of the multidimensional preservation information of different commodities based on maximum information entropy and a multivariate autoregressive model, and extract the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information of each commodity; spread spectrum modulation of the temporal features, frequency domain features, and unique identification codes using a pseudo-noise sequence to generate multiple traceability codes; a training module 13, configured to train an autoencoder based on a convolutional neural network by using one or more traceability codes of each commodity as labels and one or more preservation information of the corresponding commodity as samples; and an input module 14, configured to input the real-time traceability information of the commodity to be traced into the trained autoencoder to obtain one or more traceability codes.

[0039] Furthermore, the generation module 12 includes: a first determining unit, used to determine the minimum dimension of the multidimensional preservation information of each commodity based on the maximum information entropy and the multivariate autoregressive model; and a second determining unit, used to determine the maximum sampling time of the multidimensional preservation information of each commodity based on the maximum information entropy and the higher-order autoregressive model.

[0040] Example 3

[0041] refer to Figure 4 A third aspect of the present invention provides an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method of the first aspect of the present invention.

[0042] Electronic device 500 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of electronic device 500. The processing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. An input / output (I / O) interface 505 is also connected to bus 504.

[0043] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, hard disks; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 An electronic device 500 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 4 Each box shown can represent a device or multiple devices as needed.

[0044] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by a processing device 501, it performs the functions defined in the methods of embodiments of this disclosure. It should be noted that the computer-readable medium described in embodiments of this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0045] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more computer programs, which, when executed by the electronic device, cause the electronic device to:

[0046] Computer program code for performing the operations of embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, C++, and Python—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0047] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0048] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cold chain logistics tracking method, characterized in that, include: The system acquires unique identification codes, spatiotemporal information, and multidimensional preservation information for multiple products, including temperature, humidity, pH value, and marker gas content. Based on maximum information entropy and a multivariate autoregressive model, the minimum dimension and maximum sampling time of multidimensional preservation information for different commodities are determined respectively. Then, the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information for each commodity are extracted. The temporal and frequency domain features and the unique identification code are spread-spectrum modulated using a pseudo-noise sequence to generate multiple traceability codes. Specifically, determining the minimum dimension and maximum sampling time of multidimensional preservation information for different commodities based on maximum information entropy and a multivariate autoregressive model includes: determining the minimum dimension of multidimensional preservation information for each commodity based on maximum information entropy and a multivariate autoregressive model; and determining the maximum sampling time of multidimensional preservation information for each commodity based on maximum information entropy and a higher-order autoregressive model. The multivariate autoregressive model is expressed as follows: ,in β 0 represents the initial freshness index. y Indicates the freshness index of the product. x 1, x 2,…, x n These represent the data for each dimension in the multidimensional archive information. e The error vector represents the multidimensional preservation information; wherein, the step of using a pseudo-noise sequence to spread spectrum modulate the time-domain features, frequency-domain features, and unique identification code to generate multiple traceability codes includes: spreading the time-domain features and frequency-domain features using a direct-sequence spread spectrum method to obtain a spread spectrum code; using the maximum sampling time as the minimum period for watermark generation, with the position change rate within each minimum period being the watermark carrier, modulating the pseudo-noise sequence onto the spread spectrum code to generate a watermark; and embedding the watermarks within multiple minimum periods into the unique identification code to generate multiple traceability codes; Use one or more traceability codes for each product as tags, and one or more corresponding preservation information as samples to train an autoencoder based on a convolutional neural network. The real-time traceability information of the product to be traced is input into the trained autoencoder to obtain one or more traceability codes.

2. The cold chain logistics tracking method according to claim 1, characterized in that, Also includes: Arrange one or more trace codes in chronological order and store them in a Merkle tree.

3. The cold chain logistics tracking method according to any one of claims 1 to 2, characterized in that, The convolutional neural network is a spatiotemporal graph convolutional neural network.

4. A cold chain logistics tracking system, characterized in that, include: The acquisition module is used to acquire the unique identification code, spatiotemporal information and multidimensional preservation information of multiple products. The multidimensional preservation information includes temperature, humidity, pH value and marker gas content. The generation module is used to determine the minimum dimension and maximum sampling time of multidimensional preservation information for different commodities based on maximum information entropy and a multivariate autoregressive model, and to extract the temporal and frequency domain features of the multidimensional preservation information and its corresponding spatiotemporal information for each commodity. It then uses a pseudo-noise sequence to spread-spectrum modulate the temporal features, frequency domain features, and unique identification code to generate multiple traceability codes. Specifically, determining the minimum dimension and maximum sampling time of multidimensional preservation information for different commodities based on maximum information entropy and a multivariate autoregressive model includes: determining the minimum dimension of multidimensional preservation information for each commodity based on maximum information entropy and a multivariate autoregressive model; and determining the maximum sampling time of multidimensional preservation information for each commodity based on maximum information entropy and a higher-order autoregressive model. The multivariate autoregressive model is expressed as: ,in β 0 represents the initial freshness index. y Indicates the freshness index of the product. x 1, x 2,…, x n These represent the data for each dimension in the multidimensional archive information. e The error vector represents the multidimensional preservation information; wherein, the step of using a pseudo-noise sequence to spread spectrum modulate the time-domain features, frequency-domain features, and unique identification code to generate multiple traceability codes includes: spreading the time-domain features and frequency-domain features using a direct-sequence spread spectrum method to obtain a spread spectrum code; using the maximum sampling time as the minimum period for watermark generation, with the position change rate within each minimum period being the watermark carrier, modulating the pseudo-noise sequence onto the spread spectrum code to generate a watermark; and embedding the watermarks within multiple minimum periods into the unique identification code to generate multiple traceability codes; The training module is used to train an autoencoder based on a convolutional neural network by using one or more traceability codes for each product as tags and one or more corresponding storage information as samples. The input module is used to input the real-time traceability information of the product to be traced into the trained autoencoder to obtain one or more traceability codes.

5. An electronic device, comprising: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the cold chain logistics tracking method as described in any one of claims 1 to 3.

6. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by the processor, it implements the cold chain logistics tracking method as described in any one of claims 1 to 3.

Images