Agricultural product data storage method and device based on CRM, and storage medium

Abstract

This invention relates to the field of agricultural data storage technology, and more particularly to a method, apparatus, and storage medium for storing agricultural product data based on CRM. The method includes: receiving an agricultural product data storage request initiated by a farmer's terminal; performing preliminary classification operations on the agricultural product data; performing an activity assessment on publicly accessible attribute data and determining the storage level of the publicly accessible attribute data based on the assessment results; performing a sensitivity assessment on private attribute data to determine its sensitivity level; performing differentiated encrypted storage operations based on the sensitivity level of the private attribute data; and generating an agricultural product storage response instruction and feeding it back to the farmer's terminal. This invention realizes a differentiated hierarchical storage strategy based on the security level and activity level of agricultural product data, thereby improving data storage security and resource utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of agricultural data storage technology, and in particular to a method, apparatus and storage medium for storing agricultural product data based on CRM. Background Technology

[0002] The application of CRM (Customer Relationship Management) systems in the agricultural product sector is becoming increasingly widespread. Farmers use CRM systems to upload agricultural product data from planting, production, and sales to servers for centralized storage and management, effectively improving the informatization level and transparency of the agricultural product industry chain.

[0003] Existing CRM systems manage agricultural product data in a rather crude manner, generally adopting a uniform storage strategy without differentiating data based on its public nature, access frequency, and sensitivity. This results in delayed response times for frequently accessed data and high-cost storage resources being consumed by low-value data. Furthermore, there is a lack of hierarchical encryption and transmission mechanisms for farmers' privacy data based on dynamic assessment of the network environment, leading to a high risk of leakage under low-speed networks. Overall, security and resource utilization efficiency are insufficient. Current technologies have not yet achieved hierarchical storage of agricultural product data that is secure, controllable, and resource-efficient, taking into account the different types and dynamic characteristics of the data. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, and storage medium for storing agricultural product data based on CRM, in order to solve the problem in the existing technology that it does not realize the hierarchical storage of agricultural product data in a secure, controllable, and resource-efficient manner, taking into account the different types and dynamic characteristics of agricultural product data.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for storing agricultural product data based on CRM includes:

[0007] Receive agricultural product data storage requests initiated by farmers' terminals and perform preliminary classification operations on the agricultural product data;

[0008] Perform an activity assessment on the publicly accessible attribute data and determine the storage level of the publicly accessible attribute data based on the assessment results;

[0009] Perform a sensitivity assessment on privacy-related data to determine its sensitivity level;

[0010] Perform differentiated encrypted storage operations based on the sensitivity level of the privacy-related data;

[0011] Generate agricultural product storage response instructions and send them back to the farmer's terminal.

[0012] Furthermore, the edge server obtains the access records of the farmer's terminal's historical uploaded public attribute data from the central server of the CRM system, and determines the activity index of the public attribute data to be stored based on the access records;

[0013] The analysis method for the activity index is as follows: count the number of times the currently stored public attribute data has been accessed within the most recent first preset time period, and record it as the recent access count; count the number of times the currently stored public attribute data has been accessed within an earlier second preset time period, and record it as the historical access count; use the ratio of the recent access count to the historical access count as the access frequency ratio; obtain the time span between the current moment and the first storage moment of the data, and record it as the first storage interval; obtain the time interval between the most recent access moment of the data and the current moment, and record it as the most recent storage interval; and perform weighted fusion of the recent access count, historical access count, access frequency ratio, first storage interval, and most recent storage interval to determine the activity index.

[0014] The activity index is compared with a preset first activity threshold and a second activity threshold to divide the public attribute data into high-frequency active data, medium-frequency active data, and low-frequency inactive data.

[0015] Furthermore, the analysis method for the sensitivity index includes:

[0016] Edge servers identify the categories of sensitive information contained in privacy-related data;

[0017] The edge server determines the number of categories and the proportion of data categories;

[0018] The edge server obtains the uplink transmission rate of the current network environment;

[0019] The sensitivity index is determined based on the number of categories, the proportion of data categories, and the uplink transmission rate.

[0020] Furthermore, the categories of sensitive information include: geographic coordinates, business data, identity information, and contract information. Each category of sensitive information corresponds to a preset basic sensitivity score, and the basic sensitivity scores of each category of sensitive information satisfy the following order: geographic coordinates > business data > identity information > contract information.

[0021] The edge server counts the number of sensitive information categories contained in the privacy attribute data, denoted as the category content number, and obtains the proportion of each sensitive information category in the total amount of current privacy attribute data, denoted as the data category ratio. The sensitivity index is positively correlated with the basic sensitivity score and the data category ratio, and negatively correlated with the uplink transmission rate.

[0022] Furthermore, the edge server sensitivity index is compared with a preset first sensitivity threshold and a second sensitivity threshold, wherein the first sensitivity threshold is greater than the second sensitivity threshold, so as to classify the privacy attribute data into high-sensitivity data, medium-sensitivity data and low-sensitivity data.

[0023] Furthermore, differentiated encrypted storage operations are performed based on the sensitivity level of the privacy attribute data:

[0024] For highly sensitive data, a triple protection strategy is adopted, including data splitting, independent encryption, and separate storage;

[0025] For sensitive data, a dual protection strategy is adopted, including dynamic chunking and out-of-order transmission;

[0026] For low-sensitivity data, the edge server performs regular encryption operations on the low-sensitivity data and then transmits it to the cloud server for storage via the uplink.

[0027] Furthermore, the data is split as follows: the edge server splits the highly sensitive data into a first data segment and a second data segment. The splitting method is to classify the odd-numbered record entries into the first data segment and the even-numbered record entries into the second data segment according to the order of the record entries in the data.

[0028] The independent encryption is as follows: the edge server performs encryption operations on the first data segment and the second data segment respectively, and the key used for encryption is generated by the key derivation function;

[0029] The separate storage is as follows: the edge server uses the uplink to transmit the encrypted first data fragment to the central server for storage, and transmits the encrypted second data fragment to the cloud server for storage. At the same time, the edge server generates a key index table locally to record the location information of the first and second data fragments respectively, and transmits the key index table to the secure storage area of ​​the central server through TLS encryption.

[0030] The dynamic segmentation is as follows: the edge server calculates the number of blocks of medium-sensitive data based on the current uplink transmission rate and sensitivity index, divides the medium-sensitive data into a number of data blocks evenly according to the number of blocks, and assigns a random block sequence number to each data block.

[0031] The out-of-order transmission is as follows: the edge server transmits each data block to the cloud server in a random order according to the block sequence number through the uplink. At the same time, the edge server records the correct block sequence number order as a sequence mapping table and transmits the sequence mapping table to the central server for storage through an independent secure channel.

[0032] Furthermore, once all agricultural product data has been stored, the CRM system generates an agricultural product storage response instruction. This instruction includes the storage location information, storage timestamp, and data integrity verification value for each data segment. The data integrity verification value is calculated using the MD5 algorithm. The edge server then uses a backhaul link to send the storage response instruction back to the farmer's terminal, completing the hierarchical storage operation of the agricultural product data.

[0033] On the other hand, the present invention also provides a data storage device for agricultural products based on CRM, comprising:

[0034] The initial classification module is used to receive agricultural product data storage requests initiated by farmers' terminals and perform initial classification operations on the agricultural product data;

[0035] The hierarchical storage module is used to perform activity assessment on public attribute data and determine the storage level of public attribute data based on the assessment results.

[0036] The sensitivity assessment module is used to perform sensitivity assessments on privacy attribute data to determine its sensitivity level;

[0037] The encrypted storage module is used to perform differentiated encrypted storage operations based on the sensitivity level of privacy-related data.

[0038] The storage feedback module is used to generate agricultural product storage response instructions and send them back to the farmer's terminal.

[0039] On the other hand, the present invention also provides a storage medium storing instructions that, when run on a computer, cause the computer to execute the above-described CRM-based agricultural product data storage method.

[0040] The beneficial effects of this invention are as follows: By intelligently classifying agricultural product data into public and private attributes, and combining this with activity assessment to achieve dynamic adaptation of storage levels, high-frequency data can be responded to quickly, and low-frequency data can be archived at low cost, thus optimizing the efficiency of storage resource utilization. At the same time, a sensitivity index assessment is introduced for privacy data, taking network transmission rate as an important factor into consideration, and implementing differentiated encrypted storage based on sensitivity level, which balances data transmission security and processing efficiency, and improves the security, adaptability, and storage economy of the CRM system in the agricultural product data management scenario. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of the agricultural product data storage method based on CRM in this embodiment.

[0043] Figure 2 This is a schematic diagram of the structure of the agricultural product data storage device based on CRM in this embodiment. Detailed Implementation

[0044] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a further detailed account of the agricultural product data storage method, apparatus, and storage medium based on CRM disclosed in this invention. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated; they can be combined to achieve better technical effects. In the accompanying drawings of the following embodiments, the same reference numerals in each drawing represent the same features or components, which can be applied to different embodiments. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0045] It should be noted that the structures, proportions, sizes, etc., illustrated in the accompanying drawings are merely for illustrative purposes and to aid those skilled in the art in understanding and reading the invention. They are not intended to limit the conditions under which the invention can be implemented. Any modifications to the structure, changes in proportions, or adjustments to size, provided they do not affect the effectiveness or purpose of the invention, should fall within the scope of the technical content disclosed in the invention. The scope of the preferred embodiments of the present invention includes other implementations, wherein functions may be performed not in the order stated or discussed, including substantially simultaneously or in reverse order, depending on the functions involved. This should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0046] Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0047] In the description of the embodiments of this application, " / " means "or", and "and / or" is used to describe the relationship between related objects, indicating that there can be three relationships. For example, "A and / or B" means: A and B exist alone, B exists alone, and A and B exist simultaneously. In the description of the embodiments of this application, "multiple" refers to two or more embodiments.

[0048] Example 1

[0049] This embodiment uses the example of a corn farmer uploading agricultural product data through a CRM system to illustrate the method of the present invention in detail.

[0050] like Figure 1 As shown, the agricultural product data storage method based on CRM provided by this invention includes the following steps:

[0051] Step S1: Receive the agricultural product data storage request initiated by the farmer's terminal and perform preliminary classification operation on the agricultural product data.

[0052] The farmer terminals have been pre-registered in the CRM system. The servers that support the operation of the CRM system include three types: edge servers, central servers, and cloud servers. The edge servers are deployed close to the farmer terminals and are used to perform preliminary data processing and classification. The central servers are used to store high-frequency access data and key management data. The cloud servers are used to store low-frequency access data and archived data.

[0053] Upon receiving a data storage request for agricultural products from a farmer's terminal, the edge server first parses the request to obtain the agricultural product data to be stored. The edge server then performs a preliminary classification operation on the agricultural product data, dividing it into public attribute data and private attribute data. Public attribute data refers to data that needs to be publicly displayed to the requesting party in the CRM system, including but not limited to agricultural product images, variety names, origin information, growth cycle descriptions, and taste evaluations. Private attribute data refers to data involving core information of the farmer's production and operation that is not suitable for public disclosure, including but not limited to point cloud geographic coordinates of the planting plot, detailed production costs, yield forecast data, and supply and sales contract information.

[0054] The preliminary classification method is as follows: The edge server reads the metadata tags of agricultural product data. The metadata tags have a preset data type identifier field. When the data type identifier field takes a first preset value, the agricultural product data is determined to be public attribute data; when the data type identifier field takes a second preset value, the agricultural product data is determined to be private attribute data. For data without metadata tags, the edge server performs keyword matching analysis on the data content. If the data content contains any of the preset geographic coordinate keywords, cost keywords, or contract keywords, it is determined to be private attribute data; otherwise, it is determined to be public attribute data. The first preset value is a marker for public data, such as "public," and the second preset value is a marker for private data, such as "private." The geographic coordinate keywords include, but are not limited to, "longitude," "latitude," and "altitude." The cost keywords include, but are not limited to, "cost," "unit price," and "total price." The contract keywords include, but are not limited to, "contract number" and "contracting party."

[0055] For example, a corn farmer, Zhang San, initiates a request for agricultural product data storage via a mobile terminal. The data to be uploaded includes: textual descriptions of corn varieties, maturity cycles, and average yield per acre; real-life photos of corn at maturity; point cloud geographic coordinates of the planting plot; and detailed production cost data for this year's corn planting. The point cloud geographic coordinates of the planting plot contains longitude, latitude, and altitude information for 5,000 location points. The edge server determines the textual descriptions of the variety name and the real-life photos as public attribute data, and the point cloud geographic coordinates and cost details as private attribute data.

[0056] Step S2: Perform an activity assessment on the public attribute data and determine the storage level of the public attribute data based on the assessment results. Although the security level of the public attribute data is relatively low, the access frequency of different public attribute data varies significantly. Agricultural product data in the peak sales season has a large access volume and high response time requirements, while historical data that has passed the season is almost ignored. Therefore, the storage level of the public attribute data is further divided.

[0057] Specifically, the edge server retrieves the access records of the farmer's terminal's historical uploaded public attribute data from the central server of the CRM system. The structure of the access records is as follows: R j =(r1,r2,...,r k ,...,r t ), r k =(T k V k ,F k ), where R j r represents the public attribute data access record of the j-th farmer terminal. k This represents the record unit accessed for the kth time, and t represents the total number of times the farmer's terminal's public attribute data has been accessed. k V represents the time when the k-th visit occurs. k F represents the number of visitors on the k-th visit. k This represents the identifier of the data content accessed for the kth time.

[0058] The edge server determines the activity index of the currently stored public attribute data based on access records. The analysis method for the activity index is as follows:

[0059] The number of accesses to the currently stored public attribute data within the most recent first preset time period is counted and recorded as the recent access count. In this embodiment, the first preset time period is 30 days, which can be adjusted according to the actual length of the agricultural product sales cycle, and is generally between 15 days and 90 days. At the same time, the number of accesses to the currently stored public attribute data within an earlier second preset time period is counted and recorded as the historical access count. In this embodiment, the second preset time period is the time range of 30 days to 90 days ago, which can be adjusted according to the reference value cycle of historical agricultural product data.

[0060] The ratio of recent visits to historical visits is used as the visit frequency ratio. When the historical visits are zero, the visit frequency ratio is directly set to the third preset value. In this embodiment, the third preset value is 2, which means that a higher activity rating is given to recent data when there is no historical data for reference.

[0061] The edge server obtains the time span between the current time and the first time the data was stored, denoted as the first storage interval, in days; and obtains the time interval between the most recent access time of the data and the current time, denoted as the most recent storage interval, in days.

[0062] The expression for the activity index is: Activity Index = w1 × Recent Access Count / (Recent Access Count + Historical Access Count + 1) + w2 × 1 / (Recent Storage Interval + 1) + w3 × 1 / (First Storage Interval + 1), where w1, w2, and w3 represent preset weight coefficients, which satisfy w1 + w2 + w3 = 1, w1 > w2 > w3. In this embodiment, w1 is 0.5, w2 is 0.3, and w3 is 0.2. The principle for setting the weight coefficients is that the recent access frequency contributes the most to the activity index, so it is given the largest weight; the access interval time contributes the second most to the activity index; and the total data storage time contributes the least to the activity index. The value range of each weight coefficient is between 0 and 1, and it can be adjusted by the system administrator according to the actual business scenario. The value range of the activity index is (0, 1], and the larger the value, the more active the data.

[0063] The edge server compares the activity index with a preset first activity threshold and a second activity threshold, wherein the first activity threshold is greater than the second activity threshold. In this embodiment, the first activity threshold is 0.6 and the second activity threshold is 0.2. The first and second activity thresholds are determined based on historical statistical data. For example, data with access frequency in the top 30% usually has an activity index greater than 0.6, while data with access frequency in the bottom 20% usually has an activity index less than 0.2. The value range of the first activity threshold is 0.5 to 0.8, and the value range of the second activity threshold is 0.1 to 0.4.

[0064] If the activity index is greater than or equal to the first activity threshold, the public attribute data is determined to be high-frequency active data and allocated to the cache area of ​​the edge server for storage in order to achieve fast response to access requests.

[0065] If the activity index is less than the first activity threshold but greater than the second activity threshold, the public attribute data is determined to be medium-frequency active data and allocated to the regular storage area of ​​the central server for storage.

[0066] If the activity index is less than the second activity threshold, the publicly available attribute data is classified as low-frequency dormant data and allocated to the archive storage area of ​​the cloud server for storage, in order to free up the high-cost storage resources of the edge server and the central server.

[0067] For example, the corn variety name data uploaded by Zhang San has 45 recent visits in the last 30 days, 12 historical visits between 30 and 90 days ago, an initial storage interval of 180 days, and a recent storage interval of 3 days. The calculated activity index is 0.464. Since 0.2 < 0.464 < 0.6, this data is judged as medium-frequency active data and is allocated to the regular storage area of ​​the central server for storage. If the corn ripening period real-time photos uploaded by Zhang San have very few recent visits and their activity index is below 0.2, they are judged as low-frequency inactive data and are allocated to the archive storage area of ​​the cloud server for storage.

[0068] Step S3: Perform a sensitivity assessment on the privacy attribute data to determine its sensitivity level. The sensitivity level includes high-sensitivity data, medium-sensitivity data, and low-sensitivity data. The security level of the privacy attribute data is relatively high, but the sensitivity of different privacy attribute data also varies. Based on the differences in the sensitivity of the privacy attribute data, implement differentiated encryption protection and transmission strategies.

[0069] Specifically, the edge server obtains the data type identifier and content characteristics of the privacy attribute data to determine its sensitivity index. The analysis method of the sensitivity index includes the following sub-steps:

[0070] Sub-step S31: The edge server identifies the categories of sensitive information contained in the privacy attribute data.

[0071] The preset sensitive information categories include: geographic coordinates (such as longitude, latitude, and altitude information in point cloud data), business data (such as production costs, profit calculations, and sales prices), identity information (such as farmers' ID numbers, contact information, and bank accounts), and contract information (such as supply and sales contract terms and counterparty information). Each category of sensitive information corresponds to a preset basic sensitivity score, which ranges from 0 to 1. The basic sensitivity scores for each category of sensitive information satisfy the following order: geographic coordinates > business data > identity information > contract information. In this embodiment, the basic sensitivity score for geographic coordinates is 0.8, for business data it is 0.6, for identity information it is 0.5, and for contract information it is 0.4.

[0072] In sub-step S32, the edge server determines the number of categories and the proportion of data categories.

[0073] The edge server counts the number of sensitive information categories contained in the privacy attribute data, denoted as the category content number, and obtains the proportion of each sensitive information category in the total amount of current privacy attribute data, denoted as the data category ratio. The value of each data category ratio ranges from 0 to 1, and the sum of all data category ratios is 1.

[0074] Sub-step S33: The edge server obtains the uplink transmission rate of the current network environment.

[0075] The uplink transmission rate is obtained as follows: the edge server obtains the transmission power of the farmer terminal sending data (in watts), the distance between the farmer terminal and the base station (in meters), the total uplink bandwidth (in Hertz), the channel fading coefficient per unit distance of the uplink, and the channel noise power (in watts). The uplink transmission rate is then calculated using the formula: Uplink transmission rate = Total uplink bandwidth × log2(1 + Transmission power of farmer terminal sending data × Channel fading coefficient per unit distance of the uplink × Distance between farmer terminal and base station / Channel noise power). The transmission power of the farmer terminal sending data typically ranges from 0.01 watts to 2 watts, and the channel fading coefficient per unit distance of the uplink typically ranges from 10 watts to 10 watts. -4 Up to 10 -2 The noise power of the channel typically ranges from 10. -14 Up to 10 -10 watt.

[0076] Sub-step S34: Determine the sensitivity index based on the number of categories, the proportion of data categories, and the uplink transmission rate.

[0077] The expression for the sensitivity index is: In the formula, S represents the sensitivity index, and wc q represents the basic sensitivity score for the c-th type of sensitive information. c Let v represent the proportion of data categories containing sensitive information of type c, v0 represent the uplink transmission rate in Mbps, v0 be a preset rate baseline value, and M represent the number of categories. In this embodiment, v0 is set to 1 Mbps. The basis for setting v0 is: when the transmission rate is lower than 1 Mbps, the network is considered a low-speed network, and the risk of data leakage increases significantly; when the transmission rate is much higher than 1 Mbps, this correction term approaches zero. The meaning of the expression for the sensitivity index is that the more sensitive information categories contained in the data, the larger the proportion of each type of sensitive information, and the higher the basic sensitivity score of each sensitive category, the larger the sensitivity index. At the same time, when the uplink transmission rate is low, the data is exposed for a longer time during transmission, and the risk of being stolen is higher, so the sensitivity index increases accordingly. The value range of the sensitivity index is 0 to 2 × ∑(w c ×q c )between.

[0078] The edge server sensitivity index is compared with a preset first sensitivity threshold and a second sensitivity threshold, wherein the first sensitivity threshold is greater than the second sensitivity threshold. In this embodiment, the first sensitivity threshold is 0.7 and the second sensitivity threshold is 0.4. The first sensitivity threshold and the second sensitivity threshold are determined based on industry standard analysis of the data security level of agricultural products. For example, data with a basic sensitivity score higher than 0.7 belongs to the high sensitivity category, and data with a score lower than 0.4 belongs to the low sensitivity category. The value range of the first sensitivity threshold is 0.6 to 0.9, and the value range of the second sensitivity threshold is 0.2 to 0.5.

[0079] If the sensitivity index is greater than or equal to the first sensitivity threshold, the privacy attribute data is judged as highly sensitive data;

[0080] If the sensitivity index is between the second sensitivity threshold and the first sensitivity threshold, the privacy attribute data is classified as medium sensitive data.

[0081] If the sensitivity index is less than or equal to the second sensitivity threshold, the privacy attribute data is classified as low-sensitivity data.

[0082] For example, the point cloud geographic coordinate data of the planting plot uploaded by Zhang San contains 5,000 location point records, which belong to the sensitive information of geographic coordinates. The basic sensitivity score is 0.8, the number of categories is 1, the data category ratio is 1, the current uplink transmission rate is 744Mbps, the rate base value is 1Mbps, and the calculated sensitivity index is 0.801. Since 0.801 > 0.7, the point cloud geographic coordinate data is judged to be highly sensitive data.

[0083] The production cost details uploaded by Zhang San are sensitive information in the category of operational data. The basic sensitivity score is 0.6, and the calculated sensitivity index is 0.601. Since 0.601 is between 0.4 and 0.7, the data is judged to be medium sensitive data.

[0084] Step S4: Perform differentiated encrypted storage operations based on the sensitivity level of the privacy attribute data.

[0085] Specifically, for highly sensitive data, a triple protection strategy is adopted, including data splitting, independent encryption, and separate storage;

[0086] The data is split as follows: the edge server splits the highly sensitive data into a first data segment and a second data segment. The splitting method is to classify the record entries in the data according to the order of the record entries, and classify the record entries with odd numbers into the first data segment and the record entries with even numbers into the second data segment. If the total number of record entries in the data is odd, the first data segment has one more record than the second data segment.

[0087] The independent encryption is as follows: the edge server performs encryption operations on the first data segment and the second data segment respectively. The encryption key is generated by a key derivation function. The input parameters of the key derivation function include: the current timestamp, the device identification code of the farmer terminal, and the sensitivity index of the highly sensitive data. Specifically, the timestamp, device identification code, and sensitivity index are concatenated into a byte sequence in the order of "timestamp + device identification code + sensitivity index" and input into the key derivation function. The key derivation function can use the PBKDF2 or HKDF algorithm to obtain the first encryption key and the second encryption key. The length of the first encryption key and the second encryption key is 256 bits. The first encryption key is used to encrypt the first data segment, and the second encryption key is used to encrypt the second data segment.

[0088] The separate storage is as follows: the edge server uses the uplink to transmit the encrypted first data segment to the central server for storage, and transmits the encrypted second data segment to the cloud server for storage. At the same time, the edge server generates a key index table locally to record the location information of the first and second data segments respectively. The location information includes the server IP address, storage path, and a combination of the timestamp and device identifier used when the corresponding key is derived. The key index table is transmitted to the secure storage area of ​​the central server via TLS encryption. When it is necessary to read the complete highly sensitive data, both data segments must be obtained from the central server and the cloud server at the same time, and decrypted and reassembled using their respective keys.

[0089] For sensitive data, a dual protection strategy is adopted, including dynamic chunking and out-of-order transmission;

[0090] The dynamic segmentation is as follows: the edge server calculates the number of segments for sensitive data based on the current uplink transmission rate and sensitivity index. The expression for the number of segments is: Nb=D / (v×t) s )×(1+S / Sg), where Nb represents the number of blocks, D represents the total amount of sensitive data in Mbit, and t s This represents the preset threshold for secure transmission time of a single data block, in seconds. Sg represents the first sensitivity threshold, and the number of blocks is rounded up. In this embodiment, t s Taking 2 seconds, in a low-speed network environment, transmitting a single data block within 2 seconds can effectively reduce the risk of interception; in a high-speed network environment, its impact on the number of blocks is relatively small. s The value ranges from 0.5 seconds to 5 seconds. The expression for the number of blocks means that the larger the data volume, the lower the transmission rate, and the higher the sensitivity, the more blocks there are, and the smaller the data volume of each block is, thereby reducing the amount of information leaked after a single transmission is intercepted. When the calculated number of blocks is less than 1, the number of blocks is set to 1, that is, no blocks are formed.

[0091] The edge server evenly divides the sensitive data into a number of data blocks according to the number of blocks, and assigns a random block sequence number to each data block. The block sequence number is a non-repeating random integer sequence from 1 to the number of blocks.

[0092] The out-of-order transmission is as follows: the edge server transmits each data block to the cloud server sequentially via the uplink according to the random arrangement of the block sequence numbers. Simultaneously, the edge server records the correct block sequence number arrangement as a sequence mapping table and transmits the sequence mapping table to the central server for storage via a separate secure channel. During data block transmission, each data block carries a temporary identifier instead of its actual block sequence number, ensuring that even if data blocks are intercepted during transmission, attackers cannot determine the correct arrangement of the data blocks. The separate secure channel uses a different network port and encryption channel than the data block transmission channel. After receiving all data blocks, the cloud server temporarily stores the data blocks according to the receiving order and waits for the central server to send the sequence mapping table before restoring the data.

[0093] For low-sensitivity data, the edge server performs regular encryption operations on the low-sensitivity data and then transmits it to the cloud server for storage via the uplink. The key used for regular encryption operations is provided by the system's pre-set unified key management module and adopts the AES-128 encryption algorithm.

[0094] For example, for Zhang San's moderately sensitive data: production cost details, D=50Mbit, v=744Mbps, S=0.601, first sensitivity threshold=0.7, t s=2 seconds, the number of blocks is 1, no blocks are needed, and direct transmission is possible. This achieves the adaptability of the block strategy. When the network conditions are good, it will not excessively block to avoid increasing transmission overhead.

[0095] In a remote mountainous area with a network transmission rate of v=2Mbps, if another farmer uploads yield prediction data D=30Mbit, S=0.667, and the number of blocks is calculated to be 15, then with a transmission rate of only 2Mbps, the 30Mbit data is divided into 15 data blocks for transmission, with each block containing approximately 2Mbit of data. The transmission time for a single block is approximately 1 second, which is far less than the safe transmission time threshold of 2 seconds.

[0096] Step S5: Generate agricultural product storage response instructions and send them back to the farmer's terminal.

[0097] Specifically, once all agricultural product data has been stored, the CRM system generates an agricultural product storage response instruction. The storage response instruction includes the storage location information of each data segment, the storage timestamp, and the data integrity verification value. The data integrity verification value is calculated using the MD5 algorithm. The edge server uses a backhaul link to send the storage response instruction back to the farmer's terminal, completing the hierarchical storage operation of the agricultural product data.

[0098] After receiving the storage response command, the farmer's terminal records the storage location information and data integrity verification value locally for subsequent data query and integrity verification.

[0099] Step S6: When responding to a data query request, perform a data reorganization operation.

[0100] Specifically, when the CRM system receives a query request for stored agricultural product data, it first determines the data type of the data being queried:

[0101] If the query request targets public attribute data, the edge server determines its storage level based on the data's activity index, reads the data directly from the corresponding storage location, and returns it.

[0102] If the query request targets privacy-related data, the edge server first determines its storage method based on the sensitivity level of the data:

[0103] For highly sensitive data, the edge server obtains the key index table from the central server, reads the encrypted first data segment and the second data segment from the central server and the cloud server respectively according to the key index table, decrypts them respectively using the corresponding keys, and splices them together to restore the complete data in an alternating manner of odd and even records, and then returns the complete data to the querying party.

[0104] For medium-sensitive data, the edge server obtains the sequence mapping table from the central server, reads each data block from the cloud server, arranges each data block in the correct order recorded in the sequence mapping table, splices and restores the complete data, performs decryption operation on the restored data, and returns it to the querying party.

[0105] For low-sensitivity data, the edge server reads the encrypted data from the cloud server, performs the decryption operation, and returns the result directly.

[0106] Example 2

[0107] This embodiment uses the example of a vegetable grower uploading agricultural product data through a CRM system in a remote area to illustrate the advantages of the present invention in low transmission rate scenarios.

[0108] For example, vegetable grower Li Si's planting base is located in a remote mountainous area with weak network signal and an uplink transmission rate of 2Mbps. Li Si wants to upload point cloud geographic coordinate data of the planting plot containing 2,000 location point records through the CRM system.

[0109] The edge server classifies this data as geographic coordinate information within privacy-related data, with a base sensitivity score of 0.8 and a sensitivity index of 1.066.

[0110] Since 1.066 > 0.7, the data is judged to be highly sensitive. It is worth noting that the sensitivity index of the same type of data was 0.801 in Example 1 when the transmission rate was 744Mbps, while in this example, due to the lower transmission rate, the sensitivity index increased to 1.066. This reflects the consideration of network environment factors in this invention - the worse the network conditions, the higher the risk of data exposure during transmission, and therefore the higher the sensitivity assessment result, the more stringent the protective measures taken.

[0111] The edge server implements a triple protection strategy for highly sensitive data: the 2,000 location point records are split into a first data segment of 1,000 records and a second data segment of 1,000 records. After being encrypted with independent keys, the first data segment is stored in the central server and the second data segment is stored in the cloud server.

[0112] This embodiment demonstrates that the present invention can adaptively adjust the sensitivity assessment results and corresponding protection strategies in different network environments, providing a higher level of data protection in low-speed network environments and effectively compensating for the security risks caused by poor network conditions.

[0113] Example 3

[0114] This embodiment illustrates the activity evaluation and storage level migration mechanism of the disclosed attribute data in this invention.

[0115] Suppose that when Zhang San first uploaded a picture of an agricultural product, it was during the peak sales season for that product. The picture had been accessed 200 times in the last 30 days, 20 times in the past 30 days, the most recent storage interval was 1 day, and the first storage interval was 60 days. The activity index was calculated to be 0.6058. This data was determined to be high-frequency active data and stored in the cache of the edge server.

[0116] Three months later, the agricultural product was out of season, and market demand dropped significantly. The access frequency of the image data decreased significantly. The edge server recalculated the activity index of the data according to the preset monthly evaluation cycle. The statistical results showed that the number of accesses in the last 30 days dropped to 6, while the number of accesses from 30 to 90 days ago was updated to 150. The time interval between the most recent access time and the current time extended to 28 days, and the time span between the data's initial storage time and the current time increased to 150 days. The recalculated activity index was 0.0308, and the data was judged as low-frequency dormant data. Based on this evaluation result, the edge server automatically migrated the data from the edge server cache to the cloud server's archive storage area to free up the edge server's high-cost storage resources. At the same time, after the migration was completed, the edge server updated the data's storage location metadata to ensure that subsequent query requests could be correctly routed to the cloud server.

[0117] When the agricultural product enters the sales season again in the future, the access frequency of the image data may rise again. The edge server will detect the rebound of the activity index in the next evaluation cycle and migrate the data from the cloud server archive area to the edge server cache area or the central server regular storage area in a timely manner, so as to achieve dynamic adaptive adjustment of the storage level.

[0118] This embodiment demonstrates that the present invention achieves dynamic migration of the storage hierarchy of publicly attribute data by periodically recalculating the activity index, ensuring that storage resources are always allocated to the data that needs the fastest response, thereby improving the utilization efficiency of storage resources.

[0119] This invention divides agricultural product data into public attribute data and private attribute data, and introduces an activity evaluation mechanism for public attribute data. It dynamically allocates storage levels according to the access frequency and timeliness of the data, thereby achieving refined utilization of storage resources. High-frequency active data is stored in the edge server cache area to ensure access speed, while low-frequency inactive data is migrated to the cloud server archive area to reduce storage costs. This avoids the resource waste or response delay problems caused by unified storage in the prior art.

[0120] This invention introduces a sensitivity assessment mechanism for privacy-related data. It comprehensively considers the categories of sensitive information contained in the data, the proportion of each type of sensitive information, and the current network transmission rate to dynamically calculate a sensitivity index. Differentiated protection strategies are implemented based on the sensitivity level. High-sensitivity data employs a triple protection approach of "splitting + independent encryption + separate storage," medium-sensitivity data employs a dual protection approach of "dynamic chunking + out-of-order transmission," and low-sensitivity data employs conventional encryption. This achieves a precise match between the strength of security protection and the sensitivity of the data, avoiding resource consumption caused by over-protection or security risks caused by insufficient protection.

[0121] In processing sensitive data, this invention dynamically calculates the number of blocks based on data volume, transmission rate, and sensitivity index, enabling the block granularity to adapt to network conditions and data characteristics. When the network transmission rate is low, the number of blocks automatically increases, while the size of each block decreases, shortening the transmission time of a single block and reducing the risk of data interception. Simultaneously, out-of-order transmission and separate storage of the sequence mapping table further enhance the security of the data transmission process.

[0122] In processing highly sensitive data, this invention employs a splitting method that separates odd-numbered records from even-numbered records. The data is divided into two logically incomplete segments, each encrypted with an independent key and stored on different servers. If either single server is compromised, the complete data content cannot be obtained, thus improving the security of data storage.

[0123] Please see Figure 2 As shown, this is a data storage device for agricultural products based on CRM in this embodiment, including:

[0124] The initial classification module is used to receive agricultural product data storage requests initiated by farmers' terminals and perform initial classification operations on the agricultural product data;

[0125] The hierarchical storage module is used to perform activity assessment on public attribute data and determine the storage level of public attribute data based on the assessment results.

[0126] The sensitivity assessment module is used to perform sensitivity assessments on privacy attribute data to determine its sensitivity level;

[0127] The encrypted storage module is used to perform differentiated encrypted storage operations based on the sensitivity level of privacy-related data.

[0128] The storage feedback module is used to generate agricultural product storage response instructions and send them back to the farmer's terminal.

[0129] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the agricultural product data storage method based on CRM shown in the above embodiments is implemented.

[0130] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0131] In the above description, the disclosure of this invention is not intended to limit itself to these aspects. Rather, within the scope of the objectives of this disclosure, components can be selectively and operationally combined in any number. Furthermore, terms such as “comprising,” “encompassing,” and “having” should be interpreted by default as inclusive or open-ended, rather than exclusive or closed, unless explicitly defined as such. All technical, scientific, or other terms are to be understood by those skilled in the art, unless defined as such. Public terms found in dictionaries should not be interpreted in the context of the relevant technical documents in an overly idealistic or impractical manner, unless explicitly defined as such in this disclosure. Any modifications or alterations made by those skilled in the art based on the foregoing disclosure are within the scope of the claims.

Claims

1. A method for storing agricultural product data based on CRM, characterized in that, include: Receive agricultural product data storage requests initiated by farmers' terminals and perform preliminary classification operations on the agricultural product data; Perform an activity assessment on the publicly accessible attribute data and determine the storage level of the publicly accessible attribute data based on the assessment results; Perform a sensitivity assessment on privacy-related data to determine its sensitivity level; Perform differentiated encrypted storage operations based on the sensitivity level of the privacy-related data; Generate agricultural product storage response instructions and send them back to the farmer's terminal.

2. The method for storing agricultural product data based on CRM according to claim 1, characterized in that, The edge server obtains the access records of the farmer's terminal's historical uploaded public attribute data from the central server of the CRM system, and determines the activity index of the public attribute data to be stored based on the access records; The analysis method for the activity index is as follows: count the number of times the currently stored public attribute data has been accessed within the most recent first preset time period, and record it as the recent access count; count the number of times the currently stored public attribute data has been accessed within an earlier second preset time period, and record it as the historical access count; use the ratio of the recent access count to the historical access count as the access frequency ratio; obtain the time span between the current moment and the first storage moment of the data, and record it as the first storage interval; obtain the time interval between the most recent access moment of the data and the current moment, and record it as the most recent storage interval; and perform weighted fusion of the recent access count, historical access count, access frequency ratio, first storage interval, and most recent storage interval to determine the activity index. The activity index is compared with a preset first activity threshold and a second activity threshold to divide the public attribute data into high-frequency active data, medium-frequency active data, and low-frequency inactive data.

3. The method for storing agricultural product data based on CRM according to claim 1, characterized in that, The analysis method for the sensitivity index includes: Edge servers identify the categories of sensitive information contained in privacy-related data; The edge server determines the number of categories and the proportion of data categories; The edge server obtains the uplink transmission rate of the current network environment; The sensitivity index is determined based on the number of categories, the proportion of data categories, and the uplink transmission rate.

4. The method for storing agricultural product data based on CRM according to claim 3, characterized in that, Sensitive information categories include: geographic coordinates, business data, identity information, and contract information. Each category of sensitive information corresponds to a preset basic sensitivity score, and the basic sensitivity scores of each sensitive information category satisfy the following order: geographic coordinates > business data > identity information > contract information. The edge server counts the number of sensitive information categories contained in the privacy attribute data, denoted as the category content number, and obtains the proportion of each sensitive information category in the total amount of current privacy attribute data, denoted as the data category ratio. The sensitivity index is positively correlated with the basic sensitivity score and the data category ratio, and negatively correlated with the uplink transmission rate.

5. The method for storing agricultural product data based on CRM according to claim 4, characterized in that, The edge server sensitivity index is compared with a preset first sensitivity threshold and a second sensitivity threshold, wherein the first sensitivity threshold is greater than the second sensitivity threshold, so as to classify the privacy attribute data into high-sensitivity data, medium-sensitivity data and low-sensitivity data.

6. The method for storing agricultural product data based on CRM according to claim 1, characterized in that, Perform differentiated encrypted storage operations based on the sensitivity level of the privacy attribute data: For highly sensitive data, a triple protection strategy is adopted, including data splitting, independent encryption, and separate storage; For sensitive data, a dual protection strategy is adopted, including dynamic chunking and out-of-order transmission; For low-sensitivity data, the edge server performs regular encryption operations on the low-sensitivity data and then transmits it to the cloud server for storage via the uplink.

7. The method for storing agricultural product data based on CRM according to claim 6, characterized in that, The data is split as follows: the edge server splits the highly sensitive data into a first data segment and a second data segment. The splitting method is to classify the odd-numbered records into the first data segment and the even-numbered records into the second data segment according to the order of the records in the data. The independent encryption is as follows: the edge server performs encryption operations on the first data segment and the second data segment respectively, and the key used for encryption is generated by the key derivation function; The separate storage is as follows: the edge server uses the uplink to transmit the encrypted first data fragment to the central server for storage, and transmits the encrypted second data fragment to the cloud server for storage. At the same time, the edge server generates a key index table locally to record the location information of the first and second data fragments respectively, and transmits the key index table to the secure storage area of ​​the central server through TLS encryption. The dynamic segmentation is as follows: the edge server calculates the number of blocks of medium-sensitive data based on the current uplink transmission rate and sensitivity index, divides the medium-sensitive data into a number of data blocks evenly according to the number of blocks, and assigns a random block sequence number to each data block. The out-of-order transmission is as follows: the edge server transmits each data block to the cloud server in a random order according to the block sequence number through the uplink. At the same time, the edge server records the correct block sequence number order as a sequence mapping table and transmits the sequence mapping table to the central server for storage through an independent secure channel.

8. The method for storing agricultural product data based on CRM according to claim 7, characterized in that, Once all agricultural product data has been stored, the CRM system generates an agricultural product storage response instruction. This instruction includes the storage location information, storage timestamp, and data integrity verification value for each data segment. The data integrity verification value is calculated using the MD5 algorithm. The edge server then uses a backhaul link to send the storage response instruction back to the farmer's terminal, completing the hierarchical storage operation of the agricultural product data.

9. A CRM-based agricultural product data storage device, applied to the CRM-based agricultural product data storage method as described in any one of claims 1-8, characterized in that, include: The initial classification module is used to receive agricultural product data storage requests initiated by farmers' terminals and perform initial classification operations on the agricultural product data; The hierarchical storage module is used to perform activity assessment on public attribute data and determine the storage level of public attribute data based on the assessment results. The sensitivity assessment module is used to perform sensitivity assessments on privacy attribute data to determine its sensitivity level; The encrypted storage module is used to perform differentiated encrypted storage operations based on the sensitivity level of privacy-related data. The storage feedback module is used to generate agricultural product storage response instructions and send them back to the farmer's terminal.

10. A storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the agricultural product data storage method based on CRM as described in any one of claims 1-8.

Images