Hash encryption method for after-sales data processing of digital products

Abstract

The application discloses a hash encryption method for after-sales data processing of digital products, and relates to the technical field of network data encryption security.The method extracts a physical unclonable function response of an after-sales terminal, combines a timestamp and a fault type code to generate a dynamic salt value, calculates a hash digest after splicing the dynamic salt value with original data, and writes the hash digest into a storage chain, so that the equivalent ciphertext space can be expanded without increasing the number of hash iterations, and a pre-computation reverse search attack against low-entropy structured identification data can be resisted.For key nodes in the storage chain that involve user privacy plaintext, a chameleon hash function with a trapdoor is used to calculate a digest value, when a deletion request is received and a multi-party threshold authorization condition is met, a trapdoor private key is reconstructed and a new random number is calculated, and after an original privacy message is replaced by a desensitization placeholder, the digest value remains unchanged.The application can realize physical-level destruction of privacy data while maintaining the integrity and traceability of the storage chain, and is suitable for compliance storage and deletion processing of after-sales data of digital products.

Description

Technical Field

[0001] This invention relates to the field of network data encryption security technology, specifically to a hash encryption method for low-entropy structured identification data such as international mobile device identification codes and serial numbers, as well as privacy data such as repair work orders, applied in digital product after-sales service scenarios. In particular, it relates to a revocable evidence storage hash encryption method based on dynamic salting using a terminal physically unclonable function combined with trapdoor hash collision replacement, which falls under the category of privacy computing and data compliance technology. Background Technology

[0002] In the after-sales service system of digital products, product serial numbers, international mobile device identification codes, warranty certificates, user identity information, and repair work order data often need to be stored for a long time to cope with business needs such as quality disputes, product recalls, and warranty verification that may last for several years. Hash encryption technology is widely used to ensure the tamper-proof and integrity of the above data during storage and transmission. Its basic principle is to map an input of arbitrary length to a digest value of fixed length through a one-way hash function. Theoretically, it is difficult to reverse the original input from the digest value. In the construction of the evidence storage chain, the hash digest value of the previous node is usually used as part of the input of the next node to form a chain structure. If the data of any node is tampered with, it will cause the subsequent digest values ​​to change arithmetically, thereby achieving the tamper-proof effect.

[0003] Currently, the most common implementation schemes for hash protection of structured short identifier data in the industry mainly fall into two categories. The first category is a scheme that combines fixed salt value with standard hashing. The processing method is as follows: the system pre-generates or configures a fixed salt value, concatenates the salt value with the original plaintext before calculating the hash, performs a standard hash operation on the concatenation result, and stores the calculation result together with the salt value or only stores the digest value. Although this scheme improves security to a certain extent compared to unsalted hashing, since the number of bits of identifier data such as international mobile device identification codes and serial numbers is fixed and the character set is limited, its plaintext space is much smaller than the conventional password space. Once the salt value is fixed and obtained or guessed by an attacker, the attacker can still pre-build a customized lookup table structure for all possible identifier combinations to perform reverse lookups, thereby greatly reducing the security gains brought by salting.

[0004] The second type is the slow hash algorithm with an adjustable iteration factor. By artificially increasing the time and memory overhead of a single hash calculation, the cost of brute-force enumeration by attackers is increased. This type of solution is quite effective in traditional user password protection scenarios. However, since after-sales testing terminals generally have limited computing power and rely on battery power, using a slow hash algorithm with a high number of iterations will significantly slow down the terminal response speed and greatly increase the device's energy consumption. It is difficult to meet the efficiency requirements of the massive daily concurrent data entry of after-sales service centers, resulting in poor applicability of this type of solution on resource-constrained embedded after-sales terminals.

[0005] More importantly, both of the existing solutions mentioned above share common flaws in data deletion: due to the strict one-wayness and avalanche effect of standard hash functions, once the hash chain is established, the deletion of any underlying data requires recalculating the hash value of that node and all subsequent nodes; otherwise, the chain will break or the digest value will be distorted. Therefore, in actual business, existing solutions often have to adopt two rather extreme approaches to deal with deletion requests made by users in accordance with data compliance regulations: retaining the entire record in the long term and refusing to respond to deletion requests, or rebuilding the entire evidence storage chain. The former carries compliance risks, while the latter is costly and destroys the continuity and traceability of the original evidence storage, making it difficult to balance the compliant destruction of privacy data with the maintenance of the integrity of the evidence storage chain. Summary of the Invention

[0006] The technical problem this invention aims to solve is: in the context of digital product after-sales service, how to provide a hash encryption method that can resist pre-computation reverse lookup attacks targeting limited plaintext space with lightweight computational overhead, and achieve physical-level destruction of privacy data when responding to legitimate user deletion requests, while not compromising the integrity and traceability of the original evidence hash chain, for low-entropy structured short identifier data such as international mobile device identification codes and serial numbers, as well as privacy data such as repair work orders.

[0007] To address the aforementioned technical problems, this invention provides a hash encryption method for processing after-sales data of digital products, comprising:

[0008] Acquire raw data to be processed from the after-sales terminal, wherein the raw data includes at least one of the following: International Mobile Equipment Identity (IMEI), serial number, and repair work order text;

[0009] An excitation is applied to the physically unclonable function module in the security chip of the after-sales terminal to obtain the original response. The original response is then corrected by a fuzz extractor to obtain a stable response and auxiliary data for reconstructing the stable response.

[0010] Obtain the current timestamp and the fault type code corresponding to this service, concatenate the stable response, the timestamp, and the fault type code in a predetermined order, and input them into the key derivation function to obtain the dynamic salt value;

[0011] The dynamic salt value is concatenated with the original data and then input into a preset hash function to obtain a hash digest. The hash digest, the encrypted dynamic salt value, and the index information of the auxiliary data are written into the corresponding node of the evidence storage chain. Specifically, the encryption protection of the dynamic salt value is achieved by calling a preset symmetric encryption key in the trusted execution environment within the security chip to perform symmetric encryption operations on the dynamic salt value to obtain a ciphertext form of the dynamic salt value. This ciphertext, along with the hash digest and the index information of the auxiliary data, is written into the corresponding node of the evidence storage chain to prevent the plaintext of the dynamic salt value from being leaked and used to reverse-engineer the original data.

[0012] For key nodes in the evidence storage chain that involve user privacy plaintext, a chameleon hash function with trapdoors is used to calculate the digest value of the key node;

[0013] When a deletion request for the critical node is received, key fragment submission requests are initiated to multiple preset independent authorized entities. When the number of submitted key fragments reaches a preset threshold, the trapdoor private key of the chameleon hash function is reconstructed based on the threshold secret sharing reconstruction algorithm.

[0014] A new random number is calculated based on the trapdoor private key, the original privacy message, the pre-agreed de-identified placeholder, and the initial random number, so that the digest value calculated after the replacement is consistent with that before the replacement;

[0015] Replace the original record in the key node with the desensitized placeholder and the new random number, and destroy the trapdoor private key obtained in this reconstruction after verification.

[0016] The beneficial effects achieved by this invention are as follows:

[0017] 1. This invention extracts the physically unclonable function response of the after-sales terminal and integrates the timestamp and fault type code to form a dynamic salt value. This makes the salt value used in each hash calculation different and unpredictable. It can greatly expand the equivalent ciphertext space without increasing the number of hash iterations, fundamentally alleviating the problem that low-entropy short identifier data is easily reversed by pre-built lookup table structures. At the same time, it maintains the computational efficiency comparable to lightweight hash algorithms and is suitable for resource-constrained after-sales testing terminals.

[0018] 2. This invention introduces a chameleon hash mechanism with trapdoors at key nodes of the evidence storage chain, enabling the authorized entity to calculate a collision random number that keeps the digest value unchanged after obtaining multi-party threshold authorization. This allows the privacy plaintext to be replaced with desensitized placeholders without affecting the overall hash continuity of the evidence storage chain, achieving compatibility and unity between the physical destruction of privacy data and the maintenance of the integrity of the evidence storage chain.

[0019] 3. This invention distributes the trapdoor private key to multiple independent authorized entities based on threshold-based secret sharing. The trapdoor private key is only temporarily reconstructed when a preset threshold condition is met and is destroyed after use. This avoids the risk of the trapdoor being misused by a single entity to maliciously tamper with the evidence storage data, thereby improving the security and compliance of the overall solution. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the dynamic salting hash generation process for the physically unclonable function in the after-sales terminal of this invention.

[0021] Figure 2 This is a flowchart illustrating the construction process of the revocable evidence storage chain based on chameleon hashing in this invention.

[0022] Figure 3 This is a flowchart of the multi-threshold authorization trapdoor reconstruction and collision replacement process of the present invention. Detailed Implementation

[0023] The following is in conjunction with the appendix Figure 1-3 The specific embodiments of the present invention will be further described below. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0024] I. System Overall Architecture Implementation Example

[0025] The system upon which the hash encryption method of this invention relies comprises three parts: an after-sales service terminal layer, an after-sales cloud platform layer, and a multi-party authorization layer. The after-sales service terminal layer includes a handheld testing terminal, which internally houses a security chip. This security chip integrates a physically unclonable function module, a trusted execution environment, and a lightweight hash operation unit. The physically unclonable function module generates a device-unique physical response after an stimulus is applied. The trusted execution environment provides a secure operating environment for the generation of dynamic salt values ​​and temporary caching. The lightweight hash operation unit performs hash digest calculations. The after-sales service terminal communicates with the after-sales cloud platform layer via a network. The after-sales cloud platform layer includes a data receiving module, a dynamic salt value management module, and a data storage chain construction and storage module. The module includes a chain-structured evidence storage chain consisting of multiple evidence storage nodes. Each evidence storage node records a corresponding message field, a random number field, and a digest field. The multi-party authorization layer includes three independent key shard holders: a manufacturer's quality control department node, a legal compliance department node, and a third-party notary node. Each authorized node connects to the trapdoor reconstruction and collision calculation module in the after-sales cloud platform layer via communication. Only when a preset threshold condition is met can the key shards submitted by each authorized node be aggregated and trigger the module to perform trapdoor private key reconstruction. This architecture enables the entire process of data collection from the terminal, salting and hashing, on-chain evidence storage, and finally, triggering multi-party authorization, trapdoor reconstruction, and collision replacement after the user initiates a deletion request, to form a complete closed loop.

[0026] II. Example of Dynamic Salted Hash Generation for Physically Unclonable Functions in After-Sales Terminals

[0027] In this embodiment, the after-sales terminal security chip first receives the raw data to be processed. This raw data can be a structured short identifier or short text data such as an International Mobile Equipment Identity (IMI), serial number, or repair work order text. Subsequently, the physical non-cloning function module in the security chip generates the raw response under the action of a preset stimulus. Since the physical non-cloning function response contains environmental noise, the security chip further calls the generation algorithm of the fuzzy extractor to perform error correction processing on the raw response, outputting a stable response and auxiliary data for subsequent reconstruction. This processing method ensures that the same device can still obtain a consistent and stable response when collecting physical non-cloning function responses multiple times under different environmental conditions, providing a basis for the repeatable generation of dynamic salt values.

[0028] Next, the security chip obtains the current timestamp and the fault type code corresponding to this service, and concatenates the stable response, timestamp, and fault type code in a predetermined order to obtain a concatenated string, which is represented as follows:

[0029] ;

[0030] in This represents the seed material obtained by piecing it together. This represents the stable, physically unclonable function response after error correction by the fuzz extractor, used to introduce a device-level, unclonable physical entropy source. This represents the timestamp when the current business transaction occurred, used to introduce the unpredictability of the time dimension. This indicates the fault type code corresponding to this business operation, used to introduce differences related to the business context. This indicates a bit string concatenation operation.

[0031] Subsequently, the concatenated string is input into the key derivation function to calculate the dynamic salt value, which is represented as follows:

[0032] ;

[0033] in This represents the key derivation function, used to transform the input seed material into a fixed-length, high-entropy pseudo-random salt value. The final generated dynamic salt value is different because the stable response originates from a device-level physical non-clonable function, and the timestamp and fault type code change with business operations. Therefore, the dynamic salt values ​​generated by different evidence storage nodes are different and difficult for attackers to obtain or calculate in advance.

[0034] Furthermore, after concatenating the dynamic salt value with the original data, the data is input into a preset lightweight hash function to calculate the hash digest, which is represented as follows:

[0035] ;

[0036] in This indicates the default lightweight hash function. Represents the original plaintext data. The generated hash digest value is represented. Finally, the hash digest, the encrypted dynamic salt value, and the index information of the fuzzy extractor auxiliary data are written to the corresponding node of the evidence storage chain. Since the dynamic salt value is different for each record and depends on the device-level physical entropy source, it is difficult for attackers to reproduce the same salt value even if they obtain the same terminal. This makes it difficult for pre-lookup reverse lookup attacks targeting a fixed plaintext space to succeed. At the same time, since the above generation process does not increase the number of hash iterations, and only introduces the physical non-cloning function response and business context information in the salt value generation stage, the overall computational overhead is comparable to that of a lightweight hash algorithm, making it suitable for after-sales testing terminals with limited computing power and battery power.

[0037] III. Implementation Example of Constructing a Revocable Evidence Storage Chain Based on Chameleon Hash

[0038] In the key nodes of the evidence storage chain, namely the nodes involving user privacy plaintext, this embodiment does not directly use the standard hash function, but uses a chameleon hash function with trapdoors to calculate the digest value, so that the digest value of the node can remain unchanged after the message is legally replaced.

[0039] Specifically, the Chameleon hash function is constructed based on the discrete logarithm problem. It sets the public parameter to a large prime number, the upper order of the finite field to a preset value for the generator, and the trapdoor private key to a preset private parameter. Its corresponding public key is calculated in the following way:

[0040] ;

[0041] in Represents generator, This indicates that the trapdoor private key is only temporarily held by the authorized entity after the threshold has been reconstructed. represents a large prime number, This represents the publicly available chameleon hash public key. For both messages and random numbers, the chameleon hash function is defined as follows:

[0042] ;

[0043] in This indicates the message to be hashed. In key nodes of the evidence storage chain, before deletion, the message is the original private plaintext; after deletion, the message is replaced with a de-identified placeholder. This represents the random number corresponding to the message. , , The meaning is the same as before. This represents the chameleon hash digest value corresponding to the set of messages and random numbers. Before deletion, the message field of the key node in the evidence storage chain contains the original private plaintext, and the random number field contains the initial random number. After the legitimate deletion is completed, the message field is replaced with a desensitized placeholder, and the random number field is replaced with a new random number obtained through collision calculation. The digest value marked at the top of the node remains consistent before and after the replacement, thus achieving the legitimate replacement of the private plaintext without destroying the dependency relationship between the nodes before and after the evidence storage chain.

[0044] IV. Implementation Examples of Multi-Party Threshold Authorization Trapdoor Reconstruction and Collision Replacement

[0045] When a user exercises their right to be forgotten in accordance with data compliance regulations and requests the deletion of evidence nodes containing their private information, the after-sales platform receives a deletion request from the user after identity verification. The request specifies the original private message to be deleted and its corresponding evidence node.

[0046] Subsequently, the platform initiates key shard submission requests to three authorized entities: the manufacturer's quality control department node, the legal and compliance department node, and the third-party notary node. Subsequent operations can only be executed when the number of submitted shards reaches a preset threshold; otherwise, the process is terminated and an audit log is recorded to prevent a single entity from arbitrarily triggering privacy replacement operations. After the threshold conditions are met, the platform uses the reconstructing algorithm based on threshold secret sharing to temporarily reconstruct the chameleon hash trapdoor private key using the aggregated key shards. This private key is only valid within the memory lifecycle of this operation.

[0047] Next, based on the trapdoor private key, the original private message, and the original random number, a new random number that satisfies the collision condition is calculated, and its representation is as follows:

[0048] ;

[0049] in This indicates the original private messages to be deleted. This represents a pre-agreed de-identification placeholder, which can be a uniform padding value or a hash marker. Represents the random number in the original node. This represents the multiplicative inverse of the trapdoor private key under the preset cyclic group order. Indicates the order of the cyclic group in which the generator resides. This represents a newly calculated random number that satisfies the condition that the summary values ​​before and after the replacement are equal.

[0050] Subsequently, the original message and random number records in the corresponding nodes of the evidence storage chain are replaced with desensitized placeholders and new random numbers. The digest value recalculated after replacement is compared and verified with the digest value before replacement. If the verification results are consistent, the trapdoor private key of this temporary reconstruction is immediately destroyed, and an audit log containing the operation time, authorized subject, and node identifier is recorded. If the verification results are inconsistent, the replacement operation is rolled back and an alarm is triggered to prevent illegal tampering. Through the above processing method, this embodiment maintains the continuity and traceability of the digest value of the evidence storage chain while completing the physical-level destruction of privacy plaintext.

[0051] V. Alternative Implementation Methods

[0052] In another embodiment of the present invention, considering the potential impact of future quantum computing on the discrete logarithm problem, the chameleon hash function constructed based on the discrete logarithm problem can be replaced with a lattice cryptographic chameleon hash scheme constructed based on the small integer solution problem on lattices. The trapdoor private key and collision calculation process are adjusted accordingly to lattice basis reduction correlation operation. The aforementioned dynamic salting, multi-party threshold authorization and auditing process can remain unchanged, thereby enhancing the long-term security of the scheme while maintaining the original method structure.

[0053] In another embodiment of the present invention, for some application scenarios where the strong consistency requirement of the evidence storage chain structure is relatively low, privacy destruction can be achieved without using the chameleon hash collision replacement method. Instead, the original privacy data is stored separately in a deletable storage area in the form of encrypted ciphertext. The evidence storage chain nodes only store the standard hash digest and access index corresponding to the ciphertext. When a legitimate deletion request is triggered and the threshold authorization conditions are met, the corresponding ciphertext and decryption key fragments are directly physically destroyed, making the original privacy data unrecoverable. The digest value of the evidence storage chain node itself remains unchanged, thereby achieving a compliant deletion and evidence storage integrity compatibility effect similar to the aforementioned embodiments in the way that the ciphertext is destructible and the digest value remains unchanged.

[0054] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, any modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope defined by the claims of the present invention.

Claims

1. A hash encryption method for processing after-sales data of digital products, characterized in that, include: Acquire raw data to be processed from the after-sales terminal, wherein the raw data includes at least one of the following: International Mobile Equipment Identity (IMEI), serial number, and repair work order text; An excitation is applied to the physically unclonable function module in the security chip of the after-sales terminal to obtain the original response. The original response is then corrected by a fuzz extractor to obtain a stable response and auxiliary data for reconstructing the stable response. Obtain the current timestamp and the fault type code corresponding to this service, concatenate the stable response, the timestamp, and the fault type code in a predetermined order, and input them into the key derivation function to obtain the dynamic salt value; The dynamic salt value is concatenated with the original data and then input into a preset hash function to obtain a hash digest. The hash digest, the encrypted dynamic salt value, and the index information of the auxiliary data are written into the corresponding node of the evidence storage chain. Specifically, the encryption protection of the dynamic salt value is achieved by calling a preset symmetric encryption key in the trusted execution environment within the security chip to perform symmetric encryption operations on the dynamic salt value to obtain a ciphertext form of the dynamic salt value. This ciphertext, along with the hash digest and the index information of the auxiliary data, is written into the corresponding node of the evidence storage chain to prevent the plaintext of the dynamic salt value from being leaked and used to reverse-engineer the original data. For key nodes in the evidence storage chain that involve user privacy plaintext, a chameleon hash function with trapdoors is used to calculate the digest value of the key node. The message field of the key node is the original privacy message, and the random number field is the initial random number. When a deletion request is received for the critical node, a key fragmentation submission request is initiated to multiple preset independent authorized entities respectively; When the number of submitted key fragments reaches a preset threshold, the trapdoor private key of the chameleon hash function is reconstructed using the aggregated key fragments based on the threshold secret sharing reconstruction algorithm. Based on the trapdoor private key, the original privacy message, the pre-agreed de-identification placeholder, and the initial random number, a new random number is calculated such that the digest value calculated by the chameleon hash function under the de-identification placeholder and the new random number is equal to the digest value calculated under the original privacy message and the initial random number. Replace the original privacy message and the initial random number in the key node with the desensitized placeholder and the new random number, and destroy the trapdoor private key obtained in this reconstruction after verifying that the digest value after replacement is consistent with the digest value before replacement.

2. The method according to claim 1, characterized in that, The step of concatenating the stable response, the timestamp, and the fault type code in a predetermined order and inputting them into a key derivation function to obtain a dynamic salt value includes: concatenating the stable response, the timestamp, and the fault type code sequentially to obtain a concatenated string, and inputting the concatenated string into the key derivation function to obtain a pseudo-random dynamic salt value with a fixed length.

3. The method according to claim 1, characterized in that, The auxiliary data is stored in the corresponding node of the evidence storage chain in an indexed manner, and is used by the fuzz extractor to reconstruct the stable response consistent with the current one during subsequent verification.

4. The method according to claim 1, characterized in that, The Chameleon Hash Function is constructed based on the Discrete Logarithm Problem. The public key corresponding to its trapdoor private key is obtained by taking the modulo of the generator's private key raised to the power of the trapdoor private key with a preset large prime number. For messages and random numbers, the digest value calculated by the Chameleon Hash Function is obtained by taking the modulo of the preset large prime number with the product of the generator's message power and the public key's random number power.

5. The method according to claim 4, characterized in that, The calculation of the new random number includes: modifying the initial random number based on the multiplicative inverse of the trapdoor private key at a preset cyclic group order, combined with the difference between the original privacy message and the de-identified placeholder, to obtain the new random number.

6. The method according to claim 1, characterized in that, The preset multiple independent authorized entities include the manufacturer's quality control department node, the legal and compliance department node, and the third-party notary node.

7. The method according to claim 1, characterized in that, After verifying that the digest value after replacement is consistent with the digest value before replacement, the process also includes: recording an audit log containing the operation time, authorized entity, and node identifier; and rolling back the replacement operation and triggering an alarm when the digest value after replacement is inconsistent with the digest value before replacement.

8. The method according to claim 1, characterized in that, The trapdoor private key obtained from the reconstruction is only valid within the memory lifecycle of this deletion operation and is destroyed immediately after the digest value verification is completed.

9. A hash encryption system for processing after-sales data of digital products, characterized in that, include: The after-sales service terminal includes a security chip, which is equipped with a physically unclonable function module, a trusted execution environment, and a hash operation unit for collecting the response of the physically unclonable function, generating a dynamic salt value, and calculating a hash digest. An after-sales cloud platform is communicatively connected to the after-sales service terminal. The after-sales cloud platform includes a data receiving module, a dynamic salt value management module, an evidence storage chain construction and storage module, and a trapdoor reconstruction and collision calculation module. The evidence storage chain construction and storage module is used to store an evidence storage chain composed of multiple evidence storage nodes. The digest value of the key nodes involving user privacy plaintext in the evidence storage chain is calculated by a chameleon hash function with trapdoors. Multiple independent authorized nodes each hold a key fragment of the trapdoor private key of the chameleon hash function and connect to the trapdoor reconstruction and collision calculation module via a communication connection. When a preset threshold authorization condition is met, the node submits the key fragment to the trapdoor reconstruction and collision calculation module to reconstruct the trapdoor private key and complete the replacement of the privacy plaintext with a desensitized placeholder.

10. The system according to claim 9, characterized in that, The security chip also includes a fuzz extractor for error correction processing of the raw response collected by the physically unclonable function module to obtain a stable response and auxiliary data for reconstructing the stable response.

Images