Industry short message privacy protection system, method and server

Abstract

The invention discloses an industry short message privacy protection system, and the system comprises a client side preprocessing module which is used for carrying out the metadata disturbance, mobile phone number anonymization processing and content desensitization processing of client information and short message content; the blind transmission channel module is used for a short message service provider to carry out non-decryption pipelined sending processing according to the processed customer information and short message content; and the operator end issuing module is used for restoring the client information and the short message content and then issuing the client information and the short message content. According to the system, by means of a privacy enhancement technology, including differential privacy, zero-knowledge proof and the like, the data privacy leakage risk and service compliance challenge existing in the traditional industry short message issuing process are solved, and the relationship between privacy protection and service requirements is balanced.

Description

Technical Field

[0001] The present application relates to the field of information security, and in particular to an industry SMS privacy protection system, method, and server. Background Art

[0002] Industry SMS delivery typically involves collaboration among the customer (enterprise), SMS service provider (SP), operator, and end user. The specific process is as follows: 1. Demand initiation and content preparation: Enterprises generate SMS content based on business needs (such as verification codes and marketing notifications) and submit it to the SMS service provider (SP) through the API or management platform. The content includes the SMS template, target number list, delivery time requirements (immediate or scheduled), and priority identification. The content must comply with the operator's template specifications and avoid sensitive words or illegal information.

[0003] 2. Content Review and Template Matching: SMS service providers filter SMS content for sensitive words, review template compliance (e.g., signature, content format), and match pre-approved templates. If any content violates regulations, the message will be returned to the customer for revision.

[0004] 3. Channel selection and request delivery: The SMS service provider selects the appropriate SMS channel based on the target user's carrier (China Mobile, China Unicom, or China Telecom) and forwards the request to the carrier's gateway. Some service providers dynamically prioritize channels to improve delivery rates.

[0005] 4. Operator Delivery and Status Feedback: After receiving the SMS request from the SMS service provider, the operator re-verifies and processes the SMS, for example, by checking the sender's credentials and confirming compliance with relevant regulations. Once verification is successful, the operator delivers the SMS to the end user's phone via the base station network and returns a status report (e.g., success, failure, blacklist) to the SMS service provider in real time.

[0006] 5. Data statistics and billing: SMS service providers aggregate data such as send volume and success rate, generate billing reports, and provide them to customers. Some platforms support visual dashboards.

[0007] In the SMS delivery process, SMS service providers (SPs) may collect sensitive information such as SMS content and phone number lists for analysis or service optimization. However, the misuse of this data may lead to customer privacy leaks and business security risks. For example: SMS messages contain plain text or templated content, such as verification codes, marketing copy, and order notifications. This can leak customer business logic (such as promotional strategies) and sensitive user information (such as transaction amounts and verification codes).

[0008] The customer submits a list of numbers to be sent. It is possible to analyze the user group distribution by number location and number segment (for example, high-net-worth users are concentrated in a certain area).

[0009] Sending time and frequency: such as peak sending hours and activity cycles. This can be used to infer the customer's business rhythm (such as new product release time), user activity, or churn rate.

[0010] Therefore, there is an urgent need for a technical solution for industry SMS privacy protection that can solve the above problems. Summary of the Invention

[0011] In order to address the deficiencies of the existing technology, this application provides an industry SMS privacy protection system, method and server, which, through privacy enhancement technologies (such as differential privacy and zero-knowledge proof, etc.), solves the data privacy leakage risks and business compliance challenges in the traditional industry SMS delivery process, and balances the relationship between privacy protection and business needs.

[0012] The technical effects to be achieved by this application are achieved through the following solutions: According to a first aspect of the present application, an industry SMS privacy protection system is provided, comprising: The client-side pre-processing module is used to perform metadata perturbation, mobile phone number anonymization, and content desensitization on customer information and SMS content; The blind transmission channel module is used by SMS service providers to perform pipelined sending processing without decryption based on processed customer information and SMS content; The operator-side sending module is used to restore customer information and SMS content before sending it.

[0013] Preferably, in the client-side pre-processing module, performing metadata perturbation specifically includes: After defining sensitive fields, the metadata is divided into sensitive and non-sensitive fields; The data in the sensitive domain is divided into several levels according to the sensitivity, and differentiated noise models are used to add noise on demand for data of different sensitivity levels; Keep the original values ​​for non-sensitive fields to ensure accuracy.

[0014] Preferably, the anonymization process for mobile phone numbers is as follows: Encrypt the phone number using the operator's public key, add a routing tag, generate the phone number ciphertext, and perform hash perturbation; or After the calculation is performed using the reversible desensitization process with dynamic offset, hash perturbation is performed.

[0015] Preferably, the content desensitization processing of the SMS content is specifically as follows: Add reversible noise to sensitive words or content through local dictionaries; Through SMS content fragmentation, the information involving user privacy is encrypted and fragmented using the operator's public key; The client-side preprocessing module also generates zero-knowledge proof and submits the preprocessed customer information and SMS content to the SMS service provider and blockchain.

[0016] Preferably, the specific method for the blind transmission channel module to perform pipelined sending processing without decryption is: After verifying the validity of the zero-knowledge proof, the processed metadata is sent to the blockchain to record the operation log; The SMS service provider only stores and forwards the anonymized mobile phone number and the perturbed hash value.

[0017] Preferably, the operator-side delivery module also includes a privacy budget management platform for allocating a dynamic ε value to each customer to control the balance between noise intensity and data availability; and Dynamically adjust the privacy budget based on business needs: reduce the privacy budget in high-security scenarios, sacrificing some data accuracy; increase the privacy budget in low-sensitivity scenarios to maintain data availability.

[0018] Preferably, the specific contents of the restored customer information and SMS content are: Metadata denoising: Based on the Laplace noise distribution model, the most likely time point is calculated backwards from the generalized time interval; Mobile phone number decryption: Combine the client-side random response probability p and the reverse calculation of the operator's key or dynamic offset, filter the virtual hash value, and restore the real mobile phone number; SMS content decryption: Use the operator's private key to decrypt the content fragments and reverse the reversible noise to form the real SMS content.

[0019] Preferably, the specific steps for the operator-side sending module to send are: The operator sends the restored SMS content to the end user through the standard channel; After the operator completes the SMS delivery, it will upload the actual sending volume to the blockchain for storage.

[0020] According to a second aspect of the present application, there is provided an industry SMS privacy protection method using the above-mentioned industry SMS privacy protection system, comprising the following steps: Step 1: After the client system generates the plaintext SMS content, metadata perturbation, phone number anonymization, and content desensitization are performed on the client information and SMS content. After generating a zero-knowledge proof, the content is output and submitted to the SMS service provider and blockchain. Step 2: The SMS service provider receives the output content and verifies the validity of the zero-knowledge proof. After verification, it forwards the encrypted content to the operator. Step 3: After the operator decrypts the SMS content and customer information, it sends it to the corresponding customer through the SMS channel.

[0021] According to a third aspect of the present application, there is provided a server, comprising: a memory and at least one processor; The memory stores a computer program, and the at least one processor executes the computer program stored in the memory to implement the above-mentioned industry SMS privacy protection method.

[0022] According to one embodiment of the present application, the beneficial effects of using the SMS privacy protection system of this industry are: Perturb metadata on the client side to protect data privacy and security from the source; By combining differential privacy technology, SMS service providers can provide reliable data statistics and analysis services while protecting user privacy.

[0023] Protect individual privacy: prevent attackers from inferring specific user sending behaviors through log records; Provide quantifiable privacy guarantees: mathematical methods are used to ensure that the degree of privacy leakage is controllable; Preventing data leakage and misuse: Reduce the risk of internal and external data leakage; Regulatory compliance and trust building: meeting legal and regulatory requirements and enhancing the trust of customers and regulators; Balance between data utility and privacy: Maintaining data utility while protecting privacy and supporting multi-dimensional data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the existing technical solutions, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0025] Figure 1 This is a structural block diagram of an industry SMS privacy protection system in one embodiment of the present application; Figure 2 This is a flow chart of an industry SMS privacy protection method in one embodiment of the present application; Figure 3 This is a structural block diagram of a server in an embodiment of the present application. DETAILED DESCRIPTION

[0026] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0027] like Figure 1 As shown, the industry SMS privacy protection system in one embodiment of the present application includes: The client-side pre-processing module is used to perform metadata perturbation, mobile phone number anonymization, and content desensitization on customer information and SMS content; This module uses metadata perturbation, phone number anonymization, and content desensitization to prevent inferring user behavior through sending patterns. Specifically: Perform metadata LDP perturbation and define metadata fields such as sending time distribution, user geographic location, and sending frequency as sensitive fields. Based on the sensitive fields, metadata is divided into sensitive domains (such as sending time distribution and user geographic location) and non-sensitive domains (such as total sending volume, which can be divided according to needs). Sensitive domain data is divided into different levels of data based on sensitivity. Differentiated noise models (exponential mechanism or Laplace mechanism) are used for data of different sensitivities (time, frequency) to optimize the privacy-utility trade-off.

[0028] The differential privacy budget (ε) is dynamically allocated according to the customer's privacy level. At the same time, the customer only adds Laplace noise to the sensitive domain, and the non-sensitive domain maintains the original value to ensure accuracy.

[0029] The local perturbation algorithm is as follows: Time generalization: The exact sending time (e.g., 14:05:30) is generalized to a time interval (14:00-15:00) using the differential privacy index mechanism. Frequency noise injection: Using the Laplace mechanism, random noise that follows the Laplace distribution is added to the exact query results, making it impossible for attackers to obtain the original accurate information through data analysis, thereby achieving differential privacy protection.

[0030] If the actual transmission amount N is changed to N+Lap(Δf / ε); Where: ε is the privacy budget, Δf is the sensitivity of function f, and Lap(Δf / ε) is random noise that follows a Laplace distribution with a scale parameter of Δf / ε.

[0031] The specific methods for anonymizing mobile phone numbers are as follows: Anonymize the mobile phone number: One method is to use the operator's public key to encrypt the mobile phone number to generate the mobile phone number ciphertext; Another method is to reversibly desensitize mobile phone numbers based on dynamic offsets. First, the mobile phone number is segmented, either continuously into two segments or spaced into multiple segments. One or more segments are selected to retain the plaintext. The remaining n digits are combined to calculate a hash value. The last n digits of the hash value are used as the offset seed S. The n-digit numbers to be desensitized are then desensitized sequentially, with the offset being (the number at the position corresponding to S + the index) modulo 10. Finally, the transformed n-digit numbers are converted to a zero-width character sequence and inserted into the desensitized string. A specific example is shown below: Divide the mobile phone number 15612345678 into three parts: A (first three digits) 156; B (middle four digits) 1234; C (last four digits) 5678; Generate dynamic offset: concatenate A and C to get the string 1565678 and calculate its hash value. After converting to decimal, take the last four digits as the offset seed S (assuming S = 3457). Apply dynamic offset: Based on the offset formula: new number = (original number + S corresponding position number + S corresponding position index) % 10, offset each of the middle 4 digits B=1234 to get 4704; Embedding stego information: Convert the offset middle 4 bits 4704 into a zero-width character sequence (e.g., U+200B represents 0, U+200C represents 1, using binary encoding). 4704 → binary 0100 0111 0000 0100 → zero-width string [U+200C][U+200B][U+200B][U+200C]; Embed the zero-width string to generate a desensitized mobile phone number string.

[0032] Then, a real hash value is generated, i.e., H_real = SHA256(phone number + salt value). The salt value is pre-shared between the customer and the operator. One way is for the operator to pre-distribute it securely and the customer to store it through a hardware encryption module. Another way is to use a dynamic salt value and change it regularly to prevent long-term data leakage. A random response mechanism is used to send it to the SMS service provider. That is, after calculating the hash value of the mobile phone number, H_real is sent with probability p, and a random virtual hash H_dummy is sent with probability 1-p, which meets the ε-LDP requirements, that is, meets the "privacy-utility balance" requirements.

[0033] The hash value of the mobile phone number is randomly perturbed, and a real or virtual hash is sent to protect user privacy. The SMS service provider can only see the encrypted or anonymized mobile phone number and the perturbed hash, but cannot obtain the real information. The operator decrypts the mobile phone number, recalculates the hash (using the customer's salt value), and compares it with the received hash. If it matches, it is sent, otherwise it is discarded.

[0034] In this module, if the SMS service provider needs a mobile phone number to determine the route and select the SMS channel, the following two methods can be used: Add a routing tag. The routing tag does not contain user-level sensitive information (such as the full location), but only identifies the operator and region code. For example, a China Mobile phone number starts with CMCC_, followed by the region code; a China Unicom phone number starts with CUCC_, also followed by the region code; and a China Telecom phone number starts with CTCC_, followed by the region code.

[0035] Exposing the number prefix (public number segment information). SMS service providers can only obtain the hash value or encrypted and desensitized content and cannot reversely associate users.

[0036] The specific methods for desensitizing SMS content are as follows: Reversible noise is added to sensitive words or content through the local dictionary. That is, the noise added to the data can be restored through reverse feedback or a key to restore the original data, such as generating a reversible perturbation tag [NOISE:K1], where K1 is the pre-shared key index.

[0037] Then the content is fragmented and encrypted, that is, the SMS content is fragmented, and the information involving user privacy is encrypted and fragmented using the operator's public key, ensuring that the SMS service provider cannot parse the complete semantics but can review the basic content.

[0038] This module also generates a zero-knowledge proof, proving that the noise-added sensitive domain data satisfies the noise scale parameter b = Δ / ε. Δ is the sensitivity, which represents the maximum impact of a change in any record in the dataset on the query result. ε: privacy budget, which controls the strength of privacy protection; b=Δ / ε: The scale parameter of the Laplace distribution. The noise scale parameter b determines the size of the added noise.

[0039] The customer submits the noisy metadata hash H(D'') and the zero-knowledge proof to a blockchain (such as Hyperledger Fabric) to ensure that the data cannot be tampered with. The customer can publicly audit whether the SMS service provider has used the data in violation of regulations.

[0040] This module outputs perturbed metadata (such as generalization time, anonymized hashed mobile phone number) and desensitized and encrypted SMS content, and transmits it to the SMS service provider. It also inserts virtual requests based on a random response mechanism to make the real hash distribution indistinguishable, interfering with the statistical inference of the real sending frequency.

[0041] The blind transmission channel module is used by SMS service providers to perform pipelined sending processing without decryption based on processed customer information and SMS content; In this module, after receiving the perturbed metadata and desensitized and encrypted SMS content, the validity of the zero-knowledge proof is first verified. The SMS service provider needs to verify that differential privacy has been correctly applied to the metadata submitted by the customer, but does not need to know the original data. Specifically: When verifying zero-knowledge proofs, we use the zk-SNARKs protocol to verify that the following relationship holds: Noise∼Lap(0,Δ / ε) ∧ D S ''=D S +Noise ; Where: D S For the original sensitive data; D S '' is the sensitive domain data after noise addition, which is the sensitive data set after processing (noise addition); Noise is the added noise, which obeys Lap(0, Δ / ε); Lap(0, Δ / ε) is the Laplace distribution, centered at 0, with a scale parameter of b=Δ / ε, which meets the ε-DP requirement; ∧: logical AND, requiring that both the noise conforms to the distribution and the noise addition calculation is correct.

[0042] Noise~Lap(0,Δ / ε) means that the noise is sampled from a Laplace distribution with mean 0 and scale parameter Δ / ε. The ~ symbol means "obey a certain distribution", that is, the noise obeys a Laplace distribution with specific parameters. S ''=D S +Noise represents the sensitive domain data D after adding noise S '' is equal to the original sensitive data D S Add noise sampled from a Laplace distribution.

[0043] When using, input: original data D S , noise scale parameter b = Δ / ε. Sample noise Noise from Laplace distribution Lap(0,Δ / ε). Calculate the noised data D S ''=D S +Noise. Use the zk-SNARKs protocol to generate a zero-knowledge proof to prove that Noise is indeed sampled from Lap(0,Δ / ε) and D S''=D S +Noise established.

[0044] Output: Noised data D S '' and zero-knowledge proof π.

[0045] SMS service provider verification proof: Input: Noised data D S '', Δ, ε, zero-knowledge proof π.

[0046] Use the zk-SNARKs protocol to verify the zero-knowledge proof π; use the above formula for verification, if the verification passes, then accept D S '' is data that satisfies ε-differential privacy; otherwise, it is rejected.

[0047] After the SMS service provider processes the metadata, the operation log is recorded on the chain, and customers can trace the compliance of data usage.

[0048] This module only stores and forwards the encrypted phone number and the perturbed hash value. It cannot retrieve the actual phone number hash (the random response mechanism makes it impossible to associate the user). Therefore, the content is unreadable. The encrypted fragments require the operator's private key to decrypt. The SMS service provider only stores and forwards the ciphertext, greatly improving information security.

[0049] The operator-side sending module is used to restore customer information and SMS content before sending it.

[0050] This module also includes a privacy budget management platform, which assigns a dynamic ε value to each customer (such as through a blockchain smart contract) to control the balance between noise intensity and data availability. Customers and operators negotiate ε through smart contracts. The budget consumed by each query is recorded on the chain to prevent privacy budget exhaustion attacks while ensuring that budget consumption is transparent and traceable.

[0051] Dynamically adjust the privacy budget ε based on business needs, for example: In high-security scenarios (such as financial SMS), a smaller ε (strong privacy protection) is used, sacrificing some data accuracy. Use a larger ε in low-sensitivity scenarios (such as marketing SMS statistics) to maintain data availability.

[0052] Then, metadata denoising (i.e., restoring metadata such as time and phone number) and SMS content restoration are performed, specifically: Time interval convergence: Based on the Laplace noise distribution model, the most likely time point is calculated from the generalized time interval. Time interval convergence is used to reverse the time after the metadata is noisy to recover the metadata.

[0053] Mobile phone number hash reverse calculation: Reverse calculation based on operator keys or dynamic offsets to restore the real mobile phone number and filter virtual hash values; The encrypted mobile phone number is restored by decrypting the ciphertext with the private key to obtain the plaintext mobile phone number; The mobile phone number recovery process based on dynamic offset is as follows: first parse the zero-width characters from the desensitized string to restore the transformed n-digit number, then recalculate the hash value based on the plaintext number, take the last n digits of the hash value to obtain the same offset seed S, and perform reverse offset calculation on each digit to obtain the original digit, the original digit = (transformed digit - S corresponding position digit - S corresponding position index) % 10, and combine and splice to obtain the original complete mobile phone number.

[0054] Recalculate the hash for dynamic hash verification: H_calc=SHA256(plaintext phone number + salt value) If H_calc is equal to the hash value, it is determined to be a real request and the SMS is allowed to be sent. If H_calc is not equal to the hash value, it is determined to be a virtual request and discarded.

[0055] Decrypting the text message content is as follows: Use the operator's private key to decrypt the SMS content fragments to obtain the plaintext content containing the disturbance mark; The reversible noise is then reversed to form the actual SMS content. For example, [NOISE: K1] is replaced with the embedded perturbation token. K1 is used to extract the verification code from the client's pre-shared key library and then replaced with the content to obtain the final plaintext data. This completes the decryption of the SMS content.

[0056] After all the content is decrypted and restored, a text message will be sent to the corresponding user. The specific method is as follows: The operator sends the restored SMS content to the end user through the standard channel; After the operator completes the SMS delivery, it will upload the actual sending volume and other information to the blockchain to ensure that the real data can be traced. To a certain extent, the SMS service provider can bill the real data based on this and the real data can be protected.

[0057] The beneficial effects of using this system to send SMS content are: Content Desensitization and Differential Privacy: This technology implements differential privacy perturbation of sensitive metadata and implements "desensitization followed by encryption" for data. Reversible noise markers (such as specific position replacement symbols) are embedded in SMS content and superimposed with lightweight sharding encryption (such as the SM4 national encryption algorithm), doubly blocking SMS service providers from accessing the original data. Mobile phone number anonymization: Encrypt or dynamically shift the mobile phone number, desensitize it, and hash it to protect user privacy. At the same time, virtual hash requests are inserted to interfere with statistical inferences about the actual sending frequency. Zero-knowledge proof (ZKP) verification, which verifies privacy-preserving computing: The customer generates a ZKP to prove to the SMS service provider that the data they submitted has been correctly desensitized (e.g., the hash value corresponds to the real user), but the SMS service provider cannot obtain the original information; Blockchain-based privacy budget management: Customers and operators negotiate ε through smart contracts. Each query consumes the budget and is recorded on-chain to prevent privacy budget exhaustion attacks. Hierarchical and layered dynamic noise injection breaks through traditional global noise addition and achieves a balance between privacy and utility. Blockchain audit mechanism: The client records the hash value of the request sent to the blockchain. SP operations (such as forwarding and analysis) must be authorized by the smart contract, and violations can be traced; By deeply integrating differential privacy with lightweight encryption and dynamic budget management, we provide SMS service providers with a patentable privacy-enhancing solution. In the future, with the further integration of federated learning and differential privacy, industry SMS services can improve routing efficiency and anti-fraud capabilities while protecting user privacy.

[0058] like Figure 2 As shown, the industry SMS privacy protection method using the above industry SMS privacy protection system includes the following steps: Step 1: After the client system generates the plaintext SMS content, metadata perturbation, phone number anonymization, and content desensitization are performed on the client information and SMS content. After generating a zero-knowledge proof, the content is output and submitted to the SMS service provider and blockchain. In this step, after the client system generates the plaintext of the SMS content, it divides the metadata into sensitive and non-sensitive fields. After querying the blockchain for the consumption of the privacy budget ε, it performs dynamic noise addition on the sensitive fields. Add reversible noise marking and lightweight sharding encryption, and anonymize the mobile phone number; send it in a random response manner; and submit it to the blockchain together with zero-knowledge proof.

[0059] Step 2: The SMS service provider receives the output content and verifies the validity of the zero-knowledge proof. After verification, it forwards the encrypted content to the operator. Step 3: After the operator decrypts the SMS content and customer information, it sends it to the corresponding customer through the SMS channel.

[0060] In a specific example, for example, the zero-knowledge proof generation process on the client side mainly includes the following steps: The customer specifies their actual sending volume as N, the sending volume limit agreed upon with the SMS service provider as Limit, and also selects a random number s; Based on the sending volume N, the sending volume limit Limit, the random number s and the commitment value Com′, a zero-knowledge proof Proof′ is generated using the corresponding zero-knowledge proof algorithm. This proof can prove to the SMS service provider that the sending volume N does not exceed the limit Limit without revealing the specific value of N. The SMS service provider receives the commitment value Com′ and zero-knowledge proof Proof′ sent by the customer. The SMS service provider uses the same sending limit and verification algorithm as the customer to verify the received commitment value Com′ and zero-knowledge proof Proof′. The verification process checks the validity of the proof Proof′, confirms whether it is generated according to the correct algorithm, and whether it is consistent with the commitment value Com′ and the sending limit Limit.

[0061] In another specific example, the specific process of using the privacy budget is as follows: Query the remaining privacy budget ε_remaining based on the client's request, calculate the query sensitivity Δ to determine the privacy budget ε_used allowed to be consumed; Set the scale parameter b = Δ / ε_used to generate noise, and add noise to the real result to generate a differentially private result; Send the noisy result to the client and update the remaining budget.

[0062] like Figure 3 As shown, a server in one embodiment of the present application includes: a memory and at least one processor; The memory stores a computer program, and the at least one processor executes the computer program stored in the memory to implement the above-mentioned industry SMS privacy protection method.

[0063] According to one embodiment of the present application, the beneficial effects of using the SMS privacy protection system of this industry are: Perturb metadata on the client side to protect data privacy and security from the source; By combining differential privacy technology, SMS service providers can provide reliable data statistics and analysis services while protecting user privacy.

[0064] Protect individual privacy: prevent attackers from inferring specific user sending behaviors through log records; Provide quantifiable privacy guarantees: mathematical methods are used to ensure that the degree of privacy leakage is controllable; Preventing data leakage and misuse: Reduce the risk of internal and external data leakage; Regulatory compliance and trust building: meeting legal and regulatory requirements and enhancing the trust of customers and regulators; Data utility and privacy balance: Maintaining data utility while protecting privacy, supporting multi-dimensional data analysis, such as: Time period distribution analysis: optimize sending time periods and resource allocation; Geographical distribution: guide regional operation strategies; User behavior analysis: improve personalized push effects; Business effect: evaluate activity ROI and conversion efficiency; Group stratification: achieve precise user grouping operations; Channel monitoring: ensure service stability and availability.

[0065] It should be noted that the above detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs.

[0066] It should be noted that the terms used herein are intended only to describe specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0067] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0068] In addition, the terms "comprises" and "comprising" and any variations thereof are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, product, or apparatus.

[0069] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used herein to describe the spatial positional relationship of a device or feature to other devices or features as shown in the figures. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figures. For example, if the device in the drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be subsequently positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways, such as rotated 90 degrees or in other orientations, and the spatially relative descriptions used herein are interpreted accordingly.

[0070] In the above detailed description, reference is made to the accompanying drawings, which form a part hereof. In the drawings, similar symbols typically identify similar components, unless the context dictates otherwise. The illustrated embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be used, and other changes may be made, without departing from the spirit or scope of the subject matter presented herein.

[0071] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An industry SMS privacy protection system, characterized by: include: The client-side pre-processing module is used to perform metadata perturbation, mobile phone number anonymization, and content desensitization on customer information and SMS content; The blind transmission channel module is used by SMS service providers to perform pipelined sending processing without decryption based on processed customer information and SMS content; The operator-side sending module is used to restore customer information and SMS content before sending it.

2. The industry SMS privacy protection system according to claim 1 is characterized in that: In the client-side pre-processing module, metadata perturbation is performed specifically as follows: After defining sensitive fields, the metadata is divided into sensitive and non-sensitive fields; The data in the sensitive domain is divided into several levels according to the sensitivity, and differentiated noise models are used to add noise to the data of different sensitivity levels as needed; Keep the original values ​​for non-sensitive fields to ensure accuracy.

3. The industry SMS privacy protection system according to claim 2 is characterized in that: The specific steps for anonymizing mobile phone numbers are as follows: Encrypt the phone number using the operator's public key, add a routing tag, generate the phone number ciphertext, and perform hash perturbation; or After the calculation is performed using the reversible desensitization process with dynamic offset, hash perturbation is performed.

4. The industry SMS privacy protection system according to claim 3 is characterized in that: The content desensitization processing of SMS content is as follows: Add reversible noise to sensitive words or content through local dictionaries; Through SMS content fragmentation, the information involving user privacy is encrypted and fragmented using the operator's public key; The client-side preprocessing module also generates zero-knowledge proof and submits the preprocessed customer information and SMS content to the SMS service provider and blockchain.

5. The industry SMS privacy protection system according to claim 4 is characterized in that: The specific method for the blind transmission channel module to perform pipeline transmission processing without decryption is: After verifying the validity of the zero-knowledge proof, the processed metadata is sent to the blockchain to record the operation log; The SMS service provider only stores and forwards the anonymized mobile phone number and the perturbed hash value.

6. The industry SMS privacy protection system according to claim 5, characterized in that: The operator-side module also includes a privacy budget management platform, which is used to assign a dynamic ε value to each customer to balance noise intensity and data availability. as well as Dynamically adjust the privacy budget based on business needs: reduce the privacy budget in high-security scenarios, sacrificing some data accuracy; increase the privacy budget in low-sensitivity scenarios to maintain data availability.

7. The industry SMS privacy protection system according to claim 6, characterized in that: The specific contents of the restored customer information and SMS content are as follows: Metadata denoising: Based on the Laplace noise distribution model, the most likely time point is calculated backward from the generalized time interval; Mobile phone number decryption: Combine the client-side random response probability p and the reverse calculation of the operator's key or dynamic offset, filter the virtual hash value, and restore the real mobile phone number; SMS content decryption: Use the operator's private key to decrypt the content fragments and reverse the reversible noise to form the real SMS content.

8. The industry SMS privacy protection system according to claim 7, characterized in that: The specific steps for the operator-side delivery module to deliver are: The operator sends the restored SMS content to the end user through the standard channel; After the operator completes the SMS delivery, it will upload the actual sending volume to the blockchain for storage.

9. An industry SMS privacy protection method using the industry SMS privacy protection system according to any one of claims 1 to 8, characterized in that: The steps include: Step 1: After the client system generates the plaintext SMS content, metadata perturbation, phone number anonymization, and content desensitization are performed on the client information and SMS content. After generating a zero-knowledge proof, the content is output and submitted to the SMS service provider and blockchain. Step 2: The SMS service provider receives the output content and verifies the validity of the zero-knowledge proof. After verification, it forwards the encrypted content to the operator. Step 3: After the operator decrypts the SMS content and customer information, it sends it to the corresponding customer through the SMS channel.

10. A server, characterized in that: include: memory and at least one processor; The memory stores a computer program, and the at least one processor executes the computer program stored in the memory to implement the industry SMS privacy protection method described in claim 9.

Images