A station privacy pre-screening method based on distance threshold driving and verifiable semantic commitment

Abstract

The application provides a station site privacy preliminary screening method based on distance threshold driving and a verifiable semantic commitment, and belongs to the technical field of station site privacy. The method comprises the following steps: adapting and deriving a GeoHash parameter according to a distance threshold, constructing an adaptive neighborhood GeoHash set, generating a semantic space identifier in combination with business semantics, and generating a corresponding verifiable semantic commitment relying on a Merkle tree. The semantic commitment intersection is matched by using a key Bloom filter, and a candidate space identifier set is quickly output. The authenticity verification is completed through multi-layer verification. The overall structure adopts a verifiable semantic commitment, a key Bloom filter and a triple chain verification fusion architecture, realizes safe and efficient preliminary screening of station site space privacy, reduces the matching complexity of massive position data and reduces false screening by relying on hash intersection operation, and guarantees that the data is tamper-proof and the result is traceable and verifiable throughout the process through triple chain verification of commitment consistency, Merkle path and digital signature, and comprehensively improves the screening safety and operation efficiency of station site privacy.

Description

Technical Field

[0001] This invention belongs to the field of site privacy technology, and in particular relates to a site privacy screening method based on distance threshold driving and verifiable semantic commitment. Background Technology

[0002] Information such as site latitude and longitude, data center status, power capacity, tower type, lease validity period, antenna height, and available frequency bands are core commercially sensitive data. Directly exchanging plaintext data will lead to a series of security issues such as data leakage, commercial snooping, and compliance risks. Against this backdrop, how to efficiently complete the proximity matching and initial screening of a "small batch query point to a static site database of millions of sites" without data leaving the domain or privacy being leaked has become a key technical bottleneck restricting the efficiency of 5G co-construction and sharing.

[0003] In existing technologies, the mainstream technologies used for geospatial privacy matching mainly include: 1. GeoHash encoding technology GeoHash is a geocoding method that recursively maps two-dimensional latitude and longitude coordinates to one-dimensional Base32 strings. Proximity locations typically share the same prefix, supporting fast range retrieval. However, fixed-precision GeoHash suffers from boundary truncation under dynamic distance threshold queries. Grid distortion in high-latitude regions leads to missed detections of neighboring points, failing to guarantee 100% recall. Furthermore, it only supports pure spatial matching, without incorporating base station service attributes, resulting in a large number of invalid candidates entering subsequent processes, significantly increasing computational and manual costs.

[0004] 2. Privacy Set Intersection (PSI) Technique Privacy set intersection can compute the intersection of two sets without exposing non-intersecting elements. Typical implementations include those based on Diffie. Herman Key Exchange (DH) Schemes include PSI (Proof-of-Sight Intrusion), OT (Occursively Transmitted), and homomorphic encryption. PSI offers provable security, but it suffers from high computational complexity, numerous communication rounds, and poor scalability. When dealing with a database of millions of sites, it exhibits hourly response times and cannot meet the demands of real-time collaboration.

[0005] 3. Bloom Filter (BF) Technology Bloom filters achieve spatially efficient membership determination through multi-hash mapping, resulting in fast retrieval speeds and low storage requirements. However, traditional Bloom filters use public hash functions, making them vulnerable to membership inference attacks, statistical analysis attacks, and forgery injection attacks. They are unable to defend against malicious actors spying on site distribution patterns, and their security fails to meet industrial-grade requirements.

[0006] 4. Conventional hashing and semantic processing techniques Simple hashes or plaintext GeoHashes are easily brute-forced and reverse-engineered using rainbow tables. They lack key binding and semantic obfuscation mechanisms and cannot protect multi-dimensional business attributes such as base station type, power capacity, and data center status, resulting in uncontrollable privacy leakage risks.

[0007] In summary, existing technologies cannot simultaneously meet the requirements of high recall, business semantic fusion, strong privacy protection, resistance to forgery attacks, and large-scale efficient expansion, making them unsuitable for the unbalanced collaborative scenario of "millions of website addresses for small batches of query points." Therefore, there is an urgent need for a website privacy initial screening method that balances security, efficiency, accuracy, and business adaptability, serving as a pre-load reduction module for high-overhead homomorphic distance fine-tuning, forming a "coarse screening" approach. The "precision verification" technology closed loop supports reliable and efficient collaboration of communication infrastructure. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a site privacy screening method based on distance threshold driving and verifiable semantic commitment.

[0009] Firstly, the method includes, GeoHash parameters are derived based on distance thresholds, and an adaptive neighborhood GeoHash set is generated based on the multi-derived GeoHash parameters. Based on an adaptive neighborhood GeoHash set, semantic spatial identifiers are generated by integrating business semantics. For each generated semantic spatial identifier, a corresponding verifiable semantic commitment is generated; Construct a keyed Bloom filter and intersect it with a verifiable semantic commitment to obtain a set of candidate semantic spatial identifiers; The authenticity of the obtained set of candidate semantic spatial identifiers is verified.

[0010] Furthermore, the process of deriving GeoHash parameters based on a distance threshold and generating an adaptive neighborhood GeoHash set based on the multi-derived GeoHash parameters specifically includes: Based on the distance threshold D, the GeoHash parameters are derived to determine the encoding length L and the neighborhood expansion strategy. Based on the determined encoding length L and the neighborhood expansion strategy, an adaptive neighborhood GeoHash set is generated.

[0011] Furthermore, the generation of semantic spatial identifiers based on the adaptive neighborhood GeoHash set and by integrating business semantics specifically includes: Extract the site address business attributes and normalize them to generate normalized business attributes. The normalized business attributes are bit-compressed to generate a semantic digest. A session key is generated using the SM2 elliptic curve Diffie-Hellman key negotiation protocol; Using national cryptographic HMAC SM3, as a pseudo-random function for session key binding, concatenates each GeoHash code in the adaptive neighborhood GeoHash set with a semantic digest, and substitutes it into the pseudo-random function for calculation to obtain a semantic spatial identifier.

[0012] Furthermore, the step of generating a corresponding verifiable semantic commitment for each generated semantic spatial identifier specifically includes using the generated verifiable cryptographic commitment and constructing a Merkle tree structure to generate a verifiable semantic commitment for each semantic spatial identifier and constructing a chain-like verification system.

[0013] Furthermore, the generation of verifiable cryptographic commitments specifically includes, Generate a cryptographically secure random number for each semantic spatial identifier; A hash algorithm is used to concatenate semantic spatial identifiers with cryptographically secure random numbers to calculate a hash value, thereby obtaining a verifiable cryptographic commitment.

[0014] Furthermore, the construction of the keyed Bloom filter specifically includes, Construct a corresponding number of independent pseudo-random hash functions based on the session key; Initialize a bitmap with a preset length of bits; For each semantic space identifier, it is mapped to the corresponding number of positions in the bitmap through a corresponding number of pseudo-random hash functions, and the corresponding bit position is set to 1, thus completing the construction of the keyed Bloom filter.

[0015] Furthermore, the authenticity verification of the acquired candidate semantic spatial identifier set specifically includes a triple-chain authenticity verification to ensure the credibility of the final result; wherein the triple-chain authenticity verification includes commitment consistency verification, Merkle path verification, and digital signature verification.

[0016] Secondly, the system includes: a GeoHash set generation module, an identifier generation module, a semantic commitment generation module, a candidate set generation module, and a verification module; The GeoHash set generation module is used to derive GeoHash parameters based on a distance threshold and generate an adaptive neighborhood GeoHash set. The identifier generation module is used to generate semantic spatial identifiers based on an adaptive neighborhood GeoHash set and by integrating business semantics. The semantic commitment generation module is used to generate a corresponding verifiable semantic commitment for each generated semantic spatial identifier; The candidate set generation module is used to construct a keyed Bloom filter and intersect it with a verifiable semantic commitment to obtain a candidate semantic spatial identifier set; The verification module is used to verify the authenticity of the acquired set of candidate semantic spatial identifiers.

[0017] Thirdly, a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of any of the above-described site privacy screening methods based on distance threshold driving and verifiable semantic commitment.

[0018] Fourthly, an electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. Memory, used to store computer programs; When the processor executes a program stored in memory, it implements any of the steps described above for the site privacy initial screening method based on distance threshold driving and verifiable semantic commitment. Compared with the prior art, the present invention has the following advantages: 1. This invention employs a combined technical architecture of verifiable semantic commitment, keyed Bloom filter, and triple-chain verification to construct a fully traceable and verifiable site privacy screening system. Verifiable semantic commitments are generated through a Merkle tree structure, combined with a keyed Bloom filter to achieve efficient intersection calculation, quickly filtering candidate sets, significantly reducing the computational complexity of matching massive amounts of data, and filtering out false positives. The triple-chain verification mechanism verifies the authenticity and integrity of data layer by layer through commitment consistency verification, Merkle path verification, and digital signature verification, preventing data tampering and result falsification, and ensuring the credibility and verifiability of the initial screening results.

[0019] 2. By adaptively driving GeoHash parameter optimization through distance threshold, the distance threshold D can be flexibly set according to actual business needs, automatically derive the optimal encoding length L and neighborhood expansion strategy, and accurately generate an adaptive neighborhood GeoHash set, breaking through the technical bottlenecks of poor adaptability and insufficient spatial coverage accuracy of traditional fixed-length GeoHash encoding.

[0020] 3. This invention integrates site business attributes with national cryptographic encryption technology to generate semantic spatial identifiers. By normalizing business attributes and compressing bits to generate semantic digests, and combining SM2 key negotiation and HMAC-SM3 pseudo-random functions, spatial coding and business semantics are deeply bound together. This achieves an integrated encrypted expression of spatial information and business characteristics, effectively preventing spatial identifiers from being tampered with, forged, or reverse-engineered. It achieves dual privacy protection of business semantics and spatial location during data interaction and storage, solving the security risks of exposed business attributes and easily leaked location information in traditional site screening, and meeting the compliance requirements of data security and personal information protection.

[0021] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 The diagram illustrates a preliminary site privacy screening method based on distance threshold driving and verifiable semantic commitment according to the present invention.

[0024] Figure 2 The diagram shows a schematic of a site privacy screening system module based on distance threshold driving and verifiable semantic commitment according to the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] like Figure 1As shown, this invention proposes a site privacy screening method for communication base station sharing based on distance threshold driving and verifiable semantic commitment. It adaptively drives GeoHash parameter selection with a distance threshold D, integrates business semantics to generate irreversible semantic spatial identifiers, implements anti-forgery protection based on verifiable commitments and Merkle trees, and achieves secure and efficient intersection through a keyed Bloom filter. The entire process is unidirectional and non-interactive, with data remaining within its domain and privacy not leaked. The steps include… S1. Derive GeoHash parameters based on the distance threshold D and generate an adaptive neighborhood GeoHash set.

[0027] S1.1. Based on the distance threshold D, derive the GeoHash parameters to determine the encoding length L and the neighborhood expansion strategy.

[0028] In this embodiment, the present invention uses three times the grid width and selects the smallest integer L that satisfies 3×cell_width(L)≥D to ensure that the central grid is sufficient to cover the core area at a distance of threshold D; where cell_width(L) is the east-west width of the GeoHash grid corresponding to the encoding length L.

[0029] Optionally, the distance threshold D is customized by the business scenario, with typical values ​​including 100 meters, 300 meters, 500 meters, and 1000 meters, to meet the needs of all scenarios such as 5G base station planning, dense urban area coordination, and wide-area coverage.

[0030] In this embodiment, the GeoHash encoding length L is also strictly correlated with the grid precision; the longer L is, the smaller the grid and the higher the precision, but the number of codes increases and the retrieval efficiency decreases; the shorter L is, the larger the grid and the lower the precision, but the coverage is wide and the efficiency is high. Therefore, this invention achieves the optimal balance between precision and efficiency by inversely constraining the GeoHash encoding length L through the distance threshold D, without the need for manual parameter tuning, thus avoiding subjectivity and instability. For example, when D=300 meters, the cell_width corresponding to L=7 is ≈153 meters, and 3×153=459 meters ≥300 meters, which meets the coverage requirement; when D=500 meters, the cell_width corresponding to L=6 is ≈1253 meters, and 3×1253=3759 meters ≥500 meters, which also meets the coverage requirement.

[0031] In this embodiment, the present invention also employs a single-layer neighborhood expansion (9-grid coverage), namely a central GeoHash grid plus eight adjacent grids: East, West, South, North, Southeast, Southwest, Northeast, and Northwest. This strategy is mathematically complementary to the parameter rule of 3×cell_width(L)≥D. The central grid covers the core area at the distance threshold, while the eight neighboring grids cover the boundary areas. The combination of these two ensures that any site falling within a circular area centered on the query point and with a radius of D will necessarily match at least one GeoHash code, theoretically eliminating the possibility of missed detections and meeting the stringent accuracy requirements for initial screening of communication sites.

[0032] S1.2. Based on the determined encoding length L and the neighborhood expansion strategy, generate an adaptive neighborhood GeoHash set. The steps include: a. Obtain the original two-dimensional latitude and longitude coordinates of the station site ( , The coordinate accuracy is retained to 6 decimal places, which meets the positioning accuracy requirements of communication station sites.

[0033] b. Using the encoding length L determined in step S1, perform standard Base32 GeoHash encoding on the latitude and longitude to obtain the center GeoHash encoding. .

[0034] c. Based on The neighborhood codes in eight directions are calculated according to the GeoHash standard neighborhood algorithm to ensure the continuity of the codes and spatial proximity.

[0035] d. Merge the central code with the 8 neighbor codes, remove duplicates, and form an adaptive neighborhood GeoHash set G={ , , ..., The set size is fixed at 9, with no redundancy, no missing elements, and no duplicates.

[0036] In this embodiment, to address the GeoHash grid distortion problem in high-latitude regions, the present invention automatically offsets the distortion effect through a dual mechanism of 3 times grid width constraint + 9-grid neighborhood coverage. No additional correction algorithm or latitude compensation parameter is required, and it can maintain a stable recall rate globally, adapting to communication site scenarios in all provinces, all terrains, and all latitudes in China.

[0037] S2. Based on the adaptive neighborhood GeoHash set, semantic spatial identifiers are generated by integrating business semantics.

[0038] In this embodiment, the present invention deeply integrates multi-dimensional business semantics into the privacy screening process. By using a compact semantic digest and a key-bound pseudo-random function, a semantic spatial identifier sg that simultaneously carries spatial and business features is generated, thereby achieving dual privacy protection matching of "space + business", filtering invalid candidates from the source and reducing invalid computation.

[0039] S2.1 Extract the site address business attributes and normalize the site address business attributes to generate normalized business attributes.

[0040] S2.1.1 Extract the core business fields for each site. The core business fields include: base station type (macro base station / indoor distribution / micro base station), power capacity (available power), equipment room status (available / unavailable / under renovation), tower type (single-tube / triple-tube / angle steel / landscape), lease validity period (remaining duration), available frequency band (NR700 / NR2.6G / NR3.5G, etc.), and installation height (antenna mounting height).

[0041] S2.1.2. Discretize, normalize, and bit-code all fields.

[0042] In this embodiment, the present invention converts continuous values ​​into discrete labels, text values ​​into numerical codes, and multi-state values ​​into bit identifiers, ensuring that all business attributes can be compactly encoded and retain complete distinguishability; the normalized business attributes can accurately reflect whether the site meets the sharing conditions, providing a reliable basis for subsequent business matching.

[0043] S2.2. Perform bit compression on the normalized business attributes to generate a semantic digest (Digest).

[0044] In this embodiment, the present invention performs bit compression on the normalized business attributes to generate a fixed 4-byte (32-bit) compact semantic digest. This digest retains all the distinguishing information required for business matching with a very small amount of data, which avoids the risk of leakage caused by information redundancy and ensures the accuracy of business matching. The 4-byte digest has a very high compression ratio and expressive power, which can cover all core business attributes of the communication site, and its fixed length facilitates subsequent cryptographic calculations.

[0045] Optionally, the design of semantic summaries follows the principle of minimum information disclosure, retaining only the distinguishing features necessary for matching and eliminating irrelevant details to further reduce the risk of privacy leakage.

[0046] S2.3. Generate session key K using the SM2 elliptic curve Diffie-Hellman (ECDH) key negotiation protocol.

[0047] S2.4, Using the national cryptographic HMAC SM3, as a session key binding pseudo-random function (PRF), concatenates each encoded gh in the adaptive neighborhood GeoHash set with a semantic digest (Digest), substitutes it into the pseudo-random function for calculation, and obtains the semantic spatial identifier sg, expressed by the formula:

[0048] in, It is a key-bound, irreversible, statistically random identifier. Without the session key K, it is impossible to deduce any business information from sg.

[0049] S3. For each generated semantic space identifier, generate a corresponding verifiable semantic commitment.

[0050] In this embodiment, the present invention introduces verifiable cryptographic commitments and Merkle tree structures into the site privacy screening process, generates verifiable commitments for each semantic spatial identifier sg, constructs a chain verification system, and achieves the industrial-grade trust requirements of "data that is not forged, cannot be tampered with, is traceable, and is auditable".

[0051] In this embodiment, the present invention generates verifiable cryptographic commitments, the steps of which include: a. Generate a cryptographically secure random number r for each sg; where r is generated by a true random number generator with a length of not less than 128 bits to ensure unpredictability and non-repeatability; b. Using the national cryptographic SM3 hash algorithm, concatenate sg and r to calculate the hash value, obtaining the verifiable cryptographic commitment C, which represents...

[0052] Among them, the verifiable cryptographic commitment C realizes a strong cryptographic binding between sg and r, which cannot be separated, tampered with, or forged.

[0053] Alternatively, the verifiable cryptographic commitment C is publicly transmittable, while sg and r are kept secret locally by the holder and disclosed only as needed during candidate verification, thus ensuring both privacy and verifiability.

[0054] In this embodiment, to achieve efficient verification and global integrity assurance of large-scale site commitments, the present invention also constructs a Merkle tree with verifiable cryptographic commitment C as leaf nodes and digitally signs the root hash, forming a lightweight and scalable verification system. The steps include: a. Use the verifiable cryptographic commitments C corresponding to all site addresses as leaf nodes, and construct a binary Merkle tree in sequence. If the number of leaf nodes is insufficient, pad with zeros to ensure the integrity of the tree structure. b. Calculate the hash value of the parent node layer by layer from bottom to top. The value of the parent node is the SM3 hash result of concatenating the values ​​of the left and right child nodes, and finally obtain the unique Merkle root. c. The site holder uses its own SM2 private key to digitally sign the Merkle root, obtaining the signature result Sign(Root). The root and Sign(Root) are then made public as a verification benchmark.

[0055] S4. Construct a keyed Bloom filter (KBF) and intersect it with the verifiable semantic commitment to obtain a set of candidate sg.

[0056] S4.1 Construct a keyed Bloom filter (KBF).

[0057] S4.1.1 Construct k independent pseudo-random hash functions based on the session key K; where the hash function is represented as,

[0058] Where k is the number of hash functions, determined by the Bloom filter false positive rate formula, typically ranging from 8 to 16, to ensure that the false positive rate is below the acceptable threshold for engineering.

[0059] S4.1.2 Initialize a bitmap with a length of m bits; where m is determined by the site size and false alarm rate, and typically takes the value of 100MB~200MB for millions of sites, resulting in very small storage space.

[0060] S4.1.3 For each sg, k hash functions are used to map it to k positions in the bitmap, and the corresponding bit positions are set to 1 to complete the construction of the keyed Bloom filter KBF.

[0061] S4.2. Based on the constructed keyed Bloom filter, calculate secure communication and obtain the candidate SG set.

[0062] S4.2.1 The site holder and the querying party each construct their own KBF bitmap; among them, the separately constructed KBF bitmaps only exchange bitmap data and do not transmit any sensitive information such as sg, gh, Digest, raw data, etc. S4.2.2 Both parties perform a bitwise AND operation on the two KBF bitmaps locally to obtain the intersection bitmap; S4.2.3. Based on the reverse mapping of the intersection bitmap, the candidate SG set is obtained. The candidate set only contains potential sites that meet the requirements of spatial proximity and business matching. Its size is much smaller than the original site library, which greatly reduces the load for subsequent verification.

[0063] S5. Verify the authenticity of the obtained candidate sg set.

[0064] In this embodiment, the present invention employs a triple-chain authenticity verification method. Candidates that pass all verifications can proceed to the subsequent homomorphic distance precise verification stage, ensuring the reliability of the final result. This method includes: 1. Commitment Consistency Verification: The query requests verification materials (random number r, commitment C) corresponding to the candidate sg from the site address holder, and calculates them locally. The result is compared with C; if the result is exactly the same as C, the verification is successful, proving that sg has not been tampered with and is consistent with the commitment; if they are not equal, it is determined to be an invalid candidate and is directly eliminated.

[0065] 2. Merkle Path Verification: The site holder provides the Merkle proof path Proof corresponding to the candidate sg. The querying party calculates upwards from the leaf node based on the Proof to verify whether the public Merkle root can be derived. If the calculated root is completely consistent with the public root, the verification passes, proving that the commitment belongs to a valid Merkle tree and has not been forged. If they are inconsistent, it is determined to be an invalid candidate and is directly removed.

[0066] 3. Digital Signature Verification: The querying party uses the site holder's public SM2 key to verify the digital signature of the Merkle root, Sign (Root). If the signature is valid, the verification passes, proving that the Merkle root is authentic, tamper-proof, and traceable; if it is invalid, it is deemed an invalid candidate and is directly eliminated.

[0067] like Figure 2 As shown, the present invention also proposes a site privacy screening system based on distance threshold driving and verifiable semantic commitment, which includes a GeoHash set generation module, an identifier generation module, a semantic commitment generation module, a candidate set generation module, and a verification module.

[0068] 1. The GeoHash set generation module is used to derive GeoHash parameters based on a distance threshold D and generate an adaptive neighborhood GeoHash set, which includes: (1) Based on the distance threshold D, derive the GeoHash parameters and determine the encoding length L and neighborhood expansion strategy.

[0069] In this embodiment, the GeoHash set generation module uses three times the grid width and selects the smallest integer L that satisfies 3×cell_width(L)≥D to ensure that the central grid is sufficient to cover the core area at a distance of D; where cell_width(L) is the east-west width of the GeoHash grid corresponding to the encoding length L.

[0070] Optionally, the distance threshold D is customized by the business scenario, with typical values ​​including 100 meters, 300 meters, 500 meters, and 1000 meters, to meet the needs of all scenarios such as 5G base station planning, dense urban area coordination, and wide-area coverage.

[0071] In this embodiment, the GeoHash encoding length L is also strictly correlated with the grid precision; the longer L is, the smaller the grid and the higher the precision, but the number of codes increases and the retrieval efficiency decreases; the shorter L is, the larger the grid and the lower the precision, but the coverage is wide and the efficiency is high. Therefore, this invention achieves the optimal balance between precision and efficiency by inversely constraining the GeoHash encoding length L through the distance threshold D, without the need for manual parameter tuning, thus avoiding subjectivity and instability. For example, when D=300 meters, the cell_width corresponding to L=7 is ≈153 meters, and 3×153=459 meters ≥300 meters, which meets the coverage requirement; when D=500 meters, the cell_width corresponding to L=6 is ≈1253 meters, and 3×1253=3759 meters ≥500 meters, which also meets the coverage requirement.

[0072] In this embodiment, the GeoHash set generation module also employs a single-layer neighborhood expansion (9-grid coverage), consisting of a central GeoHash grid plus eight adjacent grids: East, West, South, North, Southeast, Southwest, Northeast, and Northwest. This strategy complements the parameter rule of 3×cell_width(L)≥D. The central grid covers the core area at the distance threshold, while the eight neighboring grids cover the boundary areas. This combination ensures that any site falling within a circular area centered on the query point and with a radius of D will match at least one GeoHash code, theoretically eliminating the possibility of missed detections and meeting the stringent accuracy requirements for initial screening of communication sites.

[0073] (2) Based on the determined encoding length L and the neighborhood expansion strategy, an adaptive neighborhood GeoHash set is generated. The steps include: a. Obtain the original two-dimensional latitude and longitude coordinates of the station site ( , The coordinate accuracy is retained to 6 decimal places, which meets the positioning accuracy requirements of communication station sites.

[0074] b. Using the encoding length L determined in step S1, perform standard Base32 GeoHash encoding on the latitude and longitude to obtain the center GeoHash encoding. .

[0075] c. Based on The neighborhood codes in eight directions are calculated according to the GeoHash standard neighborhood algorithm to ensure the continuity of the codes and spatial proximity.

[0076] d. Merge the central code with the 8 neighbor codes, remove duplicates, and form an adaptive neighborhood GeoHash set G={ , , ..., The set size is fixed at 9, with no redundancy, no missing elements, and no duplicates.

[0077] In this embodiment, to address the GeoHash grid distortion problem in high-latitude regions, the GeoHash set generation module automatically offsets the distortion effect through a dual mechanism of 3 times grid width constraint + 9-grid neighborhood coverage. No additional correction algorithm or latitude compensation parameter is required, and it can maintain a stable recall rate globally, adapting to communication site scenarios in all provinces, all terrains, and all latitudes in China.

[0078] 2. The identifier generation module is used to generate semantic spatial identifiers based on the adaptive neighborhood GeoHash set and by integrating business semantics.

[0079] In this embodiment, the identifier generation module deeply integrates multi-dimensional business semantics into the privacy screening process. By using a compact semantic digest and a key-bound pseudo-random function, it generates a semantic spatial identifier sg that simultaneously carries spatial and business features, thereby achieving dual privacy protection matching of "space + business" and filtering invalid candidates from the source, reducing invalid computation.

[0080] (1) Extract the site address business attributes and normalize the site address business attributes to generate normalized business attributes.

[0081] (1.1) Extract the core business fields for each site, including: base station type (macro station / indoor distribution / micro station), power capacity (available power), equipment room status (available / unavailable / under renovation), tower type (single pipe / triple pipe / angle steel / landscape), lease validity period (remaining duration), available frequency band (NR700 / NR2.6G / NR3.5G, etc.), and installation height (antenna mounting height).

[0082] (1.2) Discretize, normalize and bit-code all fields.

[0083] In this embodiment, the present invention converts continuous values ​​into discrete labels, text values ​​into numerical codes, and multi-state values ​​into bit identifiers, ensuring that all business attributes can be compactly encoded and retain complete distinguishability; the normalized business attributes can accurately reflect whether the site meets the sharing conditions, providing a reliable basis for subsequent business matching.

[0084] (2) The normalized business attributes are bit compressed to generate a fixed-byte compact semantic digest.

[0085] In this embodiment, the identifier generation module performs bit compression on the normalized business attributes to generate a fixed 4-byte (32-bit) compact semantic digest. This digest retains all the distinguishing information required for business matching with a very small amount of data, which avoids the risk of leakage caused by information redundancy and ensures the accuracy of business matching. The 4-byte digest has a very high compression ratio and expressive power, which can cover all core business attributes of the communication site, and its fixed length facilitates subsequent cryptographic calculations.

[0086] Optionally, the design of semantic summaries follows the principle of minimum information disclosure, retaining only the distinguishing features necessary for matching and eliminating irrelevant details to further reduce the risk of privacy leakage.

[0087] (3) Generate session key K using the SM2 elliptic curve Diffie-Hellman (ECDH) key negotiation protocol.

[0088] (4) Use national cryptographic HMAC SM3, as a session key binding pseudo-random function (PRF), concatenates each encoded gh in the adaptive neighborhood GeoHash set with a semantic digest (Digest), substitutes it into the pseudo-random function for calculation, and obtains the semantic spatial identifier sg, expressed by the formula:

[0089] in, It is a key-bound, irreversible, statistically random identifier. Without the session key K, it is impossible to deduce any business information from sg.

[0090] 3. The semantic commitment generation module is used to generate a corresponding verifiable semantic commitment for each generated semantic space identifier.

[0091] In this embodiment, the semantic commitment generation module introduces verifiable cryptographic commitments and Merkle tree structures into the site privacy screening process, generates verifiable commitments for each semantic spatial identifier sg, and constructs a chain verification system to achieve the industrial-grade trust requirements of "data that is not forged, cannot be tampered with, is traceable, and is auditable".

[0092] In this embodiment, the semantic commitment generation module generates cryptographic commitments, and the steps include: a. Generate a cryptographically secure random number r for each sg; where r is generated by a true random number generator with a length of not less than 128 bits to ensure unpredictability and non-repeatability; b. Using the national cryptographic SM3 hash algorithm, concatenate sg and r to calculate the hash value, obtaining the cryptographic commitment C, which represents,

[0093] Among them, the cryptographic commitment C implements a strong cryptographic binding between sg and r, which cannot be separated, tampered with, or forged.

[0094] Alternatively, the cryptographic commitment C is publicly transmittable, while sg and r are kept secret locally by the holder and disclosed only as needed during candidate verification, thus ensuring both privacy and verifiability.

[0095] In this embodiment, to achieve efficient verification and global integrity assurance of large-scale site commitments, the semantic commitment generation module also constructs a Merkle tree with cryptographic commitment C as the leaf node and digitally signs the root hash, forming a lightweight and scalable verification system. The steps include: a. Use the cryptographic commitments C corresponding to all site addresses as leaf nodes, and construct a binary Merkle tree in order. If the number of leaf nodes is insufficient, pad with zeros to ensure the integrity of the tree structure. b. Calculate the hash value of the parent node layer by layer from bottom to top. The value of the parent node is the SM3 hash result of concatenating the values ​​of the left and right child nodes, and finally obtain the unique Merkle root. c. The site holder uses its own SM2 private key to digitally sign the Merkle root, obtaining the signature result Sign(Root). The root and Sign(Root) are then made public as a verification benchmark.

[0096] 4. The candidate set generation module is used to construct a keyed Bloom filter (KBF) and intersect it with the verifiable semantic commitment to obtain a candidate SG set, which includes: (1) Construct a keyed Bloom filter (KBF).

[0097] a. Construct k independent pseudo-random hash functions based on the session key K; where the hash functions are represented as follows:

[0098] Where k is the number of hash functions, determined by the Bloom filter false positive rate formula, typically ranging from 8 to 16, to ensure that the false positive rate is below the acceptable threshold for engineering.

[0099] b. Initialize a bitmap of length m bits; where m is determined by the site size and false alarm rate, and typically takes the value of 100MB~200MB for sites with millions of addresses, resulting in very small storage space.

[0100] c. For each sg, map it sequentially to k positions in the bitmap using k hash functions, setting the corresponding bit position to 1, thus completing the construction of the keyed Bloom filter KBF.

[0101] (2) Calculate secure communication based on the constructed keyed Bloom filter and obtain the candidate sg set.

[0102] a. The site holder and the querying party each construct their own KBF bitmap; among them, the separately constructed KBF bitmaps only exchange bitmap data and do not transmit any sensitive information such as sg, gh, Digest, raw data, etc. b. Both parties perform a bitwise AND operation on the two KBF bitmaps locally to obtain the intersection bitmap; c. Based on the reverse mapping of the intersection bitmap, a candidate SG set is obtained; the candidate set only contains potential sites that meet the requirements of spatial proximity and business matching, and its size is much smaller than the original site library, which greatly reduces the load for subsequent verification.

[0103] 5. The verification module is used to verify the authenticity of the acquired candidate SG set.

[0104] In this embodiment, the verification module employs a triple-chain authenticity verification process. Candidates that pass all verifications can proceed to the subsequent homomorphic distance precision verification stage to ensure the reliability of the final result. This process includes: (1) Commitment consistency verification: The query direction requests the verification materials (random number r, commitment C) corresponding to the candidate sg from the site holder, and calculates them locally. The result is compared with C; if the result is exactly the same as C, the verification is successful, proving that sg has not been tampered with and is consistent with the commitment; if they are not equal, it is determined to be an invalid candidate and is directly eliminated.

[0105] (2) Merkle path verification: The site holder provides the Merkle proof path Proof corresponding to the candidate sg. The querying party calculates upward from the leaf node based on the Proof to verify whether the public Merkle root can be derived. If the calculated root is completely consistent with the public root, the verification is passed, proving that the commitment belongs to a valid Merkle tree and has not been forged. If they are inconsistent, it is determined to be an invalid candidate and is directly removed.

[0106] (3) Digital signature verification: The querying party uses the public SM2 public key of the site holder to verify the digital signature of the Merkle root, Sign (Root). If the signature is valid, the verification is successful, proving that the Merkle root is authentic, has not been tampered with, and is traceable; if it is invalid, it is determined to be an invalid candidate and is directly eliminated.

[0107] The foregoing description and accompanying drawings fully illustrate embodiments of the invention to enable those skilled in the art to practice them. Other embodiments may include structural and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Some portions and features of some embodiments may be included or substituted for portions and features of other embodiments. Embodiments of the invention are not limited to the structures described above and shown in the accompanying drawings, and various modifications and changes may be made without departing from their scope. The scope of the invention is limited only by the appended claims.

Claims

1. A station privacy triage method based on distance threshold driven and verifiable semantic commitment, characterized in that, The method includes, GeoHash parameters are derived based on distance thresholds, and an adaptive neighborhood GeoHash set is generated based on the multi-derived GeoHash parameters. Based on an adaptive neighborhood GeoHash set, semantic spatial identifiers are generated by integrating business semantics. For each generated semantic spatial identifier, a corresponding verifiable semantic commitment is generated; Construct a keyed Bloom filter and intersect it with a verifiable semantic commitment to obtain a set of candidate semantic spatial identifiers; The authenticity of the obtained set of candidate semantic spatial identifiers is verified. 2.The station privacy screening method based on distance threshold driving and verifiable semantic commitment according to claim 1, wherein, The process of deriving GeoHash parameters based on distance thresholds and generating an adaptive neighborhood GeoHash set based on the multi-derived GeoHash parameters specifically includes: Based on the distance threshold D, the GeoHash parameters are derived to determine the encoding length L and the neighborhood expansion strategy. Based on the determined encoding length L and the neighborhood expansion strategy, an adaptive neighborhood GeoHash set is generated. 3.The method of claim 1, wherein, The process of generating semantic spatial identifiers based on an adaptive neighborhood GeoHash set and integrating business semantics specifically includes: Extract the site address business attributes and normalize them to generate normalized business attributes. The normalized business attributes are bit-compressed to generate a semantic digest. A session key is generated using the SM2 elliptic curve Diffie-Hellman key negotiation protocol; Adopting national secret HMAC SM3 is used as a session key binding pseudo-random function. Each GeoHash code in the adaptive neighborhood GeoHash set is concatenated with the semantic digest, and the pseudo-random function is calculated to obtain the semantic space identifier.

4. The method of claim 1, wherein, The step of generating a corresponding verifiable semantic commitment for each generated semantic spatial identifier specifically includes using the generated verifiable cryptographic commitment and constructing a Merkle tree structure to generate a verifiable semantic commitment for each semantic spatial identifier and constructing a chain-like verification system.

5. The site privacy screening method based on distance threshold driving and verifiable semantic commitment as described in claim 4, characterized in that, The generation of verifiable cryptographic commitments specifically includes, Generate a cryptographically secure random number for each semantic spatial identifier; A hash algorithm is used to concatenate semantic spatial identifiers with cryptographically secure random numbers to calculate a hash value, thereby obtaining a verifiable cryptographic commitment.

6. The method of claim 1, wherein, The construction of the keyed Bloom filter specifically includes: Construct a corresponding number of independent pseudo-random hash functions based on the session key; Initialize a bitmap with a preset length of bits; For each semantic space identifier, it is mapped to the corresponding number of positions in the bitmap through a corresponding number of pseudo-random hash functions, and the corresponding bit position is set to 1, thus completing the construction of the keyed Bloom filter.

7. The method of claim 1, wherein, The process of verifying the authenticity of the acquired set of candidate semantic spatial identifiers specifically includes employing a triple-chain authenticity verification to ensure the credibility of the final result; wherein the triple-chain authenticity verification includes commitment consistency verification, Merkle path verification, and digital signature verification.

8. A distance threshold driven and verifiable semantic commitment based station privacy triage system, comprising: The system includes: a GeoHash set generation module, an identifier generation module, a semantic commitment generation module, a candidate set generation module, and a verification module; The GeoHash set generation module is used to derive GeoHash parameters based on a distance threshold and generate an adaptive neighborhood GeoHash set. The identifier generation module is used to generate semantic spatial identifiers based on an adaptive neighborhood GeoHash set and by integrating business semantics. The semantic commitment generation module is used to generate a corresponding verifiable semantic commitment for each generated semantic spatial identifier; The candidate set generation module is used to construct a keyed Bloom filter and intersect it with a verifiable semantic commitment to obtain a candidate semantic spatial identifier set; The verification module is used to verify the authenticity of the acquired set of candidate semantic spatial identifiers.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the site privacy screening method based on distance threshold driving and verifiable semantic commitment as described in any one of claims 1-7.

10. An electronic device, comprising: It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements the steps of the site privacy screening method based on distance threshold driving and verifiable semantic commitment as described in any one of claims 1-7.

Images