Verifiable multi-variable randomized draw system

TWI944549BActive Publication Date: 2026-11-01梁 立豐
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
TW114112158
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-03-28
Publication Date
2026-11-01
Estimated Expiration
2045-03-27

Smart Images

  • Figure TWG2TB001918804_001
    Figure TWG2TB001918804_001
  • Figure TWG2TB001918804_002
    Figure TWG2TB001918804_002
  • Figure TWG2TB001918804_003
    Figure TWG2TB001918804_003
Patent Text Reader

Abstract

This invention relates to a multivariate random selection system with verifiable characteristics. It achieves fairness, reproducibility, and tamper resistance in the selection results through seed code generation from multi-source data, hash chain encapsulation, and a replay verification module. The system supports user-interactive input, mode switching, result encapsulation, and offline verification, and is suitable for scenarios such as ticket allocation, fair lotteries, educational grouping, and Web3 governance. Compared to traditional PRNG, CSPRNG, or blockchain VRF mechanisms, it features low latency, high scalability, and no single point of trust, making it particularly suitable for applications requiring public access, high credibility, and random probability allocation.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A multivariate random sampling system with verifiable characteristics, characterized in that it comprises: a user interface module for receiving sampling parameters and data input by a user, wherein the user interface module may be a graphical user interface, a command-line interface, or remote instructions and data received via an API; a configuration file processing unit for reading a structured configuration file to construct sampling conditions; and a random data generation module for integrating user input, server environment variables, dynamic timestamps, multi-source datasets, previous sampling results, and system entropy sources, and performing hybrid seed calculation, wherein the timestamps include adjustable time units, such as milliseconds, seconds, or user-defined time intervals. It can support the extraction of one or more random samples from a preset or dynamically updated random dataset as participation factors to generate the original seed code. The module also supports random entropy salt, custom weight settings, and high-frequency random sample selection modes. A selection logic execution module is used to perform random calculations and generate selection results according to the seed code and selection logic rules. The selection logic rules may include configuration parameters for random selection based on weight ratios. The weights can be applied to the user's original input data or to samples in a preset or dynamically updated random dataset to support selection application scenarios with multiple probability conditions. The selection logic rules and encryption algorithms can... The system can be replaced, adjusted, or expanded according to application scenarios to meet different selection mechanisms and information security requirements; a result hashing module is used to perform hash calculations on the selected data to generate one or more sets of selection result hash codes for subsequent verification and comparison; a result verification module is used to generate corresponding hash verification codes and signature information, which can be HMAC, MAC, digital signature, or other algorithms with equivalent information integrity verification functions, and supports reconstructable verification mechanisms; a data encapsulation module is used to encapsulate user input values, time parameters, random data sample selection results, environmental variables, weight configuration, seed code, configuration files, parameters, and verification information involved in the selection process. The system comprises a portable package containing the code, output results, verification signature, selection result data, and other intermediate variables and dynamic data generated or involved in the calculation during the selection process. This package is used for subsequent verification, comparison, and reconstruction processes. A result output and verification module is also included to output the packaged content in file or report format, supporting third-party offline reconstruction and data consistency comparison. The system also provides a replay unit, allowing any entity to verify the selection process in an untrusted environment. It supports advanced verification operations such as error tracing, packet verification, ZKP result verification, two-way comparison, and in-situ signing, achieving a selection architecture that is fair, transparent, verifiable, and portable.

2. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the random data generation module supports user-defined input as participation factors, and the input data may include, but is not limited to: text, strings, numbers, dates, times, Boolean values, option lists, key-value pairs, encrypted hashes, image features, device information, input action features, or other data forms that can be parsed by the system, and may be combined with external data sources, system time parameters, and entropy values ​​to jointly form the original random seed code, so as to improve its uniqueness, unpredictability, and resistance to manipulation.

3. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the random data generation module supports the integration of a set of preset or dynamically generated random datasets, and extracts one or more sets of samples from the dataset as participation factors before or during the sampling process. The number of samples extracted can be fixed or determined by a random mechanism. The extraction method can be based on random order, weight ratio or user-specified conditions, and can be combined with other dynamic inputs to generate seed codes.

4. The multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the timestamp used by the random data generation module supports the setting of multiple time units, including millisecond, second, minute, or user-defined time intervals, and can be dynamically adjusted in conjunction with data update frequency or external trigger events to enhance the variability of seed code and the unpredictability of sampling results.

5. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the random data generation module further includes a bitwise operation unit, characterized in that the unit receives two or more sets of intermediate hash values ​​generated by different hash algorithms and performs a bitwise XOR operation on them to generate a hybrid seed code; the hybrid seed code is used as input to subsequent sampling logic and verification process to improve the entropy strength, collision resistance and unpredictability of the final seed code, and reduce the risks that may arise from relying on a single algorithm.

6. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the random data generation module further supports a multi-verification structure, characterized in that the module can perform phased verification on the original input, VRDF dataset and intermediate seed code during the random data generation and sampling operation stages, supporting the application of one or more sets of data verification algorithms, including but not limited to HMAC, ZKP, digital signature or MAC, etc., the purpose of which is to check the consistency of data and the correctness of calculation results in real time in the internal process, thereby enhancing the overall reproducibility, anti-manipulation and data integrity of the sampling system.

7. A multivariate random selection system with verifiable characteristics as described in claim 1, wherein the system includes a configuration file management module, characterized in that, upon system startup, it automatically detects whether a selection configuration file exists; if not, it automatically creates a preset configuration file and initializes necessary parameters; if it exists, it reads the contents of the configuration file and loads them into each unit of the selection module, and automatically checks whether any necessary fields are missing. If any are missing, they will be filled in with system preset values ​​to ensure the stability of the selection process and the consistency of parameters. The configuration file format may include, but is not limited to, JSON, YAML, or other structured data formats, and may support parameter encryption and verification tags to prevent tampering and misuse.

8. The multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the sampling logic execution module supports a custom sampling field configuration mechanism, allowing users to set one or more sets of logical field structures before sampling, including but not limited to "group", "team", "partition", "row", "column", etc.; each field structure can correspond to a specific sample subset and its exclusive sampling rules, weight configuration, exclusion conditions or sequential processing logic, and can support multi-level field nesting and condition-oriented settings to meet complex sampling needs and multi-dimensional sample management operations.

9. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein each sampling field can be independently configured with its corresponding weight parameters and selection logic. The weight can be dynamically determined based on the attributes of the sample to which the field belongs, user input, historical records, or external data, and supports random selection using cumulative arrays, probability models, or segmented weight correspondence methods. The system can perform independent or hierarchical sampling and comparison operations for different fields during the sampling process, and retain the intermediate states of the results of each level of fields, the verification chain code, and the weight mapping relationship to ensure that the sampling results are transparent, have consistent weights, and are reproducible.

10. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the sampling logic execution module further includes duplicate selection checking logic, which compares the current selected sample with the previous selection results in real time during each round of sampling and automatically excludes duplicates to ensure that each sample can be selected only once during the sampling process. The checking logic supports dynamic memory indexing, automatic filtering mechanism and retry limit, and can operate synchronously with weight logic and VRDF seed variable to ensure result uniqueness and data consistency.

11. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the sampling logic execution module includes an accumulation weight processing unit that accumulates all samples according to their specified individual weights to form an interval correspondence array, and uses one or more sets of seed codes to generate random index values ​​during sampling to correspond to weight intervals to select samples. The sampling logic execution module supports dynamic weight adjustment, adaptive screening strategies and complex stage selection processes, and can be combined with VRDF weight datasets to implement a composite probability model sampling mechanism to enhance the fairness of the sampling results and the stability of the sample distribution.

12. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the result output and verification module have a chain verification structure, which combines the hash code of the previous sampling result with the current sampling data, and generates a continuous verification code through rolling hash or similar cumulative encryption operation to form a verification chain with sequential correlation; if any sampling data or result is tampered with, it will cause the subsequent verification chain code to be inconsistent, and thus can be identified by the verification module, ensuring that the overall sampling record has immutability, traceability and integrity.

13. The multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the result output and verification module belongs to the modeling group within this system and has the ability to generate signatures, verification codes, or proof objects in real time during the sampling process. The module supports a variety of encryption verification algorithms, including but not limited to HMAC, MAC, ZKP, digital signatures, etc., and can automatically adjust the verification level and method according to the sampling stage and data sensitivity. The verification module also supports data interoperability and encapsulation compatibility with external modular verification units. Its encapsulation format has consistency and resolvability, and can support reconstruction, comparison, and verification operations in cross-system or untrusted environments, and ensure the traceability of the verification process and the non-repudiation of the results.

14. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the verification chain can be reconstructed by any third party in an independent environment based on publicly available encapsulated data to verify the consistency and integrity of the sampling process and results, and can be used to establish version correlation and time series records of sampling data.

15. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the verification chain verification structure can be combined with zero-knowledge proof technology so that each sampling result can prove the correctness of its chain logic and the previous data without revealing the actual data, thereby further improving the security level of the verification chain in terms of privacy protection and non-repudiation.

16. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the system further comprises a modular verification output unit, characterized in that the unit can automatically or according to user settings generate corresponding verification module code or package files based on the usage scenario or deployment requirements; the verification module supports independent operation, can reconstruct and compare the sampling results in untrusted environments or offline devices, and can perform verification operations through terminal execution, command-line tools or API interfaces.

17. The multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the modular verification unit automatically binds key verification conditions and environmental parameters when producing the verification module, including but not limited to the original seed code source, encapsulation parameters, algorithm version, result verification chain position, or system hash verification logic; the binding information is an embedded parameter of the verification module and cannot be modified externally to ensure the complete correspondence between the sampling process and the verification logic, and to prevent the verification module from being arbitrarily recombined or applied to non-original data sources.

18. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the modular verification unit can dynamically select the applicable verification algorithm and output format according to the type of sampling result, custom encapsulation format or verification level, including but not limited to HMAC, MAC, ZKP, digital signature, or traceable chained verification code; and after verification, output verification result code, digest comparison record, error message or process comparison report, wherein the report shall be based on the comparison result of the complete encapsulation structure, and it is prohibited to verify only a single digest field to represent the overall verification conclusion.

19. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the result output and verification module supports multiple output formats, including but not limited to JSON, CSV, or other structured data formats, and provides additional verification interfaces so that users can perform offline or online reconstruction and verification operations through command-line tools, terminal programs, Web APIs, or blockchain applications. Its encapsulation format can be JSON, CSV, XML, YAML, or other structured data formats, or it can be a transcoded, compressed, or packaged data container. Its internal data must not only retain hash digests, signature fields, or result digests for verification, but should also fully retain parameter fields, chained verification data, and original time information to support correct data reconstruction and result comparison.

20. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the result output and verification module further supports dynamically determining the output format and verification level based on user permissions, application scenarios or system preset conditions, so as to enhance the flexibility and security of data in different transmission or public environments.

21. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the encapsulated data includes time stamps of the sampling process, system version identification codes, and result verification chain position indicators, and the encapsulated structure must have field labels that can be read and parsed by machines. This structure cannot be omitted or replaced in a simplified manner, and is an indispensable data component for process reconstruction, result consistency verification, and traceability comparison. The encapsulated data can support historical restoration and version tracking at different time points, system versions, or verification environments, and effectively prevent post-selection modifications or skipping of verification mechanisms.

22. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the generation process of the encapsulated data and the sampling operation process have an inseparable data correlation. The system synchronously establishes a hash link and time snapshot of the encapsulated data during the sampling process, and generates the corresponding encapsulated block immediately after the sampling result is completed. Any subsequent data modification will cause verification failure, so as to ensure that the sampling record has non-repudiation and finality.

23. A multivariate random sampling system with verifiable characteristics as described in any of the claims 1, wherein the encapsulated result data shall not be reorganized or have its verification process simplified by front-end tools or blockchain applications after output. The verification operation shall be performed through a verification module automatically generated by the system, which supports full field reading, chain verification and parameter comparison. The front-end shall not be allowed to perform simplified verification by reading only part of the field, so as to ensure the indivisibility and verification consistency of the encapsulated data and the original sampling results.

24. The selection system that can be independently verified by a third party as described in Request 1, wherein the output of the results and the verification module support the conversion to blockchain interactive format, JavaScript object, JSON-RPC structure or other smart contract verification input format, but the encapsulation structure should still retain the necessary field labels, chain verification data and verification condition binding information after conversion, and the field structure required for verification should not be omitted or modified, so as to prevent the selection results from losing their reconstructability and verifiability when transmitted across platforms.

25. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the data encapsulation module further supports on-chain deployment scenarios, and the encapsulated data embeds necessary verification parameters and logical tags, including encapsulation time, version code, verification chain position and algorithm identification code; the encapsulation format can be directly read by smart contracts or DApps to verify the consistency of the sampling results on-chain, and prevent erroneous comparisons or bypassing of encapsulation behavior caused by simplified digest verification.

26. The multivariate random sampling system with verifiable characteristics as described in Request 1 further supports an on-chain automatic verification module, which can execute automatic comparison logic in the smart contract according to the content of the encapsulated data to verify whether the verification result is consistent with the verification code generated off-chain, and automatically record the verification time and comparison status to support decentralized auditing and on-chain process traceability.

27. The third-party independent verification sampling system as described in Request 1 further supports data verification APIs for Web3 application scenarios. The verification interface can be called by smart contracts, DApps or blockchain node systems through standard protocols to perform verification comparison of encapsulated data and return verification status.

28. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the encapsulated data can generate a record of evidence after verification, and the record can be selectively and synchronously written into a blockchain system to establish the non-repudiation, traceability and tamper-proof nature of the verification record.

29. The multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the result output and verification module can calculate a digest fingerprint based on the content of the encapsulated data after the data is generated, and combine the digest fingerprint with relevant identification parameters to form verification anchor information, which is automatically or uploaded to the blockchain system after user authorization, as a non-trusted verification anchor for subsequent encapsulation results. The existence of the digest fingerprint on the chain allows any third party to perform consistency comparison and verification of the local encapsulation report.

30. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the result output and verification module further supports the verification mechanism of on-chain historical packaged versions. The packaged data includes version number, packaged format, parameter summary and verification chain position index when it is produced, and the aforementioned information and summary fingerprint are recorded together when it is uploaded to the chain, so as to support users to track the packaged records of each version through the on-chain query interface, perform version comparison, sampling result consistency comparison and process traceability, and ensure that the results of different sampling sessions are traceable, verifiable and version identifiable.

31. The multivariate random selection system with verifiable characteristics as described in claim 1, which can be applied to various scenarios requiring randomness, fairness, verifiability, and anti-cheating mechanisms for digital selection, allocation, matching, or qualification review, including but not limited to: ranking for ticket purchase eligibility at concerts or large events, matchmaking in e-sports events, event scheduling and team drawing in sports competitions, random allocation of test questions or question sets in the education field, invigilator seating arrangement in academic tests, fair lottery for government subsidies and resource allocation, virtual item allocation and item generation logic in online games, verification of NFT forging or blind box lottery logic, Web3... The modular random generation and verification mechanism provided by this system can be applied to various scenarios, including random node voting and sorting under the DAO mechanism, risk sample clustering in financial risk control systems, random selection of online questionnaires or sampling statistics, generation of simulated participant parameters in urban traffic simulation, subject sample allocation in medical clinical trials, distributed computing task scheduling, and dynamic sample selection for artificial intelligence training datasets. This ensures that each sampling result has the characteristics of being reconstructable, verifiable, and non-repudiable.

32. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the system further supports anomaly detection and risk warning mechanisms during the sampling process. When parameter settings, data encapsulation, or result distribution are abnormal, the system will immediately issue a risk warning message and record it in the encapsulation report as a basis for subsequent verification and troubleshooting by a third party.

33. The multivariate random sampling system with verifiable characteristics as described in claim 1 further includes a replayable simulation verification module, which can replay the sampling process in any environment based on the timestamp, seed code, field configuration and algorithm flow of the encapsulated report, and generate intermediate states, corresponding verification chain codes and final results for users or third parties to perform integrity comparison and simulation.

34. The multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the result output and verification module can automatically generate a watermark identification code and a tracking code when the report file is generated, including the file generation time, version number, user ID, device code and result fingerprint summary, etc. The watermark information is stored in an embedded structure or as an attachment and can be compared synchronously during verification to prevent the forgery, imitation or reproduction of the result report.

35. A multivariate random sampling system with verifiable characteristics as described in Request 1, wherein the sampling logic execution module supports multiple sampling modes, each of which can define the sampling order, column priority order, duplicate exclusion rules, weight application scope, and hash mixing rules, etc. Users can switch modes through configuration files, APIs, or interfaces. The system will automatically construct the sampling process according to the selected mode and mark the mode type and execution record used in the output results to facilitate subsequent verification and reconstruction operations.

36. A multivariate random selection system with verifiable characteristics as described in any of claims 1, wherein the selection logic execution module further supports a streaming selection mode, characterized in that the selection process is executed in a trigger-by-transaction manner, each selection result is a single computational unit, and allows the user, host, or automated control module to send "stop," "pause," or "continue" commands during execution to intervene in the selection process in real time. After each selection operation, the system synchronously encapsulates the current seed code, weight status, field data, and timestamp, and stores the selection results of that stage in real time in a snapshot record, ensuring that even if the selection is interrupted, the state at the time of the interruption can be reconstructed, and can... Selectively continuing execution from the point of interruption, the system supports using the hash code generated from the previous result as part of the parameters for the next selection, thereby strengthening the logical coherence and verification chain between each selection, and preventing arbitrary insertion or skipping of operations. This mode is suitable for scenarios where the selection rhythm needs to be manually controlled, such as live interactive lottery, concert ticket distribution in batches, online classes using games to draw questions, or interactive applications such as blockchain smart contracts unlocking NFTs one by one. In addition to retaining the original verifiable and tamper-proof characteristics of the system, the overall process can also dynamically record the process and output data synchronously during the selection. Regardless of whether it stops in the middle, the complete verification path can be reconstructed and a corresponding encapsulated report can be generated.

37. A multivariate random selection system with verifiable characteristics as described in claim 1, wherein the system further includes a selection triggering and scheduling module, characterized in that it allows users to pre-set triggering conditions and execution schedules, including but not limited to: time intervals, external event notifications, automatic start of the selection process when the number of participants reaches a certain level or specified conditions are met, and can be equipped with pause, resume and manual termination mechanisms to support selection application scenarios in live broadcast, scheduled tasks or asynchronous execution environments.

38. The multivariate random selection system with verifiable characteristics as described in claim 1, wherein the user interface module further supports interactive selection control functions, which can receive user instructions during the selection process to pause, continue, cancel or modify selection parameters, and update the selection process status and visual results display in real time, thereby improving the flexibility of the selection operation and user participation.

39. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the system further includes a sampling triggering and scheduling module, characterized in that it can automatically schedule the execution time and order of sampling tasks according to user settings or system conditions, including but not limited to: time-specified triggering, event triggering, batch scheduling, loop sampling, and serial task triggering, and can synchronously perform sampling record sealing and data output with the packaging module and the verification module.

40. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the encapsulation module has an anonymization parameter conversion mechanism that can perform hashing, masking or zero-knowledge encapsulation processing on personal information or sensitive parameters, so that the encapsulated data does not leak the real identity or privacy information of the participants when publicly verified, and ensures that the verification process complies with data protection and privacy regulations.

41. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the participating data is subjected to one-way hashing before entering the system for processing, the system body only receives and operates on such hashing results, does not store, read or identify the original plaintext participating data, and the hash value is used as one of the participating factors in the random seed code mixing operation, thereby realizing the deidentification and irreversible security of the input parameters, and ensuring that the data processing process has the design characteristics of no personal data access.

42. A multivariate random sampling system with verifiable characteristics as described in claim 1, wherein the hash processing employs a one-way hash algorithm with salting capabilities, and may selectively add timestamps, device identifiers or other random parameters to enhance output uniqueness and security strength.

Citation Information

Patent Citations

  • Distributed lottery system and method based on verifiable delay function

    CN112527244A

  • Random lottery method, system and device based on hash algorithm and storage medium

    CN119380457A

  • Games, lotteries, and sweepstakes and tickets, systems, technologies, and methods related thereto

    TW202109324A

  • Method of generating randomness by public participation

    TWI863100B

  • System and Method for Implementing an Additional Game to Players of a Lottery Game

    US20140057692A1