General method and system for transferring international SMS to domestic SMS
Through the common methods and systems of international SMS to domestic, and technologies such as protocol analysis, semantic analysis and intelligent encryption, the compatibility and security problems in international SMS forwarding are solved, efficient and secure cross-border SMS transmission is achieved, and transmission is improved. Quality and safety.
Patent Information
- Application Number
- CN202411337890.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-25
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-25
AI Technical Summary
The existing international SMS forwarding solutions have problems such as poor protocol compatibility, low content security, and low transmission efficiency, which are difficult to meet the growing cross-border communication needs, and traditional SMS security measures are difficult to deal with complex security threats.
Provide a common method and system for international SMS to domestically transfer, and through protocol analysis, semantic analysis, encryption processing, gateway adaptability analysis and dynamic routing strategy adjustment, multi-protocol seamless conversion, deep security scoring and intelligent encryption are achieved, and the best choice is selected. Gateway, dynamically adjust routing policies to improve transmission quality and security.
It significantly improves the compatibility and security of international SMS, optimizes the efficiency of computing resource usage, improves transmission success rate and stability, reduces operation costs, and achieves seamless connection of two-way communication.
Smart Images

Figure CN118870313B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of short message processing, and in particular to a general method and system for converting international short messages to domestic short messages. Background Art
[0002] With the continuous advancement of globalization, the demand for cross-border communications is growing, and international SMS services are playing an increasingly important role in business, tourism, education and other fields. However, due to differences in communication network standards, protocol specifications and regulatory policies among countries, international SMS faces many technical challenges and security risks in the forwarding process. Existing international SMS forwarding solutions often have problems such as poor protocol compatibility, low content security and low transmission efficiency, which makes it difficult to meet the growing demand for cross-border communications.
[0003] At the same time, as the network security situation becomes increasingly severe, international SMS, as an important information carrier, has attracted much attention for its content security and transmission security. Traditional SMS security measures are often limited to simple sensitive word filtering and basic encryption, which are difficult to cope with complex and changing security threats. In addition, international SMS often faces problems such as improper gateway selection and rigid routing strategies during cross-network and cross-platform transmission, resulting in unstable transmission quality and high costs. Summary of the invention
[0004] The present application provides a general method and system for transferring international text messages to domestic messages, which is used to improve the forwarding quality and security of international text messages.
[0005] In a first aspect, the present application provides a general method for converting international text messages to domestic text messages, and the general method for converting international text messages to domestic text messages includes:
[0006] Perform protocol analysis and content reconstruction on the received international SMS to obtain the reconstructed SMS;
[0007] Performing semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score;
[0008] Based on the content security score, encrypting and digitally signing the reconstructed SMS to obtain a secure transmission message;
[0009] Performing gateway adaptability analysis on the secure transmission message to obtain a gateway selection result;
[0010] Based on the gateway selection result, real-time monitoring and historical data analysis are performed on gateway performance data to obtain a dynamic routing strategy;
[0011] According to the dynamic routing strategy, the user's reply SMS is captured in real time and the protocol is converted to obtain a forwarded SMS.
[0012] In a second aspect, the present application provides a universal system for transferring international short messages to domestic short messages, the universal system for transferring international short messages to domestic short messages comprising:
[0013] A reconstruction module is used to perform protocol analysis and content reconstruction on the received international SMS to obtain a reconstructed SMS;
[0014] An identification module, used to perform semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score;
[0015] A signature module, used to encrypt and digitally sign the reconstructed SMS based on the content security score to obtain a secure transmission message;
[0016] An analysis module, used to perform a gateway adaptability analysis on the secure transmission message to obtain a gateway selection result;
[0017] A monitoring module, used to perform real-time monitoring and historical data analysis on gateway performance data based on the gateway selection result to obtain a dynamic routing strategy;
[0018] The conversion module is used to capture and convert the user's reply SMS in real time according to the dynamic routing strategy to obtain the forwarded SMS.
[0019] In the technical solution provided by this application, through intelligent protocol parsing and content reconstruction, the method can adapt to the SMS protocol standards of different countries and regions, realize seamless conversion between multiple protocols, and greatly improve the compatibility and reliability of international SMS. By adopting deep semantic analysis and sensitive content identification technology, combined with a multi-dimensional content security scoring mechanism, it is possible to more accurately identify potential risks and effectively prevent security threats such as spam SMS and fraudulent information. Through dynamically adjusted encryption processing and digital signature mechanisms, the encryption strength can be flexibly selected according to the content security score, while ensuring transmission security, the efficiency of computing resources is optimized. Based on multi-dimensional gateway adaptability analysis and real-time performance monitoring, the method can intelligently select the optimal gateway, significantly improving the success rate and stability of SMS transmission. Through real-time monitoring of gateway performance data and historical data analysis, the routing strategy can be dynamically adjusted to effectively respond to network fluctuations, improve transmission efficiency, and reduce operating costs. Using real-time capture and protocol conversion technology, it is possible to quickly process user reply SMS, achieve seamless connection of two-way communication, and improve the forwarding quality and security of international SMS. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0021] Figure 1 This is a schematic diagram of an embodiment of a general method for converting international SMS to domestic SMS in an embodiment of the present application;
[0022] Figure 2 This is a schematic diagram of an embodiment of the international text message conversion domestic general system in the embodiment of the present application. DETAILED DESCRIPTION
[0023] The embodiments of the present application provide a general method and system for transferring international text messages to domestic text messages. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0024] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the general method for converting international SMS to domestic SMS in the embodiment of the present application includes:
[0025] Step S101, performing protocol analysis and content reconstruction on the received international SMS to obtain a reconstructed SMS;
[0026] It is understandable that the execution subject of the present application can be an international SMS to domestic general system, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.
[0027] Specifically, the received international SMS is identified by protocol. Different SMS protocols have different formats and encoding methods. After identifying the protocol type, the correct method is selected to decode the SMS and convert its character set into a format suitable for processing. Through this process, the original unprocessed content is extracted from the international SMS. The original content is segmented and the text is divided into meaningful word units. Usually combined with semantic annotation, that is, each word unit is annotated with its semantic role, such as subject, predicate, object, etc. Through word segmentation and semantic annotation, the unstructured SMS content is converted into structured data for subsequent processing. The key information in the SMS content is extracted to generate a core content summary. It contains the most important part of the SMS and removes redundant or unnecessary information. When the core content summary is generated, it is translated into the language of the target country. The translation process needs to consider the accuracy and fluency of the language to ensure that the SMS content can be correctly understood in the target country. The translated content often needs to be analyzed and segmented in length because different countries have different restrictions on the length of SMS. According to these restrictions, the translated content is divided into multiple parts, each of which meets the SMS length requirements of the target country. Encode the segmented content to make it conform to the SMS protocol of the target country. The encoding process includes generating the necessary SMS header information and control fields to ensure that the SMS can be correctly processed and transmitted in the target country's network. After the complete SMS structure is formed, it is checked for data integrity to ensure that no errors occur during the encoding and transmission process, and the content of the SMS can be transmitted to the recipient intactly. After the data integrity check is passed, the final protocol encapsulation and formatting are performed to generate a reconstructed SMS to ensure that the format of the SMS meets the requirements of both the sender and the receiver, and can be smoothly transmitted in international and domestic networks.
[0028] Step S102: perform semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score;
[0029] Specifically, the reconstructed SMS is segmented to generate a word-unit sequence. Based on the word-unit sequence, each word-unit is subjected to part-of-speech tagging and named entity recognition to construct a grammatical structure tree. The generated grammatical structure tree is subjected to dependency syntactic analysis to obtain a sentence component relationship graph. Through the sentence component relationship graph, the relationship between the components within the sentence can be better understood, and then the subject and keywords of the SMS can be extracted to generate a semantic feature vector. Multilingual sentiment analysis is performed on the semantic feature vector to obtain a sentiment polarity value. Sentiment analysis can identify the sentiment tendency of the SMS content, such as positive, negative or neutral sentiment. At the same time, according to the preset sensitive word library, the word-unit sequence is fuzzy matched to detect the frequency of occurrence of sensitive words. Fuzzy matching technology allows the detection of deformed or approximate sensitive words, making the analysis more comprehensive and accurate. The frequency of occurrence of sensitive words is weighted to obtain a sensitivity index. Considering the different weights of sensitive words reflects their importance in the SMS content. According to the sentiment polarity value and the sensitivity index, a two-dimensional scoring matrix is constructed to obtain an initial security score. The two-dimensional scoring matrix comprehensively considers the sentiment tendency and sensitivity of the content, making the initial security score more comprehensive and representative. The initial security score is compared with historical data and anomaly detection is performed to obtain a normalized security score. Historical data comparison can help identify potential anomalies and ensure the accuracy and reliability of the score. Normalization makes the score comparable at different times and under different circumstances. Based on the normalized security score and the preset security level threshold, the SMS content is graded to obtain the final content security score.
[0030] Step S103: Based on the content security score, encrypt and digitally sign the reconstructed SMS to obtain a secure transmission message;
[0031] Specifically, the content security score is compared with a threshold to determine the encryption strength level. According to the sensitivity and security requirements of the SMS content, an appropriate encryption measure is selected. According to the determined encryption strength level, a corresponding encryption algorithm is selected from a preset encryption algorithm library to obtain an encryption scheme. The reconstructed SMS is processed into blocks to divide the SMS into multiple data blocks. Block processing helps to improve the efficiency of encryption and transmission and enhance the security of data. According to the selected encryption scheme, each data block is randomly padded to ensure that the length of each data block is consistent and increase its complexity to obtain a padded data block. Based on the encryption algorithm selected in the encryption scheme, the padded data block is encrypted to obtain encrypted data. The specific method of the encryption operation depends on the selected algorithm and may include symmetric encryption or asymmetric encryption. The generation of encrypted data is to ensure the confidentiality of the SMS during transmission and prevent unauthorized access and tampering. According to the preset key management policy, a private key for digital signature is generated to obtain a signature key. After the signature key is generated, the encrypted data is hashed to obtain a message digest. The message digest is a fixed-length string that represents a unique fingerprint of the encrypted data and helps to ensure the integrity of the data. The message digest is asymmetrically encrypted according to the signature key to obtain a digital signature. The role of the digital signature is to verify the source and integrity of the data and ensure that the data has not been tampered with during transmission. The encrypted data and the digital signature are combined and encapsulated to generate an encrypted message. The encrypted message contains the encrypted data and its digital signature, providing multi-level security protection. The encrypted message is formatted to obtain a secure transmission message to ensure that the encrypted message meets the requirements of the transmission protocol and can be correctly transmitted and decoded in the network. This step may include adding message header information, setting transmission parameters, etc.
[0032] Step S104: Performing gateway adaptability analysis on the secure transmission message to obtain a gateway selection result;
[0033] Specifically, the protocol type of the secure transmission message is identified to determine its target protocol type, so as to understand the communication protocol that the message needs to use and select a suitable gateway for transmission. According to the determined target protocol type, the candidate gateways that support the protocol are screened from the preset gateway library to generate a preliminary gateway list. The historical performance data of each gateway in the preliminary gateway list is extracted to obtain a gateway performance indicator set. The performance indicators may include the response time, successful transmission rate, load status, etc. of the gateway. Based on the performance indicators, a multi-dimensional score calculation is performed on each gateway to generate a gateway score matrix. The gateway score matrix is normalized to obtain a standardized gateway score. The scoring standards of different dimensions are unified so that the scores can be directly compared. According to the preset gateway selection weight model, the standardized gateway score is weighted and calculated to obtain a comprehensive score. The setting of the weight model reflects the importance of each performance indicator in the gateway selection, and the influence of these indicators is comprehensively considered through weighted calculation. The comprehensive scores are sorted, and several gateways with the highest scores are selected to obtain a target gateway set. The target gateway set includes the gateways with the best performance in the comprehensive score as candidate gateways for subsequent transmission. According to the secure transmission message, load balancing calculation is performed on the target gateway set to determine the load allocated to each gateway and obtain the load distribution plan. The load balancing calculation ensures that the load of each gateway is even during the transmission of the message, avoiding the overload of a single gateway. The load distribution plan is verified to ensure its feasibility and stability in actual operation. Different transmission scenarios may be simulated during the verification process to test the performance of the load distribution plan in various situations. Based on the verification results, a fault-tolerant strategy is formulated to deal with possible transmission failures and abnormal situations. The fault-tolerant strategy includes the setting of backup gateways, fault switching mechanisms, etc. to ensure the continuity and reliability of message transmission. Based on the load distribution plan and fault-tolerant strategy, a gateway selection result is generated. The gateway selection result includes specific gateway selection and load distribution plans, and considers fault-tolerant mechanisms to ensure efficient, secure and stable message transmission.
[0034] The standardized gateway score is characterized and decomposed to obtain the gateway performance feature vector. Through the feature decomposition, the specific manifestations of each gateway performance indicator are clarified, and then the feature importance is sorted according to the feature vector to obtain the feature priority list. The feature priority list is subjected to hierarchical analysis to build a multi-level evaluation framework. The hierarchical analysis method decomposes complex problems and hierarchizes them, making the evaluation process more systematic and scientific. According to the multi-level evaluation framework, the gateway performance feature vectors are grouped to form a hierarchical feature set. The hierarchical feature set classifies the gateway performance indicators according to different levels, which is convenient for subsequent fuzzy evaluation and weight calculation. Each feature in the hierarchical feature set is fuzzy evaluated to generate a feature membership matrix. Fuzzy evaluation can handle the uncertainty and ambiguity in performance indicators, and quantify the evaluation results of each feature through the membership matrix. According to the feature membership matrix, the weight coefficient of each layer of evaluation indicators is calculated to obtain the hierarchical weight vector. The hierarchical weight vector is subjected to consistency check to ensure the rationality and consistency of weight distribution. In the consistency check, if the weight vector is found to be inconsistent, it will be corrected to obtain the corrected weight vector. The modified weight vector is used to perform weighted summation on the hierarchical feature set to obtain the comprehensive score of each layer. The inter-layer weight distribution is performed on the comprehensive score of each layer, and the global weight coefficient is calculated. The inter-layer weight distribution ensures the global rationality of the final score by comprehensively considering the importance of the score of each layer. According to the global weight coefficient, the comprehensive score of each layer is weighted and aggregated to obtain the final comprehensive score. The weighted aggregation process integrates the scores of different layers into an overall score, providing a comprehensive evaluation of the gateway performance.
[0035] Extract features of secure transmission messages and generate message attribute vectors. The message attribute vector contains various characteristics of the message, such as size, priority, protocol type, etc. According to the message attribute vector, perform compatibility evaluation on each gateway in the target gateway set to obtain the gateway fitness matrix. Normalize the gateway fitness matrix to obtain a standardized fitness score. Normalization standardizes the fitness scores of different gateways to make them comparable. According to the standardized fitness score, perform preliminary screening on the target gateway set to obtain a candidate gateway subset. Preliminary screening uses the fitness score to select the gateway that is most suitable for processing specific messages, thereby improving the efficiency and effect of load balancing. Perform real-time performance detection on each gateway in the candidate gateway subset to obtain the current load status of the gateway. Real-time performance detection includes measuring the current traffic, response time, processing capacity and other parameters of the gateway. Based on real-time data, build a gateway load prediction model to predict the load trend of the gateway in the short term. The load prediction model can help foresee the load changes of the gateway in the future and provide a basis for decision-making. Perform threshold analysis on the short-term load trend to obtain the load risk assessment result. Threshold analysis compares the predicted load trend with the preset load threshold to evaluate the risk level of each gateway under future load conditions. Based on the load risk assessment results, dynamic weights are assigned to the subset of candidate gateways to generate a preliminary load plan. Dynamic weight assignment takes into account the load risk and current status of each gateway to ensure the rationality and stability of load distribution. The preliminary load plan is tested to obtain the performance indicators of the plan. During the test, the actual operating environment is simulated to evaluate the performance of the load plan in different scenarios. Based on the performance indicators of the plan obtained from the test, the preliminary load plan is optimized to improve its performance and reliability. The optimization process may involve measures such as adjusting weights and redistributing loads. After optimization and adjustment, a load distribution plan is obtained. The plan comprehensively considers factors such as message attributes, gateway adaptability, current load status, load prediction and risk assessment to ensure efficient distribution and transmission of messages in the target gateway set.
[0036] Step S105: Based on the gateway selection result, the gateway performance data is monitored in real time and historical data is analyzed to obtain a dynamic routing strategy;
[0037] Specifically, a performance monitoring probe is set for each gateway in the gateway selection result to obtain a real-time performance data stream. The setting of the performance monitoring probe can continuously collect the real-time operation data of each gateway, including response time, processing capacity, traffic, etc. According to the obtained real-time performance data stream, data cleaning and outlier detection are performed on each gateway to ensure the accuracy and reliability of the data, and an effective performance indicator set is obtained. Time series analysis is performed on the effective performance indicator set to extract the periodicity and trend characteristics of gateway performance and obtain the gateway performance mode. Time series analysis can reveal the potential laws and change trends in gateway performance data. Based on the extracted gateway performance mode, a gateway performance prediction model is constructed to generate short-term performance prediction results. The prediction model can accurately predict the performance of the gateway in the short term by learning from historical data. The short-term performance prediction results are scored in multiple dimensions to obtain the gateway real-time availability score. The multi-dimensional scoring takes into account multiple key performance indicators, such as stability, response speed, processing capacity, etc., and comprehensively evaluates the real-time availability of each gateway. According to the gateway real-time availability score, historical routing data is correlated and analyzed to evaluate the effects of different routing strategies and obtain a routing effect evaluation report. Correlation analysis reveals the impact of different routing options on performance by comparing real-time scores with historical routing data. Data mining is performed on the routing effect evaluation report to identify areas that can be improved and obtain routing optimization directions. Data mining technology can deeply analyze the data in the evaluation report to find potential optimization opportunities and bottlenecks. Based on the identified routing optimization directions, multiple candidate routing strategies are designed to form a routing strategy set. Each candidate routing strategy provides multiple possible improvement plans based on different optimization goals and assumptions. The routing strategy set is tested to evaluate its effectiveness in actual operation and obtain a strategy performance matrix. The testing process simulates the actual operating environment and evaluates the performance of each strategy in various situations, including indicators such as stability, efficiency, and reliability. According to the strategy performance matrix, the strategy with the best performance is selected, and it is further optimized and adjusted to generate the final dynamic routing strategy.
[0038] Step S106: According to the dynamic routing strategy, the user's reply SMS is captured in real time and the protocol is converted to obtain a forwarded SMS.
[0039] Specifically, the dynamic routing strategy is parsed, and the reply SMS capture rules are extracted to form a capture rule set. The capture rule set includes specific conditions and methods for monitoring and capturing user reply SMS. According to the capture rule set, a gateway listener is set to monitor the user reply SMS in real time and generate the original reply data stream. The gateway listener can capture this information at the moment when the user sends the reply SMS to ensure the real-time nature of the data. The original reply data stream is cleaned and deduplicated to obtain a valid reply SMS set. Data cleaning includes removing invalid data and noise, and deduplication eliminates duplicate reply SMS to ensure that the obtained data set is accurate and unique. According to the valid reply SMS set, protocol identification is performed on each reply SMS to determine the protocol type of the reply SMS. The reply SMS protocol types are classified and counted to construct protocol conversion rules. The protocol conversion rules define how to convert SMS of different protocol types into a unified format. According to the protocol conversion rules, protocol conversion is performed on each reply SMS to generate reply content in a unified format. Semantic analysis is performed on the reply content in a unified format to extract the reply SMS summary. Semantic analysis can identify the core content and intent of the reply SMS. According to the reply SMS summary, the reply content in a unified format is prioritized to generate a processing priority list. A security check is performed on each reply in the processing priority list to generate a secure reply set. The security check includes filtering sensitive information and security assessment of the reply SMS content to ensure information security during transmission. According to the secure reply set, each reply SMS is encapsulated with the target protocol to generate a forwarded SMS. The target protocol encapsulation ensures that the reply SMS meets the protocol requirements of the target system and can be correctly received and processed.
[0040] In the embodiment of the present application, through intelligent protocol parsing and content reconstruction, the method can adapt to the SMS protocol standards of different countries and regions, realize seamless conversion between multiple protocols, and greatly improve the compatibility and reliability of international SMS. By adopting deep semantic analysis and sensitive content identification technology, combined with a multi-dimensional content security scoring mechanism, it is possible to more accurately identify potential risks and effectively prevent security threats such as spam SMS and fraudulent information. Through dynamically adjusted encryption processing and digital signature mechanisms, the encryption strength can be flexibly selected according to the content security score, while ensuring transmission security, the efficiency of computing resources is optimized. Based on multi-dimensional gateway adaptability analysis and real-time performance monitoring, the method can intelligently select the optimal gateway, significantly improving the success rate and stability of SMS transmission. Through real-time monitoring of gateway performance data and historical data analysis, the routing strategy can be dynamically adjusted to effectively respond to network fluctuations, improve transmission efficiency, and reduce operating costs. Using real-time capture and protocol conversion technology, user reply SMS can be quickly processed, and seamless connection of two-way communication can be achieved, which improves the forwarding quality and security of international SMS.
[0041] In a specific embodiment, the process of executing step S101 may specifically include the following steps:
[0042] Perform protocol identification on the received international SMS to obtain the SMS protocol type, and decode and convert the international SMS into character sets according to the SMS protocol type to obtain the original SMS content;
[0043] Perform word segmentation and semantic annotation on the original SMS content to obtain structured SMS data, and extract key information from the SMS content based on the structured SMS data to obtain a core content summary;
[0044] Translate the core content summary into the target country language to obtain the translated content, and perform SMS length analysis and segmentation based on the translated content to obtain segmented content that meets the SMS length limit of the target country;
[0045] Encode the segmented content according to the target country's SMS protocol to obtain encoded data that complies with the target country's protocol, and generate SMS header information and control fields based on the encoded data to obtain a complete SMS structure;
[0046] The complete SMS structure is verified for data integrity to obtain verified SMS data, and the final protocol encapsulation and formatting are performed based on the verified SMS data to obtain a reconstructed SMS.
[0047] Specifically, when an international SMS is received, the protocol type of the SMS is identified. Different SMS protocols (such as SMPP, CIMD, UCP, etc.) have different encoding and transmission formats. The protocol type of the SMS is determined by analyzing the header information and protocol identifier of the SMS. Set variables Indicates the protocol type of the SMS according to the protocol identification algorithm , which can be expressed as:
[0048] ;
[0049] Among them, HeaderInfo represents the header information of the SMS. Decode the international SMS and convert the character set according to the protocol type of the SMS to obtain the original SMS content. Different protocols use different encoding methods, such as GSM 7-bit, UCS2, etc. By calling the corresponding decoding function , convert the encoded SMS data into readable text content. Represents the original SMS content, which can be expressed as:
[0050] ;
[0051] Perform word segmentation and semantic annotation on the original SMS content. Word segmentation is to divide the SMS text into individual words or phrases, and semantic annotation is to add grammatical information to these words, such as part of speech and syntactic structure. Through natural language processing technology, the unstructured SMS content is converted into structured data. Set function For the word segmentation and semantic annotation process, the structured SMS data It can be expressed as:
[0052] ;
[0053] Extract key information from structured SMS data and identify important information in the SMS, such as amount, date, address, etc. Obtain a summary of the core content from structured data through information extraction algorithms. Translate the extracted core content summary into the target country language. The translation process uses a machine translation model to convert the SMS content from the source language to the target language. After the translation is completed, perform SMS length analysis and segmentation on the translated content. Different countries and communication operators have different restrictions on SMS length, usually within 160 characters (for GSM 7-bit encoding) or within 70 characters (for UCS2 encoding). Through length analysis and segmentation algorithms , the translated content Split into multiple segments that meet the SMS length restrictions of the target country , expressed as:
[0054] ;
[0055] For segment content Encode the target country's SMS protocol. The encoding process converts the segmented content into encoded data that complies with the target country's SMS protocol. Set function For the encoding process, the encoding data It can be expressed as:
[0056] ;
[0057] According to the coded data Generate SMS header information and control fields to obtain a complete SMS structure. The complete SMS structure includes the message body and related control information to ensure that the SMS can be correctly transmitted and parsed. Function to generate SMS header information It can be expressed as:
[0058] ;
[0059] Perform data integrity check on the complete SMS structure to ensure that no errors occurred during encoding and transmission. , for the complete SMS structure Perform verification and obtain the verified SMS data , expressed as:
[0060] ;
[0061] According to the verified SMS data The final protocol encapsulation and formatting are performed to obtain the reconstructed SMS. The encapsulation and formatting process converts the SMS data into a format that meets the requirements of the target protocol and is ready for transmission.
[0062] In a specific embodiment, the process of executing step S102 may specifically include the following steps:
[0063] Perform word segmentation on the reconstructed SMS to obtain a word-unit sequence, and perform part-of-speech tagging and named entity recognition on each word-unit according to the word-unit sequence to obtain a grammatical structure tree;
[0064] Perform dependency syntactic analysis on the grammatical structure tree to obtain a sentence component relationship graph, and extract the subject and keywords of the text message based on the sentence component relationship graph to obtain a semantic feature vector;
[0065] Perform multilingual sentiment analysis on the semantic feature vector to obtain the sentiment polarity value, and perform fuzzy matching on the word sequence according to the preset sensitive word library to obtain the frequency of occurrence of sensitive words;
[0066] The frequency of occurrence of sensitive words is weighted to obtain the sensitivity index, and a two-dimensional scoring matrix is constructed based on the sentiment polarity value and the sensitivity index to obtain the initial safety score;
[0067] The initial security score is compared with historical data and anomaly detection is performed to obtain a normalized security score. Based on the normalized security score and the preset security level threshold, the SMS content is graded to obtain a content security score.
[0068] Specifically, the reconstructed text message is segmented to split a sentence or a text into several independent words, which are the basis of grammatical and semantic analysis. The word segmentation algorithm in natural language processing technology, such as the word segmentation method based on the hidden Markov model, is used to accurately segment the text message content into word-unit sequences. The hidden Markov model consists of a state set, an observation set, a state transition probability matrix, an observation probability matrix, and an initial state distribution. The word segmentation problem can be regarded as a problem of inferring a state sequence from an observation sequence. Assume that the observation sequence is , the state sequence is , where each observation Corresponding to a word unit, each state Corresponds to a part of speech or word boundary. Initial state distribution:
[0069] ;
[0070] Indicates that the initial state is The probability of state transition probability matrix:
[0071] ;
[0072] in, Indicates at time Status is Under the condition of Transfer to state The probability of . Observation probability matrix:
[0073] ;
[0074] in, Indicates in status The following observation value is . Using these probability parameters, the most likely state sequence, i.e. the optimal word segmentation result, is found through the Viterbi algorithm. Based on the obtained word unit sequence, each word unit is tagged with part of speech and named entity recognition. Part of speech tagging refers to assigning a part of speech label to each word unit, such as noun, verb, adjective, etc.; named entity recognition refers to identifying specific entities in the text, such as names of people, places, and institutions. Using existing part of speech tagging tools and named entity recognition algorithms, such as Stanford NLP or spaCy, these tools can efficiently process large-scale texts and accurately perform part of speech tagging and entity recognition. After completing part of speech tagging and named entity recognition, a grammatical structure tree is constructed based on these annotation information. The grammatical structure tree is a tree structure that represents the grammatical hierarchical relationship of the sentence. Dependency syntactic analysis is performed after the grammatical structure tree is generated. Dependency syntactic analysis is a method of representing the sentence structure by describing the dependency relationship between word units. Through dependency syntactic analysis, a sentence component relationship graph is obtained, which can provide a deeper understanding of the grammatical structure of the sentence. According to the sentence component relationship graph, the subject and keywords of the text message are extracted. Topics and keywords are words or phrases that reflect the core content of SMS. Through the dependency graph, the main components of the sentence, such as the subject, predicate, and object, are identified, and keywords are extracted from them. Multilingual sentiment analysis is performed on the semantic feature vector. Sentiment analysis is an important application field in natural language processing, which is used to determine the emotional tendency of the text. Sentiment analysis can be implemented based on dictionaries or machine learning models. For example, the pre-trained sentiment analysis model BERT can be used to classify the sentiment of the text and obtain the sentiment polarity value. The sentiment polarity value indicates the sentiment tendency of the text, such as positive, negative, or neutral. At the same time, according to the preset sensitive word library, fuzzy matching is performed on the word sequence to identify sensitive words in the SMS. The sensitive word library is a dictionary containing specific sensitive words, which is used to detect whether these words are contained in the text. The frequency of sensitive words in the SMS is calculated through the fuzzy matching algorithm. The calculation formula for the frequency of sensitive words can be expressed as:
[0075] ;
[0076] in, Indicates the frequency of occurrence of sensitive words, count Indicates The number of times a sensitive word appears in a text message, N is the total number of words in the text message. The frequency of occurrence of sensitive words is weighted to obtain the sensitivity index. The sensitivity index indicates the sensitivity of the text message content and can be calculated using the following formula:
[0077] ;
[0078] in, represents the sensitivity index, Indicates The weight of sensitive words, Indicates The frequency of occurrence of sensitive words. After obtaining the sentiment polarity value and sensitivity index, a two-dimensional scoring matrix is constructed to obtain the initial security score. The initial security score is calculated based on the sentiment polarity and sensitivity of the text message. Through historical data comparison and anomaly detection, the initial security score is adjusted to obtain a normalized security score. The normalized security score is the result of standardized processing relative to historical data, which can better reflect the security of the current text message. According to the normalized security score and the preset security level threshold, the text message content is graded. The result of the grading is the content security score, which is used to evaluate the overall security of the text message. The range of the content security score can be preset to multiple levels, such as high, medium, and low, so that different treatment measures can be taken in different scenarios.
[0079] In a specific embodiment, the process of executing step S103 may specifically include the following steps:
[0080] Perform threshold comparison on the content security score to obtain the encryption strength level, and select the corresponding encryption algorithm from the preset encryption algorithm library according to the encryption strength level to obtain the encryption scheme;
[0081] The reconstructed SMS is divided into blocks to obtain a plurality of data blocks, and each data block is randomly filled according to the encryption scheme to obtain a filled data block;
[0082] Based on the encryption algorithm selected in the encryption scheme, an encryption operation is performed on the padded data block to obtain encrypted data;
[0083] According to the preset key management strategy, a private key for digital signature is generated to obtain the signature key, and a hash calculation is performed on the encrypted data to obtain a message digest;
[0084] The message digest is asymmetrically encrypted according to the signature key to obtain a digital signature, and the encrypted data and the digital signature are combined and encapsulated to obtain an encrypted message;
[0085] The encrypted message is formatted to obtain a secure transmission message.
[0086] Specifically, the content security score is compared with a threshold. The content security score is a comprehensive indicator that reflects the security of the SMS obtained through a variety of detection methods. The score is compared with the preset threshold to determine the corresponding encryption strength level. The encryption strength level can be divided into several levels, such as high, medium, and low, and each level corresponds to different encryption requirements and strategies. According to the encryption strength level, the corresponding encryption algorithm is selected from the preset encryption algorithm library. The encryption algorithm library contains a variety of encryption algorithms, such as AES (Advanced Encryption Standard), DES (Data Encryption Standard), RSA (Rivest-Shamir-Adleman), etc. According to the encryption strength level, a suitable encryption algorithm is selected. For example, for high-intensity encryption requirements, a high-security encryption algorithm such as AES-256 or RSA-2048 is selected to obtain the corresponding encryption scheme. After determining the encryption scheme, the reconstructed SMS is processed in blocks, and the long SMS is divided into multiple small data blocks suitable for encryption to improve encryption efficiency and security. Assume that the SMS is divided into n data blocks, each data block is m bytes in size, and the resulting data block sequence is Each data block is randomly filled according to the encryption scheme to prevent the content characteristics of the data block from being analyzed and cracked. The filled data block can be expressed as ,in, is the random padding data. Based on the encryption algorithm selected in the encryption scheme, the padded data block is encrypted to obtain the encrypted data. Assuming that the selected encryption algorithm is AES-256, the encryption operation of each data block can be expressed as:
[0087] ;
[0088] in, is the encrypted data block, The AES-256 encryption algorithm is used. After the encrypted data is generated, a private key for digital signature is generated according to the preset key management policy. The key management policy is a specification to ensure the safe storage and use of keys, and the signing key is generated according to the policy. Assuming that the RSA algorithm is used for digital signature, the generated private key It will be used for signature operation. Perform hash calculation on the encrypted data to get the message digest. Hash calculation is an algorithm that converts input data of any length into fixed-length output. Commonly used hash algorithms include SHA-256. The hash calculation formula can be expressed as:
[0089] ;
[0090] in, is the message digest, is the set of all encrypted data blocks. According to the signature key, the message digest is asymmetrically encrypted to obtain a digital signature. Asymmetric encryption is a technology that uses public and private keys for encryption and decryption, where the private key is used for signing and the public key is used for verification. The generation of a digital signature can be expressed as:
[0091] ;
[0092] in, For digital signature, is the RSA encryption algorithm, The encrypted data and digital signature are combined and encapsulated to obtain an encrypted message. The encrypted message contains the encrypted SMS content and signature information to ensure the integrity and non-repudiation of the data. The combined encapsulation can be expressed as:
[0093] ;
[0094] in, For encrypted messages, including all encrypted data blocks and digital signature After merging and encapsulating, the encrypted message is formatted to obtain a secure transmission message so that the encrypted message meets the requirements of the transmission protocol and ensures that no errors occur during transmission in the network. For example, the encrypted message is converted into Base64 encoding format for transmission in a text protocol.
[0095] In a specific embodiment, the process of executing step S104 may specifically include the following steps:
[0096] Identify the protocol type of the secure transmission message to obtain the target protocol type, and select candidate gateways that support the protocol from the preset gateway library according to the target protocol type to obtain a preliminary gateway list;
[0097] Extract historical performance data for each gateway in the preliminary gateway list to obtain a gateway performance indicator set, and perform multi-dimensional scoring calculations on each gateway based on the gateway performance indicator set to obtain a gateway scoring matrix;
[0098] Normalize the gateway score matrix to obtain the standardized gateway score;
[0099] According to the preset gateway selection weight model, the standardized gateway score is weighted and calculated to obtain a comprehensive score;
[0100] Sort the comprehensive scores, select several gateways with the highest scores, and obtain the target gateway set;
[0101] According to the secure transmission message, load balancing calculation is performed on the target gateway set to obtain a load distribution plan;
[0102] The load distribution scheme is verified to obtain the fault-tolerant strategy, and the gateway selection result is generated based on the load distribution scheme and the fault-tolerant strategy.
[0103] Specifically, the protocol type of the secure transmission message is identified, and the communication protocol used is parsed from the message, such as HTTP, FTP, SMTP, etc. The target protocol type is determined by parsing the message header or other protocol features. According to the target protocol type, candidate gateways that support the protocol are screened out from the preset gateway library to obtain a preliminary gateway list. The gateway library is a database containing information about all available gateways and the protocols they support. According to the target protocol type, the gateway library is queried to screen out all gateways that support the protocol to form a preliminary gateway list. Historical performance data is extracted for each gateway in the preliminary gateway list to obtain a set of gateway performance indicators. Historical performance data includes the processing speed, response time, stability, failure rate, etc. of the gateway. The performance indicators of each gateway are extracted by querying historical records and monitoring data. Assume that the performance indicator set obtained is , where each It contains multiple specific indicators, such as throughput, latency, availability, etc. After obtaining the gateway performance indicator set, a multi-dimensional score calculation is performed on each gateway based on these indicators to obtain a gateway score matrix to comprehensively evaluate the performance of each gateway. Assume that the scoring formula used is:
[0104] ;
[0105] in, Indicates The gateway is The scores on the performance indicators are is the throughput score, To score for delay, Score for availability, is the weight of each indicator. In order to ensure the comparability of scores in different dimensions, the gateway score matrix is normalized to obtain the standardized gateway score. The values in the score matrix are converted to the same dimension, such as between 0 and 1. The normalization formula can be expressed as:
[0106] ;
[0107] in, is the standardized score, and Respectively The minimum and maximum values of each performance indicator. Based on the standardized gateway score, the standardized gateway score is weighted according to the preset gateway selection weight model to obtain a comprehensive score. The gateway selection weight model is preset according to actual needs and is used to balance the weights of various performance indicators. The comprehensive score calculation formula is:
[0108] ;
[0109] in, For the The comprehensive score of the gateways, For the The weight of each performance indicator is calculated. The gateways with the highest scores are sorted according to the comprehensive scores to obtain the target gateway set. The purpose of sorting is to select the gateway with the best performance to ensure the efficiency and reliability of transmission. Assuming that the top k gateways with the highest scores are selected, the target gateway set is According to the secure transmission message, the target gateway set is load balanced and the load distribution scheme is obtained. Load balancing is to evenly distribute the transmission tasks to multiple gateways to avoid overloading a single gateway. The load distribution scheme can be implemented by a weighted polling algorithm. Assume that the load weight of each gateway is , the load distribution ratio is:
[0110] ;
[0111] Verify the load distribution plan and obtain the fault-tolerant strategy. When a gateway fails, it can switch to other gateways in time to ensure uninterrupted transmission. The fault-tolerant strategy includes fault detection and automatic switching mechanisms. The feasibility and stability of the load distribution plan are verified through simulation tests and historical data analysis. The gateway selection results are generated based on the load distribution plan and the fault-tolerant strategy. The gateway selection results include the selected target gateway and its allocated load ratio, and the specific steps and conditions for fault-tolerant switching are clarified.
[0112] In a specific embodiment, the execution step performs weighted calculation on the standardized gateway score according to a preset gateway selection weight model to obtain a comprehensive score, which may specifically include the following steps:
[0113] Perform feature decomposition on the standardized gateway score to obtain a gateway performance feature vector, and sort the feature importance according to the gateway performance feature vector to obtain a feature priority list;
[0114] Perform hierarchical analysis on the feature priority list to obtain a multi-level evaluation framework, and group the gateway performance feature vectors according to the multi-level evaluation framework to obtain a hierarchical feature set;
[0115] Perform fuzzy evaluation on each feature in the hierarchical feature set to obtain a feature membership matrix, and calculate the weight coefficient of each layer of evaluation indicators based on the feature membership matrix to obtain a hierarchical weight vector;
[0116] Perform consistency check on the hierarchical weight vector to obtain the revised weight vector, and perform weighted summation on the hierarchical feature set based on the revised weight vector to obtain the comprehensive score of each layer;
[0117] The comprehensive scores of each layer are weighted and distributed among the layers to obtain a global weight coefficient, and the comprehensive scores of each layer are weighted and aggregated according to the global weight coefficient to obtain a comprehensive score.
[0118] Specifically, the standardized gateway score is subjected to eigendecomposition. Eigendecomposition is a method to decompose complex data into basic components, which can be achieved through principal component analysis or singular value decomposition. Assume that the standardized gateway score matrix is , the gateway performance eigenvector matrix can be obtained through eigendecomposition
[0119] ;
[0120] in, is a left singular matrix, is a diagonal matrix of singular values, is a right singular matrix. After eigendecomposition, the gateway performance eigenvector is obtained Represents the performance characteristics of each gateway. Sort the characteristics by importance according to the gateway performance characteristic vector. Sort by the contribution of each characteristic to the score to determine which characteristics are more critical to the gateway performance evaluation. Calculate the variance contribution rate of each characteristic, assuming that the characteristic vector is , its importance can be expressed as:
[0121] ;
[0122] in, Features The singular values of is the singular value of all features. Sort by feature importance to get a feature priority list. The feature priority list shows the order of importance of each feature, from most important to least important. Perform a hierarchical analysis on the feature priority list to get a multi-level evaluation framework. The analytic hierarchy process is a technique that decomposes complex decision problems into multiple levels. By building a hierarchical structure, features are divided into different levels. For example, the features are divided into three levels: performance, stability, and security, and each level is further divided into specific indicators. After obtaining the multi-level evaluation framework, the gateway performance feature vectors are grouped to get a hierarchical feature set. Each level contains several features. According to the hierarchical structure, the feature vectors are grouped. Grouping to form multiple hierarchical feature sets . Perform fuzzy evaluation on each feature in the hierarchical feature set to obtain the feature membership matrix. Fuzzy evaluation is to evaluate the features using fuzzy logic, indicating the membership of each feature to a certain performance indicator. Membership matrix Indicates the degree of membership of each feature under different evaluation indicators:
[0123] ;
[0124] in, Representation characteristics In the evaluation index According to the feature membership matrix, the weight coefficient of each layer of evaluation indicators is calculated to obtain the hierarchical weight vector. The weight coefficient represents the relative importance of each feature in the evaluation and can be calculated by the fuzzy comprehensive evaluation method. Assume that the hierarchical weight vector is , and its calculation formula is:
[0125] ;
[0126] in, For the The weight coefficient of the layer evaluation index, Features The weight of the layer weight vector is checked for consistency to obtain the revised weight vector. The consistency check is an important step to ensure the rationality of weight distribution, which can be checked by the consistency ratio (CR). If the CR value is less than 0.1, the weight distribution consistency is good, otherwise the weight needs to be adjusted. The revised weight vector The reasonable relationship between the features is guaranteed. According to the modified weight vector, the layered feature set is weighted and summed to obtain the comprehensive score of each layer. The weighted summation formula is:
[0127] ;
[0128] in, For the The overall rating of the layer, is the corrected weight vector, For the Layer feature vector. Perform inter-layer weight distribution on the comprehensive scores of each layer to obtain the global weight coefficient. Inter-layer weight distribution is to uniformly distribute the comprehensive scores of each layer and calculate the global weight coefficient. Assume that the global weight coefficient is , then the global weight distribution formula of each layer score is:
[0129] ;
[0130] According to the global weight coefficient, the comprehensive scores of each layer are weighted and aggregated to obtain the comprehensive score. The weighted aggregation formula is:
[0131] ;
[0132] in, It is the final comprehensive score, indicating the overall performance of the gateway.
[0133] In a specific embodiment, the execution step performs load balancing calculation on the target gateway set according to the secure transmission message, and the process of obtaining the load distribution solution may specifically include the following steps:
[0134] Extract features of secure transmission messages to obtain message attribute vectors, and perform compatibility evaluation on each gateway in the target gateway set based on the message attribute vectors to obtain a gateway adaptability matrix;
[0135] Normalize the gateway fitness matrix to obtain a standardized fitness score, and preliminarily screen the target gateway set based on the standardized fitness score to obtain a candidate gateway subset;
[0136] Perform real-time performance detection on each gateway in the candidate gateway subset to obtain the current load status of the gateway, and build a gateway load prediction model based on the current load status of the gateway to generate short-term load trends;
[0137] Perform threshold analysis on short-term load trends to obtain load risk assessment results, and dynamically assign weights to candidate gateway subsets based on the load risk assessment results to obtain a preliminary load plan;
[0138] The preliminary load plan is tested to obtain the plan performance indicators, and the preliminary load plan is optimized and adjusted based on the plan performance indicators to obtain the load distribution plan.
[0139] Specifically, feature extraction is performed on secure transmission messages, and feature values that can represent their attributes are extracted from the messages. These feature values include message size, protocol type, priority, encryption method, etc. By parsing the message content and header, the message attribute vector is obtained. , where each Represents a characteristic attribute. According to the message attribute vector, the compatibility of each gateway in the target gateway set is evaluated to obtain the gateway adaptability matrix. The compatibility evaluation is to determine whether each gateway is suitable for processing the message, and to match it according to the message attributes and the gateway capabilities. Assume that the target gateway set is , fitness matrix It can be expressed as:
[0140] ;
[0141] in, Indicates the gateway Attributes The fitness score is usually between 0 and 1. The gateway fitness matrix is normalized to convert the fitness scores to the same scale for easy comparison and screening, and a standardized fitness score is obtained. According to the standardized fitness score, the target gateway set is preliminarily screened to obtain a candidate gateway subset. The screening criteria can set a threshold, and only gateways with a fitness score higher than the threshold can enter the candidate gateway subset. Perform real-time performance detection on each gateway in the candidate gateway subset to obtain the current load status of the gateway. Obtain the current load status by monitoring the gateway's CPU usage, memory usage, network bandwidth and other indicators. Assume that the current load status of the gateway is ,in Indicates the gateway According to the current load status of the gateway, a gateway load prediction model is constructed to generate short-term load trends. The load prediction model can use time series analysis, machine learning and other methods to predict load changes in the future. Assume that the load trend prediction result is ,in Indicates the gateway The short-term load trend of the gateway is obtained based on the load trend. The load risk assessment result is obtained based on the load trend. The threshold analysis is to determine whether the load trend exceeds the preset safety threshold to assess the load risk of the gateway. According to the load risk assessment result, the candidate gateway subset is dynamically weighted to obtain a preliminary load plan. The gateway load is adjusted according to the load risk to balance the load of each gateway. Assume that the weight allocation result is ,in Indicates that it is assigned to the gateway The load weight of the preliminary load scheme is tested to obtain the performance index of the scheme. By simulating the actual transmission environment, the effectiveness and performance index of the preliminary load scheme are tested, such as transmission delay, throughput, packet loss rate, etc. Assume that the test result is ,in Indicates the gateway According to the performance indicators of the solution, the preliminary load solution is optimized and adjusted to obtain the final load distribution solution. Optimization adjustment means to refine and optimize the load distribution solution according to the test results to improve the overall performance. The optimized load distribution solution can be expressed as:
[0142] ;
[0143] in, Indicates the final allocation to the gateway The load, For performance indicators.
[0144] In a specific embodiment, the process of executing step S105 may specifically include the following steps:
[0145] A performance monitoring probe is set for each gateway in the gateway selection result to obtain a real-time performance data stream, and data cleaning and outlier detection are performed on each gateway based on the real-time performance data stream to obtain a valid performance indicator set;
[0146] Perform time series analysis on the effective performance indicator set, extract periodicity and trend characteristics, obtain the gateway performance mode, and build a gateway performance prediction model based on the gateway performance mode to generate short-term performance prediction results;
[0147] The short-term performance prediction results are scored in multiple dimensions to obtain the gateway real-time availability score. Based on the gateway real-time availability score, historical routing data is correlated and analyzed to obtain a routing effect evaluation report.
[0148] Perform data mining on the routing effect evaluation report to obtain the routing optimization direction, and design multiple candidate routing strategies based on the routing optimization direction to obtain the routing strategy set;
[0149] The routing policy set is tested to obtain a policy performance matrix, and a dynamic routing policy is generated based on the policy performance matrix.
[0150] Specifically, performance monitoring probes are set for each gateway in the gateway selection results. These probes are used to collect real-time performance data streams of the gateway, including key indicators such as CPU usage, memory usage, network bandwidth, and response time. Through monitoring probes, gateway performance data is obtained to ensure that the system can understand the operating status of each gateway in a timely manner. According to the real-time performance data stream, data cleaning and outlier detection are performed on each gateway. Data cleaning is to remove invalid or erroneous data to ensure the accuracy of the analysis results. Outlier detection is to identify and process anomalies in the data, which may be caused by network failures, burst traffic, etc. Through data cleaning and outlier detection, a valid performance indicator set is obtained. Time series analysis is performed on the valid performance indicator set to extract periodic and trend features. Time series analysis can use methods such as ARIMA (autoregressive integrated moving average model) to identify periodic fluctuations and long-term trends in data by analyzing historical performance data. The basic formula for time series analysis is as follows:
[0151] ;
[0152] in, Indicates time The data value of is a constant term, is the autoregression coefficient, is the sliding average coefficient, Based on the gateway performance pattern, the gateway performance prediction model is constructed as the error term to generate short-term performance prediction results. The prediction model can use machine learning algorithms, such as LSTM (Long Short-Term Memory Network), to predict the gateway performance in the future. Assume that the prediction result is , which represents the performance prediction value in the future. The short-term performance prediction results are scored in multiple dimensions to obtain the gateway real-time availability score. The multi-dimensional scoring comprehensively considers multiple performance indicators, such as response time, bandwidth utilization, etc., to evaluate the real-time availability of the gateway. The scoring formula can be expressed as:
[0153] ;
[0154] in, Indicates the gateway Real-time availability ratings, For the The weight of the performance indicator, For the The performance index is The predicted value at a point in time. According to the real-time availability score of the gateway, the historical routing data is correlated and analyzed to obtain a routing effect evaluation report. Correlation analysis is to evaluate the effects of different routing strategies by analyzing the relationship between the gateway availability score and the historical routing data. Data mining is performed on the routing effect evaluation report to obtain the routing optimization direction. Data mining is to identify the direction for improvement and potential optimization strategies by analyzing the data in the evaluation report. According to the routing optimization direction, multiple candidate routing strategies are designed to obtain a routing strategy set. The candidate routing strategies can be based on different optimization directions, for example, increasing bandwidth utilization, reducing response time, etc. The routing strategy set is tested to obtain a strategy performance matrix. The test is to evaluate the performance indicators of each candidate routing strategy, such as delay, throughput, packet loss rate, etc., by simulating the actual transmission environment. Assume that the strategy performance matrix is , indicating the The strategy in The dynamic routing strategy is generated based on the strategy performance matrix. The dynamic routing strategy is to dynamically select the optimal routing strategy based on the real-time network conditions and the strategy performance matrix. The formula for generating the dynamic routing strategy can be expressed as:
[0155] ;
[0156] in, Indicates at time The optimal routing strategy is selected. For the The weight of the performance indicator, For the The strategy in scores on a performance indicator.
[0157] In a specific embodiment, the process of executing step S106 may specifically include the following steps:
[0158] Analyze the dynamic routing strategy, extract the reply SMS capture rules, obtain the capture rule set, and set the gateway listener according to the capture rule set to monitor the user reply SMS in real time to obtain the original reply data stream;
[0159] Perform data cleaning and deduplication processing on the original reply data stream to obtain a valid reply SMS set, and perform protocol identification on each reply SMS based on the valid reply SMS set to obtain the reply SMS protocol type;
[0160] Classify and count the reply SMS protocol types, build protocol conversion rules, and perform protocol conversion on each reply SMS according to the protocol conversion rules to obtain reply content in a unified format;
[0161] Perform semantic analysis on the reply content in a unified format to obtain a reply SMS summary, and prioritize the reply content in a unified format according to the reply SMS summary to obtain a processing priority list;
[0162] A security check is performed on each reply in the processing priority list to obtain a security reply set, and each reply SMS is encapsulated with a target protocol based on the security reply set to obtain a forwarded SMS.
[0163] Specifically, the dynamic routing strategy is parsed to extract the reply SMS capture rules. Capture rules refer to the rules for extracting relevant information from the user's reply SMS, usually including keyword matching, regular expressions and other methods. By parsing the dynamic routing strategy, the capture rule set is obtained. , where each Represents a capture rule. According to the capture rule set, set the gateway listener to monitor the user's reply SMS in real time. The gateway listener is a tool used to capture and process network traffic. It can monitor the user's reply SMS in real time and collect the original reply data stream. , where each Represents an original reply data. The original reply data stream is cleaned and deduplicated to ensure the accuracy and integrity of the data. Data cleaning includes removing invalid data, correcting erroneous data, etc., while deduplication is to remove duplicate reply SMS messages. Through these processing steps, a valid reply SMS set is obtained. , where each Represents a valid reply SMS. Based on the valid reply SMS set, perform protocol identification on each reply SMS. Protocol identification refers to determining the communication protocol used by the reply SMS based on its content and format. Through protocol identification, the protocol type of the reply SMS is obtained. , where each Indicates the protocol type of a reply SMS. Classify and count the reply SMS protocol types and build protocol conversion rules. Classification statistics refers to counting the number of reply SMS messages of different protocol types in order to formulate corresponding protocol conversion rules. Assume that the protocol type statistics result is , where each Indicates the statistical results of a protocol type. Based on the statistical results, the protocol conversion rules are constructed. , where each Represents a conversion rule for a protocol type. According to the protocol conversion rule, perform protocol conversion on each reply message to unify the reply content in a unified format. Protocol conversion refers to converting reply messages of different protocol types into a unified format for subsequent processing. Assume that the reply content in a unified format is , where each Represents a reply message in a unified format. Perform semantic analysis on the reply content in a unified format to extract the reply message summary. Semantic analysis is to understand the content of the reply message and extract key information through natural language processing technology. Assume that the result of semantic analysis is , where each Represents a summary of a reply message. According to the summary of the reply message, the reply content in a unified format is prioritized. Priority sorting is to determine the processing priority based on the content and importance of the reply message. Assume that the result of priority sorting is , where each Indicates the processing priority of a reply message. Perform a security check on each reply in the processing priority list to ensure the security of the reply message. The security check includes virus scanning and sensitive word filtering on the reply message to ensure that it does not contain malicious content. After the security check, the safe reply set is obtained. , where each Represents a safe reply SMS. According to the safe reply set, each reply SMS is encapsulated in the target protocol for forwarding. Target protocol encapsulation means encapsulating the safe reply SMS in the format of the target protocol for easy transmission in the network. Assume that the encapsulated forwarded SMS is , where each Represents an encapsulated reply message.
[0164] The above describes the international SMS conversion domestic general method in the embodiment of the present application. The following describes the international SMS conversion domestic general system in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the international SMS conversion domestic general system includes:
[0165] The reconstruction module 201 is used to perform protocol analysis and content reconstruction on the received international SMS to obtain a reconstructed SMS;
[0166] Identification module 202, used to perform semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score;
[0167] The signature module 203 is used to encrypt and digitally sign the reconstructed SMS based on the content security score to obtain a secure transmission message;
[0168] The analysis module 204 is used to perform gateway adaptability analysis on the secure transmission message to obtain a gateway selection result;
[0169] Monitoring module 205, for real-time monitoring of gateway performance data and historical data analysis based on gateway selection results to obtain dynamic routing strategies;
[0170] The conversion module 206 is used to capture and convert the user's reply SMS in real time according to the dynamic routing strategy to obtain a forwarded SMS.
[0171] Through the collaboration of the above components, intelligent protocol parsing and content reconstruction, this method can adapt to the SMS protocol standards of different countries and regions, realize seamless conversion between multiple protocols, and greatly improve the compatibility and reliability of international SMS. By adopting deep semantic analysis and sensitive content identification technology, combined with a multi-dimensional content security scoring mechanism, it can more accurately identify potential risks and effectively prevent security threats such as spam SMS and fraudulent information. Through dynamically adjusted encryption processing and digital signature mechanisms, the encryption strength can be flexibly selected according to the content security score, while ensuring transmission security, the efficiency of computing resources is optimized. Based on multi-dimensional gateway adaptability analysis and real-time performance monitoring, this method can intelligently select the optimal gateway, significantly improving the success rate and stability of SMS transmission. Through real-time monitoring of gateway performance data and historical data analysis, the routing strategy can be dynamically adjusted to effectively respond to network fluctuations, improve transmission efficiency, and reduce operating costs. Using real-time capture and protocol conversion technology, it can quickly process user reply SMS, realize seamless connection of two-way communication, and improve the forwarding quality and security of international SMS.
[0172] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0173] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.
[0174] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A universal method for transferring international short messages to domestic short messages, characterized in that: The general method for transferring international SMS to domestic SMS includes: Perform protocol analysis and content reconstruction on the received international SMS to obtain the reconstructed SMS; Performing semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score; Based on the content security score, the reconstructed SMS is encrypted and digitally signed to obtain a secure transmission message; specifically including: performing threshold comparison on the content security score to obtain an encryption strength level, and selecting a corresponding encryption algorithm from a preset encryption algorithm library according to the encryption strength level to obtain an encryption scheme; performing block processing on the reconstructed SMS to obtain multiple data blocks, and randomly filling each data block according to the encryption scheme to obtain a filled data block; performing encryption operation on the filled data block based on the encryption algorithm selected in the encryption scheme to obtain encrypted data; generating a private key for digital signature according to a preset key management strategy to obtain a signature key, and performing hash calculation on the encrypted data to obtain a message digest; performing asymmetric encryption on the message digest according to the signature key to obtain a digital signature, and merging and encapsulating the encrypted data and the digital signature to obtain an encrypted message; formatting the encrypted message to obtain a secure transmission message; Performing gateway adaptability analysis on the secure transmission message to obtain a gateway selection result; Based on the gateway selection result, the gateway performance data is monitored in real time and historical data is analyzed to obtain a dynamic routing strategy; specifically, the following steps are performed: a performance monitoring probe is set for each gateway in the gateway selection result to obtain a real-time performance data stream, and data cleaning and outlier detection are performed on each gateway according to the real-time performance data stream to obtain a valid performance indicator set; a time series analysis is performed on the valid performance indicator set to extract periodicity and trend characteristics to obtain a gateway performance mode, and a gateway performance prediction model is constructed according to the gateway performance mode to generate a short-term performance prediction result; a multi-dimensional scoring is performed on the short-term performance prediction result to obtain a gateway real-time availability score, and according to the gateway real-time availability score, a correlation analysis is performed on the historical routing data to obtain a routing effect evaluation report; data mining is performed on the routing effect evaluation report to obtain a routing optimization direction, and according to the routing optimization direction, multiple candidate routing strategies are designed to obtain a routing strategy set; the routing strategy set is tested to obtain a strategy performance matrix, and a dynamic routing strategy is generated according to the strategy performance matrix; According to the dynamic routing strategy, the user's reply SMS is captured in real time and converted into a protocol to obtain a forwarded SMS; specifically, the method includes: parsing the dynamic routing strategy, extracting the reply SMS capture rules, obtaining a capture rule set, and setting a gateway listener according to the capture rule set to monitor the user's reply SMS in real time to obtain an original reply data stream; performing data cleaning and deduplication processing on the original reply data stream to obtain a valid reply SMS set, and performing protocol identification on each reply SMS according to the valid reply SMS set to obtain a reply SMS protocol type; classifying and counting the reply SMS protocol types, constructing protocol conversion rules, and performing protocol conversion on each reply SMS according to the protocol conversion rules to obtain reply content in a unified format; performing semantic analysis on the reply content in the unified format to obtain a reply SMS summary, and according to the reply SMS summary, prioritizing the reply content in the unified format to obtain a processing priority list; performing a security check on each reply in the processing priority list to obtain a secure reply set, and according to the secure reply set, performing target protocol encapsulation on each reply SMS to obtain a forwarded SMS.
2. The universal method for transferring international short messages to domestic short messages according to claim 1, characterized in that: The step of performing protocol parsing and content reconstruction on the received international SMS to obtain the reconstructed SMS includes: Performing protocol identification on the received international SMS to obtain the SMS protocol type, and decoding and character set conversion on the international SMS according to the SMS protocol type to obtain the original SMS content; Performing word segmentation and semantic annotation on the original SMS content to obtain structured SMS data, and extracting key information of the SMS content based on the structured SMS data to obtain a core content summary; Translating the core content summary into the language of the target country to obtain translated content, and performing SMS length analysis and segmentation based on the translated content to obtain segmented content that complies with the SMS length limit of the target country; Encoding the segmented content according to the target country SMS protocol to obtain encoding data that complies with the target country protocol, and generating SMS header information and control fields according to the encoding data to obtain a complete SMS structure; The complete SMS structure is verified for data integrity to obtain verified SMS data, and final protocol encapsulation and formatting are performed according to the verified SMS data to obtain a reconstructed SMS.
3. The universal method for transferring international short messages to domestic short messages according to claim 1, characterized in that: The performing semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score includes: Performing word segmentation processing on the reconstructed short message to obtain a word unit sequence, and performing part-of-speech tagging and named entity recognition on each word unit according to the word unit sequence to obtain a grammatical structure tree; Performing dependency syntactic analysis on the grammatical structure tree to obtain a sentence component relationship graph, and extracting the subject and keywords of the short message based on the sentence component relationship graph to obtain a semantic feature vector; Performing multilingual sentiment analysis on the semantic feature vector to obtain a sentiment polarity value, and performing fuzzy matching on the word sequence according to a preset sensitive word library to obtain the frequency of occurrence of sensitive words; Performing weighted calculation on the frequency of occurrence of the sensitive words to obtain a sensitivity index, and constructing a two-dimensional scoring matrix based on the sentiment polarity value and the sensitivity index to obtain an initial safety score; The initial security score is compared with historical data and detected for anomalies to obtain a normalized security score, and the SMS content is graded based on the normalized security score and a preset security level threshold to obtain a content security score.
4. The universal method for transferring international short messages to domestic short messages according to claim 1, characterized in that: The performing gateway adaptability analysis on the secure transmission message to obtain a gateway selection result includes: Performing protocol type identification on the secure transmission message to obtain a target protocol type, and based on the target protocol type, screening candidate gateways supporting the protocol from a preset gateway library to obtain a preliminary gateway list; Extracting historical performance data of each gateway in the preliminary gateway list to obtain a gateway performance indicator set, and performing multi-dimensional scoring calculation on each gateway based on the gateway performance indicator set to obtain a gateway scoring matrix; Normalizing the gateway score matrix to obtain a standardized gateway score; According to a preset gateway selection weight model, weighted calculation is performed on the standardized gateway score to obtain a comprehensive score; Sorting the comprehensive scores, selecting several gateways with the highest scores, and obtaining a target gateway set; According to the secure transmission message, load balancing calculation is performed on the target gateway set to obtain a load distribution plan; The load distribution scheme is verified to obtain a fault-tolerant strategy, and a gateway selection result is generated according to the load distribution scheme and the fault-tolerant strategy.
5. The universal method for transferring international short messages to domestic short messages according to claim 4, characterized in that: The standardized gateway score is weighted and calculated according to the preset gateway selection weight model to obtain a comprehensive score, including: Performing feature decomposition on the standardized gateway score to obtain a gateway performance feature vector, and sorting feature importance according to the gateway performance feature vector to obtain a feature priority list; Performing hierarchical analysis on the feature priority list to obtain a multi-level evaluation framework, and grouping the gateway performance feature vectors according to the multi-level evaluation framework to obtain a hierarchical feature set; Performing fuzzy evaluation on each feature in the hierarchical feature set to obtain a feature membership matrix, and calculating the weight coefficient of each layer of evaluation indicators according to the feature membership matrix to obtain a hierarchical weight vector; Performing consistency check on the hierarchical weight vector to obtain a revised weight vector, and performing weighted summation on the hierarchical feature set according to the revised weight vector to obtain a comprehensive score for each layer; The comprehensive scores of each layer are weighted and allocated among the layers to obtain a global weight coefficient, and the comprehensive scores of each layer are weighted and aggregated according to the global weight coefficient to obtain a comprehensive score.
6. The universal method for transferring international short messages to domestic short messages according to claim 4, characterized in that: The step of performing load balancing calculation on the target gateway set according to the secure transmission message to obtain a load distribution scheme includes: Extracting features of the secure transmission message to obtain a message attribute vector, and performing compatibility evaluation on each gateway in the target gateway set according to the message attribute vector to obtain a gateway adaptability matrix; Normalizing the gateway fitness matrix to obtain a standardized fitness score, and preliminarily screening the target gateway set according to the standardized fitness score to obtain a candidate gateway subset; Performing real-time performance detection on each gateway in the candidate gateway subset to obtain the current load status of the gateway, and constructing a gateway load prediction model to generate a short-term load trend based on the current load status of the gateway; Performing a threshold analysis on the short-term load trend to obtain a load risk assessment result, and dynamically assigning weights to the candidate gateway subsets based on the load risk assessment result to obtain a preliminary load plan; The preliminary load plan is tested to obtain a plan performance index, and the preliminary load plan is optimized and adjusted according to the plan performance index to obtain a load distribution plan.
7. A universal system for transferring international short messages to domestic short messages, characterized in that: Used to execute the international short message conversion domestic general method according to any one of claims 1 to 6, the international short message conversion domestic general system comprises: A reconstruction module is used to perform protocol analysis and content reconstruction on the received international SMS to obtain a reconstructed SMS; An identification module, used to perform semantic analysis and sensitive content identification on the reconstructed SMS to obtain a content security score; A signature module, used to encrypt and digitally sign the reconstructed SMS based on the content security score to obtain a secure transmission message; An analysis module, used to perform a gateway adaptability analysis on the secure transmission message to obtain a gateway selection result; A monitoring module, used to perform real-time monitoring and historical data analysis on gateway performance data based on the gateway selection result to obtain a dynamic routing strategy; The conversion module is used to capture and convert the user's reply SMS in real time according to the dynamic routing strategy to obtain the forwarded SMS.
