Method for realizing real-time interaction function in 5G new call agent

Abstract

The invention discloses a method for realizing a real-time interaction function in a 5G new call agent, and relates to the technical field of 5G communication, and the method comprises the following steps: obtaining user real-time interaction demand data based on a 5G network environment, the demand data including but not limited to an information interaction instruction in a call process and a call request for a specific function; and analyzing and classifying the obtained demand data, identifying different types of interaction demands, and generating a demand classification result. Real-time interaction demand data of a user is obtained in multiple modes, click and input operation information of an interaction interface of 5G terminal equipment is collected, voice instructions in the call process are collected in real time and converted into interaction demand information in a text form, and meanwhile interaction data request information of other associated equipment is received; the multi-modal interaction intention of the user in a 5G new call agent scene is comprehensively captured, and the defect that in the prior art, the interaction demand data acquisition mode is single is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of 5G communication technology, in particular to a method for realizing real-time interaction function in 5G new call agent. BACKGROUND

[0002] With the deep popularization of 5G technology and the rapid growth of intelligent interaction demand, 5G new call agent as a new carrier of communication and intelligent service puts forward higher requirements on the richness, accuracy and response speed of real-time interaction function. In the existing 5G call agent scene, the user's interaction demand data acquisition method still has the problem of single mode, which is difficult to fully capture the diversified demand in the interaction process of the agent. For example, in a complex intelligent service scene, only relying on traditional terminal interface click, input operation information collection cannot effectively identify the interaction intention expressed by the user through voice, associated intelligent devices and other multi-modal methods, resulting in deviation of the understanding of user demand by the agent.

[0003] At the same time, when the voice instruction is converted into interaction demand information, due to the need for the agent to process more complex context association and intention reasoning, the shortcomings of existing voice recognition technology in accuracy and real-time performance are further magnified, and part of the interaction demand cannot be analyzed and responded in time and accurately, affecting the fluency of interaction between the agent and the user and the service efficiency. In addition, in the collaborative interaction scene of the agent and multiple terminals and multiple service nodes, the adaptability, security and stability of data transmission are highlighted, and the existing technology cannot meet the high-level demand of 5G new call agent for real-time interaction function, limiting the further expansion of its service capacity. SUMMARY

[0004] To solve the above technical problems, a method for realizing real-time interaction function in 5G new call agent is provided, which solves the above problems.

[0005] To achieve the above purpose, the technical scheme adopted by the present application is:

[0006] The method for realizing real-time interaction function in 5G new call agent comprises the following steps:

[0007] Based on the 5G network environment, the user's real-time interaction demand data is obtained, and the demand data includes but is not limited to information interaction instruction in the call process and calling request for specific function;

[0008] The obtained demand data is analyzed and classified, different types of interaction demand are identified, and demand classification results are generated;

[0009] According to the demand classification results, the appropriate interaction data transmission channel is selected, and the transmission channel includes but is not limited to data channel based on IMS network and special data channel newly built by 5G.

[0010] The interaction data is transmitted through a selected transmission channel, and the transmission process is monitored and managed to generate a transmission status record;

[0011] At the receiving end, the transmitted interaction data is received, checked and analyzed to restore executable interaction instructions or information;

[0012] According to the analyzed interaction instructions or information, the corresponding real-time interaction operation is executed to complete the real-time interaction function in the 5G new call agent.

[0013] Preferably, the step of obtaining user real-time interaction requirement data is specifically:

[0014] Through the interaction interface of the 5G terminal device, the click and input operation information of the user during the call process and the interaction with the agent is collected;

[0015] The voice instructions issued by the user to the agent during the call process are collected in real time, and the voice recognition technology is used to convert them into text form interaction requirement information;

[0016] Receive data request information related to this call and agent interaction sent from other associated devices.

[0017] Preferably, the step of analyzing and classifying the obtained requirement data is specifically:

[0018] The semantic analysis algorithm is used to understand the semantics of the text information in the requirement data to determine the type and intention of the requirement, and the semantic understanding adopts the TF-IDF algorithm, which is calculated by the formula:

[0019]

[0020] The importance of the word in the document In the document , where represents the number of times the word appears in the document , and is the sum of the number of times all words appear in the document , and is the total number of documents in the document set, is the number of documents containing the word ;

[0021] According to the preset requirement classification rule, the requirement data is divided into corresponding categories, including but not limited to information query type, function control type and file transmission type.

[0022] Preferably, the step of selecting an adaptive interactive data transmission channel according to the demand classification result is specifically:

[0023] A correspondence table between transmission channels and demand types is established, and the table records transmission channel information adaptive to different demand types;

[0024] According to the demand classification result, the correspondence table is queried to determine an adaptive transmission channel;

[0025] If there are multiple adaptive transmission channels, then according to the current load condition of the transmission channel and the transmission priority factor, the selection score of each channel is calculated by the formula:

[0026]

[0027] wherein is the current load rate of the channel, is the preset priority of the channel, , , is a weight coefficient, represents other factor indicators affecting channel selection, is the weight of the corresponding factor, and the transmission channel with the highest selection score is selected.

[0028] Preferably, the step of monitoring and managing the transmission process is specifically:

[0029] A data transmission monitoring point is set to collect traffic, rate, and packet loss rate parameters of the transmission data in real time;

[0030] According to the preset transmission performance indicators, the collected parameters are compared and analyzed to determine whether the transmission state is normal;

[0031] If the transmission state is abnormal, a transmission optimization strategy is started, including but not limited to adjusting the transmission rate and switching the transmission channel. Abnormality is determined by the formula:

[0032]

[0033] wherein is the actual collected parameter value, is the preset indicator value, and when exceeds the threshold value, it is determined to be abnormal.

[0034] Preferably, the step of receiving, verifying, and analyzing the interactive data transmitted at the receiving end is specifically:

[0035] The interactive data transmitted is received through the network interface of the receiving device;

[0036] The received data is checked for integrity and accuracy by using a check algorithm, and if the check fails, retransmission is requested. The check uses a CRC check algorithm according to the formula:

[0037]

[0038] wherein is the original data polynomial, denotes a shift operation, is the quotient polynomial, is the generator polynomial, is the check code polynomial;

[0039] The data that passes the check is parsed using a parsing algorithm that matches the sending end, and the interactive instructions and information therein are extracted.

[0040] Preferably, the step of performing a corresponding real-time interactive operation according to the parsed interactive instructions or information is specifically:

[0041] A mapping relationship library of interactive instructions and execution operations is established, and the library stores specific operation methods corresponding to different interactive instructions;

[0042] According to the parsed interactive instructions, the mapping relationship library is queried to determine the corresponding execution operation;

[0043] The corresponding function module or application program is called to execute the operation, thereby realizing the real-time interaction function.

[0044] Preferably, it further includes the steps of encrypting and decrypting the interactive data during real-time interaction, specifically:

[0045] In the sending end, a suitable encryption algorithm is selected to encrypt the interactive data and generate encrypted data;

[0046] In the receiving end, a corresponding decryption algorithm is used to decrypt the encrypted data and restore it to the original interactive data.

[0047] Preferably, it further includes the steps of storing and analyzing the data and records during the real-time interaction after the real-time interaction ends, specifically:

[0048] The demand data, transmission state records, and parsing result information during the interaction process are stored in a database;

[0049] Data analysis tools are used to analyze the stored data, mine user interaction behavior habits, and transmission performance bottleneck information, thereby providing a basis for subsequent interaction function optimization.

[0050] Preferably, the 5G terminal device needs to perform network adaptation and capability negotiation before real-time interaction with the intelligent agent, specifically:

[0051] Detect the signal strength, bandwidth, and latency parameters of the current 5G network to determine whether the network meets the real-time interaction requirements of the intelligent agent;

[0052] Negotiate the interaction capabilities with the intelligent agent and the peer device to determine the supported interaction functions, data formats, and transmission protocol contents.

[0053] Compared with the prior art, the present application has the following advantages:

[0054] 1. The user real-time interaction requirement data is obtained through multiple ways, not only collecting the click and input operation information of the 5G terminal device interaction interface, but also collecting the voice instructions in the call process in real time and converting them into text form interaction requirement information, while receiving the interaction data request information of other associated devices, realizing the comprehensive capture of user multi-modal interaction intent in the 5G new call intelligent agent scenario, making up for the single defect of the existing technology interaction requirement data acquisition method, and adapting to the support requirements of the intelligent agent for diversified interaction scenarios.

[0055] 2. The voice instruction is converted into text information by real-time collection and voice recognition technology, which optimizes the process of converting voice instructions into interaction requirement information, effectively improves the accuracy and real-time performance of voice recognition compared with the prior art, so that the 5G new call intelligent agent can timely and accurately identify and respond to the user's interaction requirements, ensure the smoothness and service efficiency of the interaction between the intelligent agent and the user, and enhance the interaction experience of the intelligent agent.

[0056] 3. Precise requirement data analysis and classification can make the 5G new call intelligent agent more clearly understand the characteristics of different requirements, thereby providing strong support for selecting appropriate transmission channels, making data transmission more targeted, reducing unnecessary resource waste, and improving transmission efficiency. In the transmission process, real-time monitoring and management of the transmission state can timely discover problems and start optimization strategies, not only ensuring the stability of data transmission, but also further improving the accuracy of data transmission, avoiding data loss or errors, and improving the data security of intelligent agent interaction. In addition, after the interaction is completed, the storage and analysis of data can mine user behavior habits and transmission performance bottlenecks, provide data support for optimizing the voice recognition algorithm of the intelligent agent, improving the data acquisition method, and perfecting the transmission channel selection strategy, form a closed loop of continuous optimization, and help the 5G new call intelligent agent continuously improve the overall service capability of real-time interaction function. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 The flowchart of the present application. DETAILED DESCRIPTION

[0058] The following description is used to disclose the present application to enable a person skilled in the art to implement the present application. The preferred embodiments in the following description are only as examples, and other obvious modifications can be conceived by those skilled in the art.

[0059] Referring to Figure 1 As shown in the method for implementing real-time interaction function in 5G new call intelligent agent, comprising the following steps:

[0060] Based on 5G network environment, obtain user real-time interaction demand data, the demand data includes but is not limited to information interaction instruction in call process, calling request for specific function;

[0061] The obtained demand data is parsed and classified, different types of interaction demand are identified, and demand classification result is generated;

[0062] According to the demand classification result, select the adaptive interaction data transmission channel, the transmission channel includes but is not limited to the data channel based on IMS network, the special data channel newly built by 5G;

[0063] The interaction data is transmitted through the selected transmission channel, and the transmission process is monitored and managed, and the transmission state record is generated;

[0064] In the receiving end, the interaction data transmitted is received, checked and parsed, and restored to executable interaction instruction or information;

[0065] According to the parsed interaction instruction or information, the corresponding real-time interaction operation is executed, and the real-time interaction function in 5G new call intelligent agent is completed.

[0066] Specifically, in 5G network environment, the system first collects various demands of users interacting with intelligent agent in call process, which can be information interaction instruction, such as querying some information from intelligent agent, sending message, etc., or calling request for specific function, such as starting sharing screen through intelligent agent, sending file, etc. Then, in order to process these demands more efficiently, the system will analyze and classify the obtained demand data, and determine the type of different demands. Then, according to the classification result, the most suitable channel for the demand is selected from multiple available data transmission channels, such as data channel based on IMS network, which is suitable for some general voice and text interaction, while the special data channel newly built by 5G may be more suitable for large file transmission and other demands with higher bandwidth requirement. In the data transmission process, the system will monitor and manage in real time, record the transmission state. After the data arrives at the receiving end, it will be received, checked and parsed to ensure the accuracy and integrity of the data, and finally the corresponding interaction operation is executed according to the parsed instruction or information.

[0067] This comprehensive and orderly processing flow can ensure that various interactive needs of users and intelligent agents can be responded to in a timely and accurate manner. By classifying the demand data and selecting the transmission channel, the efficiency and reliability of data transmission are improved, and transmission delay or errors caused by improper channel selection are avoided. At the same time, the monitoring and management of the transmission process and the verification and analysis of the receiving end further guarantee the quality of the data, thereby improving the overall performance and user experience of the real-time interaction function in the 5G new call intelligent agent.

[0068] The step of acquiring user real-time interaction demand data is specifically:

[0069] Through the interactive interface of the 5G terminal device, the click and input operation information of the user during the call process is collected.

[0070] Real-time collection of voice commands issued by the user to the intelligent agent during the call process, and conversion of the voice commands into text form of interaction demand information using voice recognition technology.

[0071] Receive data request information related to the current call and intelligent agent interaction sent from other associated devices.

[0072] Specifically, through the interactive interface of the 5G terminal device, the click and input operation of the user during the call process is captured, for example, the user clicks a function button or inputs text content on the interactive interface of the intelligent agent, and these operation information is collected by the system; At the same time, the system will collect the voice commands issued by the user to the intelligent agent during the call process in real time, and convert the voice content into text form of interaction demand information using advanced voice recognition technology. In addition, the system will also receive interaction data request information related to the current call and intelligent agent interaction sent from other associated devices, which can be smart watches, tablet computers, etc.

[0073] This multi-channel data acquisition method can comprehensively and accurately collect the interaction needs of users and intelligent agents. Not only the traditional operation mode of users is considered, but also the convenience of voice interaction and the possibility of collaborative work of associated devices and intelligent agents are fully utilized. This can meet the interaction needs of different users with intelligent agents in different scenarios, improve the perception ability of the system to user needs, and lay a foundation for subsequent accurate processing and response.

[0074] The step of analyzing and classifying the acquired demand data is specifically:

[0075] Use semantic analysis algorithm to understand the semantics of the text information in the demand data, determine the type and intention of the demand, and use TF-IDF algorithm for semantic understanding, through the formula:

[0076]

[0077] Calculate words in text In the document The importance of it, among which Words In the document The number of times it appears in For document The sum of the number of times all words appear in the text. It is the total number of documents in the document collection. It contains words The number of documents;

[0078] Based on the preset demand classification rules, the demand data is divided into corresponding categories, including but not limited to information query, function control, and file transfer.

[0079] Specifically, semantic analysis algorithms are used to perform in-depth semantic understanding of the textual information in the demand data. By analyzing the meaning and context of the text, the type of user's interaction with the intelligent agent and their intent are determined. For example, it can be determined whether the user wants to query information from the intelligent agent, control a function through the intelligent agent, or transfer files. Then, according to pre-defined demand classification rules, the demand data is divided into corresponding categories, such as information query, function control, and file transfer.

[0080] Accurate parsing and classification enable the system to better understand the user's interaction needs with the agent, thus providing a clear direction for subsequent processing. Different types of needs can employ different processing strategies and transmission channels, improving processing efficiency and targeting. For example, for information query needs, the agent can quickly access the corresponding information database for searching; for file transfer needs, a transmission channel with higher bandwidth can be prioritized.

[0081] The specific steps for selecting an appropriate interactive data transmission channel based on the demand classification results are as follows:

[0082] Establish a table that maps transmission channels to demand types, and record the transmission channel information that is suitable for different demand types in the table;

[0083] Based on the requirements classification results, query the corresponding relationship table to determine the suitable transmission channel;

[0084] If multiple suitable transmission channels exist, the selection is based on the current load of the transmission channels and transmission priority factors, using the formula:

[0085]

[0086] Calculate the selection score for each channel, where the current load rate of the channel, preset priority of the channel, , , weight coefficient, other factor indicators affecting channel selection, weight of the corresponding factor, the transmission channel with the highest selection score is selected.

[0087] Specifically, a pre-established correspondence table of transmission channels and demand types is established, which records detailed information of suitable transmission channels for different demand types. After obtaining the demand classification result, the system queries this correspondence table to find the suitable transmission channel. If there are multiple suitable transmission channels, the system will consider the current load situation and transmission priority of the transmission channel, etc. For example, a channel with lower current load may be more suitable for data transmission, and a channel with higher transmission priority will be selected first to ensure efficient response to user-agent interaction requirements.

[0088] By selecting the transmission channel in this way, the advantages of different transmission channels can be fully utilized to improve the efficiency and stability of data transmission. This avoids the congestion and transmission delay caused by concentrating all data on a single channel. At the same time, considering the transmission priority can ensure that important data can be transmitted first, ensuring timely response to critical user-agent interaction requirements.

[0089] The step of monitoring and managing the transmission process is specifically:

[0090] Setting up data transmission monitoring points to collect real-time transmission data flow, rate, and packet loss rate parameters;

[0091] Comparing and analyzing the collected parameters according to the preset transmission performance indicators to determine whether the transmission state is normal;

[0092] If the transmission state is abnormal, start the transmission optimization strategy, including but not limited to adjusting the transmission rate, switching the transmission channel, and the abnormality is determined by the formula:

[0093]

[0094] wherein is the actual collected parameter value, is the preset indicator value, when exceeds the threshold value, it is determined to be abnormal.

[0095] Specifically, by setting multiple monitoring points on the data transmission path, the traffic, rate, and packet loss rate of the transmitted data are collected in real time. Then, the collected parameters are compared and analyzed with the preset transmission performance indicators to determine whether the transmission state is normal. If the actual collected parameters differ greatly from the preset indicators, for example, the packet loss rate is too high or the transmission rate is too low, it is determined that the transmission state is abnormal, and the transmission optimization strategy is immediately started, such as adjusting the transmission rate, switching to other transmission channels, etc., to ensure the stable transmission of user interaction data with the intelligent agent.

[0096] Real-time monitoring and management can timely discover problems occurring in the transmission process and take effective measures to solve them. This can ensure the stability and reliability of data transmission, reduce data loss or delay caused by transmission failure, and thus improve the real-time interaction experience of users in the 5G new call intelligent agent.

[0097] The step of receiving, checking, and analyzing the interaction data transmitted at the receiving end is specifically:

[0098] The interaction data transmitted is received through the network interface of the receiving device;

[0099] The received data is checked for integrity and accuracy using a checking algorithm, and if the check fails, retransmission is requested. The CRC checking algorithm is used, according to the formula:

[0100]

[0101] wherein is the original data polynomial, denotes a shift operation, is the quotient polynomial, is the generator polynomial, is the check code polynomial;

[0102] An analysis algorithm matching the sending end is used to analyze the data that passes the check, extracting the interaction instructions and information therein.

[0103] Specifically, the receiving device receives the data related to intelligent agent interaction through its network interface, and after receiving the data, checks the integrity and accuracy of the data using a checking algorithm. For example, check if the data has been lost, damaged, etc. during transmission. If the check fails, the receiving device will request the sending end to retransmit the data. After the check passes, an analysis algorithm matching the sending end is used to analyze the data, extracting the instructions and information related to intelligent agent interaction therein.

[0104] The checksum analysis mechanism of the receiving end can ensure the accuracy of the received data and avoid interaction failure caused by data errors. The request retransmission function can compensate for data loss during transmission and ensure data integrity. The matching analysis algorithm can correctly interpret the instructions and information in the data, providing a reliable basis for the agent to perform subsequent interaction operations.

[0105] The step of executing the corresponding real-time interaction operation according to the parsed interaction instruction or information is specifically:

[0106] A mapping relationship library of interaction instructions and execution operations is established, and the mapping relationship library stores specific operation methods corresponding to different interaction instructions;

[0107] According to the parsed interaction instruction, the mapping relationship library is queried to determine the corresponding execution operation;

[0108] The corresponding function module or application program is called to execute the operation, and the real-time interaction function is realized.

[0109] Specifically, through the pre-established mapping relationship library of interaction instructions and execution operations, the mapping relationship library stores specific operation methods corresponding to different interaction instructions. When the interaction instruction or information related to the agent is parsed, the mapping relationship library is queried to find the corresponding execution operation. Then, the corresponding function module or application program is called to execute the operation, thereby realizing the real-time interaction function with the agent.

[0110] The establishment of this mapping relationship library can quickly and accurately convert the parsed instructions into actual operations of the agent. By calling the corresponding function module or application program, the modularity and scalability of the system are realized. The interaction instruction of the user and the agent can be responded in a timely manner, improving the efficiency and accuracy of real-time interaction in the 5G new call agent.

[0111] It also includes the steps of encrypting and decrypting the interaction data during real-time interaction, specifically:

[0112] In the sending end, a suitable encryption algorithm is selected to encrypt the interaction data to generate encrypted data;

[0113] In the receiving end, a corresponding decryption algorithm is used to decrypt the encrypted data to restore the original interaction data.

[0114] Specifically, in the sending end, a suitable encryption algorithm is selected to encrypt the data interacting with the agent. The encryption algorithm will convert the original data into ciphertext, so that the data can be intercepted during transmission, but it cannot be easily interpreted. In the receiving end, a decryption algorithm corresponding to the sending end is used to decrypt the encrypted data to restore the original interaction data.

[0115] The data encryption and decryption mechanism ensures the security and privacy of the data interaction between the user and the agent. In the 5G network environment, the data transmission volume is large and involves sensitive information of the user, such as personal privacy and business secrets. Through encryption processing, the data can be prevented from being stolen or tampered with during transmission, protecting the legitimate rights and interests of the user and enhancing the user's trust in the real-time interaction function of the 5G new call agent.

[0116] After the real-time interaction ends, the data and records during the interaction are stored and analyzed. The specific steps are as follows:

[0117] The demand data, transmission state records, and analysis result information during the interaction are stored in the database.

[0118] Using data analysis tools, the stored data is analyzed to mine the user's interaction behavior habits and transmission performance bottleneck information, providing a basis for subsequent interaction function optimization.

[0119] Specifically, by storing the demand data, transmission state records, and analysis result information during the interaction in the database, and then using professional data analysis tools to analyze the stored data, the user's interaction behavior habits with the agent can be mined, such as the functions the user often uses through the agent, the time regularity of interaction, etc. Transmission performance bottlenecks can also be found, such as which transmission channels are prone to problems, the main reasons for transmission delay, etc., providing direction for the iterative optimization of the agent.

[0120] The storage and analysis of interaction data can provide a strong basis for subsequent interaction function optimization. According to the user's interaction behavior habits, the functions and interfaces of the agent can be personalized optimized to improve the user experience. For transmission performance bottlenecks, appropriate improvement measures can be taken, such as optimizing the transmission channel, adjusting the transmission strategy, etc., to continuously improve the performance and quality of the real-time interaction function in the 5G new call agent.

[0121] Before the 5G terminal device interacts with the agent in real time, network adaptation and capability negotiation steps are also required, specifically:

[0122] Detect the signal strength, bandwidth, and delay parameters of the current 5G network to determine whether the network meets the requirements of real-time interaction with the agent.

[0123] Negotiate the interaction capabilities with the agent and the opposite device to determine the supported interaction functions, data formats, and transmission protocol content of each party.

[0124] Specifically, the 5G terminal device will detect the signal strength, bandwidth, and latency of the current 5G network, and determine whether the network meets the real-time interaction requirements with the agent based on these parameters. For example, if the signal strength is too weak, the bandwidth is insufficient, or the latency is too long, it may affect the quality of real-time interaction with the agent. At the same time, the terminal device will negotiate the interaction capabilities with the agent and the peer device to determine the interaction functions, data formats, and transmission protocols supported by each party. For example, each party needs to confirm whether they all support high-definition video calls through the agent, whether they use the same data compression format, etc.

[0125] Network adaptation and capability negotiation can ensure that the network environment and the capabilities of each party's device are in the best state before real-time interaction with the agent. By judging in advance whether the network meets the requirements, it can avoid problems such as lag and delay during the interaction process. Capability negotiation can ensure that the agent and the terminal device, the peer device can be compatible with each other and smoothly perform various interaction operations, improving the success rate and stability of real-time interaction in the 5G new call agent.

[0126] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements fall within the scope of the claimed present application.

Claims

A method for implementing real-time interaction functions in a 1.5G new intelligent calling agent, characterized in that, Includes the following steps: Based on the 5G network environment, real-time user interaction demand data is obtained, including but not limited to information interaction instructions during the call and requests to call specific functions. The acquired requirement data is parsed and classified to identify different types of interaction requirements and generate requirement classification results. Based on the requirements classification results, a suitable interactive data transmission channel is selected. The transmission channel includes, but is not limited to, a data channel based on the IMS network and a dedicated data channel newly built for 5G. The interactive data is transmitted through the selected transmission channel, and the transmission process is monitored and managed to generate a transmission status record. At the receiving end, the transmitted interactive data is received, verified, and parsed to restore it into executable interactive instructions or information; Based on the parsed interactive instructions or information, execute the corresponding real-time interactive operation to complete the real-time interactive function in the 5G new call intelligent agent.

2. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for obtaining real-time user interaction demand data are as follows: Through the interactive interface of 5G terminal devices, click and input operation information of users when interacting with the intelligent agent during a call is collected; The system collects voice commands issued by users to the intelligent agent in real time during the call and uses speech recognition technology to convert them into text-based interactive requirement information. Receive data request information related to the current call and intelligent agent interaction from other associated devices.

3. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for parsing and classifying the acquired demand data are as follows: Semantic analysis algorithms are used to perform semantic understanding on the textual information in the demand data to determine the type and intent of the demand. The semantic understanding uses the TF-IDF algorithm, through the formula: ; Calculate words in the text In the document The importance of it, among which Words In the document The number of times it appears in For document The sum of the number of times all words appear in the text. It is the total number of documents in the document collection. It contains words The number of documents; Based on the preset demand classification rules, the demand data is divided into corresponding categories, including but not limited to information query, function control, and file transfer.

4. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for selecting an appropriate interactive data transmission channel based on the demand classification results are as follows: Establish a table that maps transmission channels to demand types, and record the transmission channel information that is suitable for different demand types in the table; Based on the requirements classification results, query the corresponding relationship table to determine the suitable transmission channel; If multiple suitable transmission channels exist, the selection is based on the current load of the transmission channels and transmission priority factors, using the formula: ; Calculate the selection score for each channel, where The current load rate of the channel. Preset priority for the channel. , , These are the weighting coefficients. Other factors influencing channel selection, The transmission channel with the highest score is selected based on the weight of the corresponding factors.

5. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for monitoring and managing the transmission process are as follows: Set up data transmission monitoring points to collect parameters such as data flow, rate, and packet loss rate in real time. Based on the preset transmission performance indicators, the collected parameters are compared and analyzed to determine whether the transmission status is normal. If the transmission status is abnormal, a transmission optimization strategy will be activated, including but not limited to adjusting the transmission rate and switching the transmission channel. The abnormality is determined using the following formula: ;in These are the actual collected parameter values. For the preset indicator value, when If the threshold is exceeded, it is considered abnormal.

6. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for receiving, verifying, and parsing the transmitted interactive data at the receiving end are as follows: The device receives the transmitted interactive data through its network interface. The received data is checked for integrity and accuracy using a verification algorithm. If the verification fails, a retransmission is requested. The verification uses the CRC checksum algorithm, based on the formula: ;in The original data is a polynomial. Indicates a shift operation. For quotient polynomial, To generate polynomials, It is a check code polynomial; A parsing algorithm matching the sending end is used to parse the verified data and extract the interaction commands and information.

7. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, The specific steps for executing the corresponding real-time interactive operation based on the parsed interactive instructions or information are as follows: Establish a mapping library between interactive commands and execution operations, storing the specific operation methods corresponding to different interactive commands; Based on the parsed interactive instructions, query the mapping database to determine the corresponding execution operation; The corresponding functional modules or applications are invoked to perform the operation and realize the real-time interactive function.

8. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, It also includes the steps of encrypting and decrypting interactive data during real-time interaction, specifically: At the sending end, a suitable encryption algorithm is selected to encrypt the interactive data and generate encrypted data. At the receiving end, the corresponding decryption algorithm is used to decrypt the encrypted data and restore it to the original interactive data.

9. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, This also includes the steps of storing and analyzing the data and records from the interaction process after the real-time interaction ends, specifically: The required data, transmission status records, and parsing results during the interaction process are stored in the database. By using data analysis tools, we can analyze the stored data, uncover user interaction habits and transmission performance bottlenecks, and provide a basis for subsequent optimization of interactive functions.

10. The method for implementing real-time interaction function in the 5G new call intelligent agent according to claim 1, characterized in that, Before the 5G terminal device can interact with the intelligent agent in real time, it also needs to perform network adaptation and capability negotiation steps, specifically: Detect the signal strength, bandwidth, and latency parameters of the current 5G network to determine whether the network meets the requirements for real-time interaction with the intelligent agent; Negotiate interaction capabilities with intelligent agents and peer devices to determine the interactive functions, data formats, and transmission protocol content supported by each party.

Images