SIP-based IPPBX voice switching management system and method

Abstract

This application relates to the field of communication technology, specifically disclosing a SIP-based IPPBX voice switching management system and method. The method includes: authenticating users and assigning internal contact numbers through a voice management terminal; generating call paths based on a calling and called number matching routing strategy, and optimizing the SIP paths using a machine learning model; dynamically allocating media resources according to the call path and network conditions, initiating RTP stream transmission of voice data, and dynamically adjusting voice compression parameters; continuously monitoring the session status during the call and triggering corresponding processing mechanisms based on the voice session status. This application provides a secure, efficient, and intelligent IPPBX voice switching management method.

Description

Technical Field

[0001] This invention relates to the field of communication technology, specifically to a SIP-based IPPBX voice switching management system and method. Background Technology

[0002] Internet Protocol (IP) based voice communication systems, also known as IP telephony systems or IPPBX systems, offer greater flexibility, scalability, and cost-effectiveness compared to traditional circuit-switched technologies. However, existing IPPBX systems still face some problems and challenges in practical applications.

[0003] User authentication and account management are critical issues facing IPPBX systems. Ensuring that only authorized users can access the system and effectively managing user accounts are key to guaranteeing system security and reliable operation. In actual voice communication, maintaining call quality is a significant challenge. Dynamic changes in network conditions, signaling path selection, and media resource scheduling all affect call quality. Achieving high-quality, low-latency voice communication in complex and ever-changing network environments is a core issue that IPPBX systems need to address. This also places higher demands on the operation and maintenance management of IPPBX systems. Real-time monitoring of system operation status, timely detection and handling of anomalies, and system optimization and adjustment are also urgent problems to be solved.

[0004] How to build a secure, reliable, high-quality, efficient, and easy-to-manage IPPBX voice communication system has become a key technical problem that urgently needs to be solved.

[0005] In view of this, this application proposes a SIP-based IPPBX voice switching management system and method. Summary of the Invention

[0006] To achieve the above objectives, this invention provides a SIP-based IPPBX voice switching management system and method, the specific technical solution of which is as follows:

[0007] SIP-based IPPBX voice switching management methods include:

[0008] The voice management terminal receives registration requests from user terminals, verifies user identity by combining user account and device information, assigns internal contact numbers to new users, and establishes an online status table;

[0009] The voice management terminal generates call paths based on the calling and called number matching routing strategy; at the same time, it applies machine learning models combined with QoS indicators to predict network congestion points and dynamically optimize SIP paths.

[0010] Media resources are dynamically allocated based on the call path and network conditions, RTP streams are initiated and voice data is transmitted, and voice compression parameters are dynamically adjusted based on network bandwidth and packet loss rate.

[0011] During a call, the system continuously monitors the voice session status, including SIP signaling flow, call hold duration, and user hang-up signal information, and triggers corresponding processing mechanisms based on the voice session status.

[0012] Preferably, after receiving a registration request from a user terminal, the voice management terminal performs identity verification;

[0013] A certificate-based two-way authentication mechanism is adopted. When the user terminal initiates a registration request message, it exchanges certificates with the voice management terminal and verifies the validity of the certificates.

[0014] After the certificate verification is successful, the voice management terminal reviews the user account and determines whether the user account has legitimate registration permissions by matching it with the rules in the management policy.

[0015] The matching process between account attributes and management policies is abstracted as a set operation problem, namely, whether the intersection of the set of account attributes and the set of management policy rules is empty; a new user's registered account is determined to be a valid account if and only if the intersection of the set of account attributes and the set of management policy rules is not empty.

[0016] Preferably, for new user accounts that have passed the review, the voice management terminal assigns a unique internal contact number to the new user account;

[0017] The internal contact number is numbered using a segmented representation method, including the department number segment, the user type segment, and the user serial number segment;

[0018] After assigning an internal contact number, the voice management terminal creates a new user's online status table in memory. The online status table records user account, internal contact number, IP address, port number, registration time, and expiration time information, and is organized using a hash index with the user account as the primary key to enable fast querying and updating of status information.

[0019] Preferably, after receiving a call request initiated by a user, the voice management terminal performs routing strategy matching based on the calling and called numbers;

[0020] The routing strategy consists of predefined matching rules, each rule including number pattern, call permissions, and routing destination;

[0021] The matching process is abstracted into a multi-stage longest prefix matching problem, which involves searching the routing table for the routing rule that matches the longest prefix of the calling and called numbers.

[0022] Preferably, after matching a suitable routing rule, the voice management terminal generates a call path based on the routing destination;

[0023] The call path includes the call signaling path and the media transmission path; the call signaling path serves as the transmission path for voice messages; the media transmission path serves as the transmission path for voice data packets.

[0024] A network quality prediction model based on the random forest algorithm is introduced at the voice management terminal. The network quality prediction model based on the random forest algorithm is constructed, and the network quality feature vector is constructed.

[0025] In the random forest algorithm, each decision tree takes the network quality feature vector as input, and after multiple decision tree node partitioning and attribute judgment, a health score is given at the leaf node.

[0026] The path with the highest health score is selected as the optimal path for the call.

[0027] Preferably, after the voice management terminal selects the optimal path as the call path, it dynamically allocates and manages media resources; the media resources include media processors, transmission bandwidth, and codecs.

[0028] The voice management terminal calculates the media resource requirements based on the network conditions and load of each node in the call path;

[0029] After obtaining the media resource requirements of each node, the voice management terminal selects the optimal resource allocation scheme from the media resource pool and dynamically allocates the resources to each node.

[0030] Preferably, the optimal resource allocation problem is modeled as an integer programming model;

[0031] The goal of the integer programming model is to minimize the total resource allocation cost while satisfying the media resource requirements of each node; solving the integer programming model yields the optimal resource allocation scheme.

[0032] Preferably, after completing the media resource allocation, the voice management terminal notifies each node on the call path to start the corresponding media processor and codec according to the allocation scheme, and establishes a real-time transmission protocol stream to transmit voice data.

[0033] Simultaneously monitor network bandwidth and packet loss rate in real time at the voice management terminal, construct a dynamic encoding and decoding parameter adjustment algorithm, and dynamically adjust media encoding and decoding parameters;

[0034] Dynamic codec parameter adjustment algorithms find the optimal combination of codec parameters under the constraints of network bandwidth and transmission delay, so as to maximize the codec quality.

[0035] After obtaining the optimal combination of encoding and decoding parameters, the results are distributed to each node along the call path to guide the nodes in dynamically adjusting their encoding and decoding behavior.

[0036] Preferably, the voice management terminal monitors the call process based on a finite state machine model. By tracking the call hold duration, it calculates the average hold duration and variance to detect abnormal call patterns. After listening to the user's hang-up signal, it initiates the session cleanup process.

[0037] The voice management terminal performs anomaly detection periodically, including analyzing SIP message logs, checking media transmission quality, monitoring system resource usage, and verifying user account information. Based on the detection results, it automatically executes recovery measures and / or generates alarm information.

[0038] The SIP-based IPPBX voice switching management system, used in the aforementioned SIP-based IPPBX voice switching management method, includes: a user registration management module, a call path optimization module, a network resource adjustment module, and a call monitoring and adjustment module.

[0039] The user registration management module receives registration requests from user terminals through a voice management terminal, verifies user identity by combining user account and device information, assigns internal contact numbers to new users, and establishes an online status table.

[0040] The call path optimization module generates call paths based on the calling and called number matching routing strategy at the voice management end; at the same time, it uses a machine learning model combined with QoS indicators to predict network congestion points and dynamically optimize SIP paths.

[0041] The network resource adjustment module dynamically allocates media resources based on the call path and network conditions, initiates RTP streams and transmits voice data, and dynamically adjusts voice compression parameters based on network bandwidth and packet loss rate.

[0042] The call monitoring and adjustment module is used to continuously monitor the voice session status during a call, including SIP signaling flow, call hold duration and user hang-up signal information, and trigger corresponding processing mechanisms based on the voice session status.

[0043] The beneficial effects of this invention are as follows: This application ensures communication security by combining user account and device information for identity verification; assigning internal contact numbers and establishing an online status table helps manage user status and improves system organization and controllability.

[0044] This application achieves efficient routing by matching the calling and called numbers, and at the same time uses machine learning models to dynamically predict network congestion and optimize SIP paths, which can significantly improve call quality and network resource utilization efficiency.

[0045] This application dynamically allocates media resources and adjusts voice compression parameters based on real-time network conditions, which can effectively reduce packet loss and latency, ensure the stability and clarity of voice data transmission, and improve user experience.

[0046] This application continuously monitors session status and responds to event changes, effectively supporting call hold, anomaly detection, and call termination handling, thereby enhancing the system's intelligence and service continuity and reliability.

[0047] The method of this application integrates the SIP protocol and intelligent optimization mechanism to realize full-process voice exchange management from registration, routing, transmission to monitoring, effectively improving the system's communication quality, resource utilization and service intelligence level. Attached Figure Description

[0048] Figure 1 A flowchart of the SIP-based IPPBX voice switching management method provided by the present invention;

[0049] Figure 2 The user registration and number allocation flowchart provided by this invention;

[0050] Figure 3 The flowchart for route matching and call optimization provided by this invention;

[0051] Figure 4 The media resource management and adaptation flowchart provided by this invention;

[0052] Figure 5 The flowchart for call anomaly detection and recovery provided by this invention;

[0053] Figure 6 The diagram shows the structure of the SIP-based IPPBX voice switching management system provided by this invention. Detailed Implementation

[0054] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0055] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention can also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0057] Example 1:

[0058] Reference Figure 1 This is the first embodiment of the present invention, which provides a SIP-based IPPBX voice switching management method.

[0059] Step 1: The voice management terminal receives the registration request from the user terminal, verifies the user's identity by combining the user account and device information, assigns an internal contact number to the new user, and establishes an online status table; see [link / reference] Figure 2 This is a flowchart of the user registration and number allocation process for this step.

[0060] After receiving a registration request from a user terminal, the voice management terminal first performs identity verification. To improve security, a certificate-based two-way authentication mechanism is adopted. When the user terminal initiates a SIP registration message (SIP REGISTER), it must carry a digital certificate issued by a trusted CA, which contains the user's public key information. At the same time, the voice management terminal also holds a server certificate issued by a trusted CA. The two parties ensure the authenticity and trustworthiness of their identities by exchanging certificates and verifying their legality.

[0061] It should be noted that the Session Initiation Protocol (SIP) is an application layer control protocol mainly used to establish, modify, and terminate multimedia sessions between one or more participants. The SIP protocol is developed by the IETF and its design follows the principles of modularity, scalability, and compatibility with existing Internet protocols. It is widely used in IP-based voice communication (VoIP), video conferencing, instant messaging, and multimedia distribution voice management.

[0062] The fundamental function of SIP is to provide signaling and control mechanisms for communication between users. The SIP protocol defines a series of message formats and request methods to implement processes such as user location lookup, session parameter negotiation, user reachability determination, session establishment and management, and session transfer and termination. SIP uses text format for message exchange, and its structure is similar to the HTTP protocol, offering good readability and scalability. The SIP REGISTER message is used by terminal devices to register their network address information with a SIP registration server, allowing other users to locate the terminal using this identifier.

[0063] It should be noted that an IPPBX voice switching system is an enterprise-level voice communication management system based on the IP protocol, designed to enable voice call connection, management, and control between internal and external entities. Unlike traditional analog or digital PBX systems, IPPBX utilizes IP networks to carry voice signals and uses protocol stacks (such as SIP or RTP) to achieve signaling control and media transmission of voice data, thus combining the advantages of telephone switching and data network convergence.

[0064] After certificate verification, the voice management terminal further reviews the user account by checking its attributes, including account type, department, and level, and matching them with the rules in the management policy to determine whether the user account has legitimate registration permissions. This matching process between account attributes and management policies is abstracted as a set operation problem, i.e., the set of account attributes S. a With the set of management strategy rules S p Is the intersection of the two empty? Among them, S a = {a1, a2, ...} represents the set of attributes that an account possesses; S p ={p1,p2,...} represents the set of rules defined in the management strategy; a user account is considered a valid account and is allowed to register on the voice management terminal if and only if the intersection of the account attribute set and the management strategy rule set is not an empty set.

[0065] For newly approved user accounts, the voice management terminal automatically assigns a unique internal contact number according to a preset numbering rule. For example, for enterprise users, the contact number numbering rule is flexibly designed based on factors such as the enterprise's organizational structure and user classification. A common numbering rule is segmented representation, where different segments of the contact number represent different attributes or categories. For example, an 8-digit contact number N... L The numbers can be assigned according to the following rules:

[0066] N L =D1D2D3D4D5D6D7D8;

[0067] Here, D1D2 represents the department number, D3D4 represents the user type, and D5D6D7D8 represents the user's serial number within that type. This segmented numbering system allows for quick and easy identification of a user's department and type based on their contact number, while ensuring that the contact number is unique throughout the system.

[0068] To maintain the real-time status of the system, the voice management terminal immediately creates a new online status table entry for each user in memory after assigning a contact number, recording the user account, internal contact number, IP address, port number, registration time, and expiration time. The online status table is organized using a hash index, with the user account as the primary key, to ensure fast querying and updating of status information.

[0069] This step, through the adoption of certificate authentication, account attribute matching, flexible numbering rules, and an efficient status management mechanism, achieves a secure, reliable, flexible, and efficient user registration and number allocation scheme, laying a solid foundation for subsequent communication services.

[0070] Step 2: The voice management terminal generates a call path based on the calling and called number matching routing strategy; simultaneously, it applies a machine learning model combined with QoS indicators to predict network congestion points and dynamically optimize the SIP path; see [link / reference] Figure 3 This is a flowchart of the route matching and call optimization process for this step.

[0071] After receiving a call request initiated by a user, the voice management terminal performs routing policy matching based on the calling and called numbers. The routing policy consists of a set of predefined matching rules, each of which includes attributes such as number pattern, call permissions, and routing destination. The matching process is abstracted into a multi-stage longest prefix matching problem, that is, searching for the routing rule with the longest matching prefix with the calling and called numbers from the routing table.

[0072] Let the calling number of the call request be caller and the called number be callee. Let the routing table R contain n routing rules, each represented as a triple.<pa,pe,de> Where pa represents the number matching mode, pe represents the call permission, and de represents the routing destination; the routing matching algorithm is expressed as:

[0073]

[0074] Here, `route(caller, callee)` determines which routing rule to select based on the calling number `caller` and the called number `callee`; `match(number, pa)` is the matching function for the number `number` and the pattern `pa`, returning the number of matching bits when `number` matches `pa`, otherwise returning 0; `n0` is a subscript, n0∈[1,n]; `∧` represents a logical AND operation. The matching algorithm finds the routing rule with the highest matching degree between the calling and called numbers and satisfies the calling permissions, and returns its corresponding route destination.

[0075] After matching the appropriate routing rules, the voice management terminal generates a call path based on the route destination. The call path includes a call signaling path and a media transmission path. The call signaling path is used to determine the transmission path of SIP messages, and the media transmission path determines the transmission path of RTP (Real-Time Transport Protocol) data packets. The generation of the call path needs to take into account various factors such as network topology, link quality, and node load to ensure high availability and high quality of the call.

[0076] To further optimize call paths, a network quality prediction model based on the random forest algorithm was built on the voice management terminal. Random forest is an ensemble learning algorithm that improves the accuracy and robustness of predictions by constructing multiple decision trees and integrating their prediction results.

[0077] In the network quality prediction model based on the random forest algorithm, the training samples of each decision tree are randomly sampled from the original dataset through bootstrapping, and the splitting attributes of each node are also randomly selected from the candidate attribute set; this randomness helps to reduce the variance of the model and improve the generalization ability.

[0078] Let F be the network quality feature vector, F = <f1,f2,...,f m > where m is the number of features. Let m0 represent the m0th feature index, such as packet loss rate, latency, jitter, etc., where m0∈[1,m]; H is the health score of the link or node, with a value range of [0,1]; the random forest prediction model can be expressed as: Where K is the number of decision trees. Let k be the k0th decision tree, where k0∈[1,k].

[0079] Each decision tree takes the network quality feature vector F as input, and after multiple decision tree node partitions and attribute judgments, a health score is finally given at the leaf node; the final prediction result of the random forest is the average of the prediction results of all decision trees.

[0080] By performing feature engineering and data cleaning on historical call data, a large number of training samples can be obtained. These samples are then used to train the random forest model, continuously optimizing the model's parameters and hyperparameters to obtain a stable network quality prediction model.

[0081] When generating call paths, the voice management terminal inputs the network quality feature vectors of candidate paths into the network quality prediction model to obtain a health score for each path; then, the path with the highest health score is selected as the optimal path for the transmission of call signaling and media data.

[0082] The network quality prediction model can also assess the health status of each node and link in the network in real time and identify potential congestion points. For nodes or links with low predicted health scores, the voice management terminal can take proactive optimization measures, such as dynamically adjusting the transmission path of SIP messages to guide calls around congested nodes, or scheduling backup media processing resources to enhance the reliability of media transmission.

[0083] This step, through the aforementioned intelligent routing strategy and proactive optimization mechanism, can effectively improve call connection rate and call quality, reduce call failures and audio quality degradation caused by network congestion, thereby significantly improving the user's call experience; at the same time, based on machine learning prediction and optimization algorithms, it can also manage network resources more granularly and reduce network operating costs.

[0084] Step 3: Dynamically allocate media resources based on the call path and network conditions, initiate the RTP stream and transmit voice data, and dynamically adjust voice compression parameters based on network bandwidth and packet loss rate; see [link / reference] Figure 4 This is a flowchart of the media resource management and adaptation process for this step.

[0085] After generating and optimizing the call path, the voice management terminal dynamically allocates and manages media resources, including media processors, transmission bandwidth, and codecs. Media resources determine the sound quality and experience of the call.

[0086] The voice management terminal calculates the media resource requirements based on the network status and load of each node in the call path; let the call path be P, and the set of nodes on P be N. p ={n1,n2,...,n Y Let Y represent the number of nodes, and S be the network state feature vector of each node. y = y1 ,s y2 ,...,s Ym >, Ym represents the network state feature vector S y The number of dimensions includes packet loss rate, latency, and jitter metrics; the load feature vector of a node is L. y = <l y1 ,l y2 ,...,l Yn >, Yn represents the dimension of the node load feature vector, including metrics such as CPU utilization, memory usage, and number of connections, y∈[1,Y]; then the node's media resource requirement R y Calculated using the resource mapping function f:

[0087]

[0088] Among them, ||S y ​||2 is the L2 norm of the network state characteristics, reflecting the overall network state strength; φ(L y ) is the load feature vector L y The nonlinear mapping function, for example, A function representing the cross-effect between network state and load, for example, γ and δ are adjustment coefficients used to characterize the intensity of the nonlinear response.

[0089] After obtaining the resource requirements of each node, the voice management terminal comprehensively considers factors such as the node's geographical location, resource utilization, and service quality, selects the optimal resource allocation scheme from the media resource pool, and dynamically allocates it to each node; let the media resource pool be R = {r1, r2, ..., r...} Q}, where Q represents the total number of media resources, and r j Let j represent the j-th type of media resource, such as a media processor, codec, etc.; model the optimal resource allocation problem as an integer programming model:

[0090]

[0091] Where, x ij Let n be a non-negative integer decision variable, representing the decision variable assigned to node n. i The number of the j-th type of media resources; c ij To allocate resources to node n i The cost is calculated based on factors such as resource performance, energy consumption, and location; b ij For the j-th type of media resource, node n i Resource contribution; N j Let be the available quantity of the j-th type of media resource; the objective of the model is to minimize the total resource allocation cost while satisfying the media resource requirements of each node; solving this integer programming model yields the optimal resource allocation scheme, i.e.

[0092] After completing the media resource allocation, the voice management terminal notifies each node on the call path to start the corresponding media processor and codec according to the allocation scheme, and establishes an RTP (Real-Time Transport Protocol) stream to begin transmitting voice data. The RTP protocol is based on the UDP protocol and implements functions such as timing control, synchronization, and mixing in the real-time data transmission process by adding fields such as sequence number, timestamp, and synchronization source identifier to the data packets. As the initiator and controller of the RTP session, the voice management terminal is responsible for coordinating the RTP parameters of each node and monitoring and managing the RTP stream.

[0093] To adapt to dynamic changes in network conditions during a call, the voice management terminal simultaneously monitors network bandwidth and packet loss rate in real time and dynamically adjusts media codec parameters to ensure call quality and smoothness. Let the currently detected network bandwidth be B and the packet loss rate be ρ, then the dynamic codec parameter adjustment algorithm can be represented as the following optimization problem:

[0094]

[0095] in, G(·) is the encoder / decoder parameter vector, including the encoding bit rate, frame length, and FEC (forward error correction) strategy; G(·) is the encoder / decoder quality evaluation function, which comprehensively considers the impact of encoder / decoder parameters, network bandwidth, and packet loss rate on audio quality, and is constructed based on standard algorithms such as PESQ and POLQA; H(·) is the encoding bit rate function, representing the encoder parameter vector. The encoded data rate is given by D(·), where D(·) is the transmission delay function, representing the time delay under the encoder parameters. End-to-end transmission delay at packet loss rate ρ; D max For the maximum allowable transmission delay, exemplarily, D max It is usually set to 150-200 milliseconds to ensure real-time interactivity.

[0096] The goal of optimizing dynamic codec parameter adjustment algorithms is to find the optimal combination of codec parameters under constraints of network bandwidth and transmission delay. The goal is to maximize the encoder-decoder quality evaluation function G. This problem is solved using heuristic search algorithms (such as genetic algorithms, particle swarm optimization, etc.) or reinforcement learning algorithms (such as deep Q-learning, policy gradient, etc.). The optimal encoder parameter combination obtained from the solution is... The data will be sent in real time to each node along the call path, guiding the nodes to dynamically adjust their encoding and decoding behavior, thereby maintaining high call quality and user experience even when network conditions fluctuate.

[0097] This step achieves refined management and dynamic adaptation of media resources by quantifying node resource requirements, optimizing media resource allocation, applying real-time transmission protocols, and intelligent encoding / decoding control. While ensuring call quality and user experience, it maximizes resource utilization efficiency and improves the system's service capabilities and economy, which is the core competitiveness of IPPBX and voice switching. At the same time, these technical ideas and model algorithms also provide reference and guidance for the design of other multimedia communication and real-time data transmission systems.

[0098] Step 4: Continuously monitor the voice session status during the call, including SIP signaling flow, call hold duration, and user hang-up signal information, and trigger corresponding processing mechanisms based on the voice session status. See [link / reference] Figure 5This is a flowchart of the call anomaly detection and recovery process for this step.

[0099] During the call, the voice management terminal continuously monitors the status of the call session, including information such as SIP signaling flow, call hold duration, and user hang-up signal, in order to promptly detect and handle abnormal situations.

[0100] The voice management terminal uses a finite state machine model to characterize the SIP signaling process; each SIP transaction (such as call setup, call release, etc.) corresponds to a state machine instance; the state machine defines the set of states of the transaction, the state transition rules, and the processing actions in each state; a set of state machine templates are predefined on the voice management terminal according to the SIP protocol standard.

[0101] When the voice management terminal receives a SIP request message, it instantiates the corresponding state machine according to the request type, and updates the current state of the state machine as subsequent SIP messages are sent and received. By tracking the state of each SIP transaction in real time, the voice management terminal can detect abnormal events (such as timeouts, invalid responses, etc.) in a timely manner and trigger the corresponding exception handling process.

[0102] The voice management system continuously tracks the call hold duration of each call session and periodically (e.g., every 30 minutes) calculates the average hold duration and hold duration variance for all active calls. By tracking the trends of these two metrics, the voice management system identifies abnormal call patterns. For example, when the average hold duration is significantly higher than the historical average and the hold duration variance is large, it indicates that a few calls have experienced prolonged call hold periods, requiring the administrator's attention.

[0103] The voice management terminal also monitors user hang-up signals in real time. Once a hang-up signal is detected, the session cleanup process is immediately initiated, including releasing the media resources occupied by the session to avoid prolonged resource occupation and leakage.

[0104] The voice management terminal performs anomaly checks periodically. The anomaly detection steps include:

[0105] Step a: Analyze the SIP message log, check the legality and timing of the SIP messages, and identify suspicious messages or transactions;

[0106] Step b: Check media transmission quality statistics, analyze indicators such as packet loss rate and jitter during the call, and determine if there are any network problems;

[0107] Step c: Check system resource usage, monitor the usage of resources such as CPU, memory and disk, and check for resource bottlenecks;

[0108] Step d: Query user account information and business data, review user account configuration information and related business data, and check for any abnormal behavior caused by incorrect account configuration or inconsistent business data.

[0109] Based on the test results, the voice management terminal automatically executes corresponding recovery measures, including re-initiating SIP transactions, switching media transmission paths, adjusting resource allocation strategies, and correcting business data; for serious anomalies that cannot be automatically recovered, alarm information is generated and the administrator is notified to intervene manually.

[0110] This step, by real-time monitoring of SIP signaling flow, call hold duration, and user hang-up signals, and by introducing an automatic detection and recovery mechanism, achieves anomaly management throughout the entire call lifecycle, ensuring the continuity and high availability of voice services, improving the system's self-healing capabilities and operational efficiency, and laying the foundation for building high-quality voice communication services.

[0111] Example 2:

[0112] Reference Figure 6 This is the second embodiment of the present invention, which provides a SIP-based IPPBX voice switching management system.

[0113] The system includes: a user registration management module, a call path optimization module, a network resource adjustment module, and a call monitoring and adjustment module.

[0114] The user registration management module receives registration requests from user terminals via a voice management terminal, verifies user identity by combining user account and device information, assigns internal contact numbers to new users, and establishes an online status table.

[0115] The call path optimization module generates a call path based on the calling and called number matching routing strategy at the voice management end; at the same time, it uses a machine learning model combined with QoS indicators to predict network congestion points and dynamically optimize the SIP path.

[0116] The network resource adjustment module dynamically allocates media resources based on the call path and network conditions, initiates RTP streams and transmits voice data, and dynamically adjusts voice compression parameters based on network bandwidth and packet loss rate.

[0117] The call monitoring and adjustment module is used to continuously monitor the voice session status during a call, including SIP signaling flow, call hold duration and user hang-up signal information, and trigger corresponding processing mechanisms based on the voice session status.

[0118] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings or direct couplings or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0119] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the claims. All of these variations are within the protection scope of the present invention.

Claims

1. A SIP-based IPPBX voice switching management method, characterized in that, include: The voice management terminal receives registration requests from user terminals and performs identity verification by combining user account and device information. This includes: employing a certificate-based two-way authentication mechanism, where the user terminal exchanges certificates with the voice management terminal and verifies the validity of the certificates when initiating a registration request message; after successful certificate verification, the voice management terminal reviews the user account by matching it against the rules in the management policy to determine whether the user account has legitimate registration permissions; the matching process between account attributes and management policies is abstracted as a set operation problem, i.e., whether the intersection of the account attribute set and the management policy rule set is empty; the user account is deemed a legitimate account if and only if the intersection of the account attribute set and the management policy rule set is not empty; assigning an internal contact number to new users and establishing an online status table. The voice management terminal generates call paths based on the calling and called number matching routing strategy; at the same time, it uses machine learning models combined with QoS indicators to predict network congestion points and dynamically optimize call paths. Media resources are dynamically allocated based on the call path and network conditions, RTP streams are initiated and voice data is transmitted, and voice compression parameters are dynamically adjusted based on network bandwidth and packet loss rate. During a call, the system continuously monitors the voice session status, including SIP signaling flow, call hold duration, and user hang-up signal information, and triggers corresponding processing mechanisms based on the voice session status.

2. The SIP-based IPPBX voice switching management method according to claim 1, characterized in that, For user accounts that pass the review, the voice management terminal assigns a unique internal contact number to the user account; The internal contact number is numbered using a segmented representation method, including the department number segment, the user type segment, and the user serial number segment; After assigning an internal contact number, the voice management terminal creates a new user online status table in memory. The online status table records user account, internal contact number, IP address, port number, registration time, and expiration time information, and is organized using a hash index with the user account as the primary key to enable querying and updating status information.

3. The SIP-based IPPBX voice switching management method according to claim 2, characterized in that, The specific process by which the voice management terminal generates a call path based on the calling and called number matching routing strategy includes: after receiving a call request initiated by the user, the voice management terminal performs routing strategy matching based on the calling and called numbers; the routing strategy consists of predefined matching rules, each rule including number pattern, call permissions and routing destination; the matching process is abstracted into a multi-stage longest prefix matching problem, that is, searching for the routing rule with the longest matching prefix with the calling and called numbers from the routing table.

4. The SIP-based IPPBX voice switching management method according to claim 3, characterized in that, The voice management terminal generates a call path based on the calling and called numbers matching routing strategy; at the same time, it applies a machine learning model combined with QoS indicators to predict network congestion points and dynamically optimize the call path, including: after matching a suitable routing rule, the voice management terminal generates a call path based on the routing destination; The call path includes the call signaling path and the media transmission path; the call signaling path serves as the transmission path for voice messages; the media transmission path serves as the transmission path for voice data packets. A network quality prediction model based on the random forest algorithm is introduced at the voice management terminal. The network quality prediction model based on the random forest algorithm is constructed, and the network quality feature vector is constructed. In the random forest algorithm, each decision tree takes the network quality feature vector as input, and after multiple decision tree node partitioning and attribute judgment, a health score is given at the leaf node. The path with the highest health score is selected as the optimal path and used for the call.

5. The SIP-based IPPBX voice switching management method according to claim 4, characterized in that, The specific process of dynamically allocating media resources based on call path and network conditions includes: after the voice management terminal selects the optimal path as the call path, it dynamically allocates and manages media resources; the media resources include media processors, transmission bandwidth, and codecs; The voice management terminal calculates the media resource requirements based on the network conditions and load of each node in the call path; After obtaining the media resource requirements of each node, the voice management terminal selects the optimal resource allocation scheme from the media resource pool and dynamically allocates the resources to each node.

6. The SIP-based IPPBX voice switching management method according to claim 5, characterized in that, The specific process of selecting the optimal resource allocation scheme includes: modeling the optimal resource allocation problem as an integer programming model; The goal of the integer programming model is to minimize the total resource allocation cost while satisfying the media resource requirements of each node; solving the integer programming model yields the optimal resource allocation scheme.

7. The SIP-based IPPBX voice switching management method according to claim 6, characterized in that, The specific process of starting the RTP stream and transmitting voice data, and dynamically adjusting the voice compression parameters according to network bandwidth and packet loss rate includes: after completing the media resource allocation, the voice management terminal notifies each node on the call path to start the corresponding media processor and codec according to the allocation scheme, and establishes a real-time transmission protocol stream to transmit voice data. Simultaneously monitor network bandwidth and packet loss rate in real time at the voice management terminal, construct a dynamic encoding and decoding parameter adjustment algorithm, and dynamically adjust media encoding and decoding parameters; Dynamic codec parameter adjustment algorithms find the optimal combination of codec parameters under the constraints of network bandwidth and transmission delay, so as to maximize the codec quality. After obtaining the optimal combination of encoding and decoding parameters, the results are distributed to each node along the call path to guide the nodes in dynamically adjusting their encoding and decoding behavior.

8. The SIP-based IPPBX voice switching management method according to claim 7, characterized in that, The specific process of continuously monitoring the voice session status during the call and triggering the corresponding processing mechanism based on the voice session status includes: the voice management terminal monitors the call process based on the finite state machine model, tracks the call hold duration, calculates the average hold duration and variance to detect abnormal call patterns, and starts the session cleanup process after listening to the user's hang-up signal. The voice management terminal performs anomaly detection periodically, including analyzing SIP message logs, checking media transmission quality, monitoring system resource usage, and verifying user account information. Based on the detection results, it automatically executes recovery measures and / or generates alarm information to notify the administrator to intervene.

9. A SIP-based IPPBX voice switching management system, used to implement the SIP-based IPPBX voice switching management method according to any one of claims 1 to 8, characterized in that, include: The module includes a user registration management module, a call path optimization module, a network resource adjustment module, and a call monitoring and adjustment module. The user registration management module receives registration requests from user terminals via a voice management terminal and performs identity verification by combining user account and device information. This includes: employing a certificate-based two-way authentication mechanism, whereby the user terminal exchanges certificates with the voice management terminal and verifies the validity of the certificates when initiating a registration request message; after successful certificate verification, the voice management terminal reviews the user account by matching it against rules in the management policy to determine whether the user account has legitimate registration permissions; abstracting the matching process between account attributes and management policies into a set operation problem, i.e., whether the intersection of the account attribute set and the management policy rule set is empty; the user account is deemed a legitimate account if and only if the intersection of the account attribute set and the management policy rule set is not empty; assigning an internal contact number to new users and establishing an online status table. The call path optimization module generates a call path based on the calling and called number matching routing strategy at the voice management end; at the same time, it uses a machine learning model combined with QoS indicators to predict network congestion points and dynamically optimize the call path. The network resource adjustment module dynamically allocates media resources based on the call path and network conditions, initiates RTP streams and transmits voice data, and dynamically adjusts voice compression parameters based on network bandwidth and packet loss rate. The call monitoring and adjustment module is used to continuously monitor the voice session status during a call, including SIP signaling flow, call hold duration and user hang-up signal information, and trigger corresponding processing mechanisms based on the voice session status.

Images