Edge computing-based automatic routing method and system for voice over IP networks

Abstract

This invention provides an automatic routing method and system for VoIP switches based on edge computing. By decentralizing routing decision-making to each VoIP switch, a decentralized decision-making mechanism is achieved. The system consists of multiple VoIP switches and a central database. Each VoIP switch establishes a local data storage area and a local routing table, and exchanges data directly with other switches through point-to-point communication channels. When receiving a voice communication command, the switch first attempts to make a decision using its local routing table; only when the local data is insufficient or expired does it request updates from neighboring switches or the central database. The system implements a data layering module, a route learning module, and a fault detection module, and can dynamically adjust the synchronization frequency according to network conditions. This design reduces reliance on the central database, lowers communication latency, improves system reliability and adaptability, and provides a more efficient routing solution for VoIP networks.

Description

[Technical Field]

[0001] This invention relates to an automatic routing method and system architecture for network voice switches based on edge computing, and particularly to a method applicable to distributed enterprise communication systems, cross-regional telecommunications networks, and hybrid cloud / local voice communication environments. [Previous Technology]

[0002] With the rapid development of Internet technology, Voice over Internet Protocol (VoIP) has become an important part of modern communication systems. Traditional VoIP switching systems mainly rely on a central server architecture, where all routing decisions are processed by the central system. This can lead to latency, resource waste, and single point of failure risks during high traffic periods.

[0003] Currently, the interconnection between most telecommunications service networks mainly relies on fixed routing configurations, lacking dynamic adjustment capabilities and unable to automatically select the optimal call path based on current network conditions and cost structures. This often results in users incurring higher costs when communicating across networks.

[0004] Existing VoIP systems mainly rely on a central server architecture for routing decisions, which has the following problems: 1. The central system handles all routing decisions, leading to high load and high latency; 2. A single point of failure, such as a central server failure, can paralyze the entire system; 3. When the network is congested, the central system may not be able to process all routing requests in a timely manner; 4. Interconnection between telecommunications service networks mainly relies on static route configurations, which cannot be dynamically adjusted according to real-time network conditions; 5. There is a lack of ability to automatically select the best path based on the cost structure, and users often have to pay high cross-network call fees; 6. Existing technologies do not fully utilize the advantages of edge computing in reducing communication latency and improving reliability.

[0005] Edge computing, as an emerging distributed computing model, disperses data processing from central nodes to the network edge, effectively reducing communication latency and improving system reliability. However, existing technologies have not yet fully applied edge computing to network voice switching systems, especially in automatic routing. [Summary of the Invention]

[0006] The main objective of this invention is to provide an automatic routing method and system for network voice switches based on edge computing, which can decentralize the routing decision-making function to each network voice switch and realize a decentralized decision-making mechanism.

[0007] Another objective of the present invention is to provide an automatic routing method and system for network voice switches based on edge computing, which enables each network voice switch to maintain a local routing table and reduces dependence on a central database.

[0008] To achieve the above objectives, the present invention provides an automatic routing method for network voice switches based on edge computing, comprising the following steps: (A) initializing a plurality of network voice switches, loading them into the system's basic configuration, and starting the necessary communication modules and initialization processing units; (B) each of the plurality of network voice switches establishes a local data storage area to store the data required for local routing decisions; (C) the plurality of network voice switches connect to a central database and perform initial data synchronization, loading the data in the central database into the local data storage area of ​​each network voice switch, including at least the data of the plurality of network voice switches, the switch group relationship of the plurality of network voice switches' hierarchical levels, the data of each telecommunications service provider, and the fee data; (D) each network voice switch establishes a local routing table based on the data in its local data storage area and sets... (E) Establish point-to-point communication channels between each network voice switch for direct exchange of routing and status data without going through a central database; (F) Based on preset triggering conditions, each network voice switch autonomously decides whether to update data from the central database or obtain the latest routing data from neighboring network voice switches; (G) When each network voice switch receives a voice communication command, it first uses its local routing table to perform routing decisions. If the required routing data exists in the local routing table, it executes the voice communication command; (H) If the required routing data does not exist in the local routing table or the data has expired, each network voice switch selects to obtain the required routing data from a neighboring network voice switch or the central database based on the network connection status; and (I) Update the local routing table based on the latest obtained routing data and execute the voice communication command.

[0009] To achieve the above objectives, the present invention also provides an automatic routing system for network voice switches based on edge computing. This system includes: a plurality of network voice switches, each electrically connected to one or more VoIP phones, the plurality of network voice switches belonging to one or more switch groups, wherein each switch group is interconnected by a trunk line; a central database storing global routing data, each network voice switch capable of periodically synchronizing with the central database; a plurality of local data storage areas, respectively located in each network voice switch, for storing data synchronized from the central database and locally generated routing decision data; and wherein, when each network voice switch receives a voice communication command, it first independently executes routing decisions using the data in its local data storage area. If the local data is insufficient or expired, it requests data updates from the central database or obtains the required routing data from adjacent network voice switches.

[0010] The features, advantages, or similar expressions mentioned in this specification do not imply that all features and advantages achievable by the present invention should be included in any single specific embodiment of the present invention. Rather, it should be understood that expressions of features and advantages refer to specific features, advantages, or characteristics described in connection with specific embodiments that are included in at least one specific embodiment of the present invention. Therefore, the discussion of features and advantages, and similar expressions in this specification relates to the same specific embodiments, but is not essential.

[0011] These features and advantages of the present invention will become more apparent by referring to the following description and the accompanying patent claims or by utilizing the embodiments of the present invention described below. [Simplified Explanation of the Diagram]

[0031] Figure 1 shows a schematic diagram of an automatic routing system architecture for a network voice switch based on edge computing according to a preferred embodiment of the present invention.

[0032] Figure 2 shows a schematic diagram of a network voice switch architecture according to a preferred embodiment of the present invention.

[0033] Figure 3 shows a schematic diagram of the connection between the automatic routing system of the exchange and the public switched telephone network of the present invention.

[0034] Figure 4 shows a flowchart of an automatic routing method for a network voice switch based on edge computing according to a preferred embodiment of the present invention.

Implementation Method

[0035] To make the description of this disclosure more detailed and complete, illustrative descriptions of embodiments and specific examples of this invention are provided below; however, this is not the only form of implementing or utilizing the specific examples of this invention. The embodiments cover features of multiple specific examples and methods and steps for constructing and operating these specific examples, as well as their order. However, other specific examples may also be used to achieve the same or equivalent functions and order of steps.

[0036] Figure 1 shows a schematic diagram of an automatic routing system architecture for a network voice exchange based on edge computing according to a preferred embodiment of the present invention. Figure 2 shows a schematic diagram of a network voice exchange architecture according to a preferred embodiment of the present invention. Figure 3 shows a schematic diagram of the connection between the automatic routing system for the exchange and the public switched telephone network of the present invention. The following description should be taken in conjunction with Figures 1-3.

[0037] The edge computing-based automatic routing system 100 for network voice switches of the present invention mainly includes the following components: a plurality of network voice switches 101: these switches belong to one or more switch groups 103, the network voice switches 101 are connected to each other through point-to-point communication channels 102, and are electrically connected to one or more network telephones 106. Each switch group 103 is connected to each other via trunk lines 104. A central database 105: stores global routing data, and each network voice switch 101 can periodically synchronize with the central database 105.

[0038] Switch group 103 refers to a logical collection of multiple network voice switches 101 with the same attributes or located in the same geographical area, such as switches within the same branch office of an enterprise, switches within the same city area, or switches belonging to the same telecommunications operator's network. As a whole, switch group 103 has unified routing policy management, resource optimization and allocation, load balancing, and failover functions, which can improve call quality and reduce communication costs. The network voice switch 101 is located at the level closest to the user, directly connected to the VoIP phone 106, and handles call requests and executes routing decisions. The VoIP phone 106 provides the user with a call interface, storing data including user configuration, dialing history, and basic call settings; its functions include initiating call requests, receiving incoming calls, performing basic call functions (such as hold, transfer, conferencing, etc.), and transmitting voice communication commands to the network voice switch 101.

[0039] Each network voice switch 101 has a local data storage area 107 for storing data synchronized from the central database 105 and locally generated routing decision data. The data stored in the local data storage area 107 includes, but is not limited to: data of multiple network voice switches, switch group relationships of multiple network voice switches at different levels, data of various telecommunications service providers, and fee data.

[0040] Each network voice switch 101 also includes the following functional modules: Communication module 108: responsible for communication with other network voice switches and the central database. Processing unit 109: responsible for performing routing decisions and other computing tasks. Performing routing decisions refers to the process of dynamically selecting the best call path based on the call destination, current network conditions, cost standards, and service quality requirements. Local routing table 110: established based on data from the local data storage area 107, used for fast routing decisions. Resource monitoring module 111: monitors local computing resources, storage resources, and network connection status in real time. Fault detection module 112: detects the connection status with the central database 105 and switches the operating mode when the connection is interrupted. When the fault detection module 112 detects a connection interruption with the central database 105, it can automatically switch to an autonomous operating mode, in which routing decisions are made using only the local routing table 110 and routing data from adjacent network voice switches 101.

[0041] Route Learning Module 113: Optimizes the routing table based on historical routing decisions and user behavior. The route learning module 113 continuously optimizes the local routing table based on past successful routing decisions and user behavior patterns, and sends the learning results back to the central database 105 under specific conditions. These specific conditions include, but are not limited to: 1. When a better routing path than the known path in the central database is discovered (e.g., lower latency, higher call quality, or lower cost); 2. When a newly discovered routing pattern remains stable and proven effective over a period of time (e.g., 7 consecutive days); 3. When the system detects that the routing table contains information that may also be useful to other switches (e.g., new service providers or new area routing rules); 4. According to the periodic backhaul schedule set by the system administrator (e.g., once a week).

[0042] Data layering module 114: manages the storage location of routing data used at different frequencies. The data layering module 114 implements an optimized storage strategy for routing data adopted by the network voice switch 101, storing high-frequency routing data in fast-access memory and low-frequency routing data in secondary memory, and dynamically adjusting the data storage location according to the usage frequency.

[0043] Central Database 105: Stores global routing data, and each network voice switch 101 can periodically synchronize with the central database 105. The central database 105 stores all data of the switching system and implements a differential update mechanism, so that each network voice switch 101 only needs to download the data that has changed since the last synchronization, thereby saving network bandwidth and synchronization time. The data stored in the central database 105 includes: global routing table, data of multiple network voice switches, switch group relationships of multiple network voice switches at different levels, data of various telecommunications service providers, and fee data, etc.

[0044] VoIP device 115: Bridges the voice switching system and the Public Switched Telephone Network (PSTN) 116 to convert digital and analog voice signals. VoIP device 115 is a device capable of connecting a VoIP network and a traditional public switched telephone network, enabling communication between the two networks using different technical standards. The data stored in the VoIP device includes, but is not limited to: protocol conversion parameters, codec configuration data, call quality monitoring indicators, and bridging connection status information.

[0045] A plurality of analog telephones 117: A plurality of telecommunications exchanges (not shown) connected to the public switched telephone network 116. The data stored in the analog telephones 117 includes: dialing speed key settings and limited call logs. The functions of the analog telephones 117 include: initiating and receiving calls via the PSTN network, and performing basic voice communications.

[0046] The lightweight protocol used by the point-to-point communication channel 102 includes the following fields: sending switch identifier, destination switch identifier, message type, routing data version number, routing table entry, and checksum. This protocol ensures the accuracy and efficiency of data exchange. Adjacent network voice switches 101 exchange data directly through point-to-point communication, enabling rapid data exchange.

[0047] Data synchronization between the central database 105 and the network voice switch 101 adopts a multi-level strategy, adjusting the synchronization frequency and method according to different situations, including but not limited to parameters such as network load, local processing capacity, routing request frequency and internal time.

[0048] Please refer to Figure 4, which shows a flowchart of an automatic routing method for a network voice switch based on edge computing in an embodiment of the present invention. The method of the present invention mainly includes the following steps:

[0049] Step S01: Initialize multiple network voice switches 101. Load the basic system configuration into each network voice switch 101, start the necessary communication modules 108 and initialization processing unit 109, and prepare for system operation.

[0050] Step S02: Each network voice switch 101 establishes a local data storage area 107. This storage area is used to store data required for local routing decisions, or call records, etc. The local data storage area 107 refers to the data storage space set up in each network voice switch 101, used to store data synchronized from the central database 105 and locally generated routing decision data. In this step, the network voice switch 101 activates its internal data layering module 114. This module stores frequently used routing data in fast-access memory and less frequently used routing data in secondary memory, and dynamically adjusts the data storage location according to the usage frequency. The data stored locally in the network voice switch 101 includes: local routing tables, call records, QoS statistics, etc. The local storage design reduces the dependence on the central database 105 and improves the system's response speed and reliability.

[0051] Step S03: Each VoIP switch 101 connects to the central database 105 and performs initial data synchronization. Data from the central database 105 is loaded into the local data storage area 107 of each switch, including at least data on multiple VoIP switches, the switch group relationships of the multiple VoIP switches' hierarchical levels, data on each telecommunications service provider, and fee data. In this step, the central database 105 can implement a differential update mechanism, allowing each VoIP switch 101 to download only the data that has changed since the last synchronization, thereby saving network bandwidth and synchronization time.

[0052] Step S04: Each network voice switch 101 establishes a local routing table 110 based on the data in the local data storage area 107 and sets the data update frequency parameters. The local routing table 110 refers to the data structure established by each network voice switch 101 based on the data in the local data storage area 107, used to guide its routing decisions. The local routing table 110 records in detail the trunk connections between switches and the lowest-cost routing paths, which may include, but is not limited to: destination prefix, primary routing path, alternative routing path, path cost, validity period, quality of service parameters, etc. When establishing the local routing table 110, the system also sets the data update frequency parameters, which refers to the system configuration value that controls the frequency of data synchronization between the network voice switch 101 and the central database 105 or other network voice switches 101. In this step, each network voice switch 101 can set the synchronization frequency with the central database 105 to be dynamically adjusted based on parameters such as network load, local processing capacity, routing request frequency, and internal time.

[0053] In addition, each network voice switch is also equipped with a resource monitoring module (111). The resource monitoring module 111 refers to a software or hardware component built into the network voice switch 101, which can monitor local computing resources, storage resources and network connection status in real time, and dynamically adjust the local decision ratio and data synchronization strategy according to the monitoring results to achieve optimal utilization of system resources.

[0054] Step S05: A point-to-point communication channel 102 is established between each network voice switch 101. This point-to-point communication channel 102 is used to directly exchange routing data and status data without going through the central database 105. The point-to-point communication channel 102 refers to the direct communication path established between the network voice switches 101 for exchanging routing data and status data without going through the central database 105, thereby improving communication efficiency and system resilience. The point-to-point communication channel 102 established in this step uses the lightweight protocol as described above, including the sending switch identification code, the target switch identification code, the message type, the routing data version number, the routing table entry, and the checksum. This protocol design ensures the accuracy and efficiency of data exchange. The point-to-point communication channel 102 reduces communication latency, improves the immediacy and reliability of the system, and reduces the load on the central node.

[0055] Step S06: Based on preset triggering conditions, each network voice switch 101 independently decides whether to update data from the central database 105 or obtain the latest routing data from adjacent network voice switches 101. In this step, the preset triggering conditions may include, but are not limited to: the expiration of the local routing table entry's lifespan, network load falling below a preset threshold, new routing requests failing to be resolved by the local routing table, or receiving a routing update notification from another network voice switch.

[0056] Step S07: When each network voice switch 101 receives a voice communication command, it first performs a routing decision using its local routing table 110. A voice communication command is a call request initiated by a user through a VoIP device (such as a VoIP phone), containing information such as the calling number, the called number, and quality of service requirements. After receiving the command, the system first parses the called number and uses a prefix matching algorithm to search for a match in the local routing table 110. The routing decision execution involves multiple steps: evaluating the combined cost of multiple possible paths, checking the current availability of the paths, applying routing strategies, selecting the optimal path, and preparing call parameters. This routing decision process is consistent with the routing decision execution described above; it is a process of dynamically selecting the optimal call path based on the call destination, current network conditions, cost standards, and quality of service requirements. In this step, if the required routing information exists in the local routing table 110, the voice communication command can be executed directly.

[0057] In addition, during this step, each network voice switch 101 activates its internal routing learning module 113, which continuously optimizes the routing selection strategy in the local routing table 110 based on past successful routing decisions and user behavior patterns, and sends the learning results back to the central database 105 under appropriate conditions, so that the entire system can benefit from the experience of individual nodes and continuously improve the overall routing efficiency.

[0058] Step S08: When the required routing information is not found in the local routing table 110 or the information has expired, each network voice switch 101 selects to obtain the required routing information from a neighboring network voice switch 101 or the central database 105 based on the network connection status. If there is no matching information in the local routing table or the route has expired, the system will initiate a synchronization request to obtain the latest information of the specific route from the central database 105 or a neighboring network voice switch 101. The data expiration check compares the timestamp or version number of the routing entry with the validity period threshold defined by the system. In step S08, each network voice switch 101 can also apply a fault detection module 112. When a connection interruption with the central database 105 is detected, it automatically switches to an autonomous operation mode. In this mode, only the local routing table 110 and the routing information from the neighboring network voice switch 101 are used for routing decisions to ensure that the system can maintain basic communication functions even when the central database 105 is unavailable.

[0059] Step S09: Update its local routing table 110 according to the latest obtained routing data and execute voice communication commands. Regardless of the path used to obtain routing data, the system will eventually update the latest data to the local routing table 110 and execute voice communication commands, including call signaling processing, media channel settings, billing record initialization, and service quality assurance. When a user dials from one end of a telecommunications service network to the other, the switch determines the optimal path according to the local routing table 110. This process is transparent to the user and does not require changing the user's dialing habits; the system will automatically select the most cost-effective or most efficient routing path. It should be noted that the basic processing flow described in steps S07 to S09 above may be adjusted according to actual network conditions and specific application scenarios, as shown in the following embodiments.

[0060] As can be understood from the above steps, the automatic routing method for network voice switches based on edge computing of the present invention enables network voice switch 101 to fully utilize its local decision-making capabilities while maintaining appropriate coordination with the central database 105, thereby improving system efficiency, reliability, and adaptability. The edge computing mode adopted by the present invention can decentralize routing decisions from the central node to network voice switch 101, significantly reducing communication latency and improving system reliability and scalability.

[0061] Example 1: Successfully executing routing decisions using a local routing table

[0062] The following specific example illustrates how to smoothly use the local routing table to perform routing decisions and make phone calls when receiving voice communication instructions.

[0063] Assume user A is located in the Taipei branch and uses a VoIP phone to dial extension number 8001 of user B located in the Shanghai branch. The call process is as follows:

[0064] 1. The Taipei branch’s network voice exchange (referred to as exchange T) receives a voice communication command from user A, which includes the called number 8001 and service quality requirements (call priority is standard).

[0065] 2. The switch T first parses the prefix of the called number 8001 and identifies that it is an internal extension number dialed to the Shanghai branch.

[0066] 3. Switch T queries its local routing table and finds a route entry matching the prefix 8xxx. This entry contains the following information: Prefix: 8xxx Destination: Shanghai Branch Network Voice Switch (denoted as Switch S) Primary Route: Through Hong Kong Relay Node (denoted as Switch H) Alternate Route: Through Singapore Relay Node (denoted as Switch SG) Path Cost Coefficient: Primary Path 0.8, Alternate Path 1.2 Time to Live: 240 minutes remaining

[0067] 4. Switch T uses a routing learning module to analyze this routing decision, considering the following factors: historical call records show that call quality through the Hong Kong trunk node is good (average latency <100ms); the current time is 10:30 AM on a weekday, and the Hong Kong-Shanghai line load is moderate during this period; system monitoring shows that the Hong Kong trunk node is currently online and operating normally; the success rate of similar calls previously was 99.5%.

[0068] 5. Based on the above analysis, switch T chooses to use the main routing path, that is, to connect to the Shanghai network voice switch through the Hong Kong trunk node.

[0069] 6. Switch T establishes call signaling, including the following information: Source switch identification code: T; Destination switch identification code: H (Hong Kong trunk node); Final destination identification code: S (Shanghai switch); Called number: 8001; Call priority: Standard selected codec: G.711

[0070] 7. The call signaling is sent to the Hong Kong relay node through a pre-established point-to-point communication channel, then forwarded to the Shanghai exchange, and finally connected to User B's VoIP phone.

[0071] 8. The entire routing decision-making process is completed locally without accessing the central database, and the decision-making time is only 12 milliseconds, which is much lower than the average of 35 milliseconds in the central decision-making mode.

[0072] 9. After the call connection is successfully established, the switch T records the result of this routing decision for future route learning and optimization.

[0073] This embodiment demonstrates how a network voice switch can efficiently perform routing decisions using a local routing table to achieve rapid call setup. The entire process is completed at the edge node without the need for a central database, reflecting the core advantage of this invention based on edge computing.

[0074] Example 2: Routing decision process when the local routing table cannot be resolved

[0075] The following example illustrates how the system proceeds to step S08 to obtain the required information and complete the call when the local routing table cannot provide the required routing information.

[0076] Assume User A is located in the Taipei branch and uses a VoIP phone to dial an international long-distance number +44-20-7946-0958 (a company in London, UK). The call process is as follows:

[0077] 1. The Taipei branch’s network voice switch (referred to as switch T) receives a voice communication command from user A, which includes the called number +44-20-7946-0958.

[0078] 2. The switch T first parses the country code +44 (United Kingdom) and city code 20 (London) of the called number.

[0079] 3. Switch T queries its local routing table and finds no route entry matching the prefix +44-20. This may be because: previous route entries that had not been used to make calls with this prefix have expired (time-to-live has expired); or recent changes in network topology or rates have caused old route entries to be marked as invalid.

[0080] 4. Switch T detects that the local routing table cannot resolve this call request, triggering the routing data acquisition process in step S08.

[0081] 5. Switch T first assesses network connectivity: Checks connection to the central database: connection is normal, round-trip latency 125ms. Checks connection to adjacent network voice switches (e.g., Hong Kong switch H): connection is normal, round-trip latency 45ms. Checks current network load: moderate level, not exceeding the threshold.

[0082] 6. Based on the connectivity assessment, switch T decides to prioritize requesting routing information from the adjacent Hong Kong switch H because point-to-point communication has lower latency.

[0083] 7. Switch T sends a routing query request to Hong Kong switch H through a point-to-point communication channel, including the following fields: Sending switch identifier: T; Destination switch identifier: H; Message type: Routing data request; Query prefix: +44-20; Query timestamp: 2023-10-15 14:28:36

[0084] 8. After receiving the request, the Hong Kong switch H checks its local routing table and finds a suitable route entry, but the entry is also close to expiring (remaining time to live is 10 minutes).

[0085] 9. Hong Kong switch H decides to respond to Taipei switch T first, while simultaneously requesting an update from the central database. The message Hong Kong switch H replies to Taipei switch T includes: Sending switch identifier: H; Destination switch identifier: T; Message type: Routing data; Reply routing data version number: 20231015-142; Routing table entries: {Prefix: +44-20, Destination gateway: European switch E, Path cost: 1.2, Time to live: 10 minutes}; Checksum: 0x7A3F21D9

[0086] 10. After receiving routing data, Taipei switch T stores the data in its local routing table and uses the acquired routing data to perform routing decisions.

[0087] 11. Taipei switch T establishes call signaling to European switch E, and connects to the local PSTN network in the UK through this node, eventually dialing the target number.

[0088] 12. At the same time, the Hong Kong switch H obtains updated UK area code routing data from the central database, including the latest rates and optimized routes.

[0089] 13. Hong Kong switch H pushes updated routing information to Taipei switch T through a point-to-point communication channel, so that T's local routing table is also updated.

[0090] 14. The entire routing data acquisition and call setup process takes about 220 milliseconds, which is longer than the first embodiment, but still about 35% faster than the traditional fully centralized control mode.

[0091] 15. Taipei switch T and Hong Kong switch H will use the routing decision and data exchange records of this time for routing learning. In the future, similar calls can be completed directly using the local routing table without needing to query again.

[0092] This embodiment demonstrates how the system can flexibly obtain routing information from adjacent network voice switches or a central database when the local routing table cannot provide the required information, reflecting the distributed collaboration characteristics of the present invention. It is particularly noteworthy that, in one embodiment of the present invention, even when local data is insufficient, the system still prioritizes point-to-point communication and only accesses the central database when necessary, thereby maintaining the system's efficiency and autonomy.

[0093] Although various examples of the disclosed technology have been described above, it should be understood that these examples are presented by way of example and are not limiting. Similarly, various diagrams can depict the example architectures or other configurations of the disclosed technology, and can be drawn to aid in understanding the features and functionalities that may be included in the disclosed technology. The disclosed technology is not limited to the example architectures or configurations described, but the desired features can be implemented using various alternative architectures and configurations. In fact, those skilled in this art will readily understand how alternative functionalities, logical or entity partitions and configurations can be implemented to implement the desired features of the technology disclosed herein. Furthermore, regarding flowcharts, operational descriptions, and methodological solutions, the order in which the steps are presented herein should not require that the disclosed technology be implemented in the same order to perform the enumerated functionalities, unless the context otherwise requires.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Therefore, all equivalent changes and modifications made in accordance with the shape, structure, features and spirit described in the claims of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An automatic routing method for network voice switches based on edge computing, comprising: (A) initializing a plurality of network voice switches, loading them into the system's basic configuration, starting the necessary communication modules and initialization processing units; (B) each of the plurality of network voice switches establishing a local data storage area to store data required for local routing decisions; (C) the plurality of network voice switches connecting to a central database and performing initial data synchronization, loading the data in the central database into the local data storage area of ​​each network voice switch, including at least the data of the plurality of network voice switches, the switch group relationship of the hierarchy to which the plurality of network voice switches belong, data of each telecommunications service provider, and fee data; (D) each network voice switch establishing a local routing table based on the data in its local data storage area and setting data update frequency parameters; (E) establishing a point-to-point communication channel between the network voice switches for directly exchanging routing data and status data without going through... The central database; (F) Based on preset triggering conditions, each voice switch autonomously decides whether to update data from the central database or obtain the latest routing data from neighboring voice switches; (G) When each voice switch receives a voice communication command, it first uses its local routing table to perform routing decisions. If the required routing data exists in the local routing table, it executes the voice communication command; (H) If the required routing data does not exist in the local routing table or the data has expired, each voice switch selects to obtain the required routing data from a neighboring voice switch or the central database based on the network connection status; (I) It updates the local routing table based on the latest obtained routing data and executes the voice communication command; and each voice switch includes a data layering module for storing frequently used routing data in fast-access memory, storing low-frequency used routing data in secondary memory, and dynamically adjusting the data storage location according to the usage frequency.

2. The automatic routing method for a network voice switch as described in claim 1, wherein the preset triggering conditions include: the expiration of the time-to-live of a local routing table entry, the network load being lower than a preset threshold, a new routing request being unable to be resolved by the local routing table, or receiving a routing update notification from another network voice switch.

3. The automatic routing method for a network voice switch as described in claim 1, wherein the point-to-point communication channel established in step (E) uses a lightweight protocol that includes at least the following fields: sending switch identifier, destination switch identifier, message type, routing data version number, routing table entry, and checksum.

4. The automatic routing method for a network voice switch as described in claim 1, wherein each network voice switch dynamically adjusts its synchronization frequency with the central database according to the following parameters: network load, local processing capacity, routing request frequency, and internal time.

5. The automatic routing method for a network voice switch as described in claim 1, wherein each network voice switch includes a fault detection module for automatically switching to an autonomous operation mode when an interruption in connection with the central database is detected, wherein routing decisions are made using only the local routing table and routing data from neighboring network voice switches.

6. The automatic routing method for a network voice switch as described in claim 1, wherein each network voice switch implements a routing learning module, which optimizes the local routing table based on past successful routing decisions and user behavior patterns, and sends the learning results back to the central database under specific conditions.

7. An automatic routing system for network voice switches based on edge computing, the system comprising: a plurality of network voice switches, each network voice switch being electrically connected to one or more VoIP phones, the plurality of network voice switches belonging to one or more switch groups, wherein, Each of the switch groups is interconnected via a trunk line; a central database stores global routing data, and each voice switch can periodically synchronize with the central database; multiple local data storage areas are respectively set in each voice switch to store data synchronized from the central database and locally generated routing decision data; when each voice switch receives a voice communication command, it first uses the data in the local data storage area to independently perform routing decisions. If the local data is insufficient or expired, it selects to obtain the required routing data from an adjacent voice switch or the central database; and each voice switch includes a data hierarchical module to store frequently used routing data in fast-access memory, store less frequently used routing data in secondary memory, and dynamically adjust the data storage location according to the usage frequency.

8. The voice switching system as described in claim 7, wherein each of the network voice switches is equipped with a resource monitoring module that monitors local computing resources, storage resources and network connectivity in real time, and adjusts local decision-making ratios and data synchronization strategies accordingly.

9. The voice switching system as described in claim 7, wherein the central database implements a differential update mechanism, so that each network voice switch only needs to download data that has changed since the last synchronization, thereby saving network bandwidth and synchronization time.

Images