Multilink switching method based on heterogeneous network

Abstract

The invention discloses a multi-link switching method based on a heterogeneous network, relates to the technical field of multi-link switching, and is used for solving the problems of low link state detection and switching efficiency and link state misjudgment caused by a single address in the prior art. According to the method, the number of heterogeneous networks is determined, link dynamic binding and priority configuration are carried out on each heterogeneous network, the link weight and the link switching threshold value of each heterogeneous network are self-defined to serve as switching judgment logic, a dynamic routing protocol is constructed, and an optimal routing table of a link corresponding to each heterogeneous network is generated in real time; and executing routing control through the optimal routing table, performing multi-thread independent detection, binocular address detection and IP state monitoring on the link of each heterogeneous network, obtaining the link state of each link based on detection and monitoring results, and switching the link according to the link state of the link to which the heterogeneous network belongs and switching judgment logic. And a corresponding switching strategy is formulated for synchronous execution, so that efficient and accurate multi-link switching is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-link switching, and specifically to a multi-link switching method based on a heterogeneous network. Background Art

[0002] With the rapid development of information technology, the heterogeneous network environment has become the norm in modern communication. In this environment, networks of various carrier types are interdependent, such as Wi-Fi, cellular networks, satellite networks, etc. However, existing multi-link switching methods based on heterogeneous networks often have inaccurate link state detection and low efficiency when switching between links. How to achieve efficient and seamless link switching between heterogeneous networks to ensure the continuity and stability of data transmission has always been a problem and challenge faced by the industry. Summary of the Invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide a multi-link switching method based on a heterogeneous network.

[0004] The purpose of the present invention can be achieved through the following technical solutions: A multi-link switching method based on a heterogeneous network, comprising the following steps:

[0005] Step S1: Determine the number of heterogeneous networks, perform link dynamic binding and priority configuration for each heterogeneous network, and customize the link weight and link switching threshold corresponding to each heterogeneous network as the switching decision logic;

[0006] Step S2: Construct a dynamic routing protocol, generate the optimal routing table for the link corresponding to each heterogeneous network in real time according to the dynamic routing protocol, and execute the respective routing control through the optimal routing table of each link;

[0007] Step S3: Perform multi-threaded independent detection, dual destination address detection, and IP status monitoring on the link corresponding to each heterogeneous network, and obtain the link state of each link based on the detection and monitoring results;

[0008] Step S4: Develop corresponding switching strategies according to the link state of the link to which the heterogeneous network belongs and the switching decision logic, and synchronously execute the switching strategies.

[0009] Further, the process of determining the number of heterogeneous networks, performing link dynamic binding and priority configuration for each heterogeneous network includes:

[0010] Construct heterogeneous networks of different carrier types, including wireless cellular networks, WIFI networks, and Ethernet networks;

[0011] Call the preset operating system network stack interface to traverse each heterogeneous network in turn, obtain the identification code of the carrier type corresponding to each heterogeneous network, and bind different links of the corresponding heterogeneous network according to different identification codes;

[0012] Monitor the link communication quality of each link respectively, obtain the initial link binding weight of each link based on the link communication quality, update and obtain the real-time link binding weight of each link through the exponential weighted moving average algorithm based on the initial link binding weight of each link, and record the real-time link binding weight as W new ;

[0013] Set the minimum threshold of the weight when the communication intensity of the link meets the requirements, and record it as W min ;

[0014] When W new ≥W min , do nothing;

[0015] When W new <W min , reconstruct the link of the corresponding heterogeneous network;

[0016] Preset the basic priorities of heterogeneous networks of different carrier types. The architecture central server allocates resources for all links. After the link obtains communication resources, it executes its own data communication. Judge whether the data communication under each link is over-limit communication, and change the initial priority of the link according to the judgment result, and convert the initial priority into the real-time demand priority.

[0017] Furthermore, the process of judging whether the data communication under each link is over-limit communication includes:

[0018] Set the number of times of data communication for each link, obtain the real-time communication index of each data communication, and judge whether the communication resources during each data communication under each link meet the minimum resource amount for maintaining normal communication;

[0019] If so, judge that the current link does not belong to over-limit communication, and further judge whether the real-time communication index meets the requirements; when the real-time communication index of the link meets the requirements, calculate whether the remaining communication resources after the current link completes the current data communication can maintain the next data communication. If so, set the real-time demand priority of the current link to the highest demand level, if not, set the real-time demand priority of the current link to the second highest demand level; when the real-time communication index of the link does not meet the requirements, set the real-time demand priority of the current link to the lowest demand level;

[0020] If not, it is determined that the current link belongs to over-limit communication, the data communication of the current link is suspended, and the communication resource allocation at the central server is re-requested to be allocated to the current link. After the current link obtains the communication resource, the data communication continues.

[0021] Furthermore, set the number of times of data communication for each link, obtain the real-time communication metrics for each data communication, and determine whether the communication resources for each data communication under each link meet the minimum resource amount for maintaining normal communication;

[0022] If so, it is determined that the current link does not belong to over-limit communication, and further determine whether the real-time communication metrics meet the requirements; when the real-time communication metrics of the link meet the requirements, calculate whether the remaining communication resources after the current link completes the current data communication can maintain the next data communication. If so, set the real-time demand priority of the current link to the highest demand level. If not, set the real-time demand priority of the current link to the second highest demand level; when the real-time communication metrics of the link do not meet the requirements, set the real-time demand priority of the current link to the lowest demand level;

[0023] If not, it is determined that the current link belongs to over-limit communication, the data communication of the current link is suspended, and the communication resource allocation at the central server is re-requested to be allocated to the current link. After the current link obtains the communication resource, the data communication continues.

[0024] Furthermore, the process of customizing the link weight and link switching threshold corresponding to each heterogeneous network as the handover decision logic includes:

[0025] For the link corresponding to each heterogeneous network, count the number of times when the data communication under the link is at the highest demand level, the second highest demand level, and the lowest demand level respectively, assign corresponding value coefficients according to different real-time demand priorities, and accumulate and calculate to obtain the link weight of the link corresponding to the current heterogeneous network;

[0026] Set the link switching threshold corresponding to each link, including the hard handover threshold, the soft handover threshold, and the hysteresis threshold. Among them, the hard handover threshold is used as the critical condition for the current link to switch to the pre-corresponding backup link or secondary backup link;

[0027] Integrate the link weight and link switching threshold of each link of each heterogeneous network as the handover decision logic for each link, and transmit the handover decision logic to the pre-constructed cloud database for storage.

[0028] Furthermore, the process of constructing a dynamic routing protocol and generating the optimal routing table for each link corresponding to each heterogeneous network in real time according to the dynamic routing protocol includes:

[0029] Construct a dynamic routing protocol by associating the links belonging to the heterogeneous network, perform routing supervision through the dynamic routing protocol, divide the links under routing supervision by the dynamic routing protocol into several network routing segments, and the dynamic routing protocol is used to determine whether the routing forwarding of the corresponding links of the heterogeneous network meets the standards under each network routing segment;

[0030] Obtain the real-time network traffic of the link under each network routing segment, set the overload traffic when network congestion occurs, and if the real-time network traffic of the link in a certain network routing segment exceeds the overload traffic, it is determined that the routing forwarding under the corresponding network routing segment does not meet the standards, otherwise, it is determined that the routing forwarding under the corresponding network routing segment meets the standards;

[0031] When there is non-compliant routing forwarding in the network routing segment, construct the data forwarding path of the current network routing segment, connect the data forwarding path to the next network routing segment of the current link, and re-forward the data under the data forwarding path;

[0032] Until the routing forwarding of each network routing segment corresponding to the link is compliant, the optimal routing table of the link corresponding to the current heterogeneous network is constructed;

[0033] When the routing forwarding of the network routing segment meets the standards, no operation is performed.

[0034] Further, the process of performing respective routing control through the optimal routing table of each link includes:

[0035] When forwarding data packets in each link, match the corresponding destination IP for several network routing segments on each link, and send the data packets flowing and transferring in real time under each network routing segment to the preset target receiver, carrying the routing verification IPs of several corresponding network routing segments recorded in their respective optimal routing tables;

[0036] By judging the pairing relationship between the destination IP and the routing verification IP of each network routing segment, and according to the judgment result, decide whether to receive the data packets under the corresponding network routing segment, thereby completing the routing control under the corresponding link;

[0037] Before the target receiver receives the data packets of the corresponding network routing segment, judge whether the destination IP and the routing verification IP of the network routing segment are in the preset IP pairing set. If so, it means that a pairing relationship is formed between the destination IP and the routing verification IP of the current network routing segment, and the target receiver preferentially selects the routing entry with the longest subnet mask in itself and constructs a data receiving path to receive the data packets. If not, it means that a pairing relationship is not formed between the destination IP and the routing verification IP of the current network routing segment, and the data packets sent by the current network routing segment are rolled back.

[0038] Further, the process of independently detecting each link corresponding to a heterogeneous network in multiple threads, detecting dual destination addresses, and monitoring the IP status, and obtaining the link status of each link based on the detection and monitoring results includes:

[0039] Establish corresponding link supervision tasks for each link corresponding to a heterogeneous network. The link supervision tasks are used to supervise the status of the links belonging to the heterogeneous network to screen out the links in different link states. The link supervision tasks specifically include independent multi-threaded detection, dual destination address detection, and IP status monitoring;

[0040] Obtain the detection results after the independent multi-threaded detection and the dual destination address detection are each completed; obtain the monitoring results after the IP status monitoring is completed; comprehensively analyze the monitoring results and the monitoring results of the link supervision tasks under each link to obtain the link status of each link. The link status includes a fault state, a suspected fault state, and a healthy state.

[0041] Further, according to the link status of the links belonging to the heterogeneous network and the handover determination logic, formulating corresponding handover strategies, and the process of synchronously executing the handover strategies includes:

[0042] Obtain the data communication permission with the cloud database, read the handover determination logic for multi-link handover stored in the cloud database. When the link belonging to the heterogeneous network is in a healthy state, no operation is performed;

[0043] When the link belonging to the heterogeneous network is in a fault state, formulate and synchronously execute an emergency handover strategy according to the hard handover threshold and the link weight of the link. The content of the emergency handover strategy is: when the link weight of the link exceeds the hard handover threshold, switch the currently faulty link to its corresponding backup link or secondary backup link with the highest priority;

[0044] When the link belonging to the heterogeneous network is in a suspected fault state, formulate and synchronously execute a conventional handover strategy according to the soft handover threshold and the link weight of the link. The content of the conventional handover strategy is: when the link weight of the link exceeds the soft handover threshold, select any backup link for handover, and set active polling for the current link to monitor whether the link status changes to a healthy state. If so, after waiting for the occupied resources of the current link to be released, jump from the backup link to the current link. If not, maintain the operation of the backup link.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] 1. By performing link dynamic binding and priority configuration for each heterogeneous network, and customizing the link weights and link switching thresholds corresponding to each heterogeneous network as the switching decision logic, a dynamic routing protocol is constructed to generate the optimal routing table for the links corresponding to each heterogeneous network in real time, and then perform respective routing control, laying a good prerequisite for the subsequent switching of each link, facilitating the execution of subsequent switching strategies, and improving the link switching efficiency.

[0047] 2. Perform multi-threaded independent detection, dual destination address detection, and IP status monitoring on the links corresponding to each heterogeneous network. Based on the detection and monitoring results, obtain the link status of each link. According to the link status of the links belonging to the heterogeneous network and the switching decision logic, formulate corresponding switching strategies and execute them synchronously, greatly improving the accuracy of link status detection and effectively ensuring the continuity and stability of data transmission. Brief Description of the Drawings

[0048] Figure 1 It is a flowchart of the present invention. Detailed Embodiments

[0049] As Figure 1 shown, a multi-link switching method based on heterogeneous networks includes the following steps:

[0050] Step S1: Determine the number of heterogeneous networks, perform link dynamic binding and priority configuration for each heterogeneous network, and customize the link weights and link switching thresholds corresponding to each heterogeneous network as the switching decision logic;

[0051] Step S2: Construct a dynamic routing protocol, generate the optimal routing table for the links corresponding to each heterogeneous network in real time according to the dynamic routing protocol, and execute respective routing control through the optimal routing table of each link;

[0052] Step S3: Perform multi-threaded independent detection, dual destination address detection, and IP status monitoring on the links corresponding to each heterogeneous network. Based on the detection and monitoring results, obtain the link status of each link;

[0053] Step S4: According to the link status of the links belonging to the heterogeneous network and the switching decision logic, formulate corresponding switching strategies and execute the switching strategies synchronously.

[0054] It should be further noted that, in the specific implementation process, the process of determining the number of heterogeneous networks and performing link dynamic binding and priority configuration for each heterogeneous network includes:

[0055] Construct heterogeneous networks of different carrier types, where the heterogeneous networks of different carrier types specifically include wireless cellular networks, WIFI networks, and Ethernet networks. Determine the number corresponding to the heterogeneous networks of all carrier types as N, where N is a natural number greater than 0;

[0056] Call the preset operating system network stack interface to traverse each heterogeneous network in sequence, and obtain the identification codes of the carrier types corresponding to each heterogeneous network from the operating system network stack interface. Among them, the identification codes of wireless cellular networks, WIFI networks, and Ethernet networks are respectively denoted as Sign1, Sign2, and Sign3;

[0057] When the obtained identification code is Sign1, bind the first transmission link to the corresponding heterogeneous network;

[0058] When the obtained identification code is Sign2, bind the second transmission link to the corresponding heterogeneous network;

[0059] When the obtained identification code is Sign3, bind the third transmission link to the corresponding heterogeneous network;

[0060] Respectively monitor the link communication quality of the first transmission link, the second transmission link, and the third transmission link, and obtain the initial link binding weights of the first transmission link, the second transmission link, and the third transmission link based on the link communication quality. The initial link binding weight is used to represent the ratio between the actual communication intensity of the corresponding link in the actual situation and the expected communication intensity in the ideal situation during the initial period;

[0061] Link binding weight = actual communication intensity / expected communication intensity, where the link binding weight is a real number taking values in the interval (0, 1). The closer the value of the link binding weight is to 1, the more the current link's communication intensity meets the requirements. The more the value deviates from 1, the farther the current link's communication intensity is from the requirements;

[0062] Based on the initial link binding weights of each link, update and obtain the real-time link binding weights of each link through the exponential weighted moving average algorithm, and denote the real-time link binding weight as W new , denote the initial link binding weight as W current ;

[0063]

[0064] Among them, α is the smoothing adjustment factor, B instant is the instantaneous bandwidth of the link at the current moment, B Max is the maximum bandwidth of the link. The value of α is set to 0.7, and its value can be changed subsequently;

[0065] Set the minimum weight threshold for the communication intensity of the link to meet the requirements, and denote it as W min ;

[0066] When W new ≥W min , no operation is performed;

[0067] When W new <W min , reconstruct the link corresponding to the corresponding heterogeneous network.

[0068] Preset the basic priorities of heterogeneous networks of different carrier types. The order of basic priorities from high to low is as follows: Ethernet > Cellular network > WIFI network. Therefore, the order of the initial priorities of the corresponding links from high to low is: the third transmission link > the first transmission link > the second transmission link;

[0069] The architecture central server is used to allocate resources for all links. Each link requests communication resources from the central server. The central server packs the corresponding amount of communication resources to the corresponding link, and the link obtains the communication resources and then performs its own data communication;

[0070] Judge whether the data communication under each link is over-limit communication, and change the initial priority of the link according to the judgment result, and convert the initial priority into the real-time demand priority. The specific judgment process is as follows:

[0071] Set the number of times for each link to perform data communication. Based on the data communication, obtain the real-time communication metrics for each time. The real-time communication metrics specifically include communication delay, window jitter intensity, and packet loss rate. First, judge whether the communication resources for each data communication under each link meet the minimum resource amount for maintaining normal communication;

[0072] If so, judge that the current link does not belong to over-limit communication, and further judge whether the real-time communication metrics meet the requirements; specifically as follows:

[0073] Set the metric intervals corresponding to the real-time communication metrics;

[0074] Specifically include the delay safety interval corresponding to the communication delay, the jitter intensity compliance interval corresponding to the window jitter intensity, and the retransmission determination interval corresponding to the packet loss rate. If the delay value of the communication delay is within the delay safety interval, the intensity level of the window jitter intensity is within the intensity compliance interval, and the packet loss rate is within the retransmission determination interval, then judge that the real-time communication metrics of the current link meet the requirements, otherwise, judge that the real-time communication metrics do not meet the requirements;

[0075] When the real-time communication metrics of the link meet the requirements, calculate whether the remaining communication resources after the current link completes the current data communication can sustain the next data communication. If so, set the real-time demand priority of the current link to the highest demand level. If not, set the real-time demand priority of the current link to the second-highest demand level;

[0076] When the real-time communication metrics of the link do not meet the requirements, set the real-time demand priority of the current link to the lowest demand level;

[0077] If not, determine whether the current link belongs to over-limit communication, suspend the data communication of the current link, and re-request the communication resource allocation from the central server to the current link. After the current link obtains the communication resource, continue the data communication.

[0078] It should be noted that when data communication is carried out on the links belonging to the heterogeneous network, the real-time demand priorities from high to low are: the highest demand level, the second-highest demand level, and the lowest demand level. When both the real-time communication metrics and the communication resources in the link meet the corresponding requirements, the link is in the best communication state at this time, and the data communication efficiency is the highest. And the communication resource is a necessary means to maintain the data communication of the link. Judging whether the communication resource of the current data communication can meet the minimum resource amount for normal communication can control the start and stop of the data communication under the corresponding link. When the communication resource meets the demand but the real-time communication metrics do not meet the requirements, the link can still carry out data communication, but the communication efficiency is worse than that when the link is in the best communication state.

[0079] It should be further noted that in the specific implementation process, the process of customizing the link weight and link switching threshold corresponding to each heterogeneous network as the switching decision logic includes:

[0080] For the links corresponding to each heterogeneous network, count the number of times when the data communication under the link is at the highest demand level, the second-highest demand level, and the lowest demand level respectively, and assign corresponding value coefficients according to different real-time demand priorities, and accumulate and calculate to obtain the link weight of the link corresponding to the current heterogeneous network;

[0081] Specifically as follows:

[0082] The value coefficient assigned to the highest demand level is 3;

[0083] The value coefficient assigned to the second-highest demand level is 2;

[0084] The value coefficient assigned to the lowest demand level is 1;

[0085] Denote the link weight as W line Then there is the following cumulative calculation formula:

[0086] Wline = N1 × 3 + N2 × 2 + N3 × 1;

[0087] Wherein, N1, N2, and N3 are the corresponding frequencies when the link is at the highest demand level, the second-highest demand level, and the lowest demand level respectively;

[0088] Set the link switching thresholds corresponding to each link, specifically including the hard handover threshold, the soft handover threshold, and the hysteresis threshold. Among them, the hard handover threshold is used as the critical condition for the current link to switch to the pre-correspondingly constructed backup link or secondary backup link, the soft handover threshold is used as the warning condition for the current link to perform a handover, and the hysteresis threshold is used as the buffer interval for the current link to avoid frequent handovers;

[0089] Integrate the link weights and link switching thresholds of each link in each heterogeneous network as the switching decision logic for each link, and transmit the switching decision logic to the pre-constructed cloud database for storage.

[0090] It should be further noted that in the specific implementation process, the process of constructing a dynamic routing protocol and generating the optimal routing table for each link corresponding to each heterogeneous network in real time includes:

[0091] According to the carrier type of the heterogeneous network, select and construct the dynamic routing protocol for the links under the corresponding heterogeneous network. The dynamic routing protocol is used to determine whether the routing forwarding of the links corresponding to the heterogeneous network meets the standard in each network routing segment;

[0092] Associate the dynamic routing protocol constructed with the links belonging to the heterogeneous network, and then perform routing supervision on the links belonging to the heterogeneous network through the associated dynamic routing protocol, and divide the links under routing supervision under the dynamic routing protocol into several network routing segments;

[0093] Obtain the real-time network traffic of each network routing segment corresponding to the link;

[0094] Set the overload traffic corresponding to network congestion;

[0095] If the real-time network traffic of a link in a certain network routing segment exceeds the overload traffic, it is determined that the routing forwarding in the corresponding network routing segment does not meet the standard; otherwise, it is determined that the routing forwarding in the corresponding network routing segment meets the standard;

[0096] When there is non-compliant routing forwarding in a network routing segment, perform the routing optimization operation under the corresponding network routing segment. Specifically: construct the data forwarding path of the current network routing segment, and connect the data forwarding path to the next network routing segment of the current link. Under the data forwarding path, re-forward all the data that needs to be forwarded corresponding to the current network routing segment;

[0097] When the routing forwarding of each network routing segment corresponding to the link all meets the standards, the optimal routing table of the link corresponding to the current heterogeneous network is constructed. The optimal routing table is used to represent the shortest routing path when the link forwards data. When there is a faulty network routing segment on the shortest routing path, the routing information of the corresponding network routing segment is obtained, and an extended routing segment is constructed according to the routing information. The extended routing segment is used to replace the faulty network routing segment;

[0098] When the routing forwarding of the network routing segment meets the standards, no operation is performed.

[0099] It should be further noted that in the specific implementation process, the process of performing respective routing control through the optimal routing table of each link includes:

[0100] When forwarding respective data packets in each link, match respective corresponding destination IPs for several network routing segments on each link, and send the data packets that are in real-time data interaction and transfer under each network routing segment, carrying the routing verification IPs of several corresponding network routing segments recorded in their respective optimal routing tables, to a preset target receiver;

[0101] By judging the pairing relationship between the destination IP and the routing verification IP of each network routing segment, and deciding whether to receive the data packets under the corresponding network routing segment according to the judgment result, thereby completing the routing control under the corresponding link;

[0102] Before the target receiver receives the data packets of the corresponding network routing segment, judge whether the destination IP and the routing verification IP of the network routing segment are in a preset IP pairing set. If so, it means that a pairing relationship is formed between the destination IP and the routing verification IP of the current network routing segment. The target receiver preferentially selects the routing entry with the longest subnet mask in itself and constructs a data reception path to receive the data packets. If not, it means that a pairing relationship is not formed between the destination IP and the routing verification IP of the current network routing segment, and the data packets sent by the current network routing segment are rolled back.

[0103] It should be further noted that in the specific implementation process, the process of performing multi-threaded independent detection, dual destination address detection, and IP status monitoring on the links corresponding to each heterogeneous network, and obtaining the link status of each link based on the detection and monitoring results includes:

[0104] Establish corresponding link supervision tasks for the links corresponding to each heterogeneous network. The link supervision tasks are used to continuously supervise the status of the links belonging to the heterogeneous network to screen out the links in different link states. The link supervision tasks specifically include multi-threaded independent detection, dual destination address detection, and IP status monitoring;

[0105] Obtain the detection results after the multi-threaded independent detection and the dual-purpose address detection are each completed; obtain the monitoring results after the IP status monitoring is completed; comprehensively analyze the monitoring results and the monitoring results of the link supervision tasks under each link to obtain the link status of each link, where the link status includes a fault status, a suspected fault status, and a healthy status;

[0106] The specific content of the multi-threaded independent detection is as follows: Create several independent threads corresponding to each link, configure link parameters for each independent thread, complete the thread initialization of the corresponding independent thread, allow several independent threads to execute synchronously, and each independent thread is used to execute the health detection task of the link. Obtain the real-time data traffic status of the corresponding link through the health detection task, and different independent threads synchronize the data traffic status through shared memory;

[0107] The data traffic status includes a congestion status and a smooth status;

[0108] It should be noted that by creating several independent threads for each link, each independent thread is used to detect the data traffic status of the current link, preventing delays in detecting the real-time data traffic status of the link when there is only one thread and the thread is blocked.

[0109] The specific content of the dual-purpose address detection is as follows: Create a monitoring thread for each link in sequence, and assign a Ping detection target for each line. Configure two independent IP addresses for the Ping detection target. The two independent IP addresses are specifically the gateway IP and the public DNS server IP. The monitoring thread of each link periodically sends a Ping request for the Ping detection target to the two independent IP addresses at a preset periodic interval. If any independent IP address responds successfully, it is determined that the link status of the link is a healthy status. If both independent IP addresses respond failed, it is determined that the link status of the link is a suspected fault status;

[0110] The specific content of the IP status monitoring is as follows: Monitor the IP change events under the independent IP address corresponding to the link in the suspected fault status. The IP change events include but are not limited to NETLINK messages in the Linux kernel, DHCP reallocation-related information, and manual configuration change information. If the IP address status of the monitored IP change event is abnormal, the link status of the link is changed from the suspected fault status to the fault status. Otherwise, set a regular polling period, and continuously monitor the IP change events of the current independent IP address within the regular polling period. If no abnormal IP address status is still found, the link status of the link is changed from the suspected fault status to the healthy status.

[0111] It should be further noted that in the specific implementation process, according to the link state of the heterogeneous network and the handover determination logic, corresponding handover strategies are formulated. The process of synchronously executing the handover strategies includes:

[0112] Obtain the data communication permission with the cloud database, and then read the handover determination logic for multi-link handover stored in the cloud database. When the link to which the heterogeneous network belongs is in a healthy state, no operation is performed;

[0113] When the link to which the heterogeneous network belongs is in a fault state, according to the hard handover threshold in the link handover threshold and the link weight of the link, an emergency handover strategy is formulated and synchronously executed. The content of the emergency handover strategy is: when the link weight of the link exceeds the hard handover threshold, the link currently in the fault state is switched to its corresponding backup link or secondary backup link with the highest priority;

[0114] When the link to which the heterogeneous network belongs is in a suspected fault state, according to the soft handover threshold in the link handover threshold and the link weight of the link, a conventional handover strategy is formulated and synchronously executed. The content of the conventional handover strategy is: when the link weight of the link exceeds the soft handover threshold, any backup link is selected for handover, and continuous low-frequency active polling is set for the current link to monitor whether the link state changes to a healthy state. If so, after the occupied resources of the current link are released, the backup link is jumped to the current link. If not, the operation of the backup link is maintained.

[0115] When the link and the backup link or secondary backup link perform mutual handover, obtain the handover interval duration. If the handover interval duration is less than or equal to the hysteresis threshold, the mutual handover between the link and the backup link or secondary backup link is paused. Otherwise, the mutual handover between the link and the backup link or secondary backup link is allowed.

[0116] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multi-link switching method based on a heterogeneous network, characterized in that The steps include: Step S1: Determine the number of heterogeneous networks, perform link dynamic binding and priority configuration for each heterogeneous network, and customize the link weight and link switching threshold corresponding to each heterogeneous network as the switching decision logic; Step S2: Construct a dynamic routing protocol, generate the optimal routing table of the link corresponding to each heterogeneous network in real time according to the dynamic routing protocol, and execute the respective routing control through the optimal routing table of each link; Step S3: Perform multi-threaded independent detection, dual-purpose address detection, and IP status monitoring on the links corresponding to each heterogeneous network, and obtain the link status of each link based on the detection and monitoring results; Step S4: Develop corresponding handover strategies according to the link status of the links belonging to the heterogeneous network and the handover decision logic, and synchronously execute the handover strategies.

2. The multi-link switching method based on a heterogeneous network according to claim 1, characterized in that The process of determining the number of heterogeneous networks and performing link dynamic binding and priority configuration for each heterogeneous network includes: Construct heterogeneous networks of different carrier types, including wireless cellular networks, WIFI networks, and Ethernet networks; Call the preset operating system network stack interface to traverse each heterogeneous network in turn, obtain the identification code of the carrier type corresponding to each heterogeneous network, and bind different links of the corresponding heterogeneous network according to different identification codes; Monitor the link communication quality of each link separately, obtain the initial link binding weight of each link based on the link communication quality, update and obtain the real-time link binding weight of each link through the exponentially weighted moving average algorithm based on the initial link binding weight of each link, and record the real-time link binding weight as W new ; Set the minimum threshold of the weight for the communication strength of the link to meet the requirements, and denote it as W min ; When W new ≥ W min No operation is performed; When W new <W min reconstruct the links of the corresponding heterogeneous network; Preset the basic priorities of heterogeneous networks of different carrier types, and the architecture central server allocates resources for all links. After the link obtains communication resources, it executes its own data communication. Judge whether the data communication under each link is over-limit communication, and change the initial priority of the link according to the judgment result, and convert the initial priority into the real-time demand priority.

3. A multi-link switching method based on a heterogeneous network according to claim 2, characterized in that, The process of judging whether the data communication under each link is over-limit communication includes: Set the number of times of data communication for each link, obtain the real-time communication index of each data communication, and judge whether the communication resources during each data communication under each link meet the minimum resource amount for maintaining normal communication; If so, judge that the current link does not belong to over-limit communication, and further judge whether the real-time communication index meets the requirements; when the real-time communication index of the link meets the requirements, calculate whether the remaining communication resources after the current link completes the current data communication can maintain the next data communication. If so, set the real-time demand priority of the current link to the highest demand level, if not, set the real-time demand priority of the current link to the second highest demand level; when the real-time communication index of the link does not meet the requirements, set the real-time demand priority of the current link to the lowest demand level; If not, judge that the current link belongs to over-limit communication, suspend the data communication of the current link, and re-request the communication resource allocation from the central server to the current link, and continue the data communication after the current link obtains the communication resources.

4. A multi-link switching method based on a heterogeneous network according to claim 3, characterized in that The process of customizing the link weight and link switching threshold corresponding to each heterogeneous network as the switching decision logic includes: For each link corresponding to a heterogeneous network, count the number of times when the data communication under the link is at the highest demand level, the second-highest demand level, and the lowest demand level respectively. Assign corresponding value coefficients according to different real-time demand priorities, and accumulate and calculate to obtain the link weight of the link corresponding to the current heterogeneous network. Set the link switching thresholds corresponding to each link, including the hard handover threshold, the soft handover threshold, and the hysteresis threshold. Among them, the hard handover threshold is the critical condition for the current link to switch to the pre-correspondingly constructed backup link or secondary backup link. Integrate the link weights and link switching thresholds of each link of each heterogeneous network as the switching decision logic for each link, and transmit the switching decision logic to the pre-constructed cloud database for storage.

5. A multi-link switching method based on a heterogeneous network according to claim 4, characterized in that, Construct a dynamic routing protocol. The process of generating the optimal routing table for each link corresponding to a heterogeneous network in real time according to the dynamic routing protocol includes: Associate the links belonging to the heterogeneous network to construct a dynamic routing protocol, and perform routing supervision through the dynamic routing protocol. Divide the links under routing supervision by the dynamic routing protocol into several network routing segments. The dynamic routing protocol is used to determine whether the routing forwarding of the heterogeneous network corresponding link is qualified under each network routing segment. Obtain the real-time network traffic of the link under each network routing segment, and set the overload traffic when network congestion occurs. If the real-time network traffic of the link in a certain network routing segment exceeds the overload traffic, it is determined that the routing forwarding under the corresponding network routing segment is unqualified; otherwise, it is determined that the routing forwarding under the corresponding network routing segment is qualified. When there is unqualified routing forwarding in a network routing segment, construct the data forwarding path of the current network routing segment, and connect the data forwarding path to the next network routing segment of the current link. Under the data forwarding path, re-forward the data. Until the routing forwarding of each network routing segment corresponding to the link is qualified, the optimal routing table of the link corresponding to the current heterogeneous network is constructed. When the routing forwarding of the network routing segment is qualified, no operation is performed.

6. The multi-link switching method based on a heterogeneous network according to claim 5, characterized in that, The process of performing respective routing control through the optimal routing table of each link includes: When forwarding data packets on each link, match the corresponding destination IP for several network routing segments on each link, and send the data packets that are in real-time data interaction and transfer under each network routing segment, carrying the routing verification IPs of several corresponding network routing segments recorded in their respective optimal routing tables, to the preset target recipient. By judging the pairing relationship between the destination IP and the routing verification IP of each network routing segment, decide whether to receive the data packets under the corresponding network routing segment according to the judgment result, and thus complete the routing control under the corresponding link. Before the target receiver receives the data packets of the corresponding network routing segment, determine whether the destination IP of the network routing segment and the routing verification IP are in a preset IP pair set. If so, it means that a pairing relationship is formed between the destination IP and the routing verification IP of the current network routing segment. The target receiver preferentially selects the routing entry with the longest subnet mask in itself and constructs a data reception path to receive the data packets. If not, it means that no pairing relationship is formed between the destination IP and the routing verification IP of the current network routing segment, and the data packets sent by the current network routing segment are rolled back.

7. A multi-link switching method based on a heterogeneous network according to claim 6, characterized in that, The process of obtaining the link state of each link through multi-threaded independent detection, dual destination address detection, and IP status monitoring for the links corresponding to each heterogeneous network includes: Establish corresponding link supervision tasks for the links corresponding to each heterogeneous network. The link supervision tasks are used to supervise the status of the links belonging to the heterogeneous network to filter out the links in different link states. The link supervision tasks specifically include multi-threaded independent detection, dual destination address detection, and IP status monitoring; Obtain the detection results after the completion of multi-threaded independent detection and dual destination address detection respectively; obtain the monitoring result after the completion of IP status monitoring; analyze and obtain the link state of each link by synthesizing the monitoring results and the monitoring results of the link supervision tasks under each link. The link state includes a fault state, a suspected fault state, and a healthy state.

8. A multi-link switching method based on a heterogeneous network according to claim 7, characterized in that According to the link state of the links belonging to the heterogeneous network and the handover decision logic, formulate corresponding handover strategies. The process of synchronously executing the handover strategies includes: Obtain the data communication permission with the cloud database, and read the handover decision logic for multi-link handover stored in the cloud database. When the link belonging to the heterogeneous network is in a healthy state, no operation is performed; When the link belonging to the heterogeneous network is in a fault state, formulate and synchronously execute an emergency handover strategy according to the hard handover threshold and the link weight of the link. The content of the emergency handover strategy is: when the link weight of the link exceeds the hard handover threshold, switch the link currently in the fault state to one of its highest-priority backup links or secondary backup links; When the link belonging to the heterogeneous network is in a suspected fault state, formulate and synchronously execute a conventional handover strategy according to the soft handover threshold and the link weight of the link. The content of the conventional handover strategy is: when the link weight of the link exceeds the soft handover threshold, select any backup link for handover, and set active polling for the current link to monitor whether the link state changes to a healthy state. If so, after waiting for the release of the occupied resources of the current link, jump from the backup link to the current link. If not, maintain the operation of the backup link.

Images