Adaptive interconnection method and device based on DDS in cloud-end heterogeneous network

Abstract

The invention discloses a DDS-based adaptive interconnection method and device in a cloud-end heterogeneous network. The method comprises the following steps: acquiring network environment detection data information, communication domain configuration information and a candidate communication strategy information set; performing communication strategy selection processing on the network environment detection data information and the candidate communication strategy information set, calculating the adaptation degree of each network node to each candidate communication strategy according to an evaluation result, and selecting an optimal discovery / transmission strategy combination for each network node to obtain a target communication strategy information set; performing federated cross-domain interconnection processing according to the target communication strategy information set and the communication domain configuration information, including generating domain configuration, calculating cross-domain routing preference and configuring federated bridging to obtain cross-domain interconnection configuration information; and finally, carrying out network monitoring and adaptive switching processing according to the cross-domain interconnection configuration information, continuously monitoring a network state, and triggering strategy hot switching when the network is degraded to obtain target interconnection result information.

Description

Technical Field

[0001] This invention relates to the field of distributed communication technology, and in particular to an adaptive interconnection method and apparatus based on DDS in a cloud-edge heterogeneous network. Background Technology

[0002] With the rapid development of cloud computing, edge computing, and IoT technologies, cloud-edge-device collaborative architecture has become the mainstream deployment model for distributed systems. In this architecture, Data Distribution Service (DDS), as a data-centric publish / subscribe communication middleware, is widely used in scenarios such as real-time data distribution, system integration, and cross-platform communication. However, in heterogeneous cloud-edge-device network environments, different network domains may employ different network technologies, such as RoCE (Real-Time DMA over Converged Ethernet), InfiniBand (IB), and traditional Ethernet. Interoperability between these heterogeneous networks faces significant challenges. Traditional DDS communication relies on RTPS (Real-Time Publish-Subscribe Protocol) multicast for automatic discovery and data distribution within a local area network (LAN). However, in cloud architectures, cross-segment networks, and network environments that do not support multicast, the discovery and transmission mechanisms of traditional DDS are often limited due to network architecture and network security policies.

[0003] Existing heterogeneous network DDS interconnection technologies mainly include the following methods: The first type is a VPN / Overlay-based solution, which achieves cross-network segment communication by establishing a virtual private network or overlay network; the second type is a static configuration-based solution, which achieves node interconnection by pre-configuring fixed discovery server addresses and transmission parameters; the third type is a single transport stack-based solution, which uses a fixed transport protocol (such as using only TCP or only UDP) for data transmission; and the fourth type is a manual configuration-based solution, which relies on maintenance personnel to manually edit configuration files to set network parameters. However, the aforementioned existing technologies have the following drawbacks: First, VPN / Overlay-based solutions increase the system complexity of distributed communication, introducing additional encapsulation / decapsulation overhead, leading to increased communication latency. Second, statically configured solutions cannot adapt to dynamically changing network environments, requiring manual intervention when network topology or node status changes. Third, solutions based on a single transport stack struggle to simultaneously balance communication performance and cross-segment connectivity, failing to fully leverage the low latency advantage of multicast when it is available, and lacking reliable alternatives when multicast is unavailable. Furthermore, manually configured solutions are prone to errors and configuration file conflicts, making large-scale operation and maintenance difficult, and they cannot be dynamically adjusted based on network status changes during runtime. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide an adaptive interconnection method and apparatus based on DDS in cloud-edge-device heterogeneous networks, which solves the adaptive interconnection problem faced by existing DDS communication technology in cloud-edge-device heterogeneous network environments. This invention can automatically select the optimal combination of discovery / transmission strategies based on network environment detection data, realize cross-domain transparent interconnection of heterogeneous networks through federated services, and support runtime hot switching, thereby improving the adaptability, performance and reliability of DDS communication in heterogeneous network environments and reducing the complexity of system deployment and operation and maintenance.

[0005] To address the aforementioned technical problems, a first aspect of this invention discloses an adaptive interconnection method based on DDS in a cloud-edge heterogeneous network, the method comprising: S1, acquire network environment detection data, communication domain configuration information, and candidate communication strategy information set; S2, process the network environment detection data information and the candidate communication strategy information set to obtain the target communication strategy information set; S3, process the target communication strategy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information; S4, process the cross-domain interconnection configuration information to obtain the target interconnection result information.

[0006] As an optional implementation, in a first aspect of the present invention, processing the network environment detection data information and the candidate communication strategy information set to obtain the target communication strategy information set includes: S21, perform network capability assessment processing on the network environment detection data information to obtain network capability assessment information; S22, perform policy fit calculation on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set; S23, Based on the policy fit information set and the candidate communication policy information set, determine the target communication policy information set.

[0007] As an optional implementation, in the first aspect of the present invention, the step of performing network capability assessment processing on the network environment detection data information to obtain network capability assessment information includes: S211, Perform multicast support analysis on the multicast detection response data in the network environment detection data information to obtain multicast support information; S212, Perform route reachability analysis on the route detection response data in the network environment detection data information to obtain route reachability information; S213, perform link quality analysis processing on the link quality sampling data in the network environment detection data information to obtain link quality information, the link quality information including network latency value and packet loss rate value; S214, Using a network capability comprehensive evaluation calculation model, the multicast support information, the route reachability information, and the link quality information are calculated and processed to obtain network capability evaluation information; The network capability comprehensive evaluation calculation model is as follows: ; ; ; In the formula, The network capability assessment information is the first one. A comprehensive evaluation value of the network capabilities of each network node. The first in the multicast support information The multicast support value of each network node. The first in the route reachability information The route reachability value of each network node. The link quality information of the first Network latency value of each network node This is the upper limit of the network latency threshold. The link quality information of the first The packet loss rate of each network node. As the attenuation factor, , , and These are the first weight parameter, the second weight parameter, the third weight parameter, and the fourth weight parameter, respectively. For correction items, N represents the number of network nodes in the cloud-edge heterogeneous network.

[0008] As an optional implementation, in the first aspect of the present invention, the step of performing policy fit calculation processing on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set includes: S221, perform strategy parameter parsing processing on the candidate communication strategy information set to obtain the strategy parameter information set; S222, with i=1 pre-defined; S223, default j=1; S224, perform adaptation calculation on the network capability assessment information and the j-th policy parameter information in the policy parameter information set to obtain the policy adaptation value of the j-th policy in the i-th network node; S225, determine whether j is equal to M, and obtain the first determination result, where M is the total number of strategy parameter information in the strategy parameter information set; When the first judgment result is negative, increment j by 1 and execute S224; When the first judgment result is yes, determine whether i is equal to N to obtain the second judgment result, where N is the total number of network nodes; When the second judgment result is negative, increment i by 1 and execute S223; When the second judgment result is yes, all strategy fit values ​​are added to the strategy fit information set.

[0009] As an optional implementation, in a first aspect of the present invention, processing the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information includes: S31, Process the target communication policy information set and the communication domain configuration information to obtain domain configuration information and cross-domain routing information set; S32, process the cross-domain routing information set to obtain the target cross-domain routing information; S33, process the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information.

[0010] As an optional implementation, in a first aspect of the present invention, processing the target communication policy information set and the communication domain configuration information to obtain domain configuration information and cross-domain routing information set includes: S311, The target communication strategy information set is extracted and processed to obtain a strategy configuration parameter information set; S312, The communication domain configuration information is extracted and processed to obtain the domain basic configuration information; S313, Process the policy configuration parameter information set and the domain basic configuration information to obtain domain configuration information; S314. Based on the domain configuration information, obtain the cross-domain routing information set from the routing table of the Federation Service Module.

[0011] As an optional implementation, in the first aspect of the present invention, processing the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information includes: S331, Based on the domain configuration information, perform domain instantiation processing on each communication domain to obtain a domain instance information set; S332, Process the target cross-domain routing information to obtain message forwarding rule information; S333, process the domain instance information set and the message forwarding rule information to obtain cross-domain interconnection configuration information.

[0012] A second aspect of this invention discloses an adaptive interconnection device based on DDS in a cloud-edge heterogeneous network, the device comprising: The acquisition module is used to acquire network environment detection data, communication domain configuration information, and candidate communication strategy information sets. The strategy selection module is used to process the network environment detection data and the candidate communication strategy information set to obtain the target communication strategy information set; The cross-domain interconnection module is used to process the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information; The monitoring and switching module is used to process the cross-domain interconnection configuration information to obtain the target interconnection result information.

[0013] A third aspect of this invention discloses another adaptive interconnection device based on DDS in a cloud-edge heterogeneous network, the device comprising: processor; A memory coupled to the processor stores executable program code; The processor calls the executable program code stored in the memory to execute some or all of the steps of the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network disclosed in the first aspect of the present invention.

[0014] The fourth aspect of this invention discloses a computer-readable storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps of the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network disclosed in the first aspect of this invention.

[0015] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In this embodiment of the invention, network environment detection data, communication domain configuration information, and a candidate communication strategy information set are first acquired. Then, communication strategy selection processing is performed on the network environment detection data and candidate communication strategy information set, including a comprehensive evaluation of the multicast support, routing reachability, latency, and packet loss rate of network nodes. Based on the evaluation results, the adaptability of each network node to each candidate communication strategy is calculated, and the optimal discovery / transmission strategy combination is selected for each network node to obtain the target communication strategy information set. Next, federated cross-domain interconnection processing is performed based on the target communication strategy information set and communication domain configuration information, including generating domain configurations, calculating cross-domain route optimization, and configuring federated bridging to obtain cross-domain interconnection configuration information. Finally, network monitoring and adaptive switching processing are performed based on the cross-domain interconnection configuration information, continuously monitoring the network status and triggering hot policy switching when the network deteriorates to obtain the target interconnection result information. It can be seen that this application has the following technical effects: (1) It adopts an application layer orchestration method that does not rely on VPN / Overlay, avoiding the additional overhead and system complexity of system communication brought about by the unified network tunnel; (2) Through the discovery / transmission dual-dimensional adaptive strategy, the system can flexibly combine "RTPS multicast" and "InfoRepo+P2P" by domain / by network environment. When multicast is available, it prioritizes the use of RTPS decentralized discovery and RTPS / UDP transmission to ensure low latency and high throughput. When multicast is unavailable, it automatically switches to InfoRepo centralized discovery and TCP / UDP unicast transmission to achieve stable point-to-point interconnection; (3) Through multi-domain federated bridging, the same process can coexist in multiple domains, select the optimal transmission by domain, and automatically route and forward between domains, naturally adapting to the interconnection of RoCE / IB and Ethernet; (4) It maps DDS key parameters to unified domain configuration, reduces conflicts in manual configuration files, and supports / / (5) Binding facilitates deployment in NAT / ACL environments; (6) Through runtime network monitoring and hot-switching mechanisms, latency, jitter and packet loss are controllable in terms of policy and parameter dimensions, ensuring the continuity and stability of communication; (7) The configuration is structured, which can be connected with existing operation and maintenance systems, reducing deployment costs and facilitating large-scale operation and maintenance. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1This is a flowchart illustrating an adaptive interconnection method based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an adaptive interconnection device based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of another adaptive interconnection device based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0020] Specific features, structures, or characteristics described in connection with the embodiments may be included in at least one embodiment of the invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments.

[0021] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating an adaptive interconnection method based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention. Figure 1 The described adaptive interconnection method is applied to distributed communication systems, such as local servers or cloud servers used for cloud-edge-device collaborative communication, and this embodiment of the invention is not limited thereto. Figure 1 As shown, the method may include the following operations: S1, acquire network environment detection data, communication domain configuration information, and candidate communication strategy information set.

[0022] It should be noted that the network environment detection data refers to the raw network environment detection data of each network node obtained through network detection, including multicast detection response data, route detection response data, and link quality sampling data. Specifically, the multicast detection response data includes response status, response time, and the number of responding nodes; the route detection response data includes route hop count, route path, and NAT / ACL penetration status; and the link quality sampling data includes round-trip time series, packet loss count, and jitter value. The communication domain configuration information refers to pre-defined DDS communication domain configuration parameters, including domain identifier, network interface binding information, port configuration, and QoS policies. The candidate communication policy information set includes several candidate communication policy information items. Each candidate communication policy information item contains fields such as discovery mechanism type, transport protocol type, applicable scenario, and policy parameters. The candidate communication policy information set is pre-obtained from the policy configuration database.

[0023] It should be noted that DDS (Data Distribution Service) is a data-centric publish / subscribe communication middleware standard used to implement real-time data distribution in distributed systems. The network environment detection data information is automatically acquired by the capability detection module during system startup or runtime. Detection methods include multicast detection (sending probe packets to multicast addresses and detecting responses), route reachability detection (using traceroute, STUN protocol, or TCP connection testing), and link communication quality sampling (statistically calculating round-trip time, packet loss, and jitter by sending test packets). The above detection processing can use ping, traceroute, STUN protocol, or a custom detection protocol; specifically, this embodiment of the invention does not limit the specific implementation.

[0024] S2, process the network environment detection data and the candidate communication strategy information set to obtain the target communication strategy information set.

[0025] It should be noted that the above processing refers to selecting the most suitable discovery / transmission strategy combination for each network node from the candidate communication strategy information set based on network environment detection data. The candidate communication strategies include: RTPS decentralized discovery + RTPS / UDP transmission strategy (suitable for LAN environments where multicast is available), InfoRepo centralized discovery + TCP unicast transmission strategy (suitable for cross-segment environments where multicast is unavailable), InfoRepo centralized discovery + UDP unicast transmission strategy (suitable for environments requiring low latency but where multicast is unavailable), etc. The target communication strategy information set includes target communication strategy information for several network nodes, and each target communication strategy information includes the selected discovery mechanism type, transmission protocol type, relevant configuration parameters, and strategy priority information for that network node.

[0026] It should be noted that RTPS (Real-Time Publish-Subscribe Protocol) is the underlying transport protocol of DDS, supporting a decentralized automatic discovery mechanism. InfoRepo is a centralized discovery service repository provided by OpenDDS, used to provide node discovery services when multicast is unavailable.

[0027] S3, process the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information.

[0028] The cross-domain interconnection configuration information includes configuration parameters for each domain, inter-domain routing tables, message forwarding rules, and federated bridging configurations.

[0029] It should be noted that RoCE (RDMA over Converged Ethernet) is a network protocol that enables remote direct memory access over Ethernet, characterized by low latency and high bandwidth. IB (InfiniBand) is a high-performance computing network interconnect technology commonly used in data centers and high-performance computing clusters.

[0030] S4, process the cross-domain interconnection configuration information to obtain the target interconnection result information.

[0031] It should be noted that the above processing refers to continuously monitoring the health status of network links and smoothly switching communication strategies according to preset policies when the network status deteriorates. The target interconnection result information includes the currently effective communication strategy, the interconnection status of each domain, network quality indicators, and switching history.

[0032] As can be seen, implementing the DDS-based adaptive interconnection method in cloud-edge heterogeneous networks described in the embodiments of the present invention can automatically select the optimal combination of discovery / transmission strategies according to the network environment, realize cross-domain transparent interconnection of heterogeneous networks through federated services, and support runtime hot switching, thereby improving the adaptability and reliability of DDS communication in heterogeneous network environments.

[0033] In an optional embodiment, the processing of the network environment detection data information and the candidate communication strategy information set to obtain the target communication strategy information set includes: S21, perform network capability assessment processing on the network environment detection data information to obtain network capability assessment information.

[0034] It should be noted that the network capability assessment process refers to a comprehensive evaluation of the network capabilities of each network node, including assessments of dimensions such as multicast support, routing reachability, latency, and packet loss rate. The network capability assessment information includes the comprehensive network capability assessment value of each network node.

[0035] S22, perform policy fit calculation processing on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set.

[0036] It should be noted that the policy adaptability calculation process refers to calculating the degree of adaptability of each candidate communication policy to the current network environment. The policy adaptability information set includes several policy adaptability information items, and each policy adaptability information item contains a candidate communication policy identifier and a corresponding adaptability value.

[0037] S23, Based on the policy fit information set and the candidate communication policy information set, determine the target communication policy information set.

[0038] It should be noted that the determination process refers to selecting the candidate communication strategy with the highest fit value from the candidate communication strategy information set as the target communication strategy based on the fit value in the strategy fit information set. When there are multiple candidate communication strategies with the same fit value, the strategy with the higher stability value is selected first.

[0039] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the optimal combination of discovery / transmission strategies can be automatically selected according to the network environment through comprehensive evaluation of network capabilities and calculation of policy adaptability, thereby achieving adaptive strategy selection in both discovery and transmission dimensions.

[0040] In another optional embodiment, the step of performing network capability assessment processing on the network environment detection data to obtain network capability assessment information includes: S211, Perform multicast support analysis on the multicast detection response data in the network environment detection data information to obtain multicast support information.

[0041] It should be noted that the network environment detection data includes multicast probe response data. This multicast probe response data refers to the response information received after the system sends a probe packet to a preset multicast address, including the response status, response time, and number of responding nodes. The multicast support analysis process refers to determining whether the network supports multicast based on the multicast probe response data and calculating the multicast support value. A high multicast support value is considered when the response status is successful and the number of responding nodes is greater than a preset threshold; otherwise, the multicast support value is low.

[0042] It should be noted that the above multicast support analysis can be performed using threshold judgment algorithms, weighted scoring algorithms, or statistical analysis algorithms. In particular, the embodiments of the present invention do not limit the specific methods.

[0043] S212, perform route reachability analysis on the route detection response data in the network environment detection data information to obtain route reachability information.

[0044] It should be noted that the network environment detection data includes route probe response data, which refers to routing information obtained through traceroute, STUN protocol, or TCP connection testing, including route hop count, route path, and NAT / ACL penetration status. The route reachability analysis process refers to calculating the route reachability value based on the route probe response data; the fewer the route hop count and the higher the NAT / ACL penetration success rate, the higher the route reachability value.

[0045] It should be noted that the above route reachability analysis can be performed using path analysis algorithms, connectivity evaluation algorithms, or weighted scoring algorithms. Specifically, this embodiment of the invention does not limit the specific methods used.

[0046] S213, perform link quality analysis processing on the link quality sampling data in the network environment detection data information to obtain link quality information, which includes network latency value and packet loss rate value.

[0047] It should be noted that the network environment detection data includes link quality sampling data, which refers to the raw network link quality data obtained by statistically analyzing test packets, including round-trip time series, packet loss counts, and jitter values. The link quality analysis and processing refers to performing statistical analysis on the link quality sampling data to calculate indicators such as average latency, packet loss rate, and jitter standard deviation.

[0048] It should be noted that the above link quality analysis and processing can be performed using the mean statistical algorithm, the standard deviation calculation algorithm, or the sliding window statistical algorithm. Specifically, the embodiments of the present invention do not limit the specific methods.

[0049] S214, Using a network capability comprehensive evaluation calculation model, the multicast support information, the route reachability information, and the link quality information are calculated and processed to obtain network capability evaluation information; The network capability comprehensive evaluation calculation model is as follows: ; ; ; In the formula, The network capability assessment information is the first one. A comprehensive evaluation value of the network capabilities of each network node. The first in the multicast support information The multicast support value of each network node. The first in the route reachability information The route reachability value of each network node. The link quality information of the first Network latency value of each network node This is the upper limit of the network latency threshold. The link quality information of the first The packet loss rate of each network node. As the attenuation factor, , , and These are the first weight parameter, the second weight parameter, the third weight parameter, and the fourth weight parameter, respectively. For correction items, N represents the number of network nodes in the cloud-edge heterogeneous network.

[0050] It should be noted that the network capability comprehensive evaluation calculation model is used to comprehensively evaluate the network capabilities of each network node. The upper limit of the network latency threshold... The attenuation factor is typically set between 100ms and 500ms, with the specific value determined based on the real-time requirements of the application scenario. The value is typically between 1 and 10, used to control the impact of packet loss rate on the evaluation value. The higher the value, the more severe the penalty for the evaluation value due to packet loss rate. (The correction term...) It is usually taken as a decimal between -0.1 and 0.1, and is used to fine-tune the evaluation results.

[0051] It should be noted that the comprehensive network capability evaluation model described above can comprehensively evaluate the communication capabilities of network nodes from four dimensions: multicast support, route reachability, latency, and packet loss rate. The exponential decay term... It can apply non-linear penalties to packet loss rate. When the packet loss rate is high, the evaluation value drops rapidly, reflecting the sensitive impact of network quality on communication reliability and providing a quantitative basis for subsequent strategy selection.

[0052] It should be noted that the first, second, third, and fourth weight parameters mentioned above can be user-defined or obtained based on historical network data or application scenario characteristics. Specifically, this embodiment of the invention does not impose limitations. For example, , , , , , .

[0053] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network as described in the embodiments of the present invention, the serial processing of multicast detection, route detection and link quality sampling can comprehensively acquire various characteristic data of the network environment, providing complete input information for comprehensive evaluation of network capabilities.

[0054] In another optional embodiment, the step of performing policy fit calculation on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set includes: S221, perform policy parameter parsing processing on the candidate communication policy information set to obtain the policy parameter information set.

[0055] It should be noted that the strategy parameter parsing process refers to extracting the key parameters of each candidate communication strategy from the candidate communication strategy information set. The strategy parameter information set includes several strategy parameter information entries, each containing parameters such as a strategy identifier, network requirement value, theoretical bandwidth value, and stability value. The candidate communication strategy information set includes various strategy combinations such as RTPS decentralized discovery + RTPS / UDP transmission strategy, InfoRepo centralized discovery + TCP unicast transmission strategy, and InfoRepo centralized discovery + UDP unicast transmission strategy.

[0056] S222, with i=1 pre-set.

[0057] S223, default j=1.

[0058] S224, perform adaptation calculation on the network capability assessment information and the j-th policy parameter information in the policy parameter information set to obtain the policy adaptation value of the j-th policy in the i-th network node; It should be noted that the above-mentioned adaptation calculation process can be processed using a communication strategy adaptation calculation model, or it can be processed using a similarity matching algorithm, a weighted scoring algorithm, or a multi-dimensional evaluation algorithm. Specifically, the embodiments of the present invention do not limit the specifics.

[0059] The communication strategy adaptability calculation model is as follows: ; ; ; In the formula, For the first The network node pairs with the first The strategy fit value of each strategy The network capability assessment information is the first one. A comprehensive evaluation value of the network capabilities of each network node. For the first in the strategy parameter information set The network requirement value of each strategy parameter information. This represents the maximum network demand. For the first in the strategy parameter information set The theoretical bandwidth value of each strategy parameter information. For the first in the strategy parameter information set Stability values ​​of strategy parameter information, , and These are the first adaptation weight parameter, the second adaptation weight parameter, and the third adaptation weight parameter, respectively. , N is the total number of network nodes, and M is the total number of policy parameter information.

[0060] It should be noted that the communication strategy adaptability calculation model is used to calculate the adaptability of each candidate communication strategy with each network node. The strategy parameter information set includes several strategy parameter information items, each containing parameters such as strategy identifier, network requirement value, theoretical bandwidth value, and stability value. The network capability evaluation information includes a comprehensive evaluation value of the network capabilities of several network nodes. For the index of the network node (loop variable, ), The network capability assessment information is the first one. The comprehensive evaluation value of the network capabilities of each network node. In the iteration process of S223, a double loop structure is adopted: the outer loop traverses the network nodes (…). From 1 to ), inner loop traversal of candidate strategies ( From 1 to For each network node and each strategy Both use network nodes Ability Values Network demand value of this strategy Compare and calculate the fit value. The network demand value Indicates the first The candidate communication strategies corresponding to each strategy parameter information have minimum network capability requirements. For example, the RTPS+UDP strategy requires multicast support, resulting in a higher network requirement value; while the InfoRepo+TCP strategy does not rely on multicast, resulting in a lower network requirement value. The maximum network requirement value is also mentioned. The value used to normalize the difference between network capabilities and policy requirements is typically between 1.0 and 1.5, and the specific value can be set according to the range of network capability assessment values. For example, The theoretical bandwidth value Indicates the first The bandwidth that the candidate communication strategies corresponding to each strategy parameter information can achieve under ideal conditions, in Mbps. The stability value... Indicates the first The transmission stability of the candidate communication strategies corresponding to each strategy parameter information, with values ​​ranging from 0 to 1.

[0061] It should be noted that in the formula The similarity of the ratio between network capabilities and policy requirements is calculated to assess whether network capabilities meet policy requirements. Calculate normalized difference similarity to assess the absolute difference between network capabilities and policy requirements; The interaction term between bandwidth and stability is calculated to comprehensively evaluate the performance potential of the strategy. This term is normalized to ensure a balanced contribution from both bandwidth and stability factors. By weighted fusion of the three metrics, the degree of matching between network capabilities and policy requirements can be evaluated from different perspectives, enabling adaptive strategy selection that allows for flexible combinations based on domain and network environment.

[0062] It should be noted that the aforementioned adaptation weight parameters can be user-defined or obtained based on historical data or application scenario characteristics; specifically, this embodiment of the invention does not impose limitations. For example, , , .

[0063] S225, determine whether j is equal to M, and obtain the first determination result, where M is the total number of strategy parameter information in the strategy parameter information set; When the first judgment result is negative, increment j by 1 and execute S224; When the first judgment result is yes, determine whether i is equal to N to obtain the second judgment result, where N is the total number of network nodes; When the second judgment result is negative, increment i by 1 and execute S223; When the second judgment result is yes, all strategy fit values ​​are added to the strategy fit information set.

[0064] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the adaptability of each network node to each candidate communication strategy is calculated through a double loop iteration, which can comprehensively evaluate all network nodes and all candidate strategies and provide a basis for selecting the optimal strategy for each network node.

[0065] In an optional embodiment, determining the target communication policy information set based on the policy fit information set and the candidate communication policy information set includes: S231, default s=1.

[0066] S232, sort all policy fit values ​​of the s-th network node in the policy fit information set to obtain the sorted policy fit information of the s-th network node.

[0067] It should be noted that the sorting process refers to sorting the M policy fitness values ​​of the s-th network node from high to low according to the fitness values, while retaining the policy index corresponding to each fitness value. The above sorting process can be performed using algorithms such as quicksort, mergesort, or heapsort; specifically, this embodiment of the invention does not limit the specific algorithms used.

[0068] S233, obtain the policy index with the highest adaptation value from the sorted policy adaptation information of the s-th network node, and obtain the corresponding candidate communication policy information from the candidate communication policy information set according to the policy index, so as to obtain the first candidate policy information of the s-th network node.

[0069] It should be noted that the policy index refers to the position number of the candidate communication policy in the candidate communication policy information set. The complete candidate communication policy information can be obtained from the candidate communication policy information set through the policy index, including fields such as discovery mechanism type, transmission protocol type, applicable scenario and policy parameters.

[0070] S234, determine whether there are other candidate communication strategy information with the same strategy adaptability value as the first candidate strategy information of the s-th network node, and obtain the third judgment result; If the third determination result is yes, execute S235; When the third judgment result is negative, the first candidate strategy information of the s-th network node is determined as the target communication strategy information of the s-th network node, and S236 is executed.

[0071] S235, select the candidate communication strategy information with the highest stability value from all candidate communication strategy information with the same strategy adaptability value, and determine the candidate communication strategy information with the highest stability value as the target communication strategy information of the s-th network node.

[0072] It should be noted that when there are multiple candidate communication strategies with the same fit value, the strategy with the higher stability value is selected based on the stability value field in the candidate communication strategy information to ensure the reliability of communication.

[0073] S236, determine whether s is equal to N, and obtain the fourth determination result, where N is the total number of network nodes; When the fourth judgment result is negative, increment s by 1 and execute S232; When the fourth judgment result is yes, the target communication strategy information of all network nodes is added to the target communication strategy information set.

[0074] It should be noted that the target communication strategy information set includes target communication strategy information of N network nodes. Each network node has its corresponding optimal communication strategy, realizing adaptive strategy selection of "flexible combination by domain / by network environment".

[0075] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the optimal and most stable communication strategy can be selected for each network node by sorting and comparing the stability of each network node separately, thereby achieving flexible configuration by node and improving the reliability and adaptability of communication.

[0076] In another optional embodiment, the processing of the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information includes: S31, the target communication policy information set and the communication domain configuration information are processed to obtain domain configuration information and cross-domain routing information set.

[0077] It should be noted that the domain configuration generation process refers to mapping domain-level requirements to specific DDS parameters, including InfoRepo IOR address, SPDP / SEDP port, local interface binding, TTL value, buffer size, and other parameters. The domain configuration information is stored in a structured format, which can be integrated with existing operation and maintenance systems. The above domain configuration generation process can be performed using configuration template engines, parameter mapping rules, or automated configuration tools; specifically, this embodiment of the invention does not impose limitations.

[0078] It should be noted that the cross-domain routing information set includes several cross-domain routing information entries. The cross-domain routing information set is obtained from the routing table of the federation service module during the domain configuration generation process. The routing table is dynamically maintained through inter-domain probing and route announcement mechanisms.

[0079] S32, process the cross-domain routing information set to obtain the target cross-domain routing information.

[0080] It should be noted that the above processing refers to selecting the optimal route path from multiple cross-domain routes. The target cross-domain route information includes the selected route path, route parameters, and route priority.

[0081] S33, process the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information.

[0082] It should be noted that the federated bridging configuration process refers to configuring the federated service module to establish message and service forwarding channels between domains. The federated service module manages multiple DDS domains within a single process, providing cross-domain message and service forwarding for heterogeneous networks such as RoCE / IB and Ethernet, thus hiding the underlying network heterogeneity. The cross-domain interconnection configuration information includes configuration parameters for each domain, inter-domain routing tables, message forwarding rules, federated bridging endpoint configurations, and QoS mapping rules.

[0083] As can be seen, by implementing the DDS-based adaptive interconnection method in cloud-edge heterogeneous networks described in the embodiments of the present invention, through the serial processing of domain configuration generation, cross-domain route optimization and federated bridging configuration, cross-domain transparent interconnection of heterogeneous networks can be achieved, allowing the same process to coexist in multiple domains, selecting the optimal transmission according to the domain, and automatically routing and forwarding between domains.

[0084] In another optional embodiment, the processing of the target communication policy information set and the communication domain configuration information to obtain domain configuration information and cross-domain routing information set includes: S311, the target communication strategy information set is extracted and processed to obtain the strategy configuration parameter information set.

[0085] It should be noted that the above extraction process refers to extracting the discovery mechanism configuration parameters and transport protocol configuration parameters of each network node from the target communication policy information set. The discovery mechanism configuration parameters include RTPS discovery configuration or InfoRepo discovery configuration, and the transport protocol configuration parameters include UDP transport configuration or TCP transport configuration. The policy configuration parameter information set includes several policy configuration parameter information items, each containing a discovery service address, transport port, transport protocol type, and QoS parameters, corresponding to the target communication policy of a network node.

[0086] It should be noted that the above extraction process can be performed using field extraction algorithms, key-value pair parsing algorithms, structured data parsing algorithms, or regular expression matching algorithms. Specifically, the embodiments of the present invention do not limit the specific methods used.

[0087] S312, the communication domain configuration information is extracted and processed to obtain the domain basic configuration information.

[0088] It should be noted that the above extraction process refers to extracting the domain identifier, network interface binding information, and port configuration information from the communication domain configuration information. The basic domain configuration information includes parameters such as domain ID, network interface IP address, interface name, public IP address, and port range.

[0089] It should be noted that the above extraction process can be performed using field extraction algorithms, key-value pair parsing algorithms, structured data parsing algorithms, or regular expression matching algorithms. Specifically, the embodiments of the present invention do not limit the specific methods used.

[0090] S313, Process the policy configuration parameter information set and the domain basic configuration information to obtain domain configuration information.

[0091] It should be noted that the above processing refers to combining and mapping the policy configuration parameters in the policy configuration parameter information set and the domain basic configuration parameters in the domain basic configuration information according to the DDS standard format to generate complete domain configuration information.

[0092] It should be noted that the above processing can be performed using a configuration template engine, parameter mapping rules, automated configuration tools, template replacement algorithms, or parameter merging algorithms. Specifically, this embodiment of the invention does not limit the specific processing.

[0093] S314. Based on the domain configuration information, obtain the cross-domain routing information set from the routing table of the Federation Service Module.

[0094] It should be noted that the above-described acquisition process refers to querying and retrieving cross-domain routing information related to the current domain configuration from the routing table of the federated service module based on the domain identifier in the domain configuration information. The cross-domain routing information set includes several cross-domain routing entries, each containing parameters such as source domain identifier, destination domain identifier, available bandwidth value, hop count, current load value, and stability value. The routing table is dynamically maintained through inter-domain probing and route advertisement mechanisms, allowing the retrieval of cross-domain routing information related to the current domain configuration from the routing table after the domain configuration is generated.

[0095] It should be noted that the above acquisition process can be performed using routing lookup algorithms, hash table lookup algorithms, index lookup algorithms, or database lookup algorithms. Specifically, the embodiments of the present invention do not limit the specific methods.

[0096] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the policy configuration and domain configuration can be organically combined through the serial processing of policy parameter extraction, domain basic configuration extraction, parameter mapping and routing information acquisition, to generate complete DDS domain configuration information and cross-domain routing information.

[0097] In an optional embodiment, processing the cross-domain routing information set to obtain the target cross-domain routing information includes: S321, Perform routing parameter parsing processing on the cross-domain routing information set to obtain the routing parameter information set.

[0098] It should be noted that the above-mentioned routing parameter parsing process refers to extracting the key parameters of each cross-domain route from the cross-domain routing information set. The routing parameter information set includes several routing parameter information entries, each of which contains parameters such as route identifier, available bandwidth value, hop count, current load value, and stability value.

[0099] It should be noted that the above-mentioned routing parameter parsing process can be processed using field extraction algorithms, key-value pair parsing algorithms, structured data parsing algorithms, or regular expression matching algorithms. Specifically, this embodiment of the invention does not limit the specific methods used.

[0100] S322, default k=1.

[0101] S323, calculate and process the kth route parameter information in the routing parameter information set to obtain the optimization value of the kth route; It should be noted that the above calculation and processing can be performed using a cross-domain routing optimization calculation model, or it can be performed using a weighted scoring algorithm, a multi-dimensional evaluation algorithm, or a comprehensive scoring algorithm. Specifically, the embodiments of the present invention do not limit the specific calculation and processing.

[0102] The cross-domain route optimization calculation model is as follows: ; ; ; In the formula, For the first The optimization value of the cross-domain route. For the first in the routing parameter information set Available bandwidth value for each routing parameter. For the first in the routing parameter information set The hop count of the route parameter information. The maximum number of hops threshold, For the first in the routing parameter information set The current load value of the route parameter information. For load threshold, For the first in the routing parameter information set The stability value of the routing parameter information. The steepness parameter of the Sigmoid function. , , and These are the first route weight parameter, the second route weight parameter, the third route weight parameter, and the fourth route weight parameter, respectively. K is the number of routing parameter information in the routing parameter information set.

[0103] It should be noted that the cross-domain route optimization calculation model is used to select the optimal cross-domain route path during federated bridging. The route parameter information set includes several route parameter information entries, each containing parameters such as route identifier, available bandwidth value, hop count, current load value, and stability value. The available bandwidth value... The unit is Mbps, indicating the number of Mbps. The currently available bandwidth for this route. The hop count. Indicates the first The maximum number of network hops traversed by the route, the maximum hop count threshold. The value is typically between 10 and 30. (The current load value is mentioned.) The value ranges from 0 to 1, representing the first... The current load percentage of each route. The load threshold. A value typically ranges from 0.7 to 0.9, representing the load balancing trigger threshold. This stability value... The value ranges from 0 to 1, representing the first... Historical stability of the route. The steepness parameter of the Sigmoid function. The value is typically between 5 and 20, used to control the sensitivity of load balancing.

[0104] It should be noted that in the formula Calculate the bandwidth ratio to prioritize routes with higher bandwidth. Calculate the hop count normalization value to prioritize routes with fewer hops; The Sigmoid function is used for load balancing to achieve soft threshold control of load balancing. When the load approaches the threshold, the optimization priority decreases smoothly, avoiding routing oscillations caused by hard thresholds. The interaction term between bandwidth and stability is calculated to comprehensively consider the performance and reliability of routing. Through comprehensive calculation across these four dimensions, the optimal cross-domain routing path can be selected, achieving "automatic inter-domain routing and forwarding".

[0105] It should be noted that the aforementioned routing weight parameters can be user-defined or obtained based on historical routing data or network topology characteristics; specifically, this embodiment of the invention does not impose limitations. For example, , , , , , , .

[0106] S324, determine whether k is equal to K, and obtain the fifth determination result, where K is the number of routing parameter information in the routing parameter information set; When the fifth judgment result is negative, increment k by 1 and execute S323; When the fifth judgment result is yes, the cross-domain route with the highest priority value is selected from the cross-domain route information set as the target cross-domain route information based on the priority value of all routes.

[0107] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the optimal route can be selected from multiple candidate routes by iteratively calculating the optimization degree of each cross-domain route, thereby achieving efficient forwarding of cross-domain messages.

[0108] In another optional embodiment, the processing of the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information includes: S331, Based on the domain configuration information, perform domain instantiation processing on each communication domain to obtain a domain instance information set.

[0109] It should be noted that the domain instantiation process refers to creating DDS domain participant instances based on domain configuration information and configuring parameters such as discovery services, transport protocols, and QoS policies. The domain instance information set includes information such as domain participant instances, publisher / subscriber instances, and data reader / writer instances for each domain.

[0110] It should be noted that the above-mentioned domain instantiation processing can be carried out using object instantiation algorithms, factory pattern algorithms, reflection mechanism algorithms, or dynamic loading algorithms. Specifically, the embodiments of the present invention do not limit the specific methods.

[0111] S332, Process the target cross-domain routing information to obtain message forwarding rule information.

[0112] It should be noted that the above processing involves configuring inter-domain message forwarding rules. These rules include parameters such as source domain identifier, destination domain identifier, topic filtering rules, QoS mapping rules, and forwarding priority. Message forwarding rules support filtering and mapping by topic, data type, or QoS policy.

[0113] It should be noted that the above processing can be performed using rule configuration algorithms, strategy matching algorithms, rule engine algorithms, or template configuration algorithms. Specifically, the embodiments of the present invention do not limit the specific processing.

[0114] S333, process the domain instance information set and the message forwarding rule information to obtain cross-domain interconnection configuration information.

[0115] It should be noted that the above process establishes a federated bridging connection. This federated bridging connection refers to an inter-domain data channel established within the federated service module, used to achieve transparent forwarding of cross-domain messages. Federated bridging supports bidirectional forwarding, enabling the automatic forwarding of data published in one domain to subscribers in another domain.

[0116] It should be noted that the above-mentioned establishment process can be performed using connection establishment algorithms, channel creation algorithms, bridging initialization algorithms, or network connection algorithms. Specifically, the embodiments of the present invention do not limit the specific methods.

[0117] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, multi-domain coexistence and cross-domain transparent interconnection can be achieved through domain instantiation, message forwarding rule configuration and federated bridging connection establishment.

[0118] In yet another optional embodiment, processing the cross-domain interconnection configuration information to obtain the target interconnection result information includes: S41, continuously monitor and process the network links corresponding to the cross-domain interconnection configuration information to obtain network monitoring data information.

[0119] It should be noted that the network links corresponding to the cross-domain interconnection configuration information refer to the network communication links established according to the inter-domain routing paths, transmission channels, and federated bridging connections configured in the cross-domain interconnection configuration information, including intra-domain communication links and cross-domain communication links. The aforementioned continuous monitoring process refers to periodically sampling and detecting the status of the network links, collecting link quality indicators and communication statistics. The network monitoring data information refers to the network link status data continuously collected through the monitoring and hot-switching module, including current network capability assessment values, historical network capability assessment values, communication success / failure statistics, and network stability duration data.

[0120] It should be noted that the above continuous monitoring process can be performed using heartbeat detection algorithms, probe message algorithms, traffic statistics algorithms, or performance monitoring algorithms. Specifically, the embodiments of the present invention do not limit the specific methods used.

[0121] S42, process the network monitoring data information to obtain the hot switching trigger decision value.

[0122] It should be noted that the above processing refers to determining whether a hot-switch of communication strategies needs to be triggered based on network monitoring data. Specifically, the processing includes: first, performing sliding window statistical analysis on the network monitoring data to calculate the historical comprehensive network capability assessment value, communication statistics, and strategy adaptability status information; then, using a hot-switch triggering decision calculation model, comprehensively evaluating dimensions such as the network capability change rate, communication failure rate, network stability duration, and the degree of decline in strategy adaptability, to calculate the hot-switch triggering decision value. This hot-switch triggering decision value is used to determine whether a strategy switch is needed; a larger decision value indicates a stronger need to trigger a hot-switch.

[0123] It should be noted that the above processing can be performed using a hot-switching triggered decision calculation model, or it can be performed using a threshold judgment algorithm, a weighted scoring algorithm, or a multi-dimensional evaluation algorithm. Specifically, the embodiments of the present invention do not limit the specific processing.

[0124] S43, determine whether the hot switching trigger decision value is greater than the preset switching threshold, and obtain the fifth judgment result; If the fifth judgment result is yes, execute S44; If the result of the fifth judgment is negative, execute S47; It should be noted that the preset switching threshold is typically between 0.3 and 0.7, and the specific value can be set according to the communication stability requirements of the application scenario. The lower the threshold, the more sensitive the system is to network changes and the more frequent the switching; the higher the threshold, the more stable the system, but the slower the response to network degradation.

[0125] S44, execute S21 to S23 to obtain the updated target communication strategy information set; It should be noted that before executing S21 to S23, the network monitoring data information will be converted into network environment detection data information format.

[0126] It should be noted that the network monitoring data information and the network environment detection data information have similar data structures, both including multicast probe response data, route probe response data, and link quality sampling data. The above conversion process refers to organizing the latest network status data obtained from continuous monitoring according to the format of the network environment detection data information in S1, so as to re-execute the policy selection process from S21 to S23.

[0127] S45, update the cross-domain interconnection configuration information according to the updated target communication policy information set to obtain the updated cross-domain interconnection configuration information.

[0128] It should be noted that the above update process refers to a smooth switch of communication strategies without interrupting service. Specifically, firstly, a new communication channel is established based on the updated target communication strategy information set; then, the data stream is gradually migrated from the old communication channel to the new communication channel; finally, the old policy's communication channel is closed, completing the update of the cross-domain interconnection configuration information. This smooth switching method ensures communication continuity and avoids data loss during the switching process.

[0129] It should be noted that the above update process can be performed using a smooth switching algorithm, a hot migration algorithm, a seamless switching algorithm, or a progressive migration algorithm. Specifically, the embodiments of the present invention do not limit the specific methods.

[0130] S46, determine that the cross-domain interconnection configuration information is the updated cross-domain interconnection configuration information; S47, the cross-domain interconnection configuration information and the network monitoring data information are merged to obtain the target interconnection result information.

[0131] It should be noted that the above processing refers to merging the cross-domain interconnection configuration information and the network monitoring data information into target interconnection result information. The target interconnection result information includes currently effective communication policy information, interconnection status information of each domain, network quality indicator information, and handover history information, etc.

[0132] It should be noted that the above generation process can be performed using data integration algorithms, information aggregation algorithms, state aggregation algorithms, or report generation algorithms. Specifically, the embodiments of the present invention do not limit the specific methods used.

[0133] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, the communication strategy can be automatically switched when the network condition deteriorates through hot switching trigger decision and strategy switching execution, thus ensuring the continuity and stability of communication.

[0134] In an optional embodiment, processing the network monitoring data to obtain the hot-switching trigger decision value includes: S421, perform sliding window statistical processing on the network monitoring data information to obtain statistically processed network monitoring data information, historical network capability comprehensive evaluation value, communication statistics information, and policy adaptation status information.

[0135] It should be noted that the sliding window statistical processing refers to performing statistical analysis on network monitoring data within a fixed-size time window, calculating statistical quantities such as mean, variance, and trend. The statistically processed network monitoring data refers to the network monitoring data after sliding window statistical processing, including real-time data such as the current comprehensive network capability assessment value. The historical comprehensive network capability assessment value is the mean of the network capability assessment values ​​within the sliding window. The communication statistics include data such as the number of successful communications, the number of failed communications, and the duration of network stability. The policy adaptation status information includes the current policy adaptation value and the maximum policy adaptation value.

[0136] It should be noted that the current policy fit value refers to the fit value corresponding to the target communication policy currently being used by the network node. This fit value is calculated in step S223 using the communication policy fit calculation model and stored in the policy fit information set. The maximum policy fit value refers to the maximum value among all candidate communication policy fit values ​​corresponding to the current network node in the policy fit information set. When the difference between the current policy fit value and the maximum policy fit value is large, it indicates that there is a better candidate policy, and a hot switch should be considered.

[0137] It should be noted that the sliding window size is typically between 10 and 100 sampling periods, and the monitoring period is typically between 100 ms and 1000 ms. The specific values ​​can be set according to the real-time requirements of the application scenario. The above statistical processing can be performed using a heartbeat detection algorithm, a probe message algorithm, a traffic statistics algorithm, or a sliding window mean algorithm. Specifically, this embodiment of the invention does not limit the specific methods used.

[0138] S422, calculate and process the statistical network monitoring data, the historical network capability comprehensive evaluation value, the communication statistics and the policy adaptation status information to obtain the hot switching trigger decision value; It should be noted that the above calculation and processing can be performed using a hot-switching triggered decision calculation model, or it can be performed using a threshold judgment algorithm, a weighted scoring algorithm, a multi-dimensional evaluation algorithm, or a comprehensive scoring algorithm. Specifically, the embodiments of the present invention do not limit the specifics.

[0139] The hot-switching trigger decision calculation model is as follows: ; ; ; In the formula, The hot-switching trigger decision value is... This is the comprehensive evaluation value of the current network capability from the statistically analyzed network monitoring data. This is the comprehensive evaluation value of the historical network capabilities. To prevent division by zero constants, The number of communication failures within the monitoring window in the communication statistics information. This refers to the total number of communications within the monitoring window in the communication statistics information. The duration of network stability in the aforementioned communication statistics. The time decay factor, The current policy fit value in the policy fit status information. The maximum value of the policy fit in the policy fit state information. , , and These are the first switching weight parameter, the second switching weight parameter, the third switching weight parameter, and the fourth switching weight parameter, respectively.

[0140] It should be noted that the hot-switching trigger decision calculation model is used to determine whether a hot-switching of the communication strategy needs to be triggered. The statistically analyzed network monitoring data is continuously collected by the monitoring and hot-switching module and processed through a sliding window statistical method, including real-time data such as the current comprehensive network capability assessment value. The historical comprehensive network capability assessment value... This represents the average network capability evaluation value within the sliding window, the size of which is typically between 10 and 100 sampling periods. The small constant used to prevent division by zero... The value is typically between 0.001 and 0.01. This represents the number of communication failures within the monitoring window. and total number of communications Statistics are collected within a fixed time window, typically ranging from 10 to 60 seconds. This refers to the network stability duration. The unit is seconds, representing the duration for which the network state remains stable. The time decay factor... Typically, the value is between 0.01 and 0.1, used to control the degree of influence of the settling time on the decision value. The current policy fit value... This represents the fit value of the currently effective strategy, and the maximum fit value of the strategy. This represents the maximum fit value among all candidate strategies.

[0141] It should be noted that in the formula Calculate the network capability change rate to detect sudden changes in network status. When the network capability changes drastically, this value is large, triggering a switching requirement. Calculate the communication failure rate, with a value ranging from 0 to 1, to evaluate the reliability of the current strategy; This is the stabilization time decay term, with a value ranging from 0 to 1. The longer the network stabilization time, the smaller this term value, indicating a lower handover requirement. The calculation measures the degree of decline in policy adaptability, with a value ranging from 0 to 1, to assess whether a better policy exists. Through comprehensive evaluation across four dimensions, the system can intelligently determine whether a policy switch needs to be triggered, achieving "runtime hot switching" and ensuring communication continuity.

[0142] It should be noted that the aforementioned switching weight parameters can be user-defined or obtained based on historical switching data or application scenario characteristics. Specifically, this embodiment of the invention does not impose limitations. For example, , , , , , .

[0143] As can be seen, by implementing the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network described in the embodiments of the present invention, network status changes can be perceived in real time through continuous monitoring and sliding window statistics, providing accurate input data for hot switching decisions.

[0144] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of an adaptive interconnection device based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention. Figure 2 The described adaptive interconnection device is applied in distributed communication systems, such as local servers or cloud servers used for cloud-edge-device collaborative communication, and the embodiments of the present invention are not limited thereto. Figure 2 As shown, the adaptive interconnect device includes: The acquisition module 201 is used to acquire network environment detection data, communication domain configuration information, and candidate communication strategy information set; The strategy selection module 202 is used to process the network environment detection data information and the candidate communication strategy information set to obtain the target communication strategy information set; The cross-domain interconnection module 203 is used to process the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information; The monitoring and switching module 204 is used to process the cross-domain interconnection configuration information to obtain the target interconnection result information.

[0145] As can be seen, the adaptive interconnection device based on DDS in a cloud-edge heterogeneous network described in the embodiments of the present invention can automatically select the optimal combination of discovery / transmission strategies according to the network environment, realize cross-domain transparent interconnection of heterogeneous networks through federated services, and support runtime hot switching, thereby improving the adaptability and reliability of DDS communication in heterogeneous network environments.

[0146] Example 3 Please see Figure 3 , Figure 3 This is a schematic diagram of another adaptive interconnection device based on DDS in a cloud-edge heterogeneous network disclosed in an embodiment of the present invention. Figure 3 The described adaptive interconnection device is applied in distributed communication systems, such as local servers or cloud servers used for cloud-edge-device collaborative communication, and the embodiments of the present invention are not limited thereto. Figure 3 As shown, the adaptive interconnect device includes: Processor 301; A memory 302 containing executable program code is coupled to the processor 301; The processor 301 calls the executable program code stored in the memory 302 to execute some or all of the steps of the DDS-based adaptive interconnection method in the cloud-edge heterogeneous network of Embodiment 1.

[0147] As can be seen, the adaptive interconnection device based on DDS in a cloud-edge heterogeneous network described in the embodiments of the present invention can automatically select the optimal combination of discovery / transmission strategies according to the network environment, realize cross-domain transparent interconnection of heterogeneous networks through federated services, and support runtime hot switching, thereby improving the adaptability and reliability of DDS communication in heterogeneous network environments.

[0148] Example 4 This invention discloses a computer-readable storage medium storing computer instructions. When the computer instructions are invoked, they are used to execute some or all of the steps of the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network according to Embodiment 1.

[0149] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network of Embodiment 1.

[0150] The system embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0151] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electronically erasable rewritable read-only memory (EEPROM), read-only optical disc (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0152] Finally, it should be noted that the adaptive interconnection method and apparatus based on DDS in a cloud-edge heterogeneous network disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive interconnection method based on DDS in a cloud-edge heterogeneous network, characterized in that, The method includes: S1, acquire network environment detection data, communication domain configuration information, and candidate communication strategy information set; S2, process the network environment detection data information and the candidate communication strategy information set to obtain the target communication strategy information set; S3, process the target communication strategy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information; S4, process the cross-domain interconnection configuration information to obtain the target interconnection result information.

2. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 1, characterized in that, The process of processing the network environment detection data and the candidate communication strategy information set to obtain the target communication strategy information set includes: S21, perform network capability assessment processing on the network environment detection data information to obtain network capability assessment information; S22, perform policy fit calculation on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set; S23, Based on the policy fit information set and the candidate communication policy information set, determine the target communication policy information set.

3. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 2, characterized in that, The process of performing network capability assessment on the network environment detection data to obtain network capability assessment information includes: S211, Perform multicast support analysis on the multicast detection response data in the network environment detection data information to obtain multicast support information; S212, Perform route reachability analysis on the route detection response data in the network environment detection data information to obtain route reachability information; S213, perform link quality analysis processing on the link quality sampling data in the network environment detection data information to obtain link quality information; the link quality information includes network latency value and packet loss rate value; S214, Using a network capability comprehensive evaluation calculation model, the multicast support information, the route reachability information, and the link quality information are calculated and processed to obtain network capability evaluation information; The network capability comprehensive evaluation calculation model is as follows: ; ; ; In the formula, The network capability assessment information is the first one. A comprehensive evaluation value of the network capabilities of each network node. The first in the multicast support information The multicast support value of each network node. The first in the route reachability information The route reachability value of each network node. The link quality information of the first Network latency value of each network node This is the upper limit of the network latency threshold. The link quality information of the first The packet loss rate of each network node. As the attenuation factor, , , and These are the first weight parameter, the second weight parameter, the third weight parameter, and the fourth weight parameter, respectively. For correction items, N represents the number of network nodes in the cloud-edge heterogeneous network.

4. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 2, characterized in that, The step of performing policy fit calculation on the network capability assessment information and the candidate communication policy information set to obtain the policy fit information set includes: S221, perform strategy parameter parsing processing on the candidate communication strategy information set to obtain the strategy parameter information set; S222, with i=1 pre-set; S223, default j=1; S224, perform adaptation calculation on the network capability assessment information and the j-th policy parameter information in the policy parameter information set to obtain the policy adaptation value of the j-th policy in the i-th network node; S225, determine whether j is equal to M, and obtain the first determination result, where M is the total number of strategy parameter information in the strategy parameter information set; When the first judgment result is negative, increment j by 1 and execute S224; When the first judgment result is yes, determine whether i is equal to N to obtain the second judgment result, where N is the total number of network nodes; When the second judgment result is negative, increment i by 1 and execute S223; When the second judgment result is yes, all strategy fit values ​​are added to the strategy fit information set.

5. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 1, characterized in that, The process of processing the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information includes: S31, Process the target communication policy information set and the communication domain configuration information to obtain domain configuration information and cross-domain routing information set; S32, process the cross-domain routing information set to obtain the target cross-domain routing information; S33, process the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information.

6. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 5, characterized in that, The process of processing the target communication policy information set and the communication domain configuration information to obtain domain configuration information and cross-domain routing information set includes: S311, The target communication strategy information set is extracted and processed to obtain a strategy configuration parameter information set; S312, The communication domain configuration information is extracted and processed to obtain the domain basic configuration information; S313, Process the policy configuration parameter information set and the domain basic configuration information to obtain domain configuration information; S314. Based on the domain configuration information, obtain the cross-domain routing information set from the routing table of the Federation Service Module.

7. The adaptive interconnection method based on DDS in a cloud-edge heterogeneous network according to claim 5, characterized in that, The process of processing the domain configuration information and the target cross-domain routing information to obtain cross-domain interconnection configuration information includes: S331, Based on the domain configuration information, perform domain instantiation processing on each communication domain to obtain a domain instance information set; S332, Process the target cross-domain routing information to obtain message forwarding rule information; S333, process the domain instance information set and the message forwarding rule information to obtain cross-domain interconnection configuration information.

8. An adaptive interconnection device based on DDS in a cloud-edge heterogeneous network, characterized in that, The device includes: The acquisition module is used to acquire network environment detection data, communication domain configuration information, and candidate communication strategy information sets. The strategy selection module is used to process the network environment detection data and the candidate communication strategy information set to obtain the target communication strategy information set; The cross-domain interconnection module is used to process the target communication policy information set and the communication domain configuration information to obtain cross-domain interconnection configuration information; The monitoring and switching module is used to process the cross-domain interconnection configuration information to obtain the target interconnection result information.

9. An adaptive interconnection device based on DDS in a cloud-edge heterogeneous network, characterized in that, The device includes: processor; A memory coupled to the processor stores executable program code; The processor calls the executable program code stored in the memory to execute the adaptive interconnection method based on DDS in a cloud-edge heterogeneous network as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked, are used to execute the DDS-based adaptive interconnection method in a cloud-edge heterogeneous network as described in any one of claims 1-7.

Images