Protocol optimization and dynamic route selection method for network communication system

Abstract

The invention relates to the field of communication, in particular to a method for protocol optimization and dynamic routing for a network communication system. According to the invention, based on deep learning and big data analysis technologies, protocols in network communication are intelligently optimized. Through analysis of historical communication data, information such as a flow mode and a congestion condition in network communication is learned and predicted, so that communication protocol parameters are dynamically adjusted, and the efficiency and the stability of the network communication are improved. And meanwhile, a multi-objective optimization algorithm and a real-time network monitoring technology are adopted to realize dynamic routing selection. By monitoring network states such as bandwidth, delay, packet loss rate and the like in real time, an optimal routing path is calculated in combination with an objective function, and a data packet is transmitted along the path. By optimizing a data packet structure and a transmission mechanism of a protocol, redundant data is reduced, and the data transmission efficiency is improved. The method can effectively improve the transmission efficiency of network communication, reduce the time delay and packet loss rate, and improve the network stability and reliability.

Description

Technical Field

[0001] The invention relates to the field of communications, and in particular to a method for performing protocol optimization and dynamic route selection on a network communication system. Background Art

[0002] With the rapid development of network technology, network communication has become an indispensable part of modern society, and the security and efficiency of data transmission have become increasingly important research topics. However, in complex network environments, traditional communication protocols and routing selection methods often fail to meet the requirements of efficiency, stability, and security. Especially in large-scale, high-concurrency, and multi-changing network scenarios, traditional communication protocols and routing selection methods are prone to congestion, delays, packet loss, and other problems, which seriously affect the quality and efficiency of network communication. In particular, in existing dynamic routing protocols, nodes periodically broadcast HELLO information and TC information to maintain neighbor relationships and update routing tables. However, in large-scale networks, this broadcast mechanism will occupy a lot of bandwidth resources and increase energy consumption. Summary of the invention

[0003] In order to solve the technical problem, the technical solution adopted by the present invention is: a method for performing protocol optimization and dynamic routing selection on a network communication system, comprising the following steps: S01. Detect network status, including bandwidth usage, packet loss rate, round-trip delay, queue length, transmission control window, traffic direction, and network topology. S02, determine whether protocol optimization is needed, if so, perform protocol optimization, if not, execute step S03; The protocol optimization is completed through a deep learning model. The deep learning model adopts the architecture of GCN combined with TCN. The first layer is GCN and the second layer is TCN. First, all network nodes and their connection relationships are input into GCN. All network nodes and their connection relationships are represented as a graph structure through GCN to aggregate the adjacency information of each node and generate node features containing local topological features. Then, the node features generated by GCN are input into TCN as time series data. The dilated convolution operation of TCN is used to capture long-term time dependencies, thereby predicting future traffic trends. S03, determining whether dynamic routing is required, if so, executing dynamic routing, if not, updating network status information; Dynamic routing selection is achieved through distance vector routing algorithm or link state routing algorithm. Distance vector routing algorithm routing information protocol, each router maintains a vector table, records the best exit and distance to each destination, the router regularly sends update information to neighbors, and updates its own vector table based on the received information; link state routing algorithm adopts open shortest path first, each router discovers neighbors and measures the distance to neighbors, and then broadcasts link state information. All routers build a complete network topology map based on the collected link state information, and use the shortest path algorithm to calculate the shortest path to all destinations; S04. Continue to monitor the network status until the system is terminated or the exception processing is completed.

[0004] Furthermore, step S02 also includes dynamic parameter adjustment, setting a hierarchical parameter adjustment mechanism, including primary parameter adjustment and secondary parameter adjustment. When the deep learning model predicts that the future traffic trend is about to be bandwidth saturated or delay exceeded, the primary parameter adjustment is performed first, including dynamically reducing the TCP window size or enabling adaptive retransmission control, and the adaptive retransmission control is to adjust the retransmission time interval; when the primary parameter adjustment does not meet the requirements, the secondary parameter adjustment is started, including adaptive switching of the compression algorithm or segmentation management of the data packet, and the adaptive switching of the compression algorithm is realized according to the bandwidth and traffic thresholds. The segmentation management of the data packet is to set multiple data packet size thresholds, and the data packet is split when the data packet size threshold is reached.

[0005] Furthermore, a compression algorithm library is set, and multiple compression algorithms are preset, including zlib, LZ4, and Brotli. Adaptive switching of compression algorithms is achieved according to the thresholds of bandwidth and traffic. When the bandwidth is less than the set threshold but the processing speed is greater than the set threshold, LZ4 is selected. When the bandwidth is greater than the set threshold and the traffic is greater than the set threshold, Brotli is selected.

[0006] Furthermore, in order to achieve dynamic parameter adjustment, the real-time network status is collected after each parameter adjustment, and the parameter adjustment strategy is optimized based on the feedback reinforcement learning algorithm. The implementation process of the reinforcement learning algorithm is as follows: define the network status, that is, the initial collection data type, including bandwidth and packet loss rate. The action set includes the selection of compression algorithm, adjustment of TCP window size, and segmented sending of data packets. After the network communication system takes action, the reward value is calculated according to the improvement of the network status. If the delay or packet loss rate is reduced, a positive reward is obtained, otherwise a negative reward is obtained. After each action is executed, the Q value table is updated through reinforcement learning. When encountering similar states in the future, the optimal adjustment strategy is automatically selected.

[0007] Furthermore, both GCN and TCN have 2 layers.

[0008] Furthermore, the training data of the deep learning model comes from the network traffic analysis tool NetFlow and the router log system.

[0009] Furthermore, the method also optimizes the broadcast mechanism according to the node status, including the following steps: S21, each node continuously monitors its own status and receives the status of neighboring nodes. The node's own status and neighboring node status include power and traffic load. When the node status changes or receives specific trigger information, it determines whether the broadcast trigger condition is met; S22. When the broadcast trigger condition is met, the node will generate and broadcast a HELLO or TC message containing key status information, including node ID, power status, load information, neighbor node list and update timestamp; S23. Dynamically adjust the broadcast cycle according to the network change frequency and node status. When the network topology change frequency is less than the set value, extend the broadcast cycle. When the network topology change frequency is greater than the set value, shorten the broadcast cycle.

[0010] Furthermore, the fast convergence mechanism adopts a distributed computing method, where each node independently calculates the routing table based on the information provided by neighboring nodes, and transmits the updated routing information to other nodes through broadcasting.

[0011] Furthermore, when calculating the routing table, each node not only selects the optimal path as the main path, but also selects several suboptimal paths as alternative paths based on the objective function value. When the main path fails or is congested, the node switches to the alternative path for data transmission.

[0012] Furthermore, the node continuously monitors the network status and routing performance indicators and adjusts the routing strategy and weight coefficient in real time according to the monitoring results. When the delay, packet loss rate or bandwidth utilization rate exceeds the set threshold, the routing adjustment operation is triggered. First, the weight coefficient of the affected path is updated through the dynamic weighting algorithm. The adjusted weight coefficient will be input into the weighted shortest path algorithm, and the optimal path will be recalculated according to the new weight. For multiple paths with similar weights, the system adopts an equal-cost load sharing strategy to evenly distribute the traffic to avoid congestion on a single path.

[0013] Beneficial effects of the present invention: The present invention integrates a network status monitoring module inside the communication device to obtain real-time mobile network status information. The communication device integrates a routing calculation and selection module inside to calculate candidate paths and perform performance evaluation based on network status information, and select the optimal path for data transmission. Through actual application tests, the present invention can effectively improve the transmission efficiency of network communications, reduce latency and packet loss rate, and improve network stability and reliability. At the same time, the dynamic routing selection mechanism can adjust the route in real time according to the network status, further improving the utilization rate and overall performance of network resources. It is expected that the implementation of the present invention will bring users a better network usage experience and contribute to the technological progress and development in the field of network communications. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The following is a flow chart of the method described in the embodiment. DETAILED DESCRIPTION

[0015] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0016] Example 1 This embodiment discloses a method for performing protocol optimization and dynamic routing selection on a network communication system. Figure 1 As shown, the following steps are included: S01, start, system initialization, and network status detection; S02, judging whether protocol optimization is needed according to the detected network status, if so, performing protocol optimization, if not, executing step S03; S03, determining whether dynamic routing is required, if so, executing dynamic routing, if not, updating network status information; S04. Continue to monitor the network status until the system is terminated or the exception processing is completed.

[0017] In this embodiment, the network status detected in step S01 includes collecting traffic patterns and status characterization data from network devices, and the traffic patterns and status characterization data include bandwidth usage, packet loss rate, round-trip delay, queue length, transmission control window, and traffic direction. The communication protocol is optimized based on these data, and the optimization process is as follows: 1. Construction and training of deep learning model. The deep learning model adopts the architecture of GCN combined with TCN. The first layer is GCN and the second layer is TCN. Both GCN and TCN have 2 layers. First, all network nodes and their connection relationships are input into GCN. All network nodes and their connection relationships are represented as a graph structure through GCN to aggregate the adjacency information of each node and generate node features containing local topological features. Then, the node features generated by GCN are input into TCN as time series data. The dilated convolution operation of TCN is used to capture long-term time dependencies, so as to predict future traffic trends.

[0018] 2. Dynamic parameter adjustment. A hierarchical parameter adjustment mechanism is set up, including primary parameter adjustment and secondary parameter adjustment. When the future traffic trend predicted by the deep learning model is about to experience bandwidth saturation or delay excess, primary parameter adjustment is performed first, including dynamically reducing the TCP window size or enabling adaptive retransmission control, where the adaptive retransmission control is to adjust the retransmission time interval. When the primary parameter adjustment does not meet the requirements, secondary parameter adjustment is started, including adaptive switching of compression algorithms or segmentation management of data packets. Adaptive switching of compression algorithms is achieved according to bandwidth and traffic thresholds. Segmentation management of data packets is to set multiple data packet size thresholds, and data packet splitting is performed when the data packet size threshold is reached.

[0019] In this embodiment, a compression algorithm library is set, and multiple compression algorithms are preset, including zlib, LZ4, and Brotli. Adaptive switching of compression algorithms is implemented according to the thresholds of bandwidth and traffic. When the bandwidth is less than the set threshold but the processing speed is greater than the set threshold, LZ4 is selected. When the bandwidth is greater than the set threshold and the traffic is greater than the set threshold, Brotli is selected.

[0020] In order to achieve dynamic parameter adjustment, the real-time network status is collected after each parameter adjustment, and the parameter adjustment strategy is optimized based on the feedback reinforcement learning algorithm. The implementation process of the reinforcement learning algorithm is as follows: define the network status, that is, the initial collection data type, including bandwidth and packet loss rate. The action set includes the selection of compression algorithm, adjustment of TCP window size, and segmented sending of data packets. After the network communication system takes action, the reward value is calculated according to the improvement of the network status. If the delay or packet loss rate is reduced, a positive reward is obtained, otherwise a negative reward is obtained. After each action is executed, the Q value table is updated through reinforcement learning. When encountering similar states in the future, the optimal adjustment strategy is automatically selected.

[0021] In this embodiment, the training data comes from the network traffic analysis tool NetFlow and the router log system.

[0022] In this embodiment, the detection of network status also includes each node continuously monitoring its own status (such as power, traffic load, etc.) and receiving the status of neighboring nodes. The node's own status and neighboring node status include power and traffic load. When the node status changes (such as power below a threshold, traffic load increases, etc.) or receives specific trigger information (such as a neighboring node status change notification), it is determined whether the broadcast trigger condition is met; when the broadcast trigger condition is met, the node will generate and broadcast a HELLO or TC message containing key status information, including node ID, power status, load information, neighbor node list, and update timestamp; to reduce the amount of data transmitted. The broadcast cycle is dynamically adjusted according to the network change frequency and node status. When the network topology change frequency is less than the set value, the broadcast cycle is extended, and when the network topology change frequency is greater than the set value, the broadcast cycle is shortened.

[0023] In this embodiment, each node continuously monitors its own status in the following ways: power consumption, CPU / memory monitoring module, traffic monitoring module, and traffic monitoring sensor (NetFlow). The power consumption sensor is used to monitor the power status of the node; the network device has its own CPU and memory monitoring functions, and the utilization rate of these key resources is obtained in real time through the system monitoring module; NetFlow is used to collect real-time traffic load information. Because the sensor standards of different devices are not unified, and the accuracy and real-time performance of the sensors may affect the monitoring effect. To solve these problems, the system can adopt an architecture based on virtual sensors and software-defined monitoring (Software-Defined Monitoring, SDM) to enhance feasibility and adaptability. Virtual sensors simulate power consumption, load, temperature and other states by collecting existing data of the device (network traffic records, etc.), reducing dependence on physical sensors; and SDM achieves cross-device data unification and management through centralized control and dynamic configuration of monitoring strategies for different devices.

[0024] The key status information in the HELLO or TC message is interpreted as: Node ID, used by neighboring nodes to accurately identify the source of information; Battery status, showing the current battery percentage or low battery warning, so that neighboring nodes can consider the node's battery life in routing selection; Load information (CPU / memory usage) helps neighbor nodes evaluate the processing capacity of the node; Bandwidth usage, indicating the current bandwidth usage of the node to avoid data loss or delay caused by congestion; Neighbor node list, which provides the current connection relationship of nodes and helps optimize the routing calculation of network topology; Update timestamps to ensure that neighbor nodes can recognize the real-time nature of the information.

[0025] In this embodiment, detecting the network status also includes detecting the network topology. When the network topology changes, the routing table is updated through a fast convergence mechanism, thereby realizing dynamic routing selection.

[0026] In this embodiment, an objective function is constructed to evaluate the advantages and disadvantages of different paths. The design of the objective function should be able to reflect the goals of network optimization, such as minimizing delay, maximizing bandwidth utilization, improving network reliability, etc. This embodiment constructs an objective function based on round-trip delay, bandwidth, and packet loss rate: ; Where n is the number of total packets or transmission paths, representing the total number of all delay terms that need to be considered.

[0027] represents the delay value of the ith path or the ith data packet. The delay time of a specific path or data packet is expressed in milliseconds (ms). α is the weight coefficient of the delay term; β is the weight coefficient of the bandwidth utilization term, which is the ratio of the used bandwidth to the total available bandwidth; γ is the weight coefficient of the packet loss rate term, which indicates the ratio of the data packets lost during the transmission process to the total sent data packets. Indicates the bandwidth used. is the total available bandwidth, Indicates the packet loss rate.

[0028] After determining the objective function, it is necessary to select a suitable algorithm to implement dynamic routing. This embodiment implements dynamic routing through a distance vector routing algorithm or a link state routing algorithm. The distance vector routing algorithm uses a routing information protocol. Each router maintains a vector table to record the best exit and distance to each destination. The router periodically sends update information to its neighbors and updates its own vector table based on the received information. The link state routing algorithm uses an open shortest path first. Each router discovers its neighbors and measures the distance to the neighbors, and then broadcasts the link state information. All routers build a complete network topology map based on the collected link state information, and use the shortest path algorithm to calculate the shortest path to all destinations. The fast convergence mechanism uses a distributed computing method. Each node independently calculates the routing table based on the information provided by the neighboring nodes, and transmits the updated routing information to other nodes through broadcasting.

[0029] In this embodiment, when the network topology changes, the routing table is quickly updated through the fast convergence mechanism to reduce routing oscillation and data packet loss. The fast convergence mechanism is implemented through the "event-driven" trigger condition. When the topology change (such as link disconnection or new node joining) is discovered using the link state advertisement (LSA), the routing update is immediately started instead of relying on periodic updates at fixed time intervals. This event-driven mechanism allows routing changes to be propagated instantly, avoiding the delay problem of traditional periodic updates. Each node will periodically broadcast link state advertisements (LSAs) containing its own link information and connected neighbor information. When the link state changes, the node immediately generates a new LSA and broadcasts it to other nodes. The convergence mechanism adopts a distributed computing method, and each node independently calculates the routing table based on the information provided by the neighbor node, and transmits the updated routing information to other nodes through broadcasting. In a distributed environment, each node independently performs routing calculations and table updates without relying on a central control node. Each node will independently generate a topology map and parallelly calculate the shortest path to other nodes through a distributed Dijkstra algorithm. In distance vector routing algorithms (such as RIP), nodes use the distributed Bellman-Ford algorithm to iteratively calculate the optimal path based on the update messages of neighbor nodes. When topology changes occur, only the changed information is propagated instead of the entire routing table. Using differential updates can significantly reduce the amount of broadcast data and speed up information propagation. For example, if a link is disconnected, the node only needs to notify the neighbor that the link is unreachable, without having to rebroadcast all path information. In addition, using compressed broadcast technology, only key information (changed node ID, delay of changed path, etc.) is included in the broadcast, reducing the amount of data and improving propagation efficiency.

[0030] In order to cope with network failures and congestion, the present invention introduces an alternative path management mechanism. When calculating the routing table, each node not only selects the optimal path as the main path, but also selects several suboptimal paths as alternative paths according to the objective function value. When the main path fails or is congested, the node switches to the alternative path for data transmission.

[0031] The node continuously monitors the network status and routing performance indicators and adjusts the routing strategy and weight coefficient in real time according to the monitoring results. When the delay, packet loss rate or bandwidth utilization rate exceeds the set threshold (such as delay exceeding 200ms, packet loss rate exceeding 5% or bandwidth utilization rate exceeding 90%), the routing adjustment operation is triggered. First, the weight coefficient of the affected path is updated through the dynamic weighting algorithm. For example, for paths with high delay or high packet loss rate, its weight will be increased, thereby reducing its priority in routing selection. The adjusted weight coefficient will be input into the weighted shortest path algorithm (Weighted Shortest Path), and the optimal path will be recalculated according to the new weight. For multiple paths with similar weights, the system adopts the equal cost load sharing (ECMP) strategy to evenly distribute the traffic to avoid single path congestion. In addition, the system can also adaptively adjust the trigger threshold based on historical data to adapt to changes in network traffic, further improving the accuracy and flexibility of the adjustment.

[0032] The present invention is based on deep learning and big data analysis technology to intelligently optimize the protocols in network communications. By analyzing historical communication data, the flow pattern, congestion status and other information in network communications are learned and predicted, so as to dynamically adjust the communication protocol parameters and improve the efficiency and stability of network communications.

[0033] Multi-objective optimization algorithms and real-time network monitoring technology are used to achieve dynamic routing selection. By real-time monitoring of network status, such as bandwidth, delay, packet loss rate, etc., combined with objective functions (such as minimizing transmission time, maximizing transmission efficiency, etc.), the optimal routing path is calculated and the data packet is transmitted along the path. By optimizing the data packet structure and transmission mechanism of the protocol, redundant data is reduced and data transmission efficiency is improved.

[0034] The above description is only the basic principle and preferred embodiments of the present invention. Improvements and substitutions made by those skilled in the art based on the present invention belong to the protection scope of the present invention.

Claims

1. A method for performing protocol optimization and dynamic routing selection for a network communication system, characterized in that: The following steps are involved: S01. Detect network status, including bandwidth usage, packet loss rate, round-trip delay, queue length, transmission control window, traffic direction, and network topology. S02, determine whether protocol optimization is needed, if so, perform protocol optimization, if not, execute step S03; The protocol optimization is completed through a deep learning model. The deep learning model adopts the architecture of GCN combined with TCN. The first layer is GCN and the second layer is TCN. First, all network nodes and their connection relationships are input into GCN. All network nodes and their connection relationships are represented as a graph structure through GCN to aggregate the adjacency information of each node and generate node features containing local topological features. Then, the node features generated by GCN are input into TCN as time series data. The dilated convolution operation of TCN is used to capture long-term time dependencies, thereby predicting future traffic trends. S03, determining whether dynamic routing is required, if so, executing dynamic routing, if not, updating network status information; Dynamic routing selection is achieved through distance vector routing algorithm or link state routing algorithm. Distance vector routing algorithm routing information protocol, each router maintains a vector table, records the best exit and distance to each destination, the router regularly sends update information to neighbors, and updates its own vector table based on the received information; link state routing algorithm adopts open shortest path first, each router discovers neighbors and measures the distance to neighbors, and then broadcasts link state information. All routers build a complete network topology map based on the collected link state information, and use the shortest path algorithm to calculate the shortest path to all destinations; S04. Continue to monitor the network status until the end.

2. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: Step S02 also includes dynamic parameter adjustment, setting a hierarchical parameter adjustment mechanism, including primary parameter adjustment and secondary parameter adjustment. When the deep learning model predicts that the future traffic trend is about to be bandwidth saturated or delay exceeded, the primary parameter adjustment is performed first, including dynamically reducing the TCP window size or enabling adaptive retransmission control, and the adaptive retransmission control is to adjust the retransmission time interval; when the primary parameter adjustment does not meet the requirements, the secondary parameter adjustment is started, including adaptive switching of the compression algorithm or segmentation management of the data packet. The adaptive switching of the compression algorithm is realized according to the bandwidth and traffic thresholds. The segmentation management of the data packet is to set multiple data packet size thresholds, and the data packet is split when the data packet size threshold is reached.

3. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: Set the compression algorithm library and preset multiple compression algorithms, including zlib, LZ4, and Brotli. Adaptive switching of compression algorithms is implemented according to the thresholds of bandwidth and traffic. When the bandwidth is less than the set threshold but the processing speed is greater than the set threshold, LZ4 is selected. When the bandwidth is greater than the set threshold and the traffic is greater than the set threshold, Brotli is selected.

4. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: In order to achieve dynamic parameter adjustment, the real-time network status is collected after each parameter adjustment, and the parameter adjustment strategy is optimized based on the feedback reinforcement learning algorithm. The implementation process of the reinforcement learning algorithm is as follows: define the network status, that is, the initial collection data type, including bandwidth and packet loss rate. The action set includes the selection of compression algorithm, adjustment of TCP window size, and segmented sending of data packets. After the network communication system takes action, the reward value is calculated according to the improvement of the network status. If the delay or packet loss rate is reduced, a positive reward is obtained, otherwise a negative reward is obtained. After each action is executed, the Q value table is updated through reinforcement learning. When encountering similar states in the future, the optimal adjustment strategy is automatically selected.

5. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: Both GCN and TCN have 2 layers.

6. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: The training data of the deep learning model comes from the network traffic analysis tool NetFlow and the router log system.

7. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: The method also optimizes the broadcast mechanism according to the node status, including the following steps: S21, each node continuously monitors its own status and receives the status of neighboring nodes. The node's own status and neighboring node status include power and traffic load. When the node status changes or receives specific trigger information, it determines whether the broadcast trigger condition is met; S22. When the broadcast trigger condition is met, the node will generate and broadcast a HELLO or TC message containing key status information, including node ID, power status, load information, neighbor node list and update timestamp; S23. Dynamically adjust the broadcast cycle according to the network change frequency and node status. When the network topology change frequency is less than the set value, extend the broadcast cycle. When the network topology change frequency is greater than the set value, shorten the broadcast cycle.

8. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: The fast convergence mechanism adopts a distributed computing method. Each node independently calculates the routing table based on the information provided by neighboring nodes, and transmits the updated routing information to other nodes through broadcasting.

9. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: When calculating the routing table, each node not only selects the optimal path as the main path, but also selects several suboptimal paths as alternative paths based on the objective function value. When the main path fails or is congested, the node switches to the alternative path for data transmission.

10. The method for performing protocol optimization and dynamic routing selection for a network communication system according to claim 1, characterized in that: The node continuously monitors the network status and routing performance indicators and adjusts the routing strategy and weight coefficient in real time according to the monitoring results. When the delay, packet loss rate or bandwidth utilization rate exceeds the set threshold, the routing adjustment operation is triggered. First, the weight coefficient of the affected path is updated through the dynamic weighting algorithm. The adjusted weight coefficient will be input into the weighted shortest path algorithm, and the optimal path will be recalculated according to the new weight. For multiple paths with similar weights, the system adopts an equal-cost load sharing strategy to evenly distribute the traffic to avoid congestion on a single path.

Images