Network congestion processing method, apparatus, device, medium, and program product

Abstract

The application provides a network congestion processing method, device, equipment, medium and program product, which is applied to the technical field of distributed and quantum communication. The method comprises the following steps: mapping a plurality of network nodes in a physical network as quantum nodes, and establishing virtual entanglement pairs between the quantum nodes; based on a logical association relationship between the virtual entanglement pairs, sending state information of each quantum node to an associated node having a logical association relationship with the quantum node, so that each quantum node obtains a state information set containing its own state information and received state information; based on the state information set, calculating a coupling degree of network state change between quantum nodes corresponding to each virtual entanglement pair to obtain an entanglement degree coefficient; and when it is monitored that the state information of any quantum node meets a preset congestion triggering condition, performing different processing on different associated nodes according to the entanglement degree coefficient to generate a warning signal for indicating potential network congestion.

Description

Technical Field

[0001] This application relates to the financial field, specifically to the field of distributed and quantum communication technology, and more specifically to a network congestion handling method, apparatus, device, medium, and program product. Background Technology

[0002] Network congestion refers to a significant deterioration in network performance caused by the number of data packets exceeding the processing capacity of nodes. This manifests as data loss, increased latency, and a sharp drop in throughput, and in severe cases, can even lead to "congestion collapse," resulting in a reduction in overall network transmission efficiency. Currently, methods for handling network congestion mainly rely on congestion control mechanisms at the transport layer and proactive queue management techniques at the network layer. The former uses strategies such as slow start, congestion avoidance, and fast retransmission to adaptively adjust the sending rate based on packet loss and latency; the latter uses random early detection and explicit congestion notification to pre-mark or discard data packets to alleviate queue overflow. However, these methods can only passively respond when congestion has already occurred or is about to occur, lacking the ability to proactively predict future congestion. Furthermore, each network node mainly relies on local information for independent decision-making, lacking an effective state coordination mechanism between nodes. This makes it difficult to predict traffic trends from a network-wide perspective, leading to delayed responses in sudden traffic surges and hindering preventative resource scheduling. Summary of the Invention

[0003] In view of the above problems, this application provides a network congestion handling method, apparatus, device, medium and program product.

[0004] According to a first aspect of this application, a network congestion handling method is provided, comprising: mapping multiple network nodes in a physical network to quantum nodes, and establishing virtual entanglement pairs between the quantum nodes, wherein each quantum node has state information for characterizing its own network state; based on the logical association between the virtual entanglement pairs, sending the state information of each quantum node to the associated nodes with which it has a logical association, so that each quantum node obtains a set of state information containing its own state information and the received state information; based on the set of state information, calculating the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair, and obtaining an entanglement degree coefficient; and when the state information of any quantum node is detected to meet a preset congestion triggering condition, performing different processing on different associated nodes according to the entanglement degree coefficient, and generating an early warning signal for indicating potential network congestion.

[0005] According to an embodiment of this application, different processing is performed on different associated nodes based on the entanglement coefficient to generate an early warning signal for indicating potential network congestion. This includes: calculating the congestion probability for each downstream link corresponding to a different associated node based on the entanglement coefficient and through a distributed prediction model integrated by each associated node; and comparing the congestion probability with a preset congestion threshold, and generating a corresponding early warning signal if the congestion probability is greater than or equal to the preset congestion threshold.

[0006] According to an embodiment of this application, based on the entanglement coefficient, the congestion probability is calculated by a distributed prediction model integrated by each associated node, including: inputting the state information set of each associated node and the corresponding entanglement coefficient into the distributed prediction model; and obtaining the congestion probability of each associated node within a future preset time window through the distributed prediction model; wherein, the distributed prediction model assigns a higher calculation weight to associated nodes whose entanglement coefficient with the quantum node that triggers the congestion triggering condition is higher than that of other associated nodes.

[0007] According to embodiments of this application, a virtual entanglement pair includes a first virtual entanglement pair and a second virtual entanglement pair; mapping multiple network nodes in a physical network to quantum nodes and establishing virtual entanglement pairs between quantum nodes includes: obtaining a data transmission link based on the topology of the physical network; establishing a first virtual entanglement pair for adjacent quantum nodes in the data transmission link; and establishing a second virtual entanglement pair for non-adjacent quantum nodes in the data transmission link through at least one relay quantum node, wherein the relay quantum node is a quantum node located between non-adjacent quantum nodes in the data transmission link.

[0008] According to embodiments of this application, after generating an early warning signal to indicate potential network congestion, the method further includes: simultaneously enabling multiple data transmission links for data packets to be transmitted in the physical network based on the early warning signal; transmitting the data packets to be transmitted on the multiple data transmission links respectively; and determining a target transmission link from the multiple data transmission links based on feedback data from the multiple data transmission links, so as to perform subsequent transmission of data packets through the target transmission link.

[0009] According to an embodiment of this application, the method further includes: in response to identifying the priority of the data packet to be transmitted in the physical network as a target priority, determining the source quantum node and the destination quantum node corresponding to the data packet to be transmitted, wherein the source quantum node is a quantum node mapped to the network node receiving the data packet to be transmitted in the physical network, and the destination quantum node is a quantum node mapped to the target network node of the data packet to be transmitted in the physical network; establishing an entanglement protection channel between the source quantum node and the destination quantum node using a logical association relationship; and pre-allocating bandwidth resources for the data packet to be transmitted through the entanglement protection channel.

[0010] According to an embodiment of this application, pre-allocating bandwidth resources for data packets to be transmitted through an entanglement protection channel includes: obtaining the status of the entanglement protection channel through a network controller, and adjusting the bandwidth resources already allocated to the entanglement protection channel based on the load information of the physical network.

[0011] The second aspect of this application provides a network congestion processing apparatus, comprising: a construction module for mapping multiple network nodes in a physical network to quantum nodes and establishing virtual entanglement pairs between the quantum nodes, wherein each quantum node has state information characterizing its own network state; a transmission module for transmitting the state information of each quantum node to associated nodes with which it has a logical association based on the logical association relationship between the virtual entanglement pairs, so that each quantum node obtains a set of state information containing its own state information and the received state information; a calculation module for calculating the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair based on the set of state information, thereby obtaining an entanglement degree coefficient; and an early warning module for performing different processing on different associated nodes according to the entanglement degree coefficient when the state information of any quantum node is detected to meet a preset congestion triggering condition, thereby generating an early warning signal indicating potential network congestion.

[0012] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0013] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0014] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0015] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0016] Figure 1 A schematic diagram of an example environment 100 to which the method according to an embodiment of this application can be applied is shown;

[0017] Figure 2 A flowchart illustrating a network congestion handling method according to an embodiment of this application is shown schematically.

[0018] Figure 3 A flowchart illustrating the generation of early warning signals via a distributed prediction model according to an embodiment of this application is shown.

[0019] Figure 4 A schematic diagram illustrating a network congestion processing apparatus according to an embodiment of this application is shown; and

[0020] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a network congestion handling method according to an embodiment of this application. Detailed Implementation

[0021] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0022] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0023] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0024] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0025] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0026] Network congestion primarily stems from resource bottlenecks, uneven traffic distribution, and protocol design flaws. Regarding resource bottlenecks, insufficient link bandwidth cannot handle high-speed data streams, a problem common in high-bandwidth scenarios such as video transmission. When a router's buffer is full, newly arriving packets are dropped, triggering retransmission mechanisms and further exacerbating congestion. Furthermore, if the router's central processing unit cannot efficiently handle packet queuing and routing tasks on high-speed links, it can also contribute to congestion. Regarding uneven traffic distribution, burst traffic and multi-source contention are the main manifestations. Burst traffic refers to a large influx of data into the same node within a short period, such as peak server requests during e-commerce promotions. Multi-source contention refers to multiple senders simultaneously vying for the same output port, such as in peer-to-peer downloads or large-scale distributed denial-of-service attacks. Regarding protocol and design flaws, the retransmission mechanism of the Transmission Control Protocol (TCP) has vulnerabilities: packet loss leads to timeout retransmissions, and failure to adjust the sending window size in time can cause global synchronization problems, where multiple hosts simultaneously reduce their sending rates. Additionally, connectionless network service models struggle to guarantee service quality due to the lack of a central coordination mechanism.

[0027] To address the aforementioned congestion issues, existing technologies have proposed various solutions. At the transmission control protocol level, congestion control mechanisms mainly include the following technologies: slow start and congestion avoidance mechanisms, which initially allow the congestion window to grow exponentially, then switch to linear growth after reaching a threshold to balance throughput and stability; fast retransmission and recovery mechanisms, which detect packet loss through redundant acknowledgments and reduce latency caused by timeout retransmissions; and dynamic window adjustment mechanisms, which adaptively adjust the transmission rate based on round-trip time and packet loss rate. At the network level, proactive queue management technologies mainly include: random early detection technology, which randomly discards data packets when the average queue length reaches a threshold to avoid global synchronization problems caused by tail-end dropping; weighted random early detection technology, which implements differentiated packet loss probabilities based on packet priority to optimize service quality, such as ensuring low latency requirements for voice traffic; and explicit congestion notification technology, which marks congestion in the header fields of the Internet Protocol or Transmission Control Protocol to notify endpoints to reduce their transmission rates in advance, thereby reducing actual packet loss. In terms of cross-layer optimization, the data link layer can smooth burst traffic and reduce the pressure on downstream nodes through traffic shaping technology; the application layer can use content delivery networks to distribute requests or use user datagram protocol to bypass the congestion control mechanism of transmission control protocol, but it also needs to be combined with forward error correction technology.

[0028] Current network congestion management methods suffer from several core flaws. First, the packet loss-based feedback mechanism is fundamentally contradictory. Traditional transmission control protocol algorithms rely on packet loss as a congestion signal, actively reducing the transmission rate when the network approaches saturation. While this improves bandwidth utilization, frequent rate adjustments exacerbate network fluctuations. Such algorithms, in pursuing high throughput, inevitably face the problem of sacrificing low-latency performance or underutilizing bandwidth. Second, resource allocation suffers from unfairness and a "starvation" phenomenon. When different congestion control algorithms coexist, differences in resource preemption logic can lead to unfair allocation; for example, the bandwidth of some low-priority users may be occupied by other traffic, resulting in a severe decline in service quality. While queue-based service quality mechanisms can prioritize critical services, they may sacrifice the fairness of ordinary traffic. Furthermore, acknowledgment aggregation mechanisms and terminal host scheduling strategies in the network can disrupt the precise control of the algorithm, causing resource allocation to deviate from expectations. Third, responsiveness in dynamic environments is insufficient. Congestion avoidance technologies alleviate congestion through proactive packet loss, but when faced with sudden traffic surges or short-term peaks, rigid threshold settings may lead to excessive packet loss or response delays. In scenarios with rate mismatch (such as WAN and LAN interconnection), traditional queue scheduling algorithms struggle to dynamically adjust priorities. Furthermore, modern algorithms relying on network state prediction are susceptible to physical layer fluctuations (such as wireless channel interference), leading to prediction errors and misjudgments. Fourth, multi-objective optimization suffers from structural flaws. Existing algorithms cannot simultaneously achieve high throughput, low latency, and no starvation. For example, some algorithms reduce latency but exacerbate bandwidth contention, while algorithms prioritizing fairness may sacrifice throughput. In network function virtualization and software-defined network architectures, dynamic resource allocation may lag in congestion policy execution due to insufficient coordination between the control plane and data plane. Fifth, protocol and hardware evolution introduces adaptation issues. Traditional congestion control algorithms fail to adequately adapt to the flow label characteristics of next-generation Internet protocols or the network slicing requirements of fifth-generation mobile communication networks, making fine-grained resource management difficult. Additionally, excessively small router buffers exacerbate packet loss, while excessively large buffers introduce latency spikes; existing technologies struggle to strike a balance between the two.

[0029] In view of this, embodiments of this application provide a network congestion handling method, apparatus, device, medium, and program product.

[0030] The network congestion handling method provided in this application embodiment can be applied to, for example, Figure 1 The application scenario 100 is shown. For example... Figure 1As shown, data source 102 accesses the Internet through a first network node set 110, and data destination 104 accesses the Internet through a third network node set 130. The first network node set 110 and the third network node set 130 are connected via a second network node set 120. Data packets sent by data source 102 are forwarded through the first network node set 110, the second network node set 120, and the third network node set 130 before finally reaching data destination 104. The first network node set 110, the second network node set 120, and the third network node set 130 can specifically be switches or routers, and there can be at least two of each. Figure 1 The first network node set 110 shown includes a first network node 110a and a second network node 110b, the second network node set 120 includes a third network node 120a, a fourth network node 120b, a fifth network node 120c and a sixth network node 120d, and the third network node set 130 includes a seventh network node 130a and an eighth network node 130b.

[0031] In this embodiment, each of the aforementioned network nodes can be mapped to a quantum node, and virtual entanglement pairs are established between quantum nodes with logical relationships. Each quantum node has state information characterizing its own network state. When the state information of a quantum node meets a preset congestion triggering condition, other quantum nodes with logical relationships with that quantum node can collaboratively predict the congestion probability and generate an early warning signal. Based on this early warning signal, multiple candidate transmission paths can be enabled simultaneously for the data packet to be transmitted, and the target transmission path can be determined from them based on real-time feedback to complete the subsequent transmission of the data packet. The method provided in this embodiment can be executed at the data source 102, a network node set (e.g., a first network node set 110, a second network node set 120, or a third network node set 130), or the data destination 104.

[0032] In one embodiment, the target network node for the next hop determined by the data source 102 is... Figure 1 The second network node 110b can also execute the method provided in the embodiments of this application after receiving the data packet to be forwarded.

[0033] For example, the data source and data destination can be a terminal or a server. Terminals can include, but are not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. A server can be a server providing various services, such as a backend management server supporting websites browsed by users using the terminal (this is just an example). The backend management server can analyze and process received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device. A server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services such as cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and basic cloud computing services such as big data. The server can be the backend server of the aforementioned target application, used to provide backend services to the client of the target application.

[0034] It should be noted that the network congestion processing method and apparatus of this application can be used in scenarios such as the fintech field, or in any field other than the fintech field. The application field of the network congestion processing method and apparatus of this application is not limited.

[0035] The following will be based on Figure 1 The described scene, through Figure 2 and Figure 3 A network congestion handling method according to embodiments of this application will be described in detail.

[0036] Figure 2 A flowchart illustrating a network congestion handling method according to an embodiment of this application is shown.

[0037] like Figure 2 As shown, the network congestion handling method 200 according to the embodiments of this application may include steps S210 to S240.

[0038] In step S210, multiple network nodes in the physical network are mapped to quantum nodes, and virtual entanglement pairs are established between the quantum nodes, wherein each quantum node has state information for characterizing its own network state.

[0039] In embodiments of this application, authorization or confirmation from each network node can be obtained before establishing a virtual entangled pair. For example, before step S210, a request to establish a logical association can be sent to each network node. Step S210 is executed if the node agrees or confirms its participation in collaborative sensing.

[0040] The system can abstract network nodes such as routers, switches, and servers in a physical network as quantum nodes. Each quantum node can contain one or more qubits, which are used to encode the node's own network state information (such as traffic, latency, and packet loss rate). Each quantum node can correspond to a physical entity. Quantum nodes can establish logical connections through virtual entanglement pairs. This connection does not rely on actual quantum communication but is achieved through a protocol-level state synchronization mechanism to couple information between nodes.

[0041] Each quantum node can maintain its own state information, which can include network performance parameters such as real-time traffic, transmission latency, packet loss rate, and bandwidth utilization. This state information not only reflects the node's current operating status but can also include historical state sequences and trends, thus providing multi-dimensional analytical basis for subsequent congestion prediction. For example, a node's traffic surging continuously within a short period may indicate that the node is about to enter a congested state, and this trend can be transmitted to related nodes through virtual entanglement pairs.

[0042] In step S220, based on the logical association between virtual entangled pairs, the state information of each quantum node is sent to the associated node with which it has a logical association, so that each quantum node obtains a set of state information containing its own state information and the received state information.

[0043] According to the rules for establishing virtual entangled pairs, each node can synchronize its state information to one or more logically related nodes in real time or near real time. This synchronization mechanism is similar to the information transmission process in quantum teleportation protocols, where changes in the node's own state are "transmitted" to the state vectors of the related nodes.

[0044] Each quantum node can maintain a set of state information (e.g., in vector form), which can include its own state information as well as state information received from associated nodes. For example, node A can send its own flow data to its entangled nodes B and C, while node A can also receive state information from nodes B and C. In this way, each node can not only perceive its own state but also grasp the local topological state of the entire entangled network, thus providing a data foundation for subsequent collaborative judgments.

[0045] In step S230, based on the set of state information, the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair is calculated to obtain the entanglement coefficient.

[0046] In this embodiment, the entanglement coefficient can be used to reflect the synchronization or correlation of the network state changes between two nodes. For example, the correlation coefficient of the traffic fluctuations, the mutual information of the delay changes, and the co-variation trend of the packet loss rate of two nodes within a time window can be calculated. The higher the entanglement coefficient, the closer the state changes between the two nodes are related, that is, the state change of one node is often accompanied by the state change of the other node. Such node pairs have stronger linkage in the propagation of network congestion.

[0047] Exemplarily, for nodes i and j with a virtual entanglement relationship, the entanglement coefficient E(i,j) can be calculated based on their historical state sequences. Specifically, when E(i,j) is close to 1, it indicates that the state changes of node i and node j are highly coupled, for example, the traffic fluctuations are almost synchronous; when E(i,j) is close to 0, it indicates that there is no obvious correlation between the state changes of the two nodes; when E(i,j) is negative, it can indicate that the states of the two nodes are in an inverse change relationship.

[0048] It should be noted that the entanglement coefficient can not only be used to describe the coupling state at the current moment, but also can be updated in real time through a sliding window, so as to dynamically reflect the impact of network topology and service load changes on the correlation between nodes.

[0049] In step S240, when the state information of any quantum node satisfies the preset congestion trigger condition, different processing is performed on different associated nodes according to the entanglement coefficient, and a warning signal for indicating potential network congestion is generated.

[0050] When the state information of any quantum node satisfies the congestion trigger condition, for example, the real-time traffic of this node exceeds the preset traffic threshold, or the delay exceeds the maximum tolerable delay, or the packet loss rate suddenly increases and exceeds the threshold value, the system can determine that this node is a trigger node. While the trigger node issues a warning, the system is not limited to alarming the node itself, but can perform differential processing on other associated nodes that have a virtual entanglement relationship with the trigger node according to the entanglement coefficient.

[0051] Exemplarily, for associated nodes with a relatively high entanglement coefficient with the trigger node (such as E(i,j)>0.8), the system can determine that these nodes have a relatively high risk of congestion propagation and generate congestion warning signals for these nodes to perform traffic scheduling or load reduction processing in advance; for associated nodes with a medium entanglement coefficient (such as 0.4<E(i,j)≤0.8), the system can mark them as observed nodes and start more frequent state sampling to monitor whether there are early signs of congestion propagation; for associated nodes with a relatively low entanglement coefficient (such as E(i,j)≤0.4), the system can temporarily not perform active intervention and continue to maintain the normal state synchronization.

[0052] Through the above mechanism, a warning logic similar to the collapse of quantum entangled states can be realized. That is, the state change of the trigger node will cause the "state vector" of its highly entangled associated nodes to be updated synchronously, and the congestion probability prediction result will be output.

[0053] For example, in a payment clearing system scenario, if the transaction processing time of core payment gateway A increases from 200 milliseconds to 950 milliseconds within 5 seconds (exceeding the 800 millisecond threshold), the system can immediately calculate the entanglement coefficients between A and downstream clearing nodes B and C, as well as edge payment nodes D and E. Specifically, the entanglement coefficients between A and B are 0.94, with C 0.87, with D 0.32, and with E 0.28. Based on this, a "delay propagation warning" can be generated for B and C, and some payment requests from B and C can be preemptively scheduled to backup clearing channels before A experiences severe transaction backlog, thereby avoiding potential large-scale delays and congestion in interbank payments.

[0054] Understandably, by introducing virtual entanglement pair modeling and quantifying entanglement coefficients, the ability to proactively detect and defend against network congestion is significantly improved. This method does not rely on upgrading the underlying routing protocol; instead, it builds a non-local collaborative sensing mechanism on top of the existing network architecture. It can automatically identify highly coupled node pairs from massive amounts of node state data and provide differentiated congestion warnings to associated nodes based on the state changes of triggering nodes. This not only achieves a shift from passive, reactive congestion control to proactive, predictive congestion avoidance but also from single-point monitoring to full network-wide spatial analysis. Furthermore, this method has strong compatibility, seamlessly integrating with existing network monitoring systems and network and traffic scheduling platforms, with low implementation costs and easy rapid deployment.

[0055] Based on the above embodiments, in this embodiment, the virtual entanglement pair includes a first virtual entanglement pair and a second virtual entanglement pair; mapping multiple network nodes in the physical network to quantum nodes and establishing virtual entanglement pairs between quantum nodes includes: obtaining a data transmission link based on the topology of the physical network; establishing a first virtual entanglement pair for adjacent quantum nodes in the data transmission link; and establishing a second virtual entanglement pair for non-adjacent quantum nodes in the data transmission link through at least one relay quantum node, wherein the relay quantum node is a quantum node located between non-adjacent quantum nodes in the data transmission link.

[0056] The data transmission link reflects the physical or logical path that data packets take from the source node to the destination node in the network, and is the basis for establishing virtual entangled pairs.

[0057] For adjacent quantum nodes in a data transmission link, a first virtual entanglement pair can be established. This first virtual entanglement pair can be used to measure the degree of direct coupling between the state changes of adjacent quantum nodes. Because adjacent nodes are directly connected by the physical link, their state changes, such as throughput and latency, have the most direct mutual influence. Therefore, the higher the entanglement degree of the first virtual entanglement pair, the tighter the state coupling between adjacent nodes, and the higher the risk of congestion propagating along the direct link.

[0058] For non-adjacent quantum nodes in a data transmission link, a second virtual entanglement pair can be established using at least one relay quantum node. For example, in a three-hop link containing node A → relay R → node B, A and B are initially not adjacent, but by using relay R as an entanglement-swapping node, a second virtual entanglement pair can be established between A and B, thereby achieving state coupling awareness between non-adjacent nodes. The second virtual entanglement pair can then be used to measure the degree of indirect coupling across multiple hops between non-adjacent quantum nodes. The system achieves entanglement swapping through relay quantum nodes, establishing logical connections between nodes that were not originally directly related. When congestion propagates along a multi-hop path, indirect state coupling occurs between non-adjacent nodes, which can be reflected as an abnormal change in entanglement degree in the second virtual entanglement pair.

[0059] Understandably, by distinguishing between the first and second virtual entanglement pairs, it is possible to simultaneously capture the risk of congestion propagation on direct links and multi-hop paths in complex network topologies, achieving proactive perception of potential congestion without relying on modifications to the underlying routing protocol.

[0060] Figure 3 The flowchart illustrating the generation of early warning signals using a distributed prediction model according to an embodiment of this application is shown.

[0061] like Figure 3 As shown, based on the above embodiments, in this embodiment, a method 300 for generating an early warning signal to indicate potential network congestion by performing different processing on different associated nodes according to the entanglement coefficient may include steps S310 to S330.

[0062] In step S310, for the downstream links corresponding to different associated nodes, the congestion probability is calculated respectively based on the entanglement coefficient and the distributed prediction model integrated by each associated node.

[0063] In step S320, the congestion probability is compared with a preset congestion threshold to determine whether the congestion probability is greater than or equal to the preset congestion threshold.

[0064] In step S330, if the congestion probability is greater than or equal to a preset congestion threshold, a corresponding early warning signal is generated.

[0065] Distributed prediction models can be built based on time-series prediction algorithms to quantitatively assess the likelihood of downstream link congestion within a future time window. Specifically, the system can use the historical traffic sequences, real-time load states, topology weights, and entanglement coefficients of associated nodes as model inputs, and calculate and output the link congestion probability within the future time window Δt. A higher entanglement coefficient indicates a more significant impact of the triggering node's state change on the associated node, and therefore a correspondingly higher probability of congestion in the downstream links of that associated node.

[0066] Congestion probability can be used to measure the risk of downstream links becoming congested within a future period. That is, the higher the congestion probability, the greater the risk of congestion occurring in the downstream links corresponding to that associated node within the future time window. When a triggering node experiences a surge in traffic or latency, associated nodes with high entanglement typically experience sequential state deterioration, thus significantly increasing their congestion probability. Associated nodes with low entanglement show relatively stable congestion probabilities. By comparing the calculated congestion probability with a preset congestion threshold, the system can differentiate between associated nodes. When the congestion probability is greater than or equal to the threshold, a corresponding warning signal can be generated, triggering proactive defense measures such as traffic scheduling or load reduction. When the congestion probability is less than the threshold, a warning signal can be temporarily withheld, and routine monitoring of the node's state can continue.

[0067] In other embodiments, quantum optimization networks can be used to dynamically adjust the parameters of the distributed prediction model. For example, quantum optimization algorithms can be used to optimize the model's weight parameters to improve the prediction accuracy and convergence speed for time-series data. Furthermore, the system can feed back the actual results of each congestion event into the prediction model for online updates and parameter tuning, enabling the prediction accuracy to continuously improve with runtime.

[0068] Understandably, distributed prediction models do not need to aggregate the state data of all nodes to a central node. Instead, they can distribute the computational tasks to each associated node for local execution, thereby significantly reducing network bandwidth consumption and the computational pressure on the central node. This allows each associated node to independently assess the congestion risk of its downstream links based on its entanglement coefficient with the triggering node.

[0069] Based on the above embodiments, in this embodiment, the congestion probability is calculated according to the entanglement coefficient through the distributed prediction model integrated by each associated node, including: inputting the state information set of each associated node and the corresponding entanglement coefficient into the distributed prediction model; and obtaining the congestion probability of each associated node within a future preset time window through the distributed prediction model; wherein, the distributed prediction model assigns a higher calculation weight to associated nodes with an entanglement coefficient higher than a first threshold with the quantum node that triggers the congestion triggering condition.

[0070] The computational weights in distributed prediction models can be used to quantify the importance of different associated nodes in the congestion probability calculation process. Specifically, when the trigger node experiences abnormal states such as a surge in traffic or a spike in latency, associated nodes with higher entanglement coefficients are generally more susceptible to congestion propagation. Therefore, they can be assigned higher weights in the prediction model to enhance the model's sensitivity to these high-risk nodes. Conversely, associated nodes with lower entanglement coefficients with the trigger node are less likely to be affected by congestion propagation due to weaker coupling between their state changes, and thus can be assigned lower computational weights.

[0071] The positive correlation between entanglement coefficient and computational weight can be achieved through a preset mapping function. For example, the system can set the computational weight of associated nodes with entanglement coefficients higher than a first threshold to 1.5 or 2 times the baseline value, thereby significantly amplifying the congestion probability calculation results of these nodes and making it easier to reach the preset congestion threshold and trigger an early warning signal. Conversely, for associated nodes with entanglement coefficients lower than a second threshold, their computational weight can be appropriately reduced to avoid false alarms due to oversensitivity.

[0072] For each associated node, its state information set and corresponding entanglement coefficient can first be input into the distributed prediction model. The state information set reflects various operational attributes of the associated node, including current load level (absolute value of traffic intensity), latency (time taken for data packet transmission), and packet loss rate (reliability of network transmission). The entanglement coefficient reflects the degree to which the node is affected by the abnormal state of the trigger node; a higher entanglement coefficient indicates a closer correlation between the node's state change and the trigger node's state change. Internally, the state information set provides a snapshot of the current network state, while the entanglement coefficient determines the relative importance of different node state information in the prediction calculation.

[0073] After receiving input data, the model can extract the temporal features in the state information set and the correlation features in the entanglement coefficient through multi-layer nonlinear transformation, and finally output the congestion probability (such as a value between 0 and 1) within the future preset time window.

[0074] Understandably, by inputting the set of state information and the entanglement coefficients into the distributed prediction model, and assigning higher computational weights to associated nodes with high entanglement coefficients, the model can focus its limited computational resources on high-risk nodes, thereby reducing overall computational overhead while ensuring prediction accuracy and achieving accurate early warning and efficient response to network congestion.

[0075] In embodiments of this application, after generating an early warning signal to indicate potential network congestion, the method further includes: simultaneously enabling multiple data transmission links for data packets to be transmitted in the physical network based on the early warning signal; transmitting the data packets to be transmitted on the multiple data transmission links respectively; and determining a target transmission link from the multiple data transmission links based on feedback data from the multiple data transmission links, so as to perform subsequent transmission of data packets through the target transmission link.

[0076] Based on the warning signal, the system can first determine that there is a potential congestion risk in the current network and trigger a dynamic routing optimization mechanism. That is, when a link or node is predicted to be high-risk, the system is no longer limited to a single-path transmission strategy. Instead, based on the warning signal, it can simultaneously activate multiple candidate data transmission links for data packets to be transmitted to the data transmission link indicated by the warning signal, in order to simulate the characteristics of quantum superposition, that is, the data packets are in a superposition state of "multiple paths being tried simultaneously" in terms of path selection.

[0077] On multiple activated data transmission links, the system can transmit copies or fragments of the data packets to be transmitted. By sending probe data or actual service data in parallel on multiple links, the system can collect real-time transmission performance feedback for each link, including key indicators such as transmission latency, packet loss rate, bandwidth availability, and jitter. The feedback data from each link reflects the actual transmission quality of different paths in the current network environment.

[0078] After collecting feedback data from multiple data transmission links, a target transmission link can be identified from these links for subsequent data packet transmission. Specifically, the system can calculate a comprehensive performance score for each link based on the feedback data (such as a weighted score based on latency, packet loss rate, and bandwidth), and select the link with the highest score as the target transmission link. If the selected target link subsequently experiences another congestion warning, the system can repeat the above process to select a new optimal path.

[0079] Understandably, by simultaneously attempting transmission along multiple candidate paths and rapidly converging to the optimal path based on real-time feedback, the network significantly improves its information transmission efficiency and robustness in congested environments.

[0080] In embodiments of this application, the method further includes: in response to identifying the priority of the data packet to be transmitted in the physical network as a target priority, determining the source quantum node and the destination quantum node corresponding to the data packet to be transmitted, wherein the source quantum node is a quantum node mapped to the network node receiving the data packet to be transmitted in the physical network, and the destination quantum node is a quantum node mapped to the target network node of the data packet to be transmitted in the physical network; establishing an entanglement protection channel between the source quantum node and the destination quantum node using a logical association relationship; and pre-allocating bandwidth resources for the data packet to be transmitted through the entanglement protection channel.

[0081] Target priorities can include real-time services, critical transaction data (such as financial payment instructions), control signaling, and other service types sensitive to latency and packet loss. When a data packet's priority is determined to be the target priority, meaning the traffic it generates is high-priority traffic, the system can identify the source quantum node and destination quantum node corresponding to that data packet. These correspond to the entry node receiving the data packet and the final destination node of the data packet in the physical network, respectively.

[0082] After determining the source and destination quantum nodes, the system can establish an entanglement protection channel using the logical association between them (i.e., the established virtual entanglement pair). The entanglement protection channel is essentially a logically reserved resource channel, simulating the deterministic transmission characteristics of entangled channels in quantum communication. Specifically, based on the virtual entanglement pair between the source and destination nodes, the system can synchronously reserve bandwidth resources among all nodes along the path, forming a dedicated end-to-end transmission channel. Unlike the best-effort transmission of ordinary data packets, resource reservation in the entanglement protection channel is pre-completed, eliminating the need to wait for congestion to occur before scheduling.

[0083] Through entanglement protection channels, the system can pre-allocate bandwidth resources for data packets to be transmitted. This pre-allocation can be dynamically adjusted based on the entanglement coefficient between the source and destination nodes. For node pairs with high entanglement coefficients, indicating tight state coupling and a higher risk of congestion propagation, more redundant bandwidth resources can be allocated to the entanglement protection channels. Conversely, for node pairs with low entanglement coefficients, relatively fewer reserved resources can be allocated. Ultimately, the system can achieve differentiated protection for high-priority traffic within limited network bandwidth resources.

[0084] For example, in a financial transaction system, a large interbank transfer data packet is identified as high-priority traffic. The system can immediately determine the source quantum node (issuing bank gateway) and the destination quantum node (acquiring bank gateway), and establish an entanglement protection channel using the virtual entanglement pair already established between them. On this channel, the system pre-allocates 10% of the bandwidth resources specifically for this transaction and its associated subsequent transactions. When other ordinary traffic in the network experiences sudden congestion, the high-priority data packet in this entanglement protection channel can still be transmitted stably according to the pre-allocated bandwidth resources, thereby avoiding delays or failures of critical transactions due to congestion.

[0085] Understandably, by establishing entanglement protection channels for high-priority traffic and pre-allocating bandwidth resources, the system can provide deterministic transmission guarantees for critical services before or in the early stages of network congestion, thereby effectively avoiding the impact of sudden congestion on high-priority traffic.

[0086] Based on the above embodiments, in this embodiment, bandwidth resources are pre-allocated for data packets to be transmitted through an entanglement protection channel, including: obtaining the status of the entanglement protection channel through the network controller, and adjusting the bandwidth resources already allocated to the entanglement protection channel according to the load information of the physical network.

[0087] Network controllers (such as software-defined network controllers) can acquire real-time status information of entanglement-protected channels, including current bandwidth utilization, transmission latency, packet loss rate, and real-time entanglement coefficients between source and destination nodes. Based on the status of the entanglement-protected channel, the actual transmission quality of the channel in the current network environment can be determined, serving as the data basis for subsequent bandwidth resource adjustments.

[0088] Physical network load information can include global link utilization, traffic congestion levels at each node, and remaining bandwidth on critical links. When the overall network load is light, the system can appropriately reduce the reserved bandwidth ratio of the entanglement protection channel, releasing more bandwidth resources to ordinary traffic to improve overall bandwidth utilization. When the overall network load is heavy or congestion is predicted, the system can dynamically increase the reserved bandwidth of the entanglement protection channel to ensure that high-priority traffic receives sufficient transmission resources and avoid damage to critical services due to insufficient bandwidth.

[0089] By introducing a network controller, the system can implement a quantum-inspired quality of service (QoS) assurance mechanism. Specifically, the network controller can calculate the bandwidth allocation scheme in real time based on the real-time acquired entanglement protection channel status and global load information, using preset adjustment strategies (such as threshold-based adjustment algorithms or reinforcement learning-based dynamic optimization algorithms), and then distribute the scheme to each switching node for execution via the southbound interface.

[0090] Understandably, unlike the statically configured quality of service (QoS) policies in traditional networks, the bandwidth resource adjustment in this embodiment is dynamic and adaptive, capable of responding in real time to changes in network status, thereby achieving an upgrade from static QoS assurance to quantum-heuristic dynamic QoS assurance.

[0091] Based on the above-described network congestion handling method, this application also provides a network congestion handling apparatus. The following will be combined with... Figure 4 The device is described in detail.

[0092] Figure 4 A schematic block diagram of a network congestion processing apparatus according to an embodiment of this application is shown.

[0093] like Figure 4 As shown, the network congestion processing device 400 of this embodiment includes a construction module 410, a sending module 420, a calculation module 430, and an early warning module 440.

[0094] The construction module 410 is used to map multiple network nodes in the physical network to quantum nodes and establish virtual entanglement pairs between the quantum nodes, wherein each quantum node has state information to characterize its own network state. In one embodiment, the construction module 410 can be used to perform step S210 described above, which will not be repeated here.

[0095] The sending module 420 is used to send the state information of each quantum node to the associated node with which it has a logical relationship, based on the logical association between virtual entangled pairs, so that each quantum node obtains a set of state information containing its own state information and the received state information. In one embodiment, the sending module 420 can be used to perform step S220 described above, which will not be repeated here.

[0096] The calculation module 430 is used to calculate the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair based on the set of state information, and obtain the entanglement coefficient. In one embodiment, the calculation module 430 can be used to execute step S230 described above, which will not be repeated here.

[0097] The early warning module 440 is used to generate an early warning signal indicating potential network congestion by performing different processing on different associated nodes according to the entanglement coefficient when the state information of any quantum node is detected to meet the preset congestion triggering conditions. In one embodiment, the early warning module 440 can be used to execute step S240 described above, which will not be repeated here.

[0098] According to an embodiment of this application, the early warning module 440 can be specifically used to: calculate the congestion probability for each downstream link corresponding to different associated nodes based on the entanglement coefficient and through the distributed prediction model integrated by each associated node; and compare the congestion probability with a preset congestion threshold, and generate a corresponding early warning signal if the congestion probability is greater than or equal to the preset congestion threshold.

[0099] According to an embodiment of this application, the early warning module 440 can be specifically used to: input the set of state information of each associated node and the corresponding entanglement coefficient into a distributed prediction model; and obtain the congestion probability of each associated node within a future preset time window through the distributed prediction model; wherein, the distributed prediction model assigns a higher computational weight to associated nodes whose entanglement coefficient with the quantum node that triggers the congestion triggering condition is higher than that of other associated nodes.

[0100] According to an embodiment of this application, the virtual entanglement pair includes a first virtual entanglement pair and a second virtual entanglement pair; the construction module 410 can be specifically used to: obtain a data transmission link based on the topology of the physical network; establish a first virtual entanglement pair for adjacent quantum nodes in the data transmission link; and establish a second virtual entanglement pair for non-adjacent quantum nodes in the data transmission link through at least one relay quantum node, wherein the relay quantum node is a quantum node located between non-adjacent quantum nodes in the data transmission link.

[0101] According to an embodiment of this application, the early warning module 440 can also be used to: simultaneously enable multiple data transmission links for data packets to be transmitted in the physical network based on an early warning signal; transmit the data packets to be transmitted on the multiple data transmission links respectively; and determine a target transmission link from the multiple data transmission links based on feedback data from the multiple data transmission links, so as to perform subsequent transmission of data packets through the target transmission link.

[0102] According to an embodiment of this application, the early warning module 440 can also be used to: in response to identifying that the priority of the data packet to be transmitted in the physical network is the target priority, determine the source quantum node and the destination quantum node corresponding to the data packet to be transmitted, wherein the source quantum node is the quantum node mapped by the network node receiving the data packet to be transmitted in the physical network, and the destination quantum node is the quantum node mapped by the target network node of the data packet to be transmitted in the physical network; establish an entanglement protection channel between the source quantum node and the destination quantum node using a logical association relationship; and pre-allocate bandwidth resources for the data packet to be transmitted through the entanglement protection channel.

[0103] According to an embodiment of this application, the early warning module 440 can also be used to: obtain the status of the entanglement protection channel through the network controller, and adjust the bandwidth resources allocated to the entanglement protection channel according to the load information of the physical network.

[0104] According to embodiments of this application, any multiple modules among the construction module 410, transmission module 420, calculation module 430, and early warning module 440 can be merged into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the construction module 410, transmission module 420, calculation module 430, and early warning module 440 can be at least partially implemented as hardware circuitry, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any appropriate combination of any of these three implementation methods. Alternatively, at least one of the construction module 410, transmission module 420, calculation module 430, and early warning module 440 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0105] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a network congestion handling method according to an embodiment of this application.

[0106] like Figure 5 As shown, an electronic device 500 according to an embodiment of this application includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory 502 or a program loaded from a storage portion 508 into a random access memory 503. The processor 501 may include, for example, a general-purpose microprocessor, an instruction set processor and / or an associated chipset and / or a dedicated microprocessor. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for executing different steps of the method flow according to an embodiment of this application.

[0107] Random access memory 503 stores various programs and data required for the operation of electronic device 500. Processor 501, read-only memory 502, and random access memory 503 are interconnected via bus 504. Processor 501 executes various steps of the method flow according to embodiments of this application by executing programs stored in read-only memory 502 and / or random access memory 503. It should be noted that programs may also be stored in one or more memories other than read-only memory 502 and random access memory 503. Processor 501 may also execute various steps of the method flow according to embodiments of this application by executing programs stored in one or more memories.

[0108] According to embodiments of this application, the electronic device 500 may further include an input / output interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the input / output interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube, liquid crystal display, etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card, such as a local area network card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the input / output interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0109] Embodiments of this application also provide a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0110] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In embodiments of this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include the read-only memory 502, and / or random access memory 503, and / or one or more memories other than read-only memory 502 and random access memory 503 described above.

[0111] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the methods provided in the embodiments of this application.

[0112] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0113] In embodiments of this application, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of this application embodiment. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0114] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0116] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for handling network congestion, characterized in that, include: Multiple network nodes in a physical network are mapped to quantum nodes, and virtual entanglement pairs are established between the quantum nodes, wherein each quantum node has state information for characterizing its own network state; Based on the logical association between the virtual entangled pairs, the state information of each quantum node is sent to the associated node with which it has the logical association, so that each quantum node obtains a set of state information containing its own state information and the received state information. Based on the state information set, the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair is calculated to obtain the entanglement degree coefficient; and When the state information of any of the quantum nodes is detected to meet the preset congestion triggering conditions, different processing is performed on different associated nodes according to the entanglement coefficient to generate an early warning signal indicating potential network congestion.

2. The method according to claim 1, characterized in that, The step of processing different associated nodes according to the entanglement coefficient to generate an early warning signal indicating potential network congestion includes: For each downstream link corresponding to a different associated node, the congestion probability is calculated based on the entanglement coefficient using a distributed prediction model integrated by each associated node; and The congestion probability is compared with a preset congestion threshold, and if the congestion probability is greater than or equal to the preset congestion threshold, a corresponding early warning signal is generated.

3. The method according to claim 2, characterized in that, The step of calculating the congestion probability based on the entanglement coefficient and through the distributed prediction model integrated by each of the associated nodes includes: The state information sets of each associated node and the corresponding entanglement coefficients are input into the distributed prediction model; and The congestion probability of each associated node within a future preset time window is obtained through the distributed prediction model. The distributed prediction model assigns higher computational weights to associated nodes whose entanglement coefficient with the quantum node that triggers the congestion triggering condition is higher than that of other associated nodes.

4. The method according to claim 1 or 2, characterized in that, The virtual entanglement pair includes a first virtual entanglement pair and a second virtual entanglement pair; The step of mapping multiple network nodes in a physical network to quantum nodes and establishing virtual entanglement pairs between the quantum nodes includes: Based on the topology of the physical network, the data transmission links are obtained; For adjacent quantum nodes in the data transmission link, establish the first virtual entanglement pair; and For non-adjacent quantum nodes in the data transmission link, a second virtual entanglement pair is established through at least one relay quantum node, wherein the relay quantum node is a quantum node located between non-adjacent quantum nodes in the data transmission link.

5. The method according to claim 4, characterized in that, After generating the early warning signal to indicate potential network congestion, the method further includes: Based on the warning signal, multiple data transmission links are simultaneously activated for data packets to be transmitted in the physical network. The data packets to be transmitted are transmitted on the multiple data transmission links respectively; and Based on the feedback data from the multiple data transmission links, a target transmission link is determined from the multiple data transmission links so that the data packet can be subsequently transmitted through the target transmission link.

6. The method according to claim 1 or 5, characterized in that, The method further includes: In response to identifying the priority of the data packet to be transmitted in the physical network as the target priority, the source quantum node and the destination quantum node corresponding to the data packet to be transmitted are determined, wherein the source quantum node is the quantum node mapped by the network node in the physical network that receives the data packet to be transmitted, and the destination quantum node is the quantum node mapped by the target network node in the physical network that receives the data packet to be transmitted. An entanglement-protected channel is established between the source quantum node and the destination quantum node using the logical association; and The entanglement protection channel is used to pre-allocate bandwidth resources for the data packets to be transmitted.

7. The method according to claim 6, characterized in that, The step of pre-allocating bandwidth resources for the data packets to be transmitted through the entanglement protection channel includes: The network controller obtains the status of the entanglement protection channel and adjusts the bandwidth resources allocated to the entanglement protection channel based on the load information of the physical network.

8. A network congestion handling device, characterized in that, include: A construction module is used to map multiple network nodes in a physical network to quantum nodes and establish virtual entanglement pairs between the quantum nodes, wherein each quantum node has state information for characterizing its own network state; The sending module is used to send the state information of each quantum node to the associated node with which it has the logical association relationship based on the logical association relationship between the virtual entangled pairs, so that each quantum node obtains a set of state information containing its own state information and the received state information. The calculation module is used to calculate the coupling degree of network state changes between the quantum nodes corresponding to each virtual entanglement pair based on the state information set, and to obtain the entanglement degree coefficient; and The early warning module is used to generate an early warning signal indicating potential network congestion by performing different processing on different associated nodes according to the entanglement coefficient when the state information of any of the quantum nodes is detected to meet the preset congestion triggering conditions.

9. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

Images