Efficient quantum network entangled quantum source scheduling method and system based on cooperative game

Abstract

The invention discloses an efficient quantum network entangled quantum source scheduling method and system based on a cooperative game. The method comprises the steps that S1, all entangled quantum source nodes in a quantum network obtain the state of a surrounding network, and strategies of the entangled quantum source nodes are randomly initialized; s2, optimizing each entangled quantum source node; s3, selecting an optimal response strategy, and selecting a strategy for maximizing the utility of the entangled quantum source nodes in the group; s4, after the strategies of the entangled quantum source nodes are not changed any more, optimization is stopped, and a final result is recorded as actual strategy selection of the current time slot; and S5, repeating the steps S1 to S4 until all the source-destination node pairs in the quantum network establish end-to-end entangled connections and the global throughput reaches the maximum value. According to the method, the distribution problem of entangled quantum sources in the quantum network can be efficiently solved, and the network throughput is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of quantum communication networks, and specifically relates to a method and system for efficiently scheduling entangled quantum sources in quantum networks based on cooperative games. Background Art

[0002] With the rapid development of quantum computing, it is expected that commercial small-scale quantum computers with limited computing power can be built in the foreseeable future. In order to perform large-scale quantum computing, many such small quantum computers must be networked with a quantum data network (QDN) and perform computing in a distributed manner. To support distributed quantum computing, the QDN must reliably transmit data quantum bits (called qubits), which are usually photons and carry quantum state information, from one quantum computer to another.

[0003] In order to transfer a data qubit from Alice to Bob, an entangled connection (EC) must first be established between them. In other words, both Alice and Bob should possess a qubit from the same entangled qubit pair. When Alice and Bob are not directly connected via a single quantum link to establish such an EC, an entangled path (EP) from Alice to Bob should first be found, which is called the entanglement routing problem. After this, multiple entangled links (ELs) need to be created in sequence. This is a special EC created between two adjacent nodes. Finally, an operation called entanglement swap can be performed on each intermediate node on this EP to stitch these ELs together to form an EC between Alice and Bob. Since an EC will be destroyed when it is used to transmit a data qubit, in order to maximize network throughput, as many ECs as possible must be established simultaneously.

[0004] Due to the existing technical conditions, the entangled quantum source node (EPS), which is the main body of creating entangled links, is very expensive. In the actual quantum communication network, only a small number of nodes are EPS, and most nodes are ordinary nodes that do not have the ability to create EL. This requires a good scheduling strategy. At the same time, because the energy of photons in the quantum state is limited, the entanglement effect between two communication nodes has a certain probability of failure. Therefore, the creation of entangled communication links based on quantum entanglement effects is also probabilistically successful, which increases the complexity of the network. In addition, based on existing experimental physics technology, those ELs in the quantum network that have been successfully created but not used to build ECs (ELs have not been successfully created on the links connected to them) will not fail immediately in the next time slot, but will last for several time slots. During this period, these resources can be utilized by creating ELs on their adjacent links, which increases the difficulty of solving the problem.

[0005] In order to address these limitations and improve the global end-to-end throughput, a series of EPS allocation schemes in quantum communication networks have been proposed. These schemes basically use a single-element priority strategy to allocate EPS to increase the speed of establishing end-to-end entangled connections between source and destination node pairs in the network.

[0006] Some work adopts a simpler approach, namely, adhering to the principle of prioritizing the link with the highest probability of EL establishment. Each EPS will preferentially select the link with the highest probability of establishing an entangled link among its connected links. Other work focuses on existing communication resources. To enable faster utilization of the ELs already established in the network, all EPSs are deployed to select paths with the fewest number of ELs to be established, because such paths usually have a larger number of existing ELs. Still other work further considers the use of existing ELs based on the previous work. Adhering to a more direct principle, it prioritizes paths with ELs with the shortest number of sustainable time slots to utilize them as quickly as possible, regardless of the success probability of creating ELs on these links or the number of ELs to be established on the paths, thereby maximizing the utilization of existing resources in the network and improving global throughput.

[0007] Existing technical solutions suffer from the following major drawbacks: Because solutions that comprehensively consider multiple factors are too complex to be solved within conventional frameworks, they typically prioritize a single factor. This strategy fails to consider the success probability of EL creation on a link, the number of unused hops in the path between the source and destination nodes, and the sustainability of existing network resources. Instead, it statically prioritizes a single factor, limiting further improvements in the network's overall throughput. For example, the strategy prioritizing links with the highest EL establishment probability ignores existing ELs in the network and fails to maximize existing resources. The strategy prioritizing the lowest unused hop count ignores the impact of the EL establishment probability on certain links, potentially selecting links with extremely low establishment probabilities and lowering overall throughput. Furthermore, prioritizing the number of EL sustainable time slots also ignores the number of ELs required, further limiting the expected end-to-end connectivity. Consequently, none of these approaches comprehensively consider factors such as link characteristics, the distribution and sustainability of existing resources, and so on, to efficiently schedule quantum communication networks and achieve higher throughput. Summary of the Invention

[0008] To address the shortcomings of the existing technology, the present invention comprehensively considers factors such as network link characteristics (the probability of successfully creating an entangled link (EL)), the distribution of existing resources in the network, and their persistence, so as to efficiently schedule entangled quantum source nodes in the quantum communication network to achieve higher throughput. It also specifically provides a method and system for efficiently scheduling entangled quantum sources in quantum networks based on cooperative game.

[0009] In order to achieve the aforementioned object of the invention, the present invention adopts the following scheme:

[0010] One aspect of the present invention provides an efficient quantum network entangled quantum source scheduling method based on cooperative game, comprising:

[0011] Step S1: At the beginning of each time interval, all entangled quantum source nodes in the quantum network obtain the state of the surrounding network, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly acquired network state to other entangled quantum source nodes;

[0012] Step S2: For each entangled quantum source node, optimization is performed based on the best response strategy;

[0013] Step S3: After a round of independent optimization, all entangled quantum source nodes are grouped in pairs and the best response strategy is selected. The strategy that maximizes the utility of the entangled quantum source nodes in the group is selected by considering the edges connected to the nodes.

[0014] Step S4: For each round of optimization, when the strategies of all entangled quantum source nodes no longer change, the optimization stops, and the final result is recorded as the actual strategy selection of the current time slot, and then the next time slot is entered;

[0015] Step S5: In the new time slot, repeat steps S1 to S4 and perform multiple rounds of optimization until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

[0016] Another aspect of the present invention provides an efficient quantum network entangled quantum source scheduling system based on cooperative game, comprising:

[0017] The network state maintenance and initialization strategy module is used to obtain the state of the surrounding network at the beginning of each time interval, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly obtained network state to other entangled quantum source nodes;

[0018] The best response strategy optimization module is used to optimize each entangled quantum source node based on the best response strategy;

[0019] The grouping and strategy optimization module is used to group all entangled quantum source nodes into two groups after a round of independent optimization, select the best response strategy, and consider the edges connected to themselves to select the strategy that maximizes the utility of the entangled quantum source nodes in the group;

[0020] The actual strategy selection module for the current time slot is used to stop the optimization in each round when the strategies of all entangled quantum source nodes no longer change. The final result is recorded as the actual strategy selection for the current time slot, and then the next time slot is entered.

[0021] The iterative optimization module is used to perform multiple rounds of optimization in new time slots until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

[0022] Compared with the prior art, the present invention has at least the following advantages:

[0023] (1) The present invention comprehensively considers factors such as network link characteristics (the probability of successfully creating an entangled link (EL)), the distribution of existing resources in the network, and their persistence, thereby resolving various defects and resource waste caused by the existing technology's adherence to a single factor-limited strategy. It can not only focus on utilizing existing ELs in the network to prevent these resources from becoming ineffective, but also not ignore lower-cost paths in the scheduling process and links that can lead to greater ECs due to the higher probability of ELs being established;

[0024] (2) In order to overcome the difficulty of high complexity or even inability to solve the multi-factor coordinated optimization problem, the present invention adopts an idea that is different from the conventional solution framework (such as the optimization method based on linear programming), and models the problem into a cooperative game model. EPS, as a player in the game, optimizes its own choices based on the strategy of best response and performs combinatorial optimization in multiple rounds of iterations to make the game reach equilibrium faster and obtain the optimal strategy set for the current time slot, and finally complete the end-to-end communication of the source and destination node pairs in the entire network with the shortest time consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 is an example network topology diagram in the prior art;

[0027] Figure 2 This is a schematic diagram of selecting the link priority (Pro) strategy with the highest probability of establishing an entangled link in the existing technology;

[0028] Figure 3 This is a schematic diagram of selecting a path priority (Step) strategy with the minimum number of entangled links to be established in the prior art;

[0029] Figure 4 This is a schematic diagram of selecting a path priority (Remaining) strategy with the minimum number of sustainable time slots of the entangled link in the prior art;

[0030] Figure 5 It is the optimal strategy (taking multiple factors into consideration) in the example network topology diagram provided in a typical implementation case of the present invention;

[0031] Figures 6 to 9 This is a calculation process of the total expected value corresponding to each strategy in the example network topology diagram provided by a typical implementation case of the present invention. DETAILED DESCRIPTION

[0032] To make the objectives, technical solutions, and advantages of the present invention more apparent, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. Examples of these preferred embodiments are illustrated in the accompanying drawings. The embodiments of the present invention shown in and described with reference to the accompanying drawings are merely exemplary, and the present invention is not limited to these embodiments.

[0033] One aspect of the present invention provides an efficient quantum network entangled quantum source scheduling method based on cooperative game, comprising:

[0034] Step S1: At the beginning of each time interval, all entangled quantum source nodes in the quantum network obtain the state of the surrounding network, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly acquired network state to other entangled quantum source nodes;

[0035] Step S2: For each entangled quantum source node, optimization is performed based on the best response strategy;

[0036] Step S3: After a round of independent optimization, all entangled quantum source nodes are grouped in pairs and the best response strategy is selected. The strategy that maximizes the utility of the entangled quantum source nodes in the group is selected by considering the edges connected to the nodes.

[0037] Step S4: For each round of optimization, when the strategies of all entangled quantum source nodes no longer change, the optimization stops, and the final result is recorded as the actual strategy selection of the current time slot, and then the next time slot is entered;

[0038] Step S5: In the new time slot, repeat steps S1 to S4 and perform multiple rounds of optimization until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

[0039] In one embodiment, in step S1, the state of the surrounding network acquired by the entangled quantum source node includes: whether an entangled link is established on the link, and the number of sustainable time slots corresponding to the established entangled link;

[0040] The strategies for randomly initializing the entangled quantum source node itself include: selecting any link connected to itself that has not yet established an entangled link, or not selecting any link.

[0041] In one embodiment, in step S1, the state of the surrounding network acquired by the entangled quantum source node includes: whether an entangled link is established on the link, and the number of sustainable time slots corresponding to the established entangled link;

[0042] The strategies for randomly initializing the entangled quantum source node itself include: selecting any link connected to itself that has not yet established an entangled link, or not selecting any link.

[0043] In one embodiment, in step S2, the optimization based on the best response strategy is specifically as follows: fixing the strategy selection of other entangled quantum source nodes in the entire network, and independently optimizing the benefit function by adjusting its own strategy; the benefit function is the expected number of entangled connections established in the entire network after the entangled quantum source node selects a certain link to create an entangled link.

[0044] In one embodiment, there is a certain probability that an entangled quantum source node will fail to establish an entangled link on a link, and each link has its corresponding entangled link establishment success rate; the expected number of entangled connections established is equal to the product of the entangled link creation success rates of all links on the entangled link corresponding to the entangled connection.

[0045] In one embodiment, in the first round of independent optimization, each entangled quantum source node is optimized independently, that is, the strategies of all other entangled quantum source nodes are fixed, and its own strategy selection is changed to maximize the utility, and the iteration is repeated multiple times until the game reaches equilibrium and the strategy selection of all entangled quantum sources no longer changes;

[0046] The number of independent optimizations in the first round is limited, and a maximum number of iterations is set. If the maximum number of iterations has been reached and the entangled quantum sources have not converged, and the strategies of each entangled quantum source node are still changing, the independent optimization phase is terminated and the next round begins. All entangled quantum source nodes are grouped in pairs and optimized in groups. That is, the strategy selections of other groups are fixed, and the strategy selections of the two entangled quantum source nodes in the own group are changed to obtain the maximum value of the utility function.

[0047] Another aspect of the present invention provides an efficient quantum network entangled quantum source scheduling system based on cooperative game, comprising:

[0048] The network state maintenance and initialization strategy module is used to obtain the state of the surrounding network at the beginning of each time interval, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly obtained network state to other entangled quantum source nodes;

[0049] The best response strategy optimization module is used to optimize each entangled quantum source node based on the best response strategy;

[0050] The grouping and strategy optimization module is used to group all entangled quantum source nodes into two groups after a round of independent optimization, select the best response strategy, and consider the edges connected to themselves to select the strategy that maximizes the utility of the entangled quantum source nodes in the group;

[0051] The actual strategy selection module for the current time slot is used to stop the optimization in each round when the strategies of all entangled quantum source nodes no longer change. The final result is recorded as the actual strategy selection for the current time slot, and then the next time slot is entered.

[0052] The iterative optimization module is used to perform multiple rounds of optimization in new time slots until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

[0053] In one embodiment, in the network state maintenance and initialization strategy module, the state of the surrounding network obtained by the entangled quantum source node includes: whether an entangled link is established on the link, and the number of sustainable time slots corresponding to the established entangled link;

[0054] The strategies for randomly initializing the entangled quantum source node itself include: selecting any link connected to itself that has not yet established an entangled link, or not selecting any link.

[0055] In one embodiment, in the best response strategy optimization module, optimization based on the best response strategy is specifically as follows: fixing the strategy selection of other entangled quantum source nodes in the entire network, and independently optimizing the benefit function by adjusting its own strategy; the benefit function is the expected number of entangled connections established in the entire network after the entangled quantum source node selects a certain link to create an entangled link.

[0056] In one embodiment, there is a certain probability that an entangled quantum source node will fail to establish an entangled link on a link, and each link has its corresponding entangled link establishment success rate; the expected number of entangled connections established is equal to the product of the entangled link creation success rates of all links on the entangled link corresponding to the entangled connection.

[0057] In one embodiment, the grouping and policy optimization module includes:

[0058] In the first round of independent optimization, each entangled quantum source node is optimized independently, that is, the strategies of all other entangled quantum source nodes are fixed, and its own strategy selection is changed to maximize the utility. This process is repeated many times until the game reaches equilibrium and the strategy selection of all entangled quantum sources no longer changes.

[0059] The number of independent optimizations in the first round is limited, and a maximum number of iterations is set. If the maximum number of iterations has been reached and the entangled quantum sources have not converged, and the strategies of each entangled quantum source node are still changing, the independent optimization phase is terminated and the next round begins. All entangled quantum source nodes are grouped in pairs and optimized in groups. That is, the strategy selections of other groups are fixed, and the strategy selections of the two entangled quantum source nodes in the own group are changed to obtain the maximum value of the utility function.

[0060] like Figure 1 As shown in FIG, it is a schematic diagram showing the defects of the prior art. Figure 1 In the network shown in

[15] , the success probability of creating an entangled link (EL) varies on different links, and during a given time slot, each entangled photon source (EPS) can only be used to establish a single EL on one link. For the sake of simplicity, the present invention does not consider failures in entanglement exchange or limitations of links in quantum data networks (QDNs). Assume that an entangled connection (EC) is to be established for two source-destination (SD) pairs, namely (S1, D1) and (S2, D2), in two time slots, and their EPSs have been determined as follows: S1->E2->R1->D1 and S2->E2->E3->D2.

[0061] like Figure 2-4 Figure 2 shows the strategies for prioritizing the link with the highest probability of establishing an EL (Pro), prioritizing the path with the smallest number of ELs to be established (Step), and prioritizing the path with the smallest number of sustainable EL slots (Remaining). The expected number of ECs established for these three strategies is 1.62, 1.6767, and 1.7172, respectively.

[0062] like Figure 5 As shown in Figure 1, it shows the optimal strategy in the example network topology (taking multiple factors into consideration). Under the guidance of this strategy, the global throughput can reach 1.8324.

[0063] Among them, global throughput refers to the total number of ECs that have been successfully built in the current network, which is a statistics of the existing entangled connections; expectation refers to the mathematical expectation. The expectation of the number of ECs built is the mathematical expectation of the number of ECs built in the network after the next one or several time slots under the guidance of certain principles and strategies. Specifically, it is a weighted summation of various possible situations encountered, their corresponding probabilities, and the number of ECs built. It is a mathematical calculation and prediction of future benefits under the guidance of a certain strategy.

[0064] The aforementioned relationship between global throughput and the expected number of ECs built is that, if the expected number of ECs built under a certain policy is X, then if this policy is implemented, the increase in network throughput after implementation will be very close to X. Because the illustration in the text starts from the first time slot, the increase in network throughput is also the network throughput itself. Therefore, in the text related to the illustration, the global throughput and the aforementioned expected number of ECs built can be roughly regarded as equivalent.

[0065] Regarding the calculation process of global throughput and the expected number of ECs to be built, we first list several key elements. There are many nodes in the network. Only the entangled quantum source node can attempt to create an entangled link on the link connected to itself. The probability of successfully creating an entangled link on each link is different (this is related to the properties of the link itself). Figure 5-9 It includes the strategy selection, probability and expectation calculation within two time slots. The specific calculation process is shown in the attached Figure 5-9 .

[0066] In order to break through the limitations of the single-factor priority strategy in the existing technology, the network link characteristics (the probability of successfully creating EL), the distribution of existing resources in the network, and the continuity of the situation need to be solved. In the traditional solution framework (linear programming modeling and optimization using LP solvers, etc.), the problem of high problem complexity and thus the inability to obtain an effective solution needs to be solved. This paper proposes a scheduling scheme based on a cooperative game model (a type of game model).

[0067] In this solution, a game theory model is introduced. The basic concepts include players, action space, and payoff function, which roughly mean as follows:

[0068] A player is an individual or entity involved in a game or decision. In a game or decision problem, there can be one or more players, each with their own goals and interests (the interests of different players in cooperative games are the same) and taking actions based on their decisions. The action space refers to the set of possible actions each player can choose. In a game or decision problem, different players may make decisions at different points in time, and the action space describes all possible actions each player can choose. The payoff function describes the benefits or utility each player receives based on their actions and the outcome of the game. It maps each possible game outcome to each player's payoff. The payoff function can be expressed as a mathematical function or defined using a table or a set of rules.

[0069] Based on these concepts, the present invention models each EPS in the network as a player in a game model. Each EPS's choice of links connected to itself constitutes the player's action space, and the payoff function is the expected number of ECs built in the entire network after the EPS selects a link to create an EL.

[0070] Because the EPS only has a probability of successfully creating an EL on a link, the present invention does not make predictions about the impact of the creation choice at the current moment on the next moment (for example, whether an EL is built on a certain link, or whether an EC is built on a certain EP). Instead, it only performs a mathematical expectation calculation based on the link properties (the success probability of creating an EL) at the current moment.

[0071] Abstract modeling of the problem:

[0072] A network consists of multiple nodes. Some are ordinary nodes and lack the ability to create ELs, while others are EPSs, which attempt to create ELs by selecting links connected to them. The network also contains numerous source-destination pairs, which exchange information. There is a certain probability that an EPS will fail to establish an EL on a link. Each link has its own EL establishment success rate. The expected number of successful ECs is equal to the product of the EL establishment success rates of all links on the EP corresponding to that EC.

[0073] Each EPS in the network is modeled as a player in a game model. The choices made by each EPS regarding the links connected to it constitute the player's action space, and the payoff function is the expected number of ECs built in the entire network after the EPS selects a certain link to create an EL.

[0074] The basic process of the solution proposed by the present invention is as follows:

[0075] 1. At the beginning of each time slot (a time slot is simply a period of time, such as 5ms), all EPSs obtain the status of the surrounding network (for example, whether an EL has been established on the link and the number of sustainable time slots for established ELs). They randomly initialize their own policies (selecting any link connected to them that has not yet established an EL, or selecting no link at all) and synchronize their own policies and the newly acquired network status with other EPSs.

[0076] 2. For each EPS, optimize based on the best response strategy. That is, fix the strategy choices of other EPSs in the entire network and independently optimize the benefit function (i.e., the expected number of ECs built in the network after making a certain choice) by adjusting its own strategy.

[0077] 3. After a round of independent optimization, EPSs are divided into groups of two and the best response strategy is selected. Considering the edges connected to themselves, the strategy that maximizes the utility of the nodes in the group is selected.

[0078] Specifically, in the first round, each entangled quantum source node is independently optimized (specifically, the strategies of all other entangled quantum source nodes are fixed, and its own strategy selection is changed to maximize utility), and the iteration is repeated multiple times until the game reaches equilibrium and the strategy selection of all entangled quantum sources no longer changes. Taking into account practical requirements, the present invention limits the number of independent optimizations in the first round and sets a maximum number of iterations. If the maximum number of iterations has been reached and the entangled quantum sources still have not converged, and the strategies of each entangled quantum source node are still changing, the independent optimization link is terminated and the next round begins. All entangled quantum source nodes are grouped in pairs and optimized in groups. Specifically, the strategy selections of other groups are fixed, and the strategy selections of the two entangled quantum source nodes in the own group are changed to obtain the maximum value of the utility function. For example, there are 4 entangled quantum source nodes in the network, namely 1, 2, 3, and 4. In the first round, the 4 nodes are optimized independently. If they have converged before reaching the maximum number of iterations, the strategy selection of each node will be executed as the final strategy of this time slot; if they have not converged, the second round will be entered, and all nodes will be grouped into (1, 2) (1, 3) (1, 4) (2, 3) (2, 4) (3, 4). Each group will optimize independently, fix the strategy selection of other groups, change the strategy selection of the two nodes within its own group, and find the strategy corresponding to the maximum value of the effect function.

[0079] 4. For each round of optimization described above, when the strategies of each EPS no longer change, that is, when a Nash equilibrium is reached, the optimization stops and the final result is recorded as the actual strategy selection for the current time slot, and then the next time slot is entered. The strategy selection of each entangled quantum source node in this time slot is: Each entangled quantum source node is connected to several links that have not yet established entanglement links. In each time slot, each entangled quantum source node needs to choose among these links and then attempt to establish an entanglement link on the selected link.

[0080] 5. In the new time slot, the EPS maintains the latest network status, initializes its own strategy and propagates it, and performs multiple rounds of optimization until all source and destination node pairs in the network have established end-to-end entangled connections and the global throughput reaches its maximum value.

[0081] 6. The entire algorithm is performed offline, and the calculation results of all time slots are recorded uniformly as the strategy selection for the current network.

[0082] It should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions of each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. An efficient quantum network entangled quantum source scheduling method based on cooperative game, characterized in that: include: Step S1: At the beginning of each time interval, all entangled quantum source nodes in the quantum network obtain the state of the surrounding network, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly acquired network state to other entangled quantum source nodes; Step S2: For each entangled quantum source node, optimization is performed based on the best response strategy; Step S3: After a round of independent optimization, all entangled quantum source nodes are grouped in pairs and the best response strategy is selected. The strategy that maximizes the utility of the entangled quantum source nodes in the group is selected by considering the edges connected to the nodes. Step S4: For each round of optimization, when the strategies of all entangled quantum source nodes no longer change, the optimization stops, and the final result is recorded as the actual strategy selection of the current time slot, and then the next time slot is entered; Step S5: In the new time slot, repeat steps S1 to S4 and perform multiple rounds of optimization until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

2. The method for scheduling quantum network entangled quantum sources according to claim 1, characterized in that: In step S1, the state of the surrounding network obtained by the entangled quantum source node includes: whether an entangled link has been established on the link, and the number of sustainable time slots corresponding to the established entangled link; The strategies for randomly initializing the entangled quantum source node itself include: selecting any link connected to itself that has not yet established an entangled link, or not selecting any link.

3. The quantum network entangled quantum source scheduling method according to claim 1, characterized in that: In step S2, the optimization based on the best response strategy is specifically as follows: fixing the strategy selection of other entangled quantum source nodes in the entire network, and independently optimizing the benefit function by adjusting its own strategy; the benefit function is the expected number of entangled connections established in the entire network after the entangled quantum source node selects a certain link to create an entangled link.

4. The method for scheduling quantum network entangled quantum sources according to claim 3, characterized in that: There is a certain probability that an entangled quantum source node will fail to establish an entangled link on a link. Each link has its corresponding entangled link establishment success rate; the expected number of entangled connections established is equal to the product of the entangled link creation success rates of all links on the entangled link corresponding to the entangled connection.

5. The method for scheduling quantum network entangled quantum sources according to claim 1, characterized in that: The step S3 comprises: In the first round of independent optimization, each entangled quantum source node is optimized independently, that is, the strategies of all other entangled quantum source nodes are fixed, and its own strategy selection is changed to maximize the utility. This process is repeated many times until the game reaches equilibrium and the strategy selection of all entangled quantum sources no longer changes. The number of independent optimizations in the first round is limited, and a maximum number of iterations is set. If the maximum number of iterations has been reached and the entangled quantum sources have not converged, and the strategies of each entangled quantum source node are still changing, the independent optimization phase is terminated and the next round begins. All entangled quantum source nodes are grouped in pairs and optimized in groups. That is, the strategy selections of other groups are fixed, and the strategy selections of the two entangled quantum source nodes in the own group are changed to obtain the maximum value of the utility function.

6. An efficient quantum network entangled quantum source scheduling system based on cooperative game, characterized by: include: The network state maintenance and initialization strategy module is used to obtain the state of the surrounding network at the beginning of each time interval, randomly initialize the entangled quantum source node's own strategy, and synchronize its own strategy and the newly obtained network state to other entangled quantum source nodes; The best response strategy optimization module is used to optimize each entangled quantum source node based on the best response strategy; The grouping and strategy optimization module is used to group all entangled quantum source nodes into two groups after a round of independent optimization, select the best response strategy, and consider the edges connected to themselves to select the strategy that maximizes the utility of the entangled quantum source nodes in the group; The actual strategy selection module for the current time slot is used to stop the optimization in each round when the strategies of all entangled quantum source nodes no longer change. The final result is recorded as the actual strategy selection for the current time slot, and then the next time slot is entered. The iterative optimization module is used to perform multiple rounds of optimization in new time slots until all source-destination node pairs in the quantum network have established end-to-end entangled connections and the global throughput reaches the maximum value.

7. The quantum network entangled quantum source scheduling system according to claim 6, characterized in that: In the network state maintenance and initialization strategy module, the state of the surrounding network obtained by the entangled quantum source node includes: whether an entangled link has been established on the link, and the number of sustainable time slots corresponding to the established entangled link; The strategies for randomly initializing the entangled quantum source node itself include: selecting any link connected to itself that has not yet established an entangled link, or not selecting any link.

8. The quantum network entangled quantum source scheduling system according to claim 6, characterized in that: In the best response strategy optimization module, the optimization based on the best response strategy is specifically as follows: fixing the strategy selection of other entangled quantum source nodes in the entire network, and independently optimizing the benefit function by adjusting its own strategy; the benefit function is the expected number of entangled connections established in the entire network after the entangled quantum source node selects a certain link to create an entangled link.

9. The quantum network entangled quantum source scheduling system according to claim 8, characterized in that: There is a certain probability that an entangled quantum source node will fail to establish an entangled link on a link. Each link has its corresponding entangled link establishment success rate; the expected number of entangled connections established is equal to the product of the entangled link creation success rates of all links on the entangled link corresponding to the entangled connection.

10. The quantum network entangled quantum source scheduling system according to claim 6, characterized in that: The grouping and strategy optimization module includes: In the first round of independent optimization, each entangled quantum source node is optimized independently, that is, the strategies of all other entangled quantum source nodes are fixed, and its own strategy selection is changed to maximize the utility. This process is repeated many times until the game reaches equilibrium and the strategy selection of all entangled quantum sources no longer changes. The number of independent optimizations in the first round is limited, and a maximum number of iterations is set. If the maximum number of iterations has been reached and the entangled quantum sources have not converged, and the strategies of each entangled quantum source node are still changing, the independent optimization phase is terminated and the next round begins. All entangled quantum source nodes are grouped in pairs and optimized in groups. That is, the strategy selections of other groups are fixed, and the strategy selections of the two entangled quantum source nodes in the own group are changed to obtain the maximum value of the utility function.

Images