A method, device, medium and product for deterministic transmission scheduling of TSN and 5G fusion network
By establishing a centralized and distributed user management model of TSN-5G converged network, and using the D3QN algorithm to determine the optimal allocation strategy for time slots and spectrum, the transmission of TSN and 5G converged networks under limited spectrum resources and strict delay requirements is solved. The problem of low success rate has achieved efficient resource allocation and collaboration.
Patent Information
- Application Number
- CN202410248897.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-05
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-03-05
AI Technical Summary
In the prior art, the configuration decoupling of TSN and 5G converged networks is achieved in two independent stages, resulting in a low transmission success rate under limited spectrum resources and strict delay requirements, and a lack of effective joint network management and time-frequency allocation methods.
Establish a centralized network and distributed user management model for TSN-5G converged network, and transform it into a Markov decision-making process through the joint allocation of time and frequency resources, and use the D3QN algorithm to determine the optimal allocation strategy for time slots and spectrum.
The transmission success rate under limited spectrum resources and strict delay requirements has been improved, and efficient resource allocation is achieved through collaboration, and the main challenges in TSN and 5G collaboration have been overcome.
Smart Images

Figure CN118283656B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of industrial Internet of Things, and in particular to a method, device, medium and product for deterministic transmission scheduling of a TSN and 5G fused network. Background Art
[0002] Recently, deterministic transmission has attracted a lot of attention, such as the Industrial Internet of Things (IIoT), for time-sensitive industrial applications that require ultra-low latency of a few microseconds to a few milliseconds. In addition, to improve the mobility and communication range of industrial networks, wireless technologies are adopted, and high-reliability radio links between air interfaces are indispensable for critical manufacturing-related services such as equipment failure alarms and factory fire monitoring. These emerging requirements have inspired the convergence of Time-Sensitive Networks (TSN) and 5G.
[0003] Current research decouples the configuration of TSN and 5G converged networks into two independent stages: TSN configuration and 5G configuration. However, since the configuration results of the two stages are nested and need to be calculated simultaneously, it is essential to study the interaction between the two stages. In this context, it is urgent to design a method for joint network management and time-frequency allocation of TSN and 5G. Summary of the invention
[0004] The purpose of the present invention is to provide a method, device, medium and product for deterministic transmission scheduling of a TSN and 5G converged network, which improves the transmission success rate of the network under limited spectrum resources and strict delay requirements.
[0005] To achieve the above object, the present invention provides the following solution.
[0006] A method for deterministic transmission scheduling of a TSN and 5G fusion network includes: establishing a centralized network and a distributed user management model corresponding to the TSN-5G fusion network.
[0007] Based on the centralized network and distributed user management model, a random optimization problem for the joint allocation of time-frequency resources is established; the random optimization problem for the joint allocation of time-frequency resources includes: optimization objectives and constraints, the optimization objective is to maximize the number of successfully scheduled data streams and the throughput level of the device, and the constraints include: transmission delay constraints, link capacity constraints, interval injection constraints, discreteness constraints of frequency resource allocation, and reliability constraints.
[0008] Based on the stochastic optimization problem of joint allocation of time-frequency resources, the deterministic scheduling process of the TSN-5G fusion network is transformed into a Markov decision process.
[0009] The D3QN-based joint resource scheduling algorithm is used to determine the optimal allocation strategy of time slots and spectrum based on the Markov decision process.
[0010] Optionally, the TSN-5G fusion network includes: TSN equipment, 5G equipment and connection equipment.
[0011] Optionally, the transmission delay constraint is as follows.
[0012] .
[0013] in, represents f i Acceptable delay, f i represents the i-th scheduled data stream; D tran Indicates transmission delay; D prop represents the link propagation delay; D proc Indicates CPU processing delay; D queue represents the queuing delay; F represents the set of scheduled data flows.
[0014] Optionally, the link capacity constraint is as follows.
[0015] .
[0016] .
[0017] in, Indicates the total number of scheduled data flows; represents f i The total number of packets, f i represents the i-th scheduled data stream; Indicates that at time τ Transmission on the a1th TSN link, represents f i The jth data packet on ; express size; represents the maximum capacity of the a1th TSN link; Indicates that at time τ, f i Transmission conditions on the a2th 5G link; represents f i size of prop this_τ Indicates that at time τ, f i The ratio of sent data packets to total data packets; Represents the capacity of the a2th 5G link.
[0018] Optionally, the interval injection constraint is as follows.
[0019] .
[0020] .
[0021] in, Indicates that at time τ Injection of TSN links, represents f i The jth packet on f i represents the ith scheduled data stream; F represents the set of scheduled data streams; Indicates the total number of time slots; Indicates that at time τ Injection status of 5G links; Indicates that at time τ Transmitted in 5G network; Indicates that at time τ Transmitted in the TSN network.
[0022] Optionally, the discreteness constraint of the frequency resource allocation is as follows.
[0023] .
[0024] in, represents f i The final frequency resource obtained, f i represents the i-th scheduled data stream; B req Indicates the spectrum required by wireless devices; B sub Indicates the current subcarrier spacing; F indicates the set of scheduled data streams.
[0025] Optionally, the reliability constraint is as follows.
[0026] .
[0027] in, represents f i The total number of packets, f i represents the i-th scheduled data stream; represents f i The jth data packet on ; express The arrival situation of the destination.
[0028] A computer device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the TSN and 5G converged network deterministic transmission scheduling method described in any one of the above items.
[0029] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned TSN and 5G converged network deterministic transmission scheduling methods.
[0030] A computer program product comprises a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for deterministic transmission scheduling of TSN and 5G converged networks.
[0031] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the present invention discloses a method, device, medium and product for deterministic transmission scheduling of TSN and 5G converged networks. First, a centralized network and distributed user management model corresponding to the TSN-5G converged network are established; secondly, based on the centralized network and distributed user management model, a random optimization problem for joint allocation of time-frequency resources is established; the random optimization problem for joint allocation of time-frequency resources includes: optimization objectives and constraints, and the optimization objective is to maximize the number of successfully scheduled data streams and the throughput level of the device. The constraints include: transmission delay constraints, link capacity constraints, interval injection constraints, discrete constraints and reliability constraints of frequency resource allocation; thirdly, based on the random optimization problem for joint allocation of time-frequency resources, the deterministic scheduling process of the TSN-5G converged network is converted into a Markov decision process; finally, a joint resource scheduling algorithm based on D3QN is used to determine the optimal allocation strategy of time slots and spectrum based on the Markov decision process. This paper makes full use of the different resource granularities, time-varying channel characteristics and inconsistent optimization goals between TSN and 5G, and overcomes the main challenges in the collaboration between TSN and 5G by establishing a joint time-frequency resource allocation model. The D3QN algorithm is used to solve the resource allocation problem and obtain the optimal allocation strategy for time slots and spectrum. Compared with other heuristic and learning-based algorithms, this algorithm achieves efficient collaboration between TSN and 5G and improves the transmission success rate of the network under limited spectrum resources and strict delay requirements. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0033] Figure 1 A flow chart of a method for deterministic transmission scheduling in a TSN and 5G converged network provided in Example 1 of the present invention.
[0034] Figure 2Schematic diagram of the scenario of deterministic transmission scheduling method for TSN and 5G converged network.
[0035] Figure 3 This is a diagram of the internal structure of a computer device. DETAILED DESCRIPTION
[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0037] The purpose of the present invention is to provide a method, device, medium and product for deterministic transmission scheduling of a TSN and 5G converged network, aiming to improve the transmission success rate of the network under limited spectrum resources and strict delay requirements.
[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0039] Embodiment 1 provides a method for deterministic transmission scheduling in a TSN and 5G converged network.
[0040] like Figure 1 As shown, the TSN and 5G converged network deterministic transmission scheduling method in this embodiment includes steps 101 to 104.
[0041] Step 101: Establish a centralized network and distributed user management model corresponding to the TSN-5G converged network.
[0042] As an optional implementation, Figure 2 As shown in the figure, the TSN-5G converged network includes: TSN devices, 5G devices and connection devices.
[0043] Specifically, TSN equipment mainly includes: terminals, TSN switches and centralized network configuration CNC. 5G equipment mainly includes: base stations (gNB) and various important functions of the core network, including: policy control function (PCF), access and mobility management function (AMF), session management function (SMF), network open function (NEF) and unified data management (UDM). Among them, PCF provides policy rules for control plane functions and is mainly responsible for QOS service management; AMF supports user terminals with different mobility management requirements and is mainly responsible for access management and mobility management; SMF is mainly responsible for session management; NEF is mainly responsible for network capability collection and opening network capabilities to third-party applications, realizing friendly docking between network capabilities and business requirements, improving business experience, and optimizing network resource configuration; UDM is mainly responsible for storing user information. Connected devices include: device-side TSN converter (DSTT), network-side TSN converter (NW-TT) and TSN application function (TSN-AF). The entire 5G system can be regarded as a transparent logical bridge. It will not perceive specific TSN services. It only supports the TSN series of protocols through DS-TT and NW-TT on the terminal side, and realizes the protocol conversion between 5G services and TSN services. DS-TT is responsible for mapping the relevant ports, protocol data units and QoS mechanisms between user terminals and service systems, and supports TSN related data traffic scheduling features. TSN-AF supports deterministic communication of user terminals under the same user plane function (UPF). Time-sensitive information between two TSN networks is transmitted by the 5G system and the connected device relay. The unified control plane provides QOS management for multiple TSN networks and 5G systems by mastering the global network topology, device status, link capacity and other information.
[0044] The converged network operates in the following way: 1) The terminals in TSN send transmission requirements to CNC through the edge TSN switch, which can be regarded as an agent. 2) The control plane collects status information to make network configuration decisions. A centralized network and distributed user management mode is adopted. In a centralized network, the controller is responsible for distributing the gateway control list, joint resource allocation decisions and forwarding table of all TSN switches and 5G core networks. For distributed users, the terminal reports to the edge TSN switch to form an autonomous domain. Take TSN switch G1 as an example. There is a set of TSN terminal subsets U={u1, u2,…, u y}, where u1, u2, …, u y All of them are lower-level devices of TSN switch G1. TSN switch G1 maintains a list to manage devices, u yRegistering or deregistering only on switch G1 avoids directly bringing processing pressure to the control plane and generating a large amount of control data on the backbone network. 3) The data plane transmits time-sensitive data streams according to the gate control list and forwarding table from the CNC. 4) DS-TT and NW-TT fully play the role of bridge at the intersection of TSN and 5G, encapsulate or decapsulate data packets of different formats according to the network protocol, buffer data, and achieve "quasi-synchronization" of the system clock.
[0045] Step 102: Based on the centralized network and distributed user management model, a stochastic optimization problem for joint allocation of time-frequency resources is established.
[0046] Among them, the stochastic optimization problem of joint allocation of time and frequency resources includes: optimization objectives and constraints. The optimization objective is to maximize the number of successfully scheduled data streams and the throughput level of the equipment. The constraints include: transmission delay constraints, link capacity constraints, interval injection constraints, discrete constraints of frequency resource allocation, and reliability constraints.
[0047] Specifically, the optimization goal is to maximize the throughput fairness of wired devices and wireless devices while maximizing the number of successfully scheduled data flows, that is, maximizing .in, Indicates the number of successfully scheduled data flows, Indicates that the throughput of wired and wireless devices is fair, express The weight factor of express The weight factor of .
[0048] .
[0049] in, Indicates the total number of links; represents f i The total number of packets, f i represents the i-th scheduled data stream; express The size of represents f i The jth data packet on ; express The transmission situation in the kth link, if If it is sent in the kth link, the value is 1, otherwise, the value is 0; C apk represents the capacity of the kth link.
[0050] To describe the network topology, we assume Represents the link set between devices, where b1 represents the first link and b2 represents the second link. Indicates There are links, so the characteristics of data flows and links can be described by the following formula. Among them, each data flow has characteristics such as delay requirement, destination address, source address, packet number, data flow identifier, etc., and each link has characteristics such as link type (wired / wireless), link capacity, communication end node 1, end node 2, link identifier, link optional parameters, etc.
[0051] .
[0052] Among them, f i represents the i-th scheduled data stream; represents f i Acceptable delays; represents f i The destination node; represents f i The source node of represents f i The total number of packets; represents f i Identification number of k represents the kth link; Indicates the type of the kth link (wired / wireless); represents the communication end node 1 of the kth link; represents the communication end node 2 of the kth link; Indicates the identification number of the kth link; Optional parameter indicating the kth link.
[0053] As an optional implementation, the transmission delay constraint is as follows.
[0054] .
[0055] in, represents f i Acceptable delay, f i represents the i-th scheduled data stream; D tran Indicates transmission delay; D prop represents the link propagation delay; D proc Indicates CPU processing delay; D queue represents the queuing delay; F represents the set of scheduled data flows.
[0056] Specifically, the transmission delay constraint expresses that the acceptable delay of the data flow should be greater than the actual total transmission delay.
[0057] As an optional implementation, the link capacity constraint is as follows.
[0058] .
[0059] .
[0060] in, Indicates the total number of scheduled data flows; represents f i The total number of packets, f i represents the i-th scheduled data stream; Indicates that at time τ Transmission on the a1th TSN link, if at time τ If the transmission is on the a1th TSN link, the value is 1, otherwise, the value is 0. represents f i The jth data packet on ; express size; represents the maximum capacity of the a1th TSN link; Indicates that at time τ, f i Transmission situation on the a2th 5G link, if at time τ f i In a2 If the transmission is on a 5G link, the value is 1, otherwise, the value is 0; represents f i size of prop this_τ Indicates that at time τ, f i The ratio of sent data packets to total data packets; Represents the capacity of the a2th 5G link.
[0061] Specifically, the link capacity constraint expresses that in each time interval, the total amount of data transmitted on the wired link or wireless link should be less than the link capacity to avoid message congestion. TSN links are wired links; 5G links are wireless links. Data packets first pass through the wired link and then the wireless link.
[0062] As an optional implementation, the interval injection constraint is as follows.
[0063] .
[0064] .
[0065] in, Indicates that at time τ For the injection of TSN link, if If it is injected into the TSN link, the value is 1, otherwise, the value is 0. represents f i The jth packet on f i represents the ith scheduled data stream; F represents the set of scheduled data streams; Indicates the total number of time slots; Indicates that at time τ For the injection of 5G link, if If it is injected into the 5G link, the value is 1, otherwise, the value is 0; Indicates that at time τ Transmitted in 5G network; Indicates that at time τ Transmitted in the TSN network.
[0066] Specifically, the interval injection constraint expresses that TSN and 5G have different requirements for packet injection intervals. TSN requires that packets be transmitted within one time interval, while 5G allows packets to occupy two time intervals in some cases.
[0067] As an optional implementation, the discreteness constraint of frequency resource allocation is as follows.
[0068] .
[0069] in, represents f i The final frequency resource obtained, f i represents the i-th scheduled data stream; B req Indicates the spectrum required by wireless devices; B sub Indicates the current subcarrier spacing; F indicates the set of scheduled data streams.
[0070] Specifically, the discrete constraint of frequency resource allocation expresses that: unlike TSN networks where bandwidth is a continuous variable, 5G frequency allocation is discrete. Considering that the frequency resource allocation units are 15kHz, 30kHz, 60kHz, etc., the final allocated resource block must be greater than the required minimum integer.
[0071] As an optional implementation, the reliability constraint is as follows.
[0072] .
[0073] in, represents f i The total number of packets, f i represents the i-th scheduled data stream; represents f i The jth data packet on ; express If the destination is reached, If the destination is reached, the value is 1, otherwise, the value is 0.
[0074] Specifically, the reliability constraint expresses that the packet loss rate should be less than 0.5%.
[0075] Step 103: Based on the stochastic optimization problem of joint allocation of time-frequency resources, the deterministic scheduling process of the TSN-5G fusion network is transformed into a Markov decision process.
[0076] Specifically, step 103 specifically includes: Markov decision is a model, the input is state space and action space, and the output is reward and strategy formed by D3QN neural network.
[0077] (1) State space: The state space consists of two parts: the remaining capacity of the wired / wireless link at different intervals and the data flow f at step t t The transmission demand is , and use a matrix to describe .
[0078] .
[0079] Among them, s t represents the state of step t; represents the transmission demand matrix of the data flow at step t; represents the link capacity state matrix at step t; Indicates that the tth step is in the time interval T q1 When the q2th link b q2 The remaining capacity; q1=1,2,…,n; q2=1,2,…,m.
[0080] (2) Action space: The action space is described by a tuple (time, frequency). Specifically, the action space A is divided into two dimensions. Dimension 1 contains a set of waiting to be assigned to each data flow f t The time interval Г = {T1, T2, ..., Tn}. It indicates when these data streams are sent in the TSN network, corresponding to the "time" of the tuple. Dimension 2 is related to frequency allocation and corresponds to the "frequency" of the tuple. Let Indicates the wireless device connected to the gNB. Each wireless device can choose to use three actions (using indicates) to forward data flow f t : Next interval, or use two intervals at the same time. To represent the data flow f t The number of intervals occupied, ,in, Indicates the allocation to data stream f t bandwidth, Indicates the size of the subcarrier spacing (SCS).
[0081] (3) Reward: Once an action is taken, the environment provides immediate rewards based on the state transition results. Among them, r t is the reward for step t, v and are weight factors, is the data stream f t The size of is the interval T in the tth step n The throughput fairness index of the wired device or wireless device, , For the tth step in the time interval T n When the kth link b k The remaining capacity, is the kth link b k total capacity.
[0082] Step 104: Determine the optimal allocation strategy of time slots and spectrum based on the Markov decision process using the joint resource scheduling algorithm based on D3QN.
[0083] Specifically, step 104 specifically includes: using a deep neural network to achieve state perception. In the current state, the agent randomly selects actions based on a greedy strategy, interacts with the environment, and generates reward values for the next state and the current action. The efficient use of data is achieved based on the experience replay mechanism, and the neural network is updated based on the data extracted from the experience replay pool. The optimal solution for the unloading decision can be obtained before the iteration limit reaches or falls below a certain threshold. The algorithm uses two neural networks, and its parameters are periodically synchronized. One network is used to select the optimal action, and the other network is used to estimate the Q value of the action to avoid overestimation. In addition, the hidden layer of the neural network outputs the value function and the advantage function respectively. After collecting data for only one discrete action, Dueling DQN can estimate the Q value more accurately and select a more appropriate action. DoubleDQN selects the target Q value through the action selected by the target Q value, thereby eliminating the problem of overestimation of the Q value. DuelingDouble DQN (D3QN) combines the advantages of Dueling DQN and Double DQN.
[0084] (1) Action exploration based on greedy strategy: The deep neural network takes the current state of the agent as input and outputs the state-action value function of all possible actions in this state. Only one action can be selected each time the agent interacts with the environment. To prevent falling into the local optimal solution, the action selection strategy is based on the ε-greedy strategy, that is, in the current state, the agent has a probability of 1-ε to execute the optimal action, and a probability of ε to randomly select an arbitrary action to execute. The corresponding action selection formula is as follows.
[0085] .
[0086] Among them, a t is the t-th action; randow action is a random action; is the current neural network about s t and a t The state-action value function; x is a random variable, which is regenerated at each step. When a round is over, the exploration rate ε is updated. The update formula is , ε fin is the minimum exploration rate; ε1 is the exploration rate before the update; ε2 is the exploration rate after the update. fin When , stop updating ε.
[0087] After the action selection is completed, the agent will perform this action in the environment, and then the environment will return to the state s at step t+1 t+1 and the reward r at step t t , then the current state, current action, next state and whether to end the flag will be stored in the memory playback library. The five-tuple is represented as: (s t , a t ,s t+1 , r t , done). Done indicates state S t Whether it is in the end state.
[0088] (2) Experience replay mechanism: The samples randomly generated by the greedy strategy are stored in the memory replay pool. During the iterative training process, the memory replay pool stores the experience values of the agent in various states and different behaviors in interacting with the environment. These experience values are randomly sampled from the replay memory pool, and the experience values at different times are mixed, which reduces the correlation between the data. At the same time, the sampled samples are used for model training and parameter updating, which improves data utilization and training efficiency.
[0089] (3) Update the neural network: Improve the training stability and algorithm convergence by introducing a target neural network with the same structure as the current neural network but different parameters. The current neural network evaluates and estimates the value function obtained by the action taken by the agent in the current state. After each certain number of iterations of the target neural network, the parameters θ of the current neural network are copied and updated.
[0090] The current neural network and the target neural network randomly extract samples from the memory playback library in a random sampling manner to generate the corresponding current Q value and target Q value. The current Q value is: Q(s t , a t ,θ); the target Q value is: Among them, a t+1 is the t+1th step; is the parameter of the target network; γ is the attenuation coefficient.
[0091] The mean square error function Loss(θ) is introduced and the mean square error function is as follows.
[0092] .
[0093] .
[0094] Where E is the total number of samples; z is the sequence number of the sample; s z is the state of the zth sample; a z The action taken for the zth sample; y z is the corrected Q value of the zth sample; r z is the reward value of the zth sample; is the state-action value function of the current neural network; is the state-action value function of the target neural network; For the zth sample taking action a z The new state to which it is transferred later; In the new state Next, select the action.
[0095] The parameters of the current neural network are updated by minimizing the loss function.
[0096] At certain step lengths, the parameters of the current neural network are copied to the target neural network to update the parameters of the target neural network.
[0097] Finally, the above process is repeated until the iteration limit reaches the set value, and the optimal resource allocation deployment can be obtained.
[0098] Embodiment 2 provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the TSN and 5G converged network deterministic transmission scheduling method in Embodiment 1.
[0099] Embodiment 3 provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the TSN and 5G converged network deterministic transmission scheduling method in Embodiment 1.
[0100] Embodiment 4 provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the TSN and 5G converged network deterministic transmission scheduling method in Embodiment 1.
[0101] Embodiment 5 provides a computer device, which may be a database, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the TSN and 5G converged network deterministic transmission scheduling method in Example 1 is implemented.
[0102] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.
[0103] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.
[0104] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0105] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A method for deterministic transmission scheduling in a TSN and 5G fusion network, characterized in that: The method comprises: Establish a centralized network and distributed user management model corresponding to the TSN-5G converged network; Based on the centralized network and distributed user management model, a random optimization problem of joint allocation of time-frequency resources is established; the random optimization problem of joint allocation of time-frequency resources includes: optimization objectives and constraints, the optimization objective is to maximize the number of successfully scheduled data flows and the throughput level of the device, and the constraints include: transmission delay constraints, link capacity constraints, interval injection constraints, discreteness constraints of frequency resource allocation, and reliability constraints; Based on the stochastic optimization problem of joint allocation of time-frequency resources, the deterministic scheduling process of the TSN-5G fusion network is converted into a Markov decision process; Determine the optimal allocation strategy of time slots and spectrum based on the Markov decision process using a joint resource scheduling algorithm based on D3QN; The link capacity constraint includes: Where |F| represents the total number of scheduled data flows; F represents the set of scheduled data flows; f i.pnum represents f i The total number of packets, f i represents the i-th scheduled data stream; Indicates that at time τ Transmission on the a1th TSN link, represents f i The jth data packet on ; express size; represents the maximum capacity of the a1th TSN link; Indicates that at time τ, f i Transmission situation on the a2 5G link; f i .size means f i size of prop this_τ Indicates that at time τ, f i The ratio of sent data packets to total data packets; C a2 .avai represents the capacity of the a2th 5G link; The interval injection constraints include: in, Indicates that at time τ Injection of TSN links, represents f i The jth packet on f i represents the ith scheduled data stream; F represents the set of scheduled data streams; |T| represents the total number of time slots; Indicates that at time τ Injection status of 5G links; Indicates that at time τ Transmitted in 5G network; Indicates that at time τ Transmitted in the TSN network.
2. The method for deterministic transmission scheduling of TSN and 5G converged network according to claim 1, characterized in that: The TSN-5G fusion network includes: TSN equipment, 5G equipment and connection equipment.
3. The method for deterministic transmission scheduling of TSN and 5G converged network according to claim 1, characterized in that: The transmission delay constraint includes: f i .latency>D tran +D prop +D proc +D queue ; Among them, f i .latency means f i Acceptable delay, f i represents the i-th scheduled data flow; D tran Indicates transmission delay; D prop represents the link propagation delay; D proc Indicates CPU processing delay; D queue represents the queuing delay; F represents the set of scheduled data flows.
4. The method for deterministic transmission scheduling of TSN and 5G converged network according to claim 1, characterized in that: The discreteness constraints of the frequency resource allocation include: in, represents f i The final frequency resource obtained, f i represents the i-th scheduled data stream; B req Indicates the spectrum required by wireless devices; B sub Indicates the current subcarrier spacing; F indicates the set of scheduled data streams.
5. The method for deterministic transmission scheduling of TSN and 5G converged network according to claim 1, characterized in that: The reliability constraints include: Among them, f i .pnum means f i The total number of packets, f i represents the i-th scheduled data stream; represents f i The jth data packet on ; express The arrival situation of the destination.
6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the TSN and 5G converged network deterministic transmission scheduling method described in any one of claims 1-5.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the TSN and 5G converged network deterministic transmission scheduling method described in any one of claims 1 to 5 are implemented.
8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the TSN and 5G converged network deterministic transmission scheduling method described in any one of claims 1 to 5 are implemented.
