A method, device and computer-readable storage medium for determining a route

Abstract

The present application discloses a method for determining a route, comprising: determining network parameters for each network layer in a flexible network architecture; wherein the network parameters are used to characterize the quality of the network layer; determining a local incentive for each network layer based on the network parameters and an incentive factor; and determining a target route from an initial node to a destination node based on the local incentive and an AI engine for each network layer. The present application also discloses a route determination device and a computer-readable storage medium.

Description

Technical Field

[0001] The present application relates to information determination technology in the field of communications, and in particular to a route determination method, device, and computer-readable storage medium. Background Art

[0002] Compared to traditional networks, flexible networks enable smooth expansion and contraction of end-to-end network capabilities across any link through elastic scaling of network node processing power and transmission bandwidth. This adapts to the agile service demands of future sixth-generation mobile networks (6G), supports deep cloud-network integration, and enables "cloud on the network, network in the cloud." Flexible networks primarily consist of elastic optical networks, software-defined networks (SDN), and network function virtualization (NFV). Currently, related technical solutions primarily employ methods such as reinforcement learning to optimize quality of service (QoS) or routing within a small number of network elements in SDN networks. However, these optimizations do not address large-scale flexible network applications and cannot achieve cross-layer network optimization. Summary of the Invention

[0003] To solve the above technical problems, the embodiments of the present application hope to provide a routing determination method, device and computer-readable storage medium, which solves the problem that the routing optimization solutions in the related art cannot achieve the optimization of large-scale flexible networks and cross-layer networks.

[0004] The technical solution of this application is achieved as follows:

[0005] A route determination method, the method comprising:

[0006] Determining network parameters of each network layer in the network architecture of the flexible network; wherein the network parameters are used to characterize the quality of the network layer;

[0007] determining a local excitation for each network layer based on the network parameters and the excitation factor of each network layer;

[0008] Based on the local excitation of each network layer and the AI ​​engine of each network layer, a target route from an initial node to a destination node is determined.

[0009] In the above solution, the network parameters of each network layer in the network architecture of the flexible network are determined, including:

[0010] Obtaining the delay, packet loss rate and link factor of the optical link of the optical transmission network layer in the network architecture of the flexible network;

[0011] Obtaining the link delay, link packet loss rate, and link delay jitter of the IP network layer in the network architecture of the flexible network;

[0012] Obtaining a link delay and a link bit error rate of a wireless access network layer in a network architecture of the flexible network;

[0013] Obtain the link delay and link bit error rate of the core network layer in the network architecture of the flexible network.

[0014] In the above solution, determining the local excitation of each network layer based on the network parameters and excitation factors of each network layer includes:

[0015] Obtaining a first excitation factor of the optical transmission network layer, a second excitation factor of the IP network layer, a third excitation factor of the radio access network layer, and obtaining a fourth excitation factor of the core network layer;

[0016] Calculating a first local excitation of the optical transmission network layer based on a delay of an optical link of the optical transmission network layer, a packet loss rate of the optical link, a link factor, and the first excitation factor;

[0017] Calculating a second local excitation of the IP network layer based on the link delay, the link packet loss rate, the link delay jitter and the second excitation factor of the IP network layer;

[0018] Calculating a third local excitation of the radio access network network layer based on the link delay, the link bit error rate and the third excitation factor of the radio access network network layer;

[0019] Based on the link delay of the core network network layer, the link bit error rate and the fourth excitation factor, the fourth local excitation of the core network network layer is calculated; wherein the local excitation includes the first local excitation, the second local excitation, the third local excitation and the fourth local excitation.

[0020] In the above solution, obtaining the link factor of the optical transmission network layer includes:

[0021] Determining a current modulation mode, a current transmission distance, and a number of nodes of the optical transmission network layer;

[0022] A link factor of the optical transmission network layer is determined based on a current modulation mode of the optical transmission network layer, a current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer.

[0023] In the above solution, determining the link factor of the optical transmission network layer based on the current modulation mode, the current transmission distance, and the number of nodes includes:

[0024] Determining the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path;

[0025] The link factor is determined based on the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes of the current optical path.

[0026] In the above solution, the obtaining of the first excitation factor of the optical transmission network layer, the second excitation factor of the IP network layer, the third excitation factor of the radio access network layer, and the fourth excitation factor of the core network layer includes:

[0027] respectively determining the system capacity and target capacity of the optical transmission network layer, the system capacity and target capacity of the IP network layer, the system capacity and target capacity of the radio access network layer, and the system capacity and target capacity of the core network layer;

[0028] Determining, based on the system capacity and the target capacity, a load factor of the optical transmission network layer, a load factor of the IP network layer, a load factor of the radio access network layer, and a load factor of the core network layer;

[0029] Determine global incentives;

[0030] For each network layer, the first excitation factor, the second excitation factor, the third excitation factor, and the fourth excitation factor are determined based on the load factor and the global excitation.

[0031] In the above solution, determining the global incentive includes:

[0032] Obtain end-to-end link delay and end-to-end packet loss rate;

[0033] determining a gateway load factor between the network layers based on a system capacity of each network layer and the target capacity;

[0034] The global incentive is determined based on the end-to-end link delay, the end-to-end packet loss rate, and the gateway load factor.

[0035] In the above solution, the determination of the target route from the initial node to the destination node based on the local excitation of each network layer and the AI ​​engine of each network layer includes:

[0036] Using a deep learning algorithm, an AI engine of each network layer determines a to-be-screened route from the initial node to the destination node based on the first local stimulus, the second local stimulus, the third local stimulus, and the fourth local stimulus;

[0037] The target route is determined from the routes to be screened based on a size relationship between the packet delay of the routes to be screened and a target delay threshold.

[0038] A route determination device, comprising: a processor, a memory, and a communication bus;

[0039] The communication bus is used to realize the communication connection between the processor and the memory;

[0040] The processor is used to execute the route determination program in the memory to implement the steps of the route determination method described above.

[0041] A computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the above-mentioned route determination method.

[0042] The route determination method, device, and computer-readable storage medium provided in the embodiments of the present application can determine the network parameters of each network layer in the network architecture of the flexible network for characterizing the quality of the network layer, determine the local excitation of each network layer based on the network parameters and excitation factors of each network layer, and determine the target route from the initial node to the destination node based on the local excitation of each network layer and the AI ​​engine of each network layer. In this way, the network parameters of each network layer of the flexible network are used when determining the target route, which solves the problem that the route optimization scheme in the related art cannot realize the optimization of large-scale flexible networks and cross-layer networks, and realizes the route optimization of large-scale flexible networks and cross-layer networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a route determination method provided in an embodiment of the present application;

[0044] Figure 2 A flowchart of another route determination method provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of a distributed AI architecture of a flexible network corresponding to a routing determination method provided in an embodiment of the present application;

[0046] Figure 4 A schematic diagram of the principles of a DQN algorithm used in a routing determination method provided in another embodiment of the present application;

[0047] Figure 5 A flowchart of another route determination method provided in an embodiment of the present application;

[0048] Figure 6A schematic diagram of the relationship between the modulation format level and transmission distance corresponding to the optical transmission network layer of a route determination method provided in an embodiment of the present application;

[0049] Figure 7 A schematic diagram of the structure of a route determination device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0051] It should be understood that the “embodiments of the present application” or “the aforementioned embodiments” mentioned throughout the specification mean that the specific features, structures or characteristics related to the embodiments are included in at least one embodiment of the present application. Therefore, “in the embodiments of the present application” or “in the aforementioned embodiments” appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. In the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments.

[0052] Unless otherwise specified, when an electronic device performs any step in the embodiments of the present application, the processor of the electronic device may perform the step. It is also worth noting that the embodiments of the present application do not limit the order in which the electronic device performs the following steps. In addition, the methods used to process data in different embodiments may be the same method or different methods. It should also be noted that any step in the embodiments of the present application can be independently executed by the electronic device, that is, when the electronic device performs any step in the following embodiments, it can be independent of the execution of other steps.

[0053] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0054] The present invention provides a method for determining a route, which can be applied to a route determining device. Figure 1 As shown, the method includes the following steps:

[0055] Step 101: Determine the network parameters of each network layer in the network architecture of the flexible network.

[0056] Among them, network parameters are used to characterize the quality of the network layer.

[0057] In an embodiment of the present application, the network architecture of the flexible network can be a distributed artificial intelligence (AI) architecture; wherein, the elastic optical network in the flexible network transmits bandwidth through variable granularity, and for the same end-to-end optical path, different bit rates can be used according to the actual connection environment and connection requirements to achieve high spectral efficiency. SDN and NFV define and control the network in the form of software programming, and have the characteristics of separation of control plane and forwarding plane and open programmability, which can use network resources more flexibly and efficiently. The above technologies are centrally controlled through a distributed AI architecture to form a flexible network with elastic scalability of end-to-end link capacity.

[0058] In an embodiment of the present application, the network parameters of the network layer may refer to parameters that can ensure the quality of the network; in a feasible implementation method, the network parameters may include at least one of the following: link delay, link bit error rate, link factor, link packet loss rate and link delay jitter and other parameters.

[0059] Step 102: Determine the local excitation of each network layer based on the network parameters and excitation factors of each network layer.

[0060] In an embodiment of the present application, after obtaining the network parameters of each network layer of the flexible network, the routing determination device can determine the excitation factor of each network layer of the flexible network, and then calculate the network parameters and excitation factor of each network layer to obtain local excitation of each network layer.

[0061] Step 103: Based on the local excitation of each network layer and the AI ​​engine of each network layer, determine the target route from the initial node to the destination node.

[0062] In an embodiment of the present application, the route determination device can obtain a target route from the initial node to the destination node of the flexible network by performing calculations based on the local excitations of each network layer through an AI engine set in each network layer. Among them, the calculation based on the local excitations of each network layer by the AI ​​engine can be implemented by using a deep learning algorithm. It should be noted that in this embodiment, because the network parameters of each network layer are used when determining the local excitations of each network layer of the flexible network, the target route determined by using the local excitations of each network layer fully takes into account the conditions of each network layer; therefore, the optimization of routes and QoS optimization across network layers can be better achieved.

[0063] The route determination method provided in the embodiment of the present application can determine the network parameters of each network layer in the network architecture of the flexible network for characterizing the quality of the network layer, determine the local excitation of each network layer based on the network parameters and excitation factors of each network layer, and determine the target route from the initial node to the destination node based on the local excitation of each network layer and the AI ​​engine of each network layer. In this way, the network parameters of each network layer of the flexible network are used when determining the target route, which solves the problem that the route optimization scheme in the related art cannot realize the optimization of large-scale flexible networks and cross-layer networks, and realizes the route optimization of large-scale flexible networks and cross-layer networks.

[0064] Based on the above embodiments, the present application provides a method for determining a route. Figure 2 As shown, the method includes the following steps:

[0065] Step 201: A route determination device obtains a time delay, a packet loss rate, and a link factor of an optical link of an optical transmission network layer in a network architecture of a flexible network.

[0066] In the embodiment of the present application, the routing determination method can be applied to Figure 3In the distributed AI architecture of the flexible network shown, the distributed AI architecture of the flexible network may include an optical transmission network layer, an IP network layer, a radio access network (Radio Access Network, RAN) network layer and a core network (Core Network, CN) network layer. Among them, the optical transmission network layer mainly includes variable bandwidth-wavelength selective switch (Bandwidth Variable-Wavelength Selective Switch, BV-WSS) and variable bandwidth transmitter (Bandwidth Variable Transponder, BVT); the IP network layer mainly includes switches and routers; the wireless access network network layer mainly includes terminals and base stations; the core network network layer mainly includes various network elements; in a feasible implementation method, if it is a 4G core network, the core network network layer may include mobility management entity (Mobility Management Entity, MME), serving gateway (Serving Gate Way, SGW), public data network gateway (Public Data Network, Gate Way, PGW), home user service (Home Subscriber Server, HSS), policy and charging rules function (Policy and Charging Rules Function, PCRF); if it is a 5G core network, the core network network layer may include authentication management function (Authentication Management Function, AMF), service management function (Service Management Function, SMF), user plane function (User Plane Function, UPF), unified data management function (Unified Data Management, UDM), Policy Control Function (PCF), Network Repository Function (NRF), Network Slice Selection Function (NSSF). Figure 3 As shown in the figure, each network layer of the distributed AI architecture of the flexible network has its own AI engine deployed, and routing optimization and QoS optimization have been performed.

[0067] The route determination device can directly obtain the delay of each optical link and the bit error rate of each optical link in the optical transmission network layer.

[0068] Step 202: The route determination device obtains the link delay, link packet loss rate, and link delay jitter of the Internet Protocol (IP) network layer in the network architecture of the flexible network.

[0069] Among them, the routing determination device can directly obtain the delay of each link in the IP network layer, the packet loss rate of each link, and the delay jitter of each link. It should be noted that because the IP layer is widely distributed and numerous, the IP layer can be further subdivided into subgroups based on business volume and region; at the same time, the routing determination device can set up an AI engine at the group level (i.e. for each network layer). An AI engine at the Operation Support Systems (OSS) level is set up to calculate the global optimization of the entire network.

[0070] Step 203: The route determination device obtains the link delay and link bit error rate of the radio access network layer in the network architecture of the flexible network.

[0071] The route determination device may directly obtain the time delay and bit error rate of each link in the network layer of the radio access network.

[0072] Step 204: The route determination device obtains the link delay and link bit error rate of the core network layer in the network architecture of the flexible network.

[0073] The route determination device can directly obtain the delay and bit error rate of each link in the core network layer.

[0074] It should be noted that the link factor of each network layer in the distributed AI architecture of the flexible network can be composed of the load factor of each network layer and the global factor.

[0075] Step 205: The route determination device obtains a first excitation factor of the optical transmission network layer, a second excitation factor of the IP network layer, a third excitation factor of the wireless access network layer, and a fourth excitation factor of the core network layer.

[0076] In an embodiment of the present application, the excitation factors of each network layer, i.e., the first excitation factor, the second excitation factor, the third excitation factor, and the fourth excitation factor, can be obtained by performing calculations based on the same parameters of different network layers according to the same method; that is, the first excitation factor, the second excitation factor, the third excitation factor, and the fourth excitation factor can be calculated using the same calculation formula.

[0077] Step 206: The route determination device calculates a first local excitation of the optical transmission network layer based on the delay of the optical link of the optical transmission network layer, the packet loss rate of the optical link, the link factor, and the first excitation factor.

[0078] In the embodiment of the present application, the delay of the optical link of the optical transmission network layer, the packet loss rate of the optical link, the link factor and the first excitation factor can be calculated according to a certain formula to obtain the first local excitation; in a feasible implementation method, the first local excitation can be obtained by using the formula r OP =1 / Delay+1 / Loss+1 / L link +r g_op Calculated; where r OP represents the first local excitation, Delay represents the link delay of each optical link in the optical transmission network layer, Loss represents the packet loss rate of each optical link in the optical transmission network layer, L link represents the link factor of the optical transmission network layer, r g_op Represents the first excitation factor; r in the above formula OP , Delay, Loss and L link It can exist in the form of a vector. And, for the optical transmission network layer, the state space of the corresponding deep learning network (Deep Q-learning Network, DQN) is [C, Delay, Loss, L link ], where C represents the capacity of each optical network element. It should be noted that the optimization function of the optical transmission network layer is to maximize the total network throughput under the constraints of delay and bit error rate.

[0079] Step 207: The route determination device calculates a second local excitation of the IP network layer based on the link delay, link packet loss rate, link delay jitter, and the second excitation factor of the IP network layer.

[0080] In the embodiment of the present application, the link delay, link packet loss rate, link delay jitter of the IP network layer and the second excitation factor can be calculated according to a certain formula to obtain the second local excitation; in a feasible implementation method, the second local excitation can be obtained by using the formula r IP =1 / Delay+1 / Loss+1 / Jitter+r g_ip Calculated; where r IP represents the second local excitation, Delay represents the delay of each link in the IP network layer, Loss represents the packet loss rate of each link in the IP network layer, Jitter represents the delay jitter of the IP network layer, and r g_ip Represents the second excitation factor; r in the above formula IP, Delay, Loss, and Jitter can be expressed as vectors. Furthermore, the corresponding DQN state space for the IP network layer is [C, Delay, Loss, Jitter], where C represents the capacity of each switch. It should be noted that the optimization function for the IP network layer is to maximize network throughput under the constraints of delay and bit error rate.

[0081] Step 208: The route determination device calculates a third local excitation of the radio access network network layer based on the link delay, the link bit error rate, and the third excitation factor of the radio access network network layer.

[0082] In the embodiment of the present application, the link delay of the wireless access network layer, the link bit error rate and the third excitation factor can be calculated according to a certain formula to obtain the third local excitation; in a feasible implementation method, the third local excitation can be calculated using the formula r RAN =exp(-Delay)+exp(-Ber)+r g_ran Calculated; where r RAN represents the third local excitation, Delay represents the delay of each link in the wireless access network layer, Ber represents the bit error rate of each link in the wireless access network layer, r g_ran Represents the third excitation factor; r in the above formula RAN , Delay and Ber can exist in the form of vectors. And, for the wireless access network layer, the corresponding DQN state space is [C, RSRP uplink ,RSRQ uplink , Delay, Ber] and [C downlink , RSRP downlink ,RSRQ downlink , Delay, Ber], where C represents the current capacity of each base station, RSRP uplink Indicates the pilot received signal level and RSRP of each link uplink downlink Indicates the pilot received signal level and RSRQ of each link downlink uplink Indicates the pilot received signal quality and RSRQ of each link uplink downlink Represents the quality of the pilot received signal on each downlink link. It should be noted that the optimization function at the radio access network layer is to maximize network throughput under latency and bit error rate constraints. In other words, the Q function maximizes network throughput when selecting the optimal base station for each terminal. It should be noted that the formula used to calculate the third local excitation uses an exponential operation to strengthen the excitation constraints and avoid negative impacts caused by resource constraints and link instability at the radio access network layer.

[0083] Step 209: The route determination device calculates a fourth local excitation of the core network layer based on the link delay, the link bit error rate, and the fourth excitation factor of the core network layer.

[0084] The local excitation includes a first local excitation, a second local excitation, a third local excitation and a fourth local excitation.

[0085] In the embodiment of the present application, the link delay of the core network layer, the link bit error rate and the fourth excitation factor can be calculated according to a certain formula to obtain the fourth local excitation; in a feasible implementation method, the fourth local excitation can be obtained by using the formula r CN =1 / Delay+1 / Ber+r g_cn Calculated; where r CN represents the fourth local excitation, Delay represents the delay of each link in the core network layer, Ber represents the bit error rate of each link in the core network layer, r g_cn Represents the fourth excitation factor; r in the above formula CN , Delay, and Ber can exist in vector form. Furthermore, for the core network layer, the corresponding DQN state space is [C, Delay, Loss], where C represents the capacity of each core network element. It should be noted that the optimization function for the core network layer is to maximize network throughput under the constraints of delay and bit error rate; that is, when the Q function selects the optimal node for each link, the core network's network throughput is maximized.

[0086] Step 210: The route determination device adopts a deep learning algorithm to determine the route to be screened from the initial node to the destination node based on the first local excitation, the second local excitation, the third local excitation, and the fourth local excitation through the AI ​​engine of each network layer.

[0087] In the embodiment of the present application, the AI ​​engine of each network layer is composed of DQN. The main principle can be: DQN is a type of deep reinforcement learning (DRL), which is a combination of deep learning and Q learning. Figure 4As shown in the figure, DQN is mostly used in scenarios that require interaction with the environment. That is, given a state of the environment (State), the program selects a corresponding behavior (Action) according to a certain strategy (Policy), and after executing this Action, the environment will change, that is, the state will be converted to a new state S'. After each action is executed, the program will receive an incentive value (Reward), and the program will adjust its strategy according to the size of the incentive value so that after all steps are executed, that is, when the state reaches the terminal state (Terminal), the sum of the rewards obtained is maximized.

[0088] Among them, Deep Neural Networks (DNN) can provide a batch of quadruple (s, a, r, s'), where s' is the next state after s executes a. A batch of such quadruple needs to be cached in the experience pool for training. Since each time an action is performed, the next state can be transferred and a reward is obtained, so we can obtain such a quadruple after each action is performed, or we can put this quadruple directly into the experience pool. The dual DNN method is used to randomly extract quadruple data for parameter update, mainly to prevent overfitting during quadruple training. The DNN objective function iteration method is: Where α represents the normalized step size, and the DQN objective function can be defined as: In the above formula, r represents the incentive; in DQN, a dual DNN approach is used, and after each τ step, the objective function of the DNN is

[0089] Step 211: The route determination device determines a target route from the routes to be screened based on a relationship between the packet delay of the routes to be screened and a target delay threshold.

[0090] In an embodiment of the present application, after determining that multiple routes to be screened are obtained, the packet delay of the optimal route among the routes to be screened can be calculated first. If the packet delay of the optimal route is outside the target delay threshold range, we consider that the route is not suitable for the route from the initial node to the destination node; then the packet delay of the second-best route among the routes to be screened can be calculated. If the packet delay of the second-best route is outside the target delay threshold range, we still consider that the second-best route is not suitable for the route from the initial node to the destination node; then the packet delay of the Nth-best route among the routes to be screened can be calculated. If the packet delay of the Nth-best route is within the target delay threshold range, the Nth-best route is considered to be the target route required in the end. Of course, if the packet delay of the optimal route determined at the beginning is within the target delay threshold range, the optimal route is considered to be the target route. Among them, the routes to be screened can be considered as a path set; when the path calculated by the AI ​​engine exceeds the path set, by setting σ = -|V|, where V will be determined according to the size of the invalid path set and the subnet size. Moreover, the target delay threshold can be determined based on historical usage data. It should be noted that adopting this routing screening solution can further prevent the situation where the network topology self-loops and cannot reach the destination node.

[0091] It should be noted that, for explanations of the same or corresponding steps in the embodiment of the present application and other embodiments, reference can be made to the descriptions in other embodiments and will not be repeated here.

[0092] The route determination method provided in the embodiment of the present application uses the network parameters of each network layer of the flexible network when determining the target route, which solves the problem that the route optimization scheme in the related art cannot achieve the optimization of large-scale flexible networks and cross-layer networks, and realizes the route optimization of large-scale flexible networks and cross-layer networks.

[0093] Based on the above embodiments, the embodiments of the present application provide a method for determining a route, referring to Figure 5 As shown, the method includes the following steps:

[0094] Step 301: A route determination device obtains a time delay and a packet loss rate of an optical link of an optical transmission network layer in a network architecture of a flexible network.

[0095] Step 302: The route determination device determines the current modulation mode, current transmission distance, and number of nodes of the optical transmission network layer.

[0096] Step 303: The route determination device determines the link factor of the optical transmission network layer based on the current modulation mode of the optical transmission network layer, the current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer.

[0097] Step 303 can be implemented by the following steps:

[0098] Step 303a: The route determination device determines the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes of the current optical path.

[0099] In the embodiment of the present application, the modulation mode and transmission distance of the optical transmission network layer are as follows: Figure 6 The corresponding relationship is shown. Each optical link in the optical transmission network layer, while meeting service QoS requirements, uses the shortest distance and the corresponding highest modulation method, utilizing the fewest nodes and minimizing network node energy consumption, achieving the highest transmission efficiency. Therefore, the link factor for each optical link can be calculated based on the transmission distance, the actual modulation method used, and the number of nodes. At this point, the spectral efficiency corresponding to the current modulation method of the optical transmission network layer, the spectral efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path can be directly obtained.

[0100] Step 303b: The route determination device determines a link factor based on the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path.

[0101] In an embodiment of the present application, the link factor of the optical transmission network layer can be obtained by calculating the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path through a certain formula; wherein the spectrum efficiency corresponding to the current transmission distance can refer to the maximum spectrum efficiency that can be used for the current transmission distance, and the number of nodes in the current optical path refers to the number of nodes used in the current optical path; in this case, the above formula can be: L link =Mod select / Mod max +1 / N node , among which, Mod select Indicates the spectrum efficiency corresponding to the current modulation mode, Mod max Indicates the maximum spectrum efficiency that can be used for the current transmission distance, N node Indicates the number of nodes used by the current lightpath.

[0102] Step 304: The route determination device obtains the link delay, link packet loss rate, and link delay jitter of the IP network layer in the network architecture of the flexible network.

[0103] Step 305: The route determination device obtains the link delay and link bit error rate of the radio access network layer in the network architecture of the flexible network.

[0104] Step 306: The route determination device obtains the link delay and link bit error rate of the core network layer in the network architecture of the flexible network.

[0105] Step 307: The route determination device determines the system capacity and target capacity of the optical transmission network layer, the system capacity and target capacity of the IP network layer, the system capacity and target capacity of the wireless access network layer, and the system capacity and target capacity of the core network layer.

[0106] In an embodiment of the present application, the routing determination device can directly obtain the system capacity and target capacity of the optical transmission network layer, directly obtain the system capacity and target capacity of the IP network layer, directly obtain the system capacity and target capacity of the wireless access network layer, and directly obtain the system capacity and target capacity of the core network layer.

[0107] Step 308: The route determination device determines the load factor of the optical transmission network layer, the load factor of the IP network layer, the load factor of the wireless access network layer, and the load factor of the core network layer based on the system capacity and the target capacity.

[0108] In the embodiment of the present application, the load factor of the optical transmission network layer, the load factor of the IP network layer, the load factor of the wireless access network layer, and the load factor of the core network layer can be determined using the same method based on their respective system capacities and target capacities; that is, for the optical transmission network layer, the IP network layer, the wireless access network layer, and the core network layer, the load factor can be obtained by performing a certain operation on the system capacity and the target capacity. In a feasible implementation, the wireless access network layer is used as an example for explanation: assuming that the maximum capacity of the subnet of the wireless access network layer is C ran , when the system capacity of the wireless access network layer is greater than x*C ran (x∈[0,1]), Ф ran_load =C0-C ran ; Where C0 represents the system capacity of the wireless access network layer, Ф ran_load Indicates the load factor of the wireless access network layer. x is a constant between 0 and 1, which can be determined based on specific application scenarios and historical data. When the system capacity is less than or equal to x*C ran Time φ ran_load =0; It should be noted that the calculation method of the load factor of other network layers is the same as the calculation method of the load factor of the wireless access network layer, and the specific calculation process is not repeated here.

[0109] Step 309: The routing determination device determines a global stimulus.

[0110] Step 309 can be implemented by the following steps:

[0111] It should be noted that for the wireless access network layer, global excitation can be used g Represents, and the global incentive r gThis can be achieved by taking the following steps 309a-309c:

[0112] Step 309a: The route determination device obtains the end-to-end link delay and the end-to-end packet loss rate.

[0113] Step 309b: The route determination device determines the gateway load factor between network layers based on the system capacity and target capacity of each network layer.

[0114] In the embodiment of the present application, the gateway link load factor can be calculated by performing certain operations on the system capacity and the target capacity. In a feasible implementation, the gateway load factor ξ can be calculated as follows: gateway_load :Assume that the maximum capacity of the subnet of the wireless access network layer is C gateway , when the system capacity of the wireless access network layer is greater than x*C gateway (x∈[0,1]), ξ gateway_load =C gateway -C0, where C0 represents the current capacity and x is a constant between 0 and 1, which can be determined based on the specific application scenario and historical data; when the system capacity is less than or equal to x*C gateway When gateway_load =0.

[0115] Step 309c: The route determination device determines a global incentive based on the end-to-end link delay, the end-to-end packet loss rate, and the gateway load factor.

[0116] In the embodiment of the present application, r can be calculated using the following formula: g :r g =1 / Delay+1 / Loss+ξ gateway_load , where Delay represents the end-to-end link delay and Loss represents the end-to-end packet loss rate.

[0117] It should be noted that the global incentive for the optical transport network layer, IP network layer and core network layer can be represented by σ; wherein σ can be determined based on additional factors such as the management and optimization of the overall flexible network.

[0118] Step 310: The route determination device determines, for each network layer, a first incentive factor, a second incentive factor, a third incentive factor, and a fourth incentive factor based on the load factor and the global incentive.

[0119] In the embodiment of the present application, the first excitation factor, the second excitation factor, the third excitation factor and the fourth excitation factor are all obtained by performing certain operations on the load factor and the global excitation; in a feasible implementation, the first excitation factor can be calculated using the following formula: g_op =Фop_load +σ, where r g_op represents the first excitation factor, Φ op_load represents the load factor of the optical transmission network layer, σ represents the global factor; the second excitation factor can be calculated using the following formula: g_ip =Ф ip_load +σ, where r g_ip represents the second excitation factor, Φ ip_load Indicates the load factor of the IP network layer; the third excitation factor can be calculated using the following formula: g_ran =Ф ran_load +r g , where r g_ran Represents the third excitation factor, Φ ran_load represents the load factor of the wireless access network layer, r g Represents global incentive; the fourth incentive factor can be calculated using the following formula: g_cn =Ф cn_load +σ, where r g_cn Represents the fourth excitation factor, Φ cn_load Indicates the load factor of the core network layer.

[0120] Step 311: The route determination device calculates a first local excitation of the optical transmission network layer based on the delay of the optical link of the optical transmission network layer, the packet loss rate of the optical link, the link factor, and the first excitation factor.

[0121] Step 312: The route determination device calculates a second local excitation of the IP network layer based on the link delay, link packet loss rate, link delay jitter, and the second excitation factor of the IP network layer.

[0122] Step 313: The route determination device calculates a third local excitation of the radio access network network layer based on the link delay, the link bit error rate, and the third excitation factor of the radio access network network layer.

[0123] Step 314: The route determination device calculates a fourth local excitation of the core network layer based on the link delay, the link bit error rate, and the fourth excitation factor of the core network layer.

[0124] The local excitation includes a first local excitation, a second local excitation, a third local excitation and a fourth local excitation.

[0125] Step 315: The route determination device adopts a deep learning algorithm to determine the route to be screened from the initial node to the destination node based on the first local excitation, the second local excitation, the third local excitation, and the fourth local excitation through the AI ​​engine of each network layer.

[0126] Step 316: The route determination device determines a target route from the routes to be screened based on the relationship between the packet delay of the routes to be screened and the target delay threshold.

[0127] It should be noted that, for explanations of the same or corresponding steps in the embodiment of the present application and other embodiments, reference can be made to the descriptions in other embodiments and will not be repeated here.

[0128] The route determination method provided in the embodiment of the present application uses the network parameters of each network layer of the flexible network when determining the target route, which solves the problem that the route optimization scheme in the related art cannot achieve the optimization of large-scale flexible networks and cross-layer networks, and realizes the route optimization of large-scale flexible networks and cross-layer networks.

[0129] Based on the above embodiments, the embodiments of the present application provide a route determination device, which can be applied to Figures 1-2 In the route determination method provided in the embodiment corresponding to 5, reference is made to Figure 7 As shown, the route determination device 4 may include: a processor 41, a memory 42 and a communication bus 43, wherein:

[0130] The communication bus 43 is used to realize the communication connection between the processor 41 and the memory 42;

[0131] The processor 41 is configured to execute the routing determination program in the memory 42 to implement the following steps:

[0132] Determine the network parameters of each network layer in the network architecture of the flexible network; wherein the network parameters are used to characterize the quality of the network layer;

[0133] Determine the local excitation of each network layer based on the network parameters and excitation factors of each network layer;

[0134] Based on the local incentives of each network layer and the AI ​​engine of each network layer, the target route from the initial node to the destination node is determined.

[0135] In other embodiments of the present application, the processor 41 is configured to execute the routing determination program in the memory 42 to determine the network parameters of each network layer in the network architecture of the flexible network, so as to implement the following steps:

[0136] Obtain the delay, packet loss rate, and link factor of the optical link in the optical transmission network layer of the flexible network architecture;

[0137] Obtain the link delay, link packet loss rate, and link delay jitter of the IP network layer in the network architecture of the flexible network;

[0138] Obtain the link delay and link bit error rate of the wireless access network layer in the network architecture of the flexible network;

[0139] Obtain the link delay and link bit error rate of the core network layer in the network architecture of the flexible network.

[0140] In other embodiments of the present application, the processor 41 is configured to execute the routing determination program in the memory 42 based on the network parameters and incentive factors of each network layer, and determine the local incentive of each network layer to implement the following steps:

[0141] Obtaining a first excitation factor of the optical transmission network layer, a second excitation factor of the IP network layer, a third excitation factor of the wireless access network layer, and a fourth excitation factor of the core network layer;

[0142] A first local excitation of the optical transmission network layer based on the delay of the optical link of the optical transmission network layer, the bit error rate of the optical link, the link factor and the first excitation factor;

[0143] Calculate the second local excitation of the IP network layer based on the link delay, link packet loss rate, link delay jitter and the second excitation factor of the IP network layer;

[0144] Calculating a third local excitation of the wireless access network network layer based on the link delay, the link bit error rate and the third excitation factor of the wireless access network network layer;

[0145] Calculate the fourth local excitation of the core network layer based on the link delay, link bit error rate and fourth excitation factor of the core network layer;

[0146] The local excitation includes a first local excitation, a second local excitation, a third local excitation and a fourth local excitation.

[0147] In other embodiments of the present application, the processor 41 is configured to execute the route determination program in the memory 42 to obtain the link factor of the optical transport network layer, so as to implement the following steps:

[0148] Determine the current modulation mode, current transmission distance and number of nodes of the optical transmission network layer;

[0149] A link factor of the optical transmission network layer is determined based on a current modulation mode of the optical transmission network layer, a current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer.

[0150] In other embodiments of the present application, the processor 41 is configured to execute the routing determination program in the memory 42, based on the current modulation mode of the optical transmission network layer, the current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer, to determine the link factor of the optical transmission network layer, so as to implement the following steps:

[0151] Determine the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path;

[0152] The link factor is determined based on the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path.

[0153] In other embodiments of the present application, the first processor 51 is configured to execute the route determination program in the first memory 52 to obtain the first excitation factor of the optical transmission network layer, the second excitation factor of the IP network layer, the third excitation factor of the radio access network layer, and the fourth excitation factor of the core network layer, and may further implement the following steps:

[0154] Determine the system capacity and target capacity of the optical transmission network layer, the system capacity and target capacity of the IP network layer, the system capacity and target capacity of the wireless access network layer, and the system capacity and target capacity of the core network layer respectively;

[0155] Determine the load factors of the optical transmission network layer, the IP network layer, the wireless access network layer, and the core network layer based on the system capacity and the target capacity respectively;

[0156] Determine global incentives;

[0157] For each network layer, a first excitation factor, a second excitation factor, a third excitation factor, and a fourth excitation factor are determined based on the load factor and the global excitation.

[0158] In other embodiments of the present application, the processor 41 is configured to execute the global stimulus determination of the routing determination program in the memory 42, and may further implement the following steps:

[0159] Obtain end-to-end link delay and end-to-end packet loss rate;

[0160] Determine the gateway load factor between network layers based on the system capacity and target capacity of each network layer;

[0161] The global incentive is determined based on the end-to-end link delay, end-to-end packet loss rate, and gateway load factor.

[0162] In other embodiments of the present application, the processor 41 is configured to execute the routing determination program in the memory 42 based on the local excitation of each network layer and the AI ​​engine of each network layer to determine the target route from the initial node to the destination node, and may further implement the following steps:

[0163] Using a deep learning algorithm, the AI ​​engine of each network layer determines the route to be screened from the initial node to the destination node based on the first local incentive, the second local incentive, the third local incentive, and the fourth local incentive.

[0164] Based on the relationship between the packet delay of the routes to be screened and the target delay threshold, a target route is determined from the routes to be screened.

[0165] It should be noted that the specific description of the steps performed by the processor can be referred to Figures 1-2 The route determination method provided in the embodiment corresponding to 5 is not repeated here.

[0166] The routing determination device provided in the embodiment of the present application uses the network parameters of each network layer of the flexible network when determining the target route, solving the problem that the routing optimization scheme in the related art cannot achieve the optimization of large-scale flexible networks and cross-layer networks, and realizes the routing optimization of large-scale flexible networks and cross-layer networks.

[0167] Based on the above embodiments, the embodiments of the present application provide a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement Figures 1-2 The steps of the route determination method provided in the embodiment corresponding to 5.

[0168] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0169] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0170] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0171] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0172] The above description is merely a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application.

Claims

1. A method for determining a route, characterized in that: The method comprises: Determining network parameters of each network layer in a network architecture of a flexible network; wherein the flexible network includes an optical transmission network layer, an Internet Protocol (IP) network layer, a core network layer, and a wireless access network layer, and the network parameters are used to characterize the quality of the network layers; Based on the network parameters and excitation factors of each network layer, a local excitation of each network layer is determined; wherein the excitation factor of each network layer is determined based on the load factor of each network layer and the global excitation; the load factor of each network layer is determined based on the determined system capacity and target capacity of each network layer respectively; the global excitation of the optical transmission network layer, the IP network layer and the core network layer is determined based on the management factors and optimization factors of the overall flexible network; the global excitation of the wireless access network layer is determined based on the obtained end-to-end link delay, end-to-end packet loss rate and gateway load factor; the gateway load factor is determined based on the system capacity and the target capacity; Based on the local excitation of each network layer and the artificial intelligence AI engine of each network layer, a target route from an initial node to a destination node is determined.

2. The method according to claim 1, characterized in that The network parameters of each network layer in the network architecture of the flexible network are determined, including: Obtaining the delay, packet loss rate and link factor of the optical link of the optical transmission network layer in the network architecture of the flexible network; Obtaining the link delay, link packet loss rate, and link delay jitter of the IP network layer in the network architecture of the flexible network; Obtaining a link delay and a link bit error rate of a wireless access network layer in a network architecture of the flexible network; Obtain the link delay and link bit error rate of the core network layer in the network architecture of the flexible network.

3. The method according to claim 2, characterized in that The determining of the local excitation of each network layer based on the network parameters and the excitation factor of each network layer includes: Obtaining a first excitation factor of the optical transmission network layer, a second excitation factor of the IP network layer, a third excitation factor of the radio access network layer, and a fourth excitation factor of the core network layer; Calculating a first local excitation of the optical transmission network layer based on the delay of the optical link of the optical transmission network layer, the packet loss rate of the optical link, the link factor, and the first excitation factor; Calculating a second local excitation of the IP network layer based on the delay of the link of the IP network layer, the packet loss rate of the link, the delay jitter of the link, and the second excitation factor; Calculating a third local excitation of the radio access network network layer based on a time delay of a link of the radio access network network layer, a bit error rate of the link, and the third excitation factor; Based on the link delay of the core network network layer, the bit error rate of the link and the fourth excitation factor, the fourth local excitation of the core network network layer is calculated; wherein, the local excitation includes the first local excitation, the second local excitation, the third local excitation and the fourth local excitation.

4. The method according to claim 2, characterized in that Obtaining the link factor of the optical transmission network layer includes: Determining a current modulation mode, a current transmission distance, and a number of nodes of the optical transmission network layer; A link factor of the optical transmission network layer is determined based on a current modulation mode of the optical transmission network layer, a current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer.

5. The method according to claim 4, characterized in that The determining the link factor of the optical transmission network layer based on the current modulation mode of the optical transmission network layer, the current transmission distance of the optical transmission network layer, and the number of nodes of the optical transmission network layer includes: Determining the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes in the current optical path; The link factor is determined based on the spectrum efficiency corresponding to the current modulation mode of the optical transmission network layer, the spectrum efficiency corresponding to the current transmission distance, and the number of nodes of the current optical path.

6. The method according to claim 3, characterized in that The obtaining of the first excitation factor of the optical transmission network layer, the second excitation factor of the IP network layer, the third excitation factor of the radio access network layer, and the fourth excitation factor of the core network layer includes: respectively determining the system capacity and target capacity of the optical transmission network layer, the system capacity and target capacity of the IP network layer, the system capacity and target capacity of the radio access network layer, and the system capacity and target capacity of the core network layer; Determining, based on the system capacity and the target capacity, a load factor of the optical transmission network layer, a load factor of the IP network layer, a load factor of the radio access network layer, and a load factor of the core network layer; Determine global incentives; For each network layer, the first excitation factor, the second excitation factor, the third excitation factor, and the fourth excitation factor are determined based on the load factor and the global excitation.

7. The method according to claim 3, characterized in that The determining of a target route from an initial node to a destination node based on the local excitation of each network layer and the AI ​​engine of each network layer includes: Using a deep learning algorithm, an AI engine of each network layer determines a to-be-screened route from the initial node to the destination node based on the first local stimulus, the second local stimulus, the third local stimulus, and the fourth local stimulus; The target route is determined from the routes to be screened based on a size relationship between the packet delay of the routes to be screened and a target delay threshold.

8. A route determination device, characterized in that: The device includes: a processor, a memory and a communication bus; The communication bus is used to realize the communication connection between the processor and the memory; The processor is used to execute the route determination program in the memory to implement the steps of the route determination method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the route determination method according to any one of claims 1 to 7.

Images