A multi-task path determination method, apparatus and network device

Abstract

An embodiment of the present application provides a multi-task path determination method, apparatus, and network device. The method includes: obtaining a plurality of tasks to be executed; determining, among pre-determined network nodes, a first path with the smallest cost increment when executing the first task; determining first remaining resources of the pre-determined network nodes according to resources required for executing the first task on the first path; for the i-th task among the plurality of tasks, determining, among the pre-determined network nodes, an i-th path with the smallest cost increment when executing the i-th task according to the (i-1)-th remaining resources, and determining the i-th remaining resources of the pre-determined network nodes according to resources required for executing the i-th task on the i-th path. Therefore, in an embodiment of the present application, in a scenario where multiple tasks need to be executed, the cost of each task executed individually can reach a local optimum, and network resources can be utilized more fully.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technologies, and in particular, to a method, an apparatus, and a network device for determining a multi-task path. Background Art

[0002] Based on the reconfigurable architecture of the space-earth integrated network using Software Defined Networking (SDN) and Network Function Virtualization (NFV) technologies, research on resource management technologies can be carried out to achieve the purpose of optimizing resource allocation and improving resource utilization. Currently, a satellite network virtualized resource management model named the TRS model (where T represents task, R represents resources, and S represents service) has been proposed. Among them, after mathematical modeling according to the TRS model, corresponding algorithms can be used to calculate the tasks that need to be processed by the space-earth integrated network. The calculation result is a path including one or more nodes, and this path makes the total cost of task execution as small as possible.

[0003] However, for the scenario where multiple tasks need to be executed simultaneously, currently, it is impossible to make the cost of each task executed individually reach a local optimum, and it is impossible to make more full use of network resources. Summary of the Invention

[0004] Embodiments of the present application provide a method, an apparatus, and a network device for determining a multi-task path, so that in a scenario where multiple tasks need to be executed, the cost of each task executed individually reaches a local optimum, and network resources are more fully utilized.

[0005] In a first aspect, embodiments of the present application provide a method for determining a multi-task path, and the method includes:

[0006] Obtain multiple tasks to be executed;

[0007] In the pre-determined network nodes, determine a first path with the smallest cost increment when executing the first task, where the first path includes at least one of the pre-determined network nodes;

[0008] Determine the first remaining resources of the pre-determined network nodes according to the resources required for executing the first task on the first path;

[0009] For the i-th task among the multiple tasks, according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determine the i-th path with the minimum cost increment when executing the i-th task, and determine the i-th remaining resources of the pre-determined network nodes according to the resources required to execute the i-th task on the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and i takes each integer from 2 to n, and n represents the number of the multiple tasks.

[0010] Optionally, the step of determining the i-th path with the minimum cost increment when executing the i-th task in the pre-determined network nodes according to the (i - 1)-th remaining resources includes:

[0011] Based on Dijkstra's algorithm, and according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determine the i-th path with the minimum cost increment when executing the i-th task.

[0012] Optionally, the step of determining the i-th path with the minimum cost increment when executing the i-th task in the pre-determined network nodes based on Dijkstra's algorithm and according to the (i - 1)-th remaining resources includes:

[0013] From the pre-determined network nodes, determine the start network node and the destination network node of the multiple tasks;

[0014] Obtain the path from the start network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network nodes include the network nodes in the pre-determined network nodes other than the start network node, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes;

[0015] According to the (i - 1)-th remaining resources, select the first target path with the minimum cost increment when executing the i-th task from the first candidate paths;

[0016] When the destination network node is at the end of the first target path, determine the first target path as the i-th path.

[0017] Optionally, based on Dijkstra's algorithm and according to the (i-1)-th remaining resources, in the pre-determined network nodes, to determine the i-th path with the minimum cost increment when executing the i-th task, further includes: when the node at the end of the first target path is not the destination network node, obtaining a path formed after adding the j2-th second candidate network node to the end of the first target path, as the j2-th second candidate path, where the second candidate network nodes include the network nodes in the pre-determined network nodes except the network nodes on the first target path, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes;

[0018] According to the (i-1)-th remaining resources, select the second target path with the minimum cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths;

[0019] When the end of the second target path is the destination network node, determine the second target path as the i-th path.

[0020] Optionally, the step of selecting the first target path with the minimum cost increment when executing the i-th task from the first candidate paths according to the (i-1)-th remaining resources includes:

[0021] According to the (i-1)-th remaining resources, calculate the cost increment of each first candidate path when executing the i-th task, and store it as the target parameter of the first candidate path in a pre-established first set;

[0022] Obtain the first minimum value in the first set, and determine the first candidate path to which the first minimum value belongs as the first target path.

[0023] Optionally, the step of selecting the second target path with the minimum cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths includes:

[0024] According to the (i-1)-th remaining resources, calculate the cost increment of each second candidate path when executing the i-th task, as the target parameter of the second candidate path;

[0025] In the case that the target parameter of the second candidate path is less than the target parameter of the first candidate path associated with the second candidate path, in the first set, replace the target parameter of the first candidate path with the target parameter of the second candidate path associated with the first candidate path, where the first candidate path and the second candidate path with the same end node are associated;

[0026] Obtain the second minimum value in the first set, and determine the path to which the second minimum value belongs as the second target path.

[0027] Optionally, before selecting, according to the (i - 1)-th remaining resource, the first target path with the smallest cost increment when executing the i-th task from the first candidate paths, the method further includes:

[0028] Store the starting network node in a pre-established second set;

[0029] Store the network nodes in the pre-determined network nodes except the starting network node in a pre-established third set;

[0030] After selecting the first target path with the smallest cost increment when executing the i-th task, the method further includes:

[0031] Delete the network node at the end of the first target path from the third set, and store the network node at the end of the first target path in the second set;

[0032] After selecting the second target path with the smallest cost increment when executing the i-th task, the method further includes:

[0033] Delete the network node at the end of the second target path from the third set, and store the network node at the end of the second target path in the second set.

[0034] Optionally, calculating the cost increment of each of the first candidate paths when executing the i-th task according to the (i - 1)-th remaining resource includes:

[0035] Determine the reconstruction cost when the end node on the j1-th first candidate path executes the i-th task according to the (i - 1)-th remaining resource and the resources required to execute the i-th task

[0036] According to the first preset formula Calculate the first parameter of the j1-th first candidate path where T 00 = 0, T (i-1)d represents the delay of the (i - 1)-th path when executing the (i - 1)-th task, represents the delay of the j1-th first candidate path when executing the i-th task;

[0037] According to the second preset formula: Calculate the cost increment of the j1-th first candidate path when executing the i-th task Among them, a and b are pre-determined weight values.

[0038] Optionally, calculating the cost increment of each of the second candidate paths when executing the i-th task according to the (i - 1)-th remaining resource includes:

[0039] Calculating a first cost increment of the first target path when executing the i-th task according to the (i - 1)-th remaining resource;

[0040] Calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task according to the (i - 1)-th remaining resource, as the j2-th second cost increment;

[0041] Determining the sum of the first cost increment and the j2-th second cost increment as the cost increment of the j2-th second candidate path when executing the i-th task.

[0042] Optionally, calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task according to the (i - 1)-th remaining resource includes:

[0043] Determining the reconstruction cost of the j2-th second candidate network node when executing the i-th task according to the (i - 1)-th remaining resource and the resources required for executing the i-th task

[0044] According to a third preset formula:

[0045] Calculating a second parameter of the j2-th second candidate path Among them, T (i-1)d represents the delay of the (i - 1)-th path when executing the (i - 1)-th task, represents the delay of the first target path when executing the i-th task, represents the delay from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task, j M represents that the end node of the first target path is the j M -th first candidate network node;

[0046] According to a fourth preset formula: Calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task Among them, a and b are pre-determined weight values.

[0047] Optionally, the method further includes:

[0048] Calculate the latency of the j1-th first candidate path when executing the i-th task And store it as the third parameter of the first candidate path in a pre-established fourth set;

[0049] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the method further includes:

[0050] Determine the second candidate path associated with the first candidate path whose target parameter in the first set is replaced, as the first path to be processed;

[0051] Calculate the latency of the first path to be processed when executing the i-th task, as the fourth parameter of the first path to be processed;

[0052] Replace the third parameter of the first candidate path whose target parameter in the first set is replaced in the fourth set with the fourth parameter of the first path to be processed associated with the first candidate path.

[0053] Optionally, the method further includes:

[0054] According to the fourth preset formula Calculate the fifth parameter of the j1-th first candidate path And store it in a pre-established fifth set, where T 00 = 0;

[0055] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the method further includes:

[0056] Determine the second candidate path associated with the first candidate path whose target parameter in the first set is replaced, as the first path to be processed;

[0057] According to the fifth preset formula Calculate the sixth parameter of the first path to be processed

[0058] Replace the fifth parameter of the first candidate path whose target parameter in the first set is replaced in the fifth set with the sixth parameter of the first path to be processed associated with the first candidate path;

[0059] Where j M Indicates that the end node of the first target path is the j M -th first candidate network node.

[0060] Optionally, before determining the first path with the smallest cost increment when performing the first task among the pre-determined network nodes, the method further includes:

[0061] Sort the multiple tasks.

[0062] In a second aspect, an embodiment of the present application further provides a multi-task path determination device, and the device includes:

[0063] A task acquisition module, configured to acquire multiple tasks to be executed;

[0064] A path determination module, configured to determine a first path with the smallest cost increment when performing the first task among the pre-determined network nodes, where the first path includes at least one of the pre-determined network nodes;

[0065] A remaining resource determination module, configured to determine the first remaining resource of the pre-determined network node according to the resources required for the first task executed according to the first path;

[0066] The path determination module is further configured to, when i takes each integer from 2 to n, determine an i-th path with the smallest cost increment when performing the i-th task among the pre-determined network nodes according to the (i - 1)-th remaining resource;

[0067] The remaining resource determination module is further configured to determine the i-th remaining resource of the pre-determined network node according to the resources required for the i-th task executed according to the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and n represents the number of the multiple tasks.

[0068] In a third aspect, an embodiment of the present application further provides a network device, including a memory, a transceiver, and a processor: The memory is used to store a computer program; the transceiver is used to transmit and receive data under the control of the processor; the processor is used to execute the steps in the multi-task path determination method described in the first aspect above.

[0069] In a fourth aspect, an embodiment of the present application further provides a processor-readable storage medium, and the processor-readable storage medium stores a computer program, and the computer program is used to cause the processor to execute the multi-task path determination method described in the first aspect above.

[0070] In an embodiment of the present application, a plurality of tasks to be executed can be obtained, so as to determine, among pre-determined network nodes, a first path with the smallest cost increment when executing the first task, where the first path includes at least one pre-determined network node; further, according to the resources required to execute the first task along the first path, the first remaining resources of the pre-determined network nodes are determined; for the i-th task among the plurality of tasks, according to the (i-1)-th remaining resources, among the pre-determined network nodes, a i-th path with the smallest cost increment when executing the i-th task is determined, and according to the resources required to execute the i-th task along the i-th path, the i-th remaining resources of the pre-determined network nodes are determined, where the i-th path includes at least one pre-determined network node, and i takes each integer from 2 to n, and n represents the number of the plurality of tasks.

[0071] It can be seen that the embodiment of the present application adopts a greedy algorithm to sequentially find the optimal path for each single task according to the principle of the smallest cost increment for a single task among a plurality of tasks, and after finding the optimal path for each single task, calculate the remaining resources of the network nodes, so that the next task finds a path according to the remaining resources. Therefore, in the embodiment of the present application, in a scenario where there are multiple tasks to be executed, the cost of each task executed alone can reach a local optimum, and the network resources can be utilized more fully. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0073] Figure 1 A flowchart of a method for determining a multi-task path provided by the prior art;

[0074] Figure 2 A flowchart of steps of a method for determining a multi-task path provided by an embodiment of the present application;

[0075] Figure 3 A schematic diagram of the usage scenario of the method for determining a multi-task path provided by an embodiment of the present application;

[0076] Figure 4 A flowchart of steps of the algorithm for the path with the smallest cost increment for a single task in an embodiment of the present application;

[0077] Figure 5 A schematic diagram of the path finding process for multiple tasks in an embodiment of the present application;

[0078] Figure 6Schematic diagram of the connection status of multiple network nodes in the embodiments of the present application;

[0079] Figure 7 Schematic diagram of the change of the total cost of the paths found by multitasking with the total resource demand of the multitasking in the embodiments of the present application;

[0080] Figure 8 Block diagram of the multi-path determination device provided in the embodiments of the present application;

[0081] Figure 9 Block diagram of the electronic device provided in the embodiments of the present application. Detailed implementation manners

[0082] In the embodiments of the present application, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0083] In the embodiments of the present application, the term "plurality" means two or more, and other quantifiers are similar thereto.

[0084] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0085] The embodiments of the present application provide a multitasking path determination method, device and network device, so that in a scenario where multiple tasks need to be executed, the cost of each task executed alone reaches a local optimum and satellite resources are more fully utilized.

[0086] Among them, the method and the device are based on the same inventive concept. Since the principles of the method and the device for solving problems are similar, the implementation of the device and the method can be referred to each other, and the repeated parts will not be described again.

[0087] In addition, the technical solutions provided by the embodiments of the present application can be applied to a variety of systems, especially 5G systems. For example, the applicable systems can be the global system of mobile communication (GSM) system, the code division multiple access (CDMA) system, the wideband code division multiple access (WCDMA) general packet radio service (GPRS) system, the long term evolution (LTE) system, the LTE frequency division duplex (FDD) system, the LTE time division duplex (TDD) system, the long term evolution advanced (LTE-A) system, the universal mobile telecommunication system (UMTS), the worldwide interoperability for microwave access (WiMAX) system, the 5G new radio (NR) system, the space-ground integrated information network, etc. Both terminal devices and network devices are included in these various systems. The core network part can also be included in the system, such as the evolved packet system (EPS), the 5G system (5GS), etc.

[0088] The terminal device involved in the embodiments of the present application may be a device that provides voice and / or data connectivity to users, such as a handheld device with wireless connection capabilities, or other processing devices connected to a wireless modem, etc. In different systems, the name of the terminal device may also be different. For example, in a 5G system, the terminal device may be referred to as a user equipment (UE). The wireless terminal device can communicate with one or more core networks (CNs) via a radio access network (RAN). The wireless terminal device can be a mobile terminal device, such as a mobile phone (or a "cellular" phone) and a computer with a mobile terminal device. For example, it can be a portable, pocket-sized, handheld, computer-integrated, or vehicle-mounted mobile device that exchanges voice and / or data with the radio access network. For example, devices such as personal communication service (PCS) phones, cordless phones, session initiated protocol (SIP) phones, wireless local loop (WLL) stations, and personal digital assistants (PDAs). The wireless terminal device can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile, a remote station, an access point, a remote terminal device, an access terminal device, a user terminal device, a user agent, a user device, which is not limited in the embodiments of the present application.

[0089] The network device involved in the embodiments of the present application can be a base station, which can include multiple cells that provide services to terminals. Depending on the specific application scenarios, the base station can also be referred to as an access point, or it can be a device in the access network that communicates with wireless terminal devices through one or more sectors on the air interface, or other names. The network device can be used to mutually replace the received air frames and Internet Protocol (IP) packets, and act as a router between the wireless terminal device and the rest of the access network, where the rest of the access network can include an Internet Protocol (IP) communication network. The network device can also coordinate the management of the attributes of the air interface. For example, the network device involved in the embodiments of the present application can be a network device (Base Transceiver Station, BTS) in the Global System for Mobile communications (GSM) or Code Division Multiple Access (CDMA), or it can be a network device (NodeB) in Wide-band Code Division Multiple Access (WCDMA), or it can also be an evolved network device (evolutional Node B, eNB or e-NodeB) in the Long Term Evolution (LTE) system, a 5G base station (gNB) in the 5G network architecture (next generation system), or it can be a Home evolved Node B (HeNB), a relay node, a femto, a pico, etc. The embodiments of the present application do not limit this. In some network structures, the network device can include a centralized unit (centralized unit, CU) node and a distributed unit (distributed unit, DU) node, and the centralized unit and the distributed unit can also be arranged separately geographically.

[0090] A network device and a terminal device can each use one or more antennas for Multi-Input Multi-Output (MIMO) transmission. The MIMO transmission can be Single User MIMO (SU-MIMO) or Multiple User MIMO (MU-MIMO). According to the form and number of antenna combinations, the MIMO transmission can be 2D-MIMO, 3D-MIMO, FD-MIMO or massive-MIMO, or it can be diversity transmission, precoding transmission, beamforming transmission, etc.

[0091] Among them, in order to facilitate the understanding of the multi-task path determination method provided by the embodiments of the present application, the related technologies of the prior art are first described as follows:

[0092] The reconfigurable architecture of the space-ground integrated network based on SDN and NFV technologies provides a basis for the research of network resource management technology. The TRS model is a virtualized resource management model in the space-ground integrated network. By unifying the representation of tasks, resources, and services, a graph theory model of resource management is established. Among them, through the objective function, the graph theory model can be further transformed into a specific mathematical model. Therefore, the TRS model is inseparable from the reconfigurable network architecture. It can be said that without the reconfigurable network architecture, the TRS model cannot be established.

[0093] Among them, the goal of the TRS model is to select a path containing one or more satellite nodes for each task between the starting point and the ending point of communication. For each satellite node in the path, its on-board processing resources such as computing resources and storage resources can meet the requirements of the task. If the resources of the satellite node do not meet the task requirements, resource reconstruction can be used to mutually transform the resources.

[0094] Specifically, resource reconstruction transformation does not refer to resource exchange in the actual sense. Among them, computing and communication are essentially computing power. For the implementation of specific tasks, the purpose of meeting the task requirements can be achieved through the optimal allocation of different resources.

[0095] The objective function in the TRS model is represented by minP = min(a*T d + b*P rec ). Among them, P represents the total cost when the task is executed. The goal of resource management is to minimize P; T d represents the maximum delay among multiple tasks executed simultaneously, T d = max(T 1d , T 2d , T 3d , …), T idDenote the latency from the start to the end of the \(i\)-th sub-task. \(P\) rec Denote the sum of the reconstruction costs of each satellite during the task execution, that is \(P\) rec =\(\sum\) i \(P\) i_rec . When the \(i\)-th satellite needs to meet the task requirements through resource reconstruction, the corresponding reconstruction cost \(P\) i_rec is generated, where \(P\) i_rec =\(\beta\) i *\(\Delta R\) i , \(\beta\) i denotes the reconstruction cost coefficient of satellite \(i\), \(\Delta R\) i denotes the resource conversion amount when satellite \(i\) performs resource reconstruction, and \(a\), \(b\) are the weights of the latency cost and the reconstruction cost, satisfying \(a + b=1\).

[0096] As Figure 1 shown, a multi-task path determination method provided in the prior art is as described in Steps 1 to 6 below:

[0097] Step 1: Multiple tasks form a set \(L\), that is, the complex task is decomposed into multiple sub-task sets \(L = \{l_1, l_2,\cdots, l_k\}\), where the number of tasks is \(k\). Starting from the task source node, then perform Step 2;

[0098] Step 2: Confirm the task source node \(T\) and the service end node \(S\), and define two sets, namely set \(V\) and \(U\), representing the set of satellite nodes that have been searched and the set of satellite nodes that execute tasks. Initially, \(V\) only contains the starting satellite, and \(U\) is initially empty. Denote the number of nodes in \(U\) as \(x\).

[0099] Step 3: Gradually search for the satellites directly connected to it starting from the satellites in set \(V\) and add them to set \(V\). If \(P\) T(l1) \(\leq P\) Rh , and \(x\lt k\), then add this satellite to set \(U\). If \(x\gt k\), then perform Step 4. If \(P\) T(l1) \(\geq P\) Rh , then after performing Step 5, continue to perform Step 4, where \(P\) T(l1) represents the required ability of the \(l_1\)-th sub-task, and \(P\) Rh represents the execution ability of the \(h\)-th satellite;

[0100] Step 4: Determine whether the cost value of any node in set \(U\) is greater than the cost value of this node (i.e., satellite \(h\)) for executing the task. If so, replace this node with the node with the highest cost value in set \(U\). If not, continue to search for the nodes in this resource pool. If the search is completed, perform Step 6. Otherwise, go to Step 3.

[0101] Step 5: If the task requirement ability \(P\) T(l1) \(\gt P\) Rh , and \(P\) T(l1) \(\lt P\)total (P total represents the overall ability evaluation coefficient of the reconfigurable resource), it means that the satellite resource can adjust its resource structure to enable it to serve the task, and the required adjusted resource ability weight is ΔP = (P T(l1) -P Rh ). After adjustment, calculate its corresponding task cost value; if P T(l1) >P total , it means that the satellite resource ability is insufficient to serve the task. After this step is completed, return to step three.

[0102] Step Six: Determine whether the number of nodes x in V is equal to the number of meta-tasks k. If so, calculate the cost value of the resource aggregation within the set U, and the set U is the resource aggregation flow for executing this task; if not, it means the task execution fails.

[0103] As Figure 2 shown, Figure 2 shows a schematic flowchart of a multi-task path determination method provided by an embodiment of the present application, which may include the following steps 201 to 204:

[0104] Step 201: Obtain multiple tasks to be executed.

[0105] It should be noted that the multiple tasks described in the embodiments of the present application include multiple tasks that can be executed in parallel. Therefore, in the embodiments of the present application, although when determining the execution path of multiple tasks, the optimal path of a single task is searched in a certain order successively, in the actual data communication process, all tasks are executed simultaneously.

[0106] Optionally, before step 201, the method further includes:

[0107] Sort the multiple tasks.

[0108] For example, the multiple tasks can be sorted in ascending or descending order according to the resource requirements.

[0109] Step 202: Among the pre-determined network nodes, determine the first path with the smallest cost increment when executing the first task.

[0110] Wherein, the first path includes at least one of the pre-determined network nodes.

[0111] In addition, the pre-determined network nodes include network nodes within a pre-determined preset range area. For example, if satellite nodes within a certain spatial area are pre-determined to be available, these satellite nodes can be used as the network nodes in step 202, that is, these satellite nodes can be used to execute the multiple tasks described in the aforementioned step 201.

[0112] In addition, when the amount of available resources of a network node is less than the resource requirement for executing a certain task, the resources of the network node can be mutually transformed through resource reconstruction. For example, when the network node has more available storage resources but less available computing resources, a part of the available storage resources can be transformed into computing resources to meet the resource requirements for executing the task.

[0113] It should also be noted that in the embodiments of the present application, the cost increment of a path includes a resource reconstruction cost increment and a delay cost increment. Moreover, the weight a of the resource reconstruction cost increment and the weight b of the delay cost increment can be determined in advance, so as to multiply the value of the resource reconstruction cost increment by a and add the value of the delay cost increment multiplied by b to obtain the total cost increment (i.e., the cost increment of the path), where a + b = 1.

[0114] Step 203: Determine the first remaining resources of the predetermined network node according to the resources required for the first task to be executed on the first path.

[0115] The first remaining resources include the available resources of the network nodes other than the network nodes on the first path in the predetermined network node, and the remaining available resources of the network nodes on the first path after executing the first task.

[0116] Step 204: For the i-th task among the multiple tasks, determine the i-th path with the smallest cost increment when executing the i-th task in the predetermined network node according to the (i - 1)-th remaining resources, and determine the i-th remaining resources of the predetermined network node according to the resources required for the i-th task to be executed on the i-th path.

[0117] The i-th path includes at least one of the predetermined network nodes, and i takes each integer from 2 to n, where n represents the number of the multiple tasks.

[0118] For example, when n = 3, step 204 specifically includes:

[0119] Determine the second path with the smallest cost increment when executing the second task in the predetermined network node according to the first remaining resources;

[0120] Determine the second remaining resources of the predetermined network node according to the resources required for the second task to be executed on the second path;

[0121] Determine the third path with the smallest cost increment when executing the third task in the predetermined network node according to the second remaining resources;

[0122] Determine the third remaining resources of the predetermined network node according to the resources required for the third task to be executed on the third path.

[0123] It can be seen from this that in the embodiments of the present application, after each task finds the optimal single-task path, the remaining resources of the network nodes are calculated, so that the next task finds a path according to the remaining resources.

[0124] In addition, in the embodiments of the present application, each task may pass through the same path or different paths, that is, a network node may need to process multiple tasks simultaneously or only process one task.

[0125] In addition, if the path with the smallest cost increment is determined for each task execution, the total cost when all tasks are processed simultaneously can be calculated, and this total cost is the sum of the cost increments of all tasks.

[0126] It can be seen from the above steps 201 to 204 that the embodiments of the present application adopt a greedy algorithm to sequentially find the optimal single-task path for multiple tasks according to the principle of the minimum cost increment of a single task. After each task finds the optimal single-task path, the remaining resources of the satellite are calculated, so that the next task finds a path according to the remaining resources. Therefore, in the embodiments of the present application, in a scenario where multiple tasks need to be executed, the cost of each task executed separately can reach a local optimum, and the satellite resources can be utilized more fully.

[0127] Optionally, according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determining the i-th path with the smallest cost increment for executing the i-th task includes:

[0128] Based on the Dijkstra algorithm and according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determining the i-th path with the smallest cost increment for executing the i-th task.

[0129] Similarly, step 202 can also specifically include: Based on the Dijkstra algorithm, in the pre-determined network nodes, determining the first path with the smallest cost increment for executing the first task.

[0130] Among them, the Dijkstra algorithm is a typical shortest path algorithm used to calculate the shortest path from one node to other nodes. Its main feature is to expand layer by layer (breadth-first search idea) with the starting point as the center until the end point is reached. Therefore, in the embodiments of the present application, the process of finding the path with the minimum cost increment of a single task is implemented based on the Dijkstra algorithm. Therefore, the cost increment of the path of each task found is theoretically optimal.

[0131] Optionally, based on the Dijkstra algorithm and according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determining the i-th path with the smallest cost increment for executing the i-th task includes:

[0132] Determine the starting network node and the destination network node of the multiple tasks from the pre-determined network nodes;

[0133] Obtain the path from the starting network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network nodes include the network nodes in the pre-determined network nodes other than the starting network node, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes;

[0134] According to the (i - 1)-th remaining resource, select the first target path with the smallest cost increment when executing the i-th task from the first candidate paths;

[0135] When the end node of the first target path is the destination network node, determine the first target path as the i-th path.

[0136] Optionally, the method for determining the i-th path with the smallest cost increment when executing the i-th task in the pre-determined network nodes based on Dijkstra's algorithm and according to the (i - 1)-th remaining resource further includes:

[0137] When the node at the end of the first target path is not the destination network node, obtain the path formed after adding the j2-th second candidate network node to the end of the first target path as the j2-th second candidate path, where the second candidate network nodes include the network nodes in the pre-determined network nodes other than the network nodes on the first target path, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes;

[0138] According to the (i - 1)-th remaining resource, select the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths in the first candidate paths other than the first target path;

[0139] When the end node of the second target path is the destination network node, determine the second target path as the i-th path.

[0140] Optionally, the method for determining the i-th path with the smallest cost increment when executing the i-th task in the pre-determined network nodes based on Dijkstra's algorithm and according to the (i - 1)-th remaining resource further includes:

[0141] When the node at the end of the second target path is not the destination network node, obtain the path formed after adding the j3-th third candidate network node to the end of the second target path as the third candidate path, where the third candidate network node includes the network nodes in the pre-determined network nodes except the network nodes on the second target path, and j3 takes each integer from 1 to m3, and m3 represents the number of the third candidate network nodes;

[0142] Select the third target path with the smallest cost increment when executing the i-th task from the third candidate path, the paths other than the first target path and the second target path in the first candidate path and the second candidate path.

[0143] For the process of determining the path with the smallest cost increment when executing a certain task based on Dijkstra's algorithm, the following target example is used to illustrate as follows:

[0144] Currently, there are three tasks to be executed, namely the first task, the second task, and the third task, and the starting network node of these tasks is node 0, and the destination network node is node 10. The pre-determined network nodes include nodes 0 to 10. Then the process of determining the first path with the smallest cost increment when executing the first task is described as follows:

[0145] First, since the starting network node of the task is node 0, 10 first candidate paths can be formed by combining node 0 with nodes 1 to 10 respectively, that is, node 0-node 1, node 0-node 2, node 0-node 3, node 0-node 4, node 0-node 5, node 0-node 6, node 0-node 7, node 0-node 8, node 0-node 9, node 0-node 10. Thus, the first target path with the smallest cost increment when executing the first task can be selected from these first candidate paths;

[0146] For example, if the selected first target path is: node 0 - node 3, and the end node of this first target path is not the destination network node (i.e., node 10), then based on the first target path, nodes 1, 2, 4 - 10 can be added respectively to obtain 9 second candidate paths, namely node 0 - node 3 - node 1, node 0 - node 3 - node 2, node 0 - node 3 - node 4, node 0 - node 3 - node 5, node 0 - node 3 - node 6, node 0 - node 3 - node 7, node 0 - node 3 - node 8, node 0 - node 3 - node 9, node 0 - node 3 - node 10. Thus, from these second candidate paths and the first candidate paths other than the aforementioned first target path (i.e., node 0 - node 1, node 0 - node 2, node 0 - node 4, node 0 - node 5, node 0 - node 6, node 0 - node 7, node 0 - node 8, node 0 - node 9, node 0 - node 10), the second target path with the smallest cost increment when executing the first task can be selected;

[0147] For example, if the selected second target path is: node 0 - node 3 - node 2, and the end node of this second target path is not the destination network node (i.e., node 10), then based on the second target path, nodes 1, 4 - 10 can be added respectively to obtain 8 third candidate paths, namely node 0 - node 3 - node 2 - node 1, node 0 - node 3 - node 2 - node 4, node 0 - node 3 - node 2 - node 5, node 0 - node 3 - node 2 - node 6, node 0 - node 3 - node 2 - node 7, node 0 - node 3 - node 2 - node 8, node 0 - node 3 - node 2 - node 9, node 0 - node 3 - node 2 - node 10. Thus, from these third candidate paths, the first candidate paths other than the aforementioned first target path, and the first candidate paths other than the second target path, the third target path with the smallest cost increment when executing the first task can be selected;

[0148] Until the end node of the path with the smallest cost increment when executing the first task is node 10, the path found at this time is the first path for executing the first task.

[0149] Similarly, for the aforementioned second task and third task, the method for determining the path with the smallest cost increment can refer to the process of finding the first path for executing the first task described above, which will not be elaborated here. The difference is that when finding paths for the second task and third task, the remaining resources of the network nodes after executing the previous tasks need to be considered.

[0150] Optionally, according to the (i - 1)-th remaining resource, selecting the first target path with the smallest cost increment when executing the i-th task from the first candidate paths includes:

[0151] Calculate the cost increment of each first candidate path when executing the i-th task according to the (i - 1)-th remaining resource, and store it as the target parameter of the first candidate path in a pre-established first set;

[0152] Obtain the first minimum value in the first set, and determine the first candidate path to which the first minimum value belongs as the first target path.

[0153] For example, in the aforementioned target example, the first candidate paths include 10 paths: node 0 - node 1, node 0 - node 2, node 0 - node 3, node 0 - node 4, node 0 - node 5, node 0 - node 6, node 0 - node 7, node 0 - node 8, node 0 - node 9, node 0 - node 10. Then, the cost increment of each first candidate path when executing the first task can be calculated and stored in the first set. For example, the first set is {ΔP 01 、ΔP 02 、ΔP 03 、ΔP 04 、ΔP 05 、ΔP 06 、ΔP 07 、ΔP 08 、ΔP 09 、ΔP 010}. For example, the minimum value in the first set is ΔP 03 , then the selected first target path is: node 0 - node 3. Among them, ΔP 01 、ΔP 02 、ΔP 03 、ΔP 04 、ΔP 05 、ΔP 06 、ΔP 07 、ΔP 08 、ΔP 09 、ΔP 010 respectively represent the cost increments of the 1st to 9th first candidate paths when executing the first task.

[0154] Optionally, according to the (i - 1)-th remaining resource, select the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths among the first candidate paths other than the first target path, including:

[0155] Calculate the cost increment of each second candidate path when executing the i-th task according to the (i - 1)-th remaining resource, so as to be used as the target parameter of the second candidate path;

[0156] In the case where the target parameter of the second candidate path is less than the target parameter of the first candidate path associated with the second candidate path, replace the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, where the first candidate path and the second candidate path with the same end node are associated;

[0157] Obtain the second minimum value in the first set, and determine the path to which the second minimum value belongs as the second target path.

[0158] For example, in the foregoing target example, the second candidate paths include 9 paths, namely node 0 - node 3 - node 1, node 0 - node 3 - node 2, node 0 - node 3 - node 4, node 0 - node 3 - node 5, node 0 - node 3 - node 6, node 0 - node 3 - node 7, node 0 - node 3 - node 8, node 0 - node 3 - node 9, node 0 - node 3 - node 10. Then, the cost increment of each second candidate path when executing the first task can be calculated respectively. For example, the cost increments of these 9 second candidate paths are respectively:

[0159] ΔP 03 +ΔP 31 、ΔP 03 +ΔP 32 、ΔP 03 +ΔP 34 、ΔP 03 +ΔP 35 、ΔP 03 +ΔP 36 、ΔP 03 +ΔP 37 、ΔP 03 +ΔP 38 、ΔP 03 +ΔP 39 、ΔP 03 +ΔP 310 ;

[0160] Among them, the second candidate path of node 0 - node 3 - node 1 is associated with the first candidate path of node 0 - node 1. Therefore, it is necessary to compare the size of ΔP 03 +ΔP 31 and ΔP 01 . If ΔP 03 +ΔP 31 is less than ΔP 01 , then it is necessary to replace ΔP 01 in the foregoing first set with ΔP 03 +ΔP 31 ;

[0161] The second candidate path of Node 0 - Node 3 - Node 2 is associated with the first candidate path of Node 0 - Node 2. Therefore, it is necessary to compare ΔP 03 +ΔP 32 with ΔP 02 . If ΔP 03 +ΔP 32 is less than ΔP 02 , then it is necessary to replace ΔP 02 in the aforementioned first set with ΔP 03 +ΔP 32 ;

[0162] Similarly, it is necessary to compare the sizes of ΔP 03 +ΔP 34 and ΔP 04 , the sizes of ΔP 03 +ΔP 35 and ΔP 05 , the sizes of ΔP 03 +ΔP 36 and ΔP 06 , the sizes of ΔP 03 +ΔP 37 and ΔP 07 , the sizes of ΔP 03 +ΔP 38 and ΔP 08 , the sizes of ΔP 03 +ΔP 39 and ΔP 09 , the sizes of ΔP 03 +ΔP 310 and ΔP 010 to replace the larger one in the aforementioned first set with the smaller one.

[0163] After comparing the target parameters of the aforementioned second candidate path with those of the associated first candidate path and updating the aforementioned first set, for example, the updated first set is {ΔP 03 +ΔP 31 , ΔP 03 +ΔP 32 , ΔP 03 , ΔP 04 , ΔP 05 , ΔP 06 , ΔP 07 , ΔP 08 , ΔP 09 , ΔP 010}. For example, if the minimum value in the first set at this time is ΔP 03 +ΔP 32 , then the selected second target path is: Node 0 - Node 3 - Node 2.

[0164] Among them, by comparing the target parameters of the associated second candidate path with those of the first candidate path to update the first set, and then selecting the second target path from the updated first set, the numerical comparison process in the process of selecting the second target path can be reduced, thereby improving the path finding speed.

[0165] Similarly, optionally, selecting the third target path with the smallest cost increment when executing the i-th task from the paths other than the first target path and the second target path among the third candidate path, the first candidate path, and the second candidate path includes:

[0166] Calculating the cost increment when each of the third candidate paths executes the i-th task according to the (i - 1)-th remaining resource as the target parameter of the third candidate path;

[0167] In the case where the target parameter of the third candidate path is less than the target parameter of the first candidate path associated with the third candidate path, replacing the target parameter of the first candidate path in the first set with the target parameter of the third candidate path associated with the first candidate path;

[0168] In the case where the target parameter of the third candidate path is less than the target parameter of the second candidate path associated with the third candidate path, replacing the target parameter of the second candidate path in the first set with the target parameter of the third candidate path associated with the second candidate path;

[0169] Among them, the first candidate path and the second candidate path with the same end node are associated;

[0170] Obtaining the second minimum value in the first set and determining the path to which the second minimum value belongs as the third target path;

[0171] Among them, the first candidate path and the third candidate path with the same end node are associated, and the second candidate path and the third candidate node with the same end are associated.

[0172] For example, in the aforementioned target example, the third candidate paths include 8 paths, namely node 0 - node 3 - node 2 - node 1, node 0 - node 3 - node 2 - node 4, node 0 - node 3 - node 2 - node 5, node 0 - node 3 - node 2 - node 6, node 0 - node 3 - node 2 - node 7, node 0 - node 3 - node 2 - node 8, node 0 - node 3 - node 2 - node 9, node 0 - node 3 - node 2 - node 10. Then, the cost increment when each of the third candidate paths executes the first task can be calculated respectively. For example, the cost increments of these 8 third candidate paths are respectively:

[0173] ΔP03 +ΔP 32 +ΔP 21 、ΔP 03 +ΔP 32 +ΔP 24 、ΔP 03 +ΔP 32 +ΔP 25 、ΔP 03 +ΔP 32 +ΔP 26 、ΔP 03 +ΔP 32 +ΔP 27 、ΔP 03 +ΔP 32 +ΔP 28 、ΔP 03 +ΔP 32 +ΔP 29 、ΔP 03 +ΔP 32 +ΔP 210 ;

[0174] Among them, as described above, after obtaining the first target path, the first set is updated to: {ΔP 03 +ΔP 31 、ΔP 03 +ΔP 32 、ΔP 03 、ΔP 04 、ΔP 05 、ΔP 06 、ΔP 07 、ΔP 08 、ΔP 09 、ΔP 010};

[0175] And, the third candidate path of node 0 - node 3 - node 2 - node 1 is associated with the second candidate path of node 0 - node 3 - node 1. Therefore, it is necessary to compare the magnitudes of ΔP 03 +ΔP 32 +ΔP 21 with ΔP 03 +ΔP 31 . If ΔP 03 +ΔP 32 +ΔP 21 is less than ΔP 03 +ΔP 31 , then it is necessary to replace ΔP 03 +ΔP 31 in the aforementioned first set with ΔP 03 +ΔP 32 +ΔP 21 ;

[0176] The third candidate path, Node 0 - Node 3 - Node 2 - Node 4, is associated with the first candidate path, Node 0 - Node 4. Therefore, it is necessary to compare ΔP 03 + ΔP 32 + ΔP 24 with ΔP 04 . If ΔP 03 + ΔP 32 + ΔP 24 is less than ΔP 04 , then it is necessary to replace ΔP 03 + ΔP 32 + ΔP 24 in the aforementioned first set with ΔP 04 ;

[0177] Similarly, it is necessary to compare ΔP 03 + ΔP 32 + ΔP 25 with ΔP 05 , the size of ΔP 03 + ΔP 32 + ΔP 26 with ΔP 06 , the size of ΔP 03 + ΔP 32 + ΔP 27 with ΔP 07 , the size of ΔP 03 + ΔP 32 + ΔP 28 with ΔP 08 , the size of ΔP 03 + ΔP 32 + ΔP 29 with ΔP 09 , the size of ΔP 03 + ΔP 32 + ΔP 210 with ΔP 010 , so as to replace the larger one in the aforementioned updated first set with the smaller one to achieve the second update of the first set.

[0178] For example, the first set after the second update is {ΔP 03 + ΔP 32 + ΔP 21 , ΔP 03 + ΔP 32 , ΔP 03 , ΔP 03 + ΔP 32 + ΔP 24 , ΔP 05 , ΔP 06 , ΔP 07 , ΔP 08 , ΔP 09, ΔP 010}, for example, the minimum value in the first set at this time is ΔP 03 +ΔP 32 +ΔP 24 , then the selected third target path is: node 0 - node 3 - node 2 - node 4.

[0179] Optionally, before selecting the first target path with the smallest cost increment when executing the i-th task from the first candidate paths according to the i-1 remaining resources, the method further includes:

[0180] Storing the starting network node in a pre-established second set;

[0181] Storing the network nodes other than the starting network node in the pre-determined network nodes in a pre-established third set;

[0182] After selecting the first target path with the smallest cost increment when executing the i-th task, the method further includes:

[0183] Deleting the network node at the end of the first target path from the third set and storing the network node at the end of the first target path in the second set;

[0184] After selecting the second target path with the smallest cost increment when executing the i-th task, the method further includes:

[0185] Deleting the network node at the end of the second target path from the third set and storing the network node at the end of the second target path in the second set.

[0186] For example, in the foregoing target example, the third set is initially {node 1, node 2, node 3, node 4, node 5, node 6, node 7, node 8, node 9, node 10}, and the second set is initially {node 0}. Then, after selecting node 0 - node 3 as the first target path, the third set is updated to {node 1, node 2, node 4, node 5, node 6, node 7, node 8, node 9, node 10}, and the second set is updated to {node 0, node 3}; after selecting node 0 - node 3 - node 2 as the second target path, the third set is updated to {node 1, node 4, node 5, node 6, node 7, node 8, node 9, node 10}, and the second set is updated to {node 0, node 3, node 2}.

[0187] Optionally, calculating the cost increment of each of the first candidate paths when executing the i-th task according to the i-1 remaining resources includes:

[0188] Determine the reconstruction cost when the end node on the j1-th first candidate path executes the i-th task according to the (i - 1)-th remaining resource and the resources required to execute the i-th task

[0189] According to the first preset formula Calculate the first parameter of the j1-th first candidate path Wherein T 00 = 0, T (i-1)d represents the delay of the (i - 1)-th path when executing the (i - 1)-th task, and represents the delay of the j1-th first candidate path when executing the i-th task;

[0190] According to the second preset formula: Calculate the cost increment of the j1-th first candidate path when executing the i-th task Wherein, a and b are pre-determined weight values.

[0191] Wherein, the foregoing represents the maximum value of the delay upper limit of the previously routed (i - 1) tasks on their respective paths and the delay passed by the previous node (i.e., the starting network node) on this forwarding path (i.e., T 00 ). In addition, the above first parameter represents the delay cost increment of the first candidate path.

[0192] It should be noted here that when a network node can meet the resource requirements of the i-th task without resource reconstruction, its reconstruction cost is 0; when a network node cannot meet the resource requirements of the i-th task even after reconstructing resources (i.e., cannot execute the i-th task), its reconstruction cost is positive infinity. Among them, whether a certain network node can meet the resource requirements of the task needs to be compared according to the currently remaining resources of the network node and the resources required by the task.

[0193] For example, in the foregoing target example, the starting network node is node 0, and the first first candidate path is node 0 - node 1. Then, the process of calculating the cost increment of the first first candidate path when executing the i-th task is as follows:

[0194] First, determine the reconstruction cost P of node 1 when executing the i-th task according to the remaining resources of node 1 and the resources required to execute the i-th task 1_rec ;

[0195] Secondly, calculate the maximum value T' of the delay upper limit of the previously routed (i - 1) tasks on their respective paths and the delay passed by the previous node (i.e., the starting network node) on this forwarding path (i.e., T 00 ) d01 = max(T 1d,T 2d ,…,T (i-1)d ,T 00 );

[0196] Again, compare T' d01 and T 01 , so as to obtain the delay cost increment ΔT 01 from node 0 to node 1, that is

[0197] Finally, when executing the i-th task, the cost increment of the first first candidate path (i.e., from node 0 to node 1) can be calculated: ΔP 01 = a * ΔT 01 + b * P 1_rec .

[0198] Optionally, according to the i-1 remaining resources, calculate the cost increment of each of the second candidate paths when executing the i-th task, including:

[0199] According to the i-1 remaining resources, calculate the first cost increment of the first target path when executing the i-th task;

[0200] According to the i-1 remaining resources, calculate the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task, as the j2-th second cost increment;

[0201] Determine the sum of the first cost increment and the j2-th second cost increment as the cost increment of the j2-th second candidate path when executing the i-th task.

[0202] It should be noted here that the first target path is selected from the first candidate paths. Therefore, the calculation process of the cost increment of the first target path when executing the i-th task is the same as the aforementioned "calculation process of the cost increment of the first candidate path when executing the i-th task", which will not be elaborated here.

[0203] Among them, based on the first target path, adding a second candidate network node to obtain a second candidate path. Therefore, the cost increment of the second candidate path when executing the i-th task = the cost increment of the first target path when executing the i-th task + the cost increment from the end node of the first target path to the end node of the second candidate path (i.e., the second candidate network node).

[0204] For example, in the aforementioned target example, the starting network node is node 0, and the first second candidate path is node 0 - node 3 - node 1. Then, when calculating the cost increment of the first second candidate path when executing the i-th task, it is ΔP 03 + ΔP 31 . Among them, ΔP03 is the cost increment of the first target path when executing the i-th task, ΔP 31 represents the cost increment from node 3 to node 1 when executing the i-th task.

[0205] Optionally, according to the (i - 1)-th remaining resource, calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task includes:

[0206] Determining the reconstruction cost of the j2-th second candidate network node when executing the i-th task according to the (i - 1)-th remaining resource and the resources required for executing the i-th task

[0207] According to the third preset formula:

[0208] Calculating the second parameter of the j2-th second candidate path where T (i-1)d represents the delay of the (i - 1)-th path when executing the (i - 1)-th task, represents the delay of the first target path when executing the i-th task, represents the delay from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task, j M represents that the end node of the first target path is the j M th first candidate network node;

[0209] According to the fourth preset formula: Calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task where a and b are pre-determined weight values.

[0210] For example, the process of calculating the cost increment from node 3 to node 1 when executing the i-th task is as follows:

[0211] First, determining the reconstruction cost P of node 1 when executing the i-th task according to the remaining resources of node 1 and the resources required for executing the i-th task 1_rec ;

[0212] Secondly, calculating the maximum value T′ of the delay upper limit of the previously routed (i - 1) tasks on their respective paths and the delay (i.e., T 03 ) passed by the previous node (i.e., node 3) on this forwarding path d31 = max(T 1d , T 2d , …, T (i-1)d , T03 )

[0213] Again, compare T' d31 and T 03 + e 31 to obtain the delay cost increment ΔT 31 from node 3 to node 1, that is

[0214] Finally, the cost increment ΔP 31 from node 3 to node 1 when executing the i-th task can be calculated as: ΔP 31 = a * ΔT 1_rec .

[0215] Optionally, the method further includes:

[0216] Calculating the delay of the j1-th first candidate path when executing the i-th task and storing it as the third parameter of the first candidate path in a pre-established fourth set;

[0217] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the method further includes:

[0218] Determining the second candidate path associated with the first candidate path whose target parameter in the first set is replaced as the first path to be processed;

[0219] Calculating the delay of the first path to be processed when executing the i-th task as the fourth parameter of the first path to be processed;

[0220] Replacing the third parameter of the first candidate path whose target parameter in the first set is replaced in the fourth set with the fourth parameter of the first path to be processed associated with the first candidate path.

[0221] It should be noted here that the first path to be processed is the second candidate path associated with the first candidate path whose target parameter in the first set is replaced. Therefore, the first path to be processed is essentially the second candidate path. In the embodiments of the present application, the first candidate path and the second candidate path with the same end node are associated. Therefore, there is also an association stored between the first path to be processed and the first candidate path.

[0222] For example, in the foregoing target example, the first candidate paths include: node 0 - node 1, node 0 - node 2, node 0 - node 3, node 0 - node 4, node 0 - node 5, node 0 - node 6, node 0 - node 7, node 0 - node 8, node 0 - node 9, node 0 - node 10. For these 10 paths, the delays when these 10 paths execute the i-th task can be stored in the fourth set. For example, the fourth set is {T 01 、T 02 、T 03 、T 04 、T 05 、T 06 、T 07 、T 08 、T 09 、T 010};

[0223] Among them, if the first candidate paths in which the target parameter in the foregoing first set is replaced are: node 0 - node 1, node 0 - node 2, the delays when the second candidate paths associated with these two first candidate paths execute the i-th task can be calculated, that is, calculate the delays when the two paths node 0 - node 3 - node 1 and node 0 - node 3 - node 2 execute the i-th task: T 03 +T 31 、T 03 +T 32 , and replace T 01 in the foregoing fourth set with T 03 +T 31 , and replace T 02 with T 03 +T 32 . Therefore, the updated fourth set obtained is:

[0224] {T 03 +T 31 、T 03 +T 32 、T 03 、T 04 、T 05 、T 06 、T 07 、T 08 、T 09 、T 010}。

[0225] Optionally, the method further includes:

[0226] When j1 takes each integer from 1 to m1, calculate the fifth parameter of the j1-th first candidate path according to the fourth preset formula and store it in the pre-established fifth set, where T and00 = 0;

[0227] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the method further includes:

[0228] Determine the second candidate path associated with the first candidate path whose target parameter in the first set is replaced, as the first path to be processed;

[0229] According to the fifth preset formula Calculate the sixth parameter of the first path to be processed

[0230] Replace the fifth parameter of the first candidate path whose target parameter in the first set is replaced, in the fifth set, with the sixth parameter of the first path to be processed associated with the first candidate path;

[0231] Wherein, j M Indicates that the end node of the first target path is the jth M first candidate network node.

[0232] For example, in the foregoing target example, the first candidate paths include: node 0 - node 1, node 0 - node 2, node 0 - node 3, node 0 - node 4, node 0 - node 5, node 0 - node 6, node 0 - node 7, node 0 - node 8, node 0 - node 9, node 0 - node 10. For these 10 paths, the fifth parameters of these 10 paths can be calculated and stored in the fifth set. For example, the fifth set is {T' d01 , T' d02 , T' d03 , T' d04 , T' d05 , T' d06 , T' d07 , T' d08 , T' d09 , T' d010};

[0233] Wherein, if the first candidate paths in the foregoing first set whose target parameters are replaced are: node 0 - node 1, node 0 - node 2, then the sixth parameters of the second candidate paths (i.e., the first paths to be processed) associated with these two first candidate paths can be calculated, that is, calculate the sixth parameters of the two paths node 0 - node 3 - node 1 and node 0 - node 3 - node 2: T' d31 , T' d32 , and replace T' d01 in the foregoing fourth set with T' d31 , and replace T' d02 with T'd32 , so the updated fifth set is as follows:

[0234] {T′ d31 , T′ d32 , T′ d03 , T′ d04 , T′ d05 , T′ d06 , T′ d07 , T′ d08 , T′ d09 , T′ d010}.

[0235] In summary, the specific implementation of the multi-task path determination method in the embodiments of the present application can be described as follows:

[0236] The multi-path determination method in the embodiments of the present invention mainly includes three parts. The first part is to calculate the cost increment when the task is transmitted from one network node to another network node and the current task is executed; the second part is the process of finding the path with the minimum cost increment for a single task; the third part is the process of calculating the total cost of all tasks.

[0237] Specifically, the first part includes the following content:

[0238] When the current task is sent from node c to node z, the total cost increment of all tasks is ΔP cz , ΔP cz = a * ΔT cz + b * P z_rec . Where P z_rec represents the reconstruction cost when node z executes the current task. If node z cannot successfully execute the task, then P z_rec is positive infinity; ΔT cz represents the change in the maximum delay in the multi-task, and the calculation formula is where T′ dcz = max(T 1d , T 2d , …, T (i-1)d , T 0c ), represents the maximum value of the delay upper limit of the previous i - 1 tasks that have found paths on their respective paths and the delay passed by the previous node (i.e., node c) on this forwarding path. T 1d represents the delay of the first path when the first task is executed, T 2d represents the delay of the second path when the second task is executed, T (i-1)d represents the delay of the (i - 1)-th path when the (i - 1)-th task is executed, T 0c represents the delay from the task start point to node c when the current task is executed, and e cz represents the delay from node c to node z when the current task is executed. Therefore, (T0c +e cz ) represents the time delay from the starting node to node z when executing the current task. If this time delay is greater than T′ dcz , then the maximum time delay needs to be updated, generating a corresponding time delay cost increment.

[0239] Among them, the content in the first part of the step can be written into a function, and corresponding calculations are performed when the function needs to be called.

[0240] The content of the second part is the determination process of the minimum cost increment path of a single task based on the Dijkstra algorithm, as Figure 4 shown, specifically including the following (1) to (7):

[0241] (1) Determine the starting point and ending point of the task, and initialize sets M, N, and U, that is, define sets M, N, and U. Set M records the traversed nodes, that is, the nodes for which the minimum cost increment path from the starting point to this point has been found. Set N records the un-traversed nodes, and set U records T in the first part mentioned above 0c .

[0242] Initially, set M includes the starting point, set N contains other nodes except the starting point, and the value of set U is the time delay cost of the direct connection link from the starting point to other nodes, T 00 = 0. When there is no direct connection link between the starting point and a certain node, the value of T 0c is infinite.

[0243] (2) Calculate the one-way link cost increment from the starting point of the task to other nodes, and record the results in set P, and update set V, that is:

[0244] Define set P to record the cost increment from the starting point to other nodes when executing the current task, that is, set P initially records ΔP 0c , where 0 represents the starting point and c represents any node in set N.

[0245] Define set V to record T′ in the first part mentioned above dcz , that is, set V initially records T′ d0c , where 0 represents the starting point and c represents any node in set N.

[0246] (3) Traverse set N to find the node c with the minimum cost in P, that is, find a node c such that the value corresponding to node c in set P is the smallest, that is, among the un-traversed nodes, the cost increment from the starting point node to this node is the smallest.

[0247] (4) If node c is not the ending point, it means that the traversal is not complete yet, and go to step (5). If node c is the ending point, go to step (7).

[0248] (5) With node c as the intermediate node, update the total cost from the starting node to other nodes in set P in the set, and at the same time update the values in set U and set V, that is:

[0249] For other nodes f in set N, calculate the cost from the starting point to node f passing through node c in the middle, that is, ΔP 0c +ΔP cf , ΔP cf Calculate according to the content of the first part mentioned above. If ΔP 0c +ΔP cf is smaller than the value ΔP 0f corresponding to node f in set P, then use ΔP 0c +ΔP cf to update the value ΔP 0f in set P, and use T′ cf obtained during the calculation of ΔP dcf to update the value T′ d0f corresponding to node f in set V, and use (T 0c +e cf ) to update the value T 0f corresponding to node f in set U.

[0250] (6) Remove node c from set N and add node f to set M. Go to step (3).

[0251] (7) If node c is the end point, it means that the minimum cost increment path from the starting point to the end point has been found, then the cost and delay upper limit from the task starting point to the task end point can be output, that is, the cost of this path is the cost from the starting point to the end point corresponding in set P, and the delay upper limit is the delay upper limit corresponding in set V.

[0252] Among them, the content of the second part mentioned above can be written into a function, and then the path finding for each task in multiple tasks can be carried out by means of function call.

[0253] As Figure 5 shown, the third part includes the following steps (8) to (12):

[0254] (8) Determine the task calculation order. The rule of the calculation order is not limited. For example, it can be arranged from small to large according to the total task resource requirements.

[0255] (9) Select tasks according to the calculation order, and find the path through the single-task minimum cost increment path algorithm (that is, find the path by using the content described in the second part and the first part mentioned above).

[0256] (10) Judge whether all tasks have been calculated. If they have been calculated, go to step (12), otherwise go to the previous step (11).

[0257] (11) Calculate the remaining resources of the computing node, that is, the resources remaining after the computing network node finishes processing the previous task, and go to step (9).

[0258] (12) Output the total cost of multiple tasks, that is, calculate the total cost when all tasks are processed simultaneously, and the value of the cost is the sum of the cost increments calculated for all tasks.

[0259] In addition, the multi-task path determination method of the embodiment of the present application is adopted to simulate the process in which multiple tasks need to be sent from one ground node to another ground node at a certain static time slice.

[0260] As Figure 3 shown, assume that a certain task needs to be sent from the starting ground station, relayed by a satellite, and finally reach the destination ground station. For a task with high resource requirements, the task can be decomposed into multiple subtasks with lower requirements at the starting ground station (node 0), so as to improve the utilization rate of network resources. The subtasks are merged at the terminating ground station (node 10).

[0261] Among them, the situation of simultaneously sending multiple tasks is the same as that of task decomposition. As Figure 1 shown, the goal of the simulation is to find a path for each task between node 0 and node 10 so that the total cost of multiple tasks is minimized. Each task may pass through the same path or different paths. A satellite may need to process multiple tasks simultaneously or only process one task.

[0262] For the sake of simplicity, the mobility of the satellite is not considered in the simulation process, and only the data communication process at a static time slice is considered. Assume that the task completion time is relatively fast, so the changes in the inter-satellite link and the satellite-ground link during the task execution process are not considered. Assume that the connection state of the network at a certain moment is as Figure 6 shown. Among them, nodes numbered 1-9 represent satellites in the space-based network, node 0 represents the ground station accessed by users, and node 10 represents the ground station where the task finally arrives.

[0263] Since the processing capacity of the ground station is relatively strong, the situation where nodes 0 and 10 cannot meet the task requirements is not considered. In specific implementation, it is set that the computing and storage resources of nodes 0 and 10 are infinite. During the simulation process, it is set that the reconstruction cost weight values of each satellite are all 1. Then all tasks are executed according to the content of the foregoing first to third parts, and the total execution cost of multiple tasks can be calculated.

[0264] Among them, the change of the total cost of the paths found for multiple tasks by using the multi-task path determination method of the embodiment of the present application with the total resource demand of multiple tasks can be as Figure 7 shown. Among them,Figure 7 The abscissa represents the total resource requirements of multiple tasks, and the ordinate represents the total cost of multiple tasks.

[0265] In summary, in the embodiments of the present application, complex problems are decomposed, and the multi-task pathfinding process is decomposed into a single-task pathfinding process. When performing pathfinding for each task, the link cost unifies the calculation of the delay cost and the reconstruction cost. During the calculation process, there is no need to further optimize the preliminary calculation results by means of "shuffling" as in the prior art described above, and the logic is relatively clear. Among them, since the embodiments of the present application decompose complex problems, effective support can be provided when the TRS model is further expanded. In addition, the embodiments of the present application adopt a greedy algorithm, and for the process of pathfinding for each task, the optimal solution has been achieved. Based on the single-task pathfinding solution, the optimal algorithm for multi-task pathfinding can be further explored.

[0266] In addition, in order to more clearly understand the method of the embodiments of the present application, the following is an introduction by way of examples:

[0267] Currently, if there are three tasks to be executed, namely the first task, the second task, and the third task, and the starting network nodes of these tasks are node 0 and the destination network node is node 10, and the pre-determined network nodes include nodes 0 to 10, then first sort the first task, the second task, and the third task in ascending order according to the required resource amount. If the sorting is: the first task, the second task, and the third task, then first determine the process of the first path with the smallest cost increment when executing the first task from the node as follows:

[0268] First, create a set M to store the nodes that have found paths, and create a set N to store the nodes that have not found paths. Then the initial set M is {node 0}, and the set N is {node 1, node 2, node 3, node 4, node 5, node 6, node 7, node 8, node 9, node 10};

[0269] Secondly, since the starting network node of the task is node 0, the cost increments of the 10 paths of node 0-node 1, node 0-node 2, node 0-node 3, node 0-node 4, node 0-node 5, node 0-node 6, node 0-node 7, node 0-node 8, node 0-node 9, and node 0-node 10 when performing the first task can be calculated and stored in the set P. For example, the set P is {ΔP 01 、ΔP 02 、ΔP 03 、ΔP 04 、ΔP 05 、ΔP 06 、ΔP 07 、ΔP 08 、ΔP 09, ΔP 010};

[0270] And, calculate the latency when each of these 10 paths executes the first task and store it in set U. For example, at this time, set U is: {T 01 , T 02 , T 03 , T 04 , T 05 , T 06 , T 07 , T 08 , T 09 , T 010};

[0271] And, calculate the fifth parameter of these 10 paths according to the formula and store it in set V. For example, at this time, set V is {T' , T' d01 , T' d02 , T' d03 , T' d04 , T' d05 , T' d06 , T' d07 , T' d08 , T' d09 , T' d010};

[0272] Again, select the minimum value in set P. If it is ΔP 03 , then the path found for the first task is from node 0 to node 3. Among them, node 3 is not the destination node. Therefore, it is necessary to calculate the cost increment when these 9 paths, namely node 0 - node 3 - node 1, node 0 - node 3 - node 2, node 0 - node 3 - node 4, node 0 - node 3 - node 5, node 0 - node 3 - node 6, node 0 - node 3 - node 7, node 0 - node 3 - node 8, node 0 - node 3 - node 9, node 0 - node 3 - node 10, execute the first task, denoted as ΔP 03 +ΔP 31 , ΔP 03 +ΔP 32 , ΔP 03 +ΔP 34 , ΔP 03 +ΔP 35 , ΔP 03 +ΔP 36 , ΔP 03 +ΔP 37 , ΔP 03 +ΔP 38 , ΔP 03 +ΔP 39 , ΔP 03 +ΔP310 ;

[0273] Again, compare ΔP 03 +ΔP 31 with ΔP 01 , the magnitudes of ΔP 03 +ΔP 32 and ΔP 02 , the magnitudes of ΔP 03 +ΔP 34 and ΔP 04 , the magnitudes of ΔP 03 +ΔP 35 and ΔP 05 , the magnitudes of ΔP 03 +ΔP 36 and ΔP 06 , the magnitudes of ΔP 03 +ΔP 37 and ΔP 07 , the magnitudes of ΔP 03 +ΔP 38 and ΔP 08 , the magnitudes of ΔP 03 +ΔP 39 and ΔP 09 , the magnitudes of ΔP 03 +ΔP 310 and ΔP 010 , and replace the larger one in the current set P with the smaller one;

[0274] Among them, if ΔP 03 +ΔP 31 is less than ΔP 01 , and ΔP 03 +ΔP 32 is less than ΔP 02 , then the first updated set P is obtained: {ΔP 03 +ΔP 31 , ΔP 03 +ΔP 32 , ΔP 03 , ΔP 04 , ΔP 05 , ΔP 06 , ΔP 07 , ΔP 08 , ΔP 09 , ΔP 010};

[0275] And synchronously update the set U to {T 03 +T 31 , T 03 +T 32 , T 03 , T 04 , T 05, T 06 , T 07 , T 08 , T 09 , T 010}, update the set V to {T′ d31 , T′ d32 , T′ d03 , T′ d04 , T′ d05 , T′ d06 , T′ d07 , T′ d08 , T′ d09 , T′ d010};

[0276] Again, remove node 3 from the set N and add node 3 to the set M;

[0277] Again, from the set P after the first update, find the minimum value until the end node of the path to which the minimum value of the set P belongs is the destination network node, and obtain the first path with the smallest cost increment when executing the first task.

[0278] Thereafter, referring to the foregoing process and according to the remaining resources of the nodes, respectively determine the second path with the smallest cost increment when executing the second task and the third path with the smallest cost increment when executing the third task among nodes 0 to 10.

[0279] The above introduces the multi-task path determination method provided by the embodiments of the present application. Next, the multi-task path determination device provided by the embodiments of the present application will be introduced with reference to the accompanying drawings.

[0280] See Figure 8 , the embodiments of the present application also provide a multi-task path determination device, and the device includes:

[0281] A task acquisition module 801 for acquiring multiple tasks to be executed;

[0282] A path determination module 802 for determining the first path with the smallest cost increment when executing the first task among the pre-determined network nodes, where the first path includes at least one of the pre-determined network nodes;

[0283] A remaining resource determination module 803 for determining the first remaining resources of the pre-determined network nodes according to the resources required to execute the first task by the first path;

[0284] The path determination module 802 is further configured to, for the i-th task among the multiple tasks, determine the i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes according to the (i - 1)-th remaining resources;

[0285] The remaining resource determination module 803 is further configured to determine the i-th remaining resource of the pre-determined network node according to the resources required for the i-th task to be executed on the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and i takes each integer from 2 to n, and n represents the number of the multiple tasks.

[0286] Optionally, the path determination module 802 includes:

[0287] A path determination sub-module, configured to determine, based on Dijkstra's algorithm and according to the (i - 1)-th remaining resource, the i-th path with the minimum cost increment when executing the i-th task among the pre-determined network nodes.

[0288] Optionally, the path determination sub-module is specifically configured to:

[0289] Determine the start network node and the destination network node of the multiple tasks from the pre-determined network nodes;

[0290] Obtain the path from the start network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network nodes include the network nodes other than the start network node among the pre-determined network nodes, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes;

[0291] According to the (i - 1)-th remaining resource, select the first target path with the minimum cost increment when executing the i-th task from the first candidate paths;

[0292] When the end of the first target path is the destination network node, determine the first target path as the i-th path.

[0293] Optionally, the path determination sub-module is further configured to:

[0294] When the node at the end of the first target path is not the destination network node, obtain the path formed after adding the j2-th second candidate network node to the end of the first target path as the j2-th second candidate path, where the second candidate network nodes include the network nodes other than the network nodes on the first target path among the pre-determined network nodes, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes;

[0295] According to the (i - 1)-th remaining resource, select the second target path with the minimum cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths;

[0296] When the destination network node is at the end of the second target path, determine the second target path as the i-th path.

[0297] Optionally, when the path determination sub-module selects the first target path with the smallest cost increment when executing the i-th task from the first candidate paths according to the remaining resources of the (i - 1)-th, it specifically is used for:

[0298] According to the remaining resources of the (i - 1)-th, calculate the cost increment of each first candidate path when executing the i-th task, and store it as the target parameter of the first candidate path in a pre-established first set;

[0299] Obtain the first minimum value in the first set, and determine the first candidate path to which the first minimum value belongs as the first target path.

[0300] Optionally, when the path determination sub-module selects the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths according to the remaining resources of the (i - 1)-th, it specifically is used for:

[0301] According to the remaining resources of the (i - 1)-th, calculate the cost increment of each second candidate path when executing the i-th task to be used as the target parameter of the second candidate path;

[0302] In the case where the target parameter of the second candidate path is less than the target parameter of the first candidate path associated with the second candidate path, replace the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, where the first candidate path and the second candidate path with the same end node are associated;

[0303] Obtain the second minimum value in the first set, and determine the path to which the second minimum value belongs as the second target path.

[0304] Optionally, the device further includes:

[0305] A first storage module, configured to store the starting network node in a pre-established second set;

[0306] A second storage module, configured to store the network nodes other than the starting network node in the pre-determined network nodes in a pre-established third set;

[0307] The first update module is configured to delete the network node at the end of the first target path from the third set and store the network node at the end of the first target path in the second set; delete the network node at the end of the second target path from the third set and store the network node at the end of the second target path in the second set.

[0308] Optionally, when calculating the cost increment of each of the first candidate paths for executing the i-th task according to the (i - 1)-th remaining resource, the path determination sub-module is specifically configured to:

[0309] Determine the reconstruction cost when the end node on the j1-th first candidate path executes the i-th task according to the (i - 1)-th remaining resource and the resources required for executing the i-th task

[0310] According to the first preset formula Calculate the first parameter of the j1-th first candidate path Where T 00 = 0, T (i-1)d represents the time delay of the (i - 1)-th path when executing the (i - 1)-th task, represents the time delay of the j1-th first candidate path when executing the i-th task;

[0311] According to the second preset formula: Calculate the cost increment of the j1-th first candidate path when executing the i-th task Where a and b are pre-determined weight values.

[0312] Optionally, when calculating the cost increment of each of the second candidate paths for executing the i-th task according to the (i - 1)-th remaining resource, the path determination sub-module is specifically configured to:

[0313] Calculate the first cost increment of the first target path when executing the i-th task according to the (i - 1)-th remaining resource;

[0314] Calculate the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task according to the (i - 1)-th remaining resource, and use it as the j2-th second cost increment;

[0315] Determine the sum of the first cost increment and the j2-th second cost increment as the cost increment of the j2-th second candidate path when executing the i-th task.

[0316] Optionally, when calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task according to the (i-1)-th remaining resource, the path determination sub-module is specifically configured to:

[0317] Determine the reconstruction cost when the j2-th second candidate network node executes the i-th task according to the (i-1)-th remaining resource and the resources required to execute the i-th task

[0318] According to the third preset formula:

[0319] Calculate the second parameter of the j2-th second candidate path Wherein, T (i-1)d represents the time delay of the (i-1)-th path when executing the (i-1)-th task, represents the time delay of the first target path when executing the i-th task, represents the time delay from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task;

[0320] According to the fourth preset formula: Calculate the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task Wherein, a and b are pre-determined weight values;, j M represents that the end node of the first target path is the j M -th first candidate network node.

[0321] Optionally, the apparatus further includes:

[0322] A third storage module, configured to calculate the time delay of the j1-th first candidate path when executing the i-th task and store it as the third parameter of the first candidate path in a pre-established fourth set;

[0323] An update object determination module, configured to determine a second candidate path associated with the first candidate path in which the target parameter in the first set is replaced, as a first path to be processed;

[0324] A first update parameter calculation module, configured to calculate the time delay of the first path to be processed when executing the i-th task, as the fourth parameter of the first path to be processed;

[0325] A second update module, configured to replace the third parameter of the first candidate path in which the target parameter in the first set is replaced in the fourth set with the fourth parameter of the first path to be processed associated with the first candidate path.

[0326] Optionally, the device further comprises:

[0327] A fourth storage module is used to store Calculate the fifth parameter of the j1th first candidate path and stored in the pre-established fifth set, where T 00 =0;

[0328] An update object determination module is used to determine a second candidate path associated with a first candidate path whose target parameter is replaced in the first set, as a first path to be processed;

[0329] The second update parameter calculation module is used to calculate the update parameter according to the fifth preset formula Calculate the sixth parameter of the first path to be processed

[0330] A third updating module, configured to replace the fifth parameter of the first candidate path whose target parameter is replaced in the first set in the fifth set with the sixth parameter of the first to-be-processed path associated with the first candidate path;

[0331] Among them, j M Indicates that the end node of the first target path is the jth M The first candidate network node.

[0332] Optionally, the device further comprises:

[0333] A sorting module is used to sort the multiple tasks.

[0334] As can be seen from the above, the embodiment of the present application adopts a greedy algorithm to find the optimal path for a single task for multiple tasks in turn according to the principle of the minimum cost increment of a single task, and after each task finds the optimal path for a single task, calculate the remaining resources of the satellite, so that the next task can find a path based on the remaining resources. Therefore, in the scenario where there are multiple tasks to be executed, the embodiment of the present application can make the cost of executing each task individually reach the local optimum and make more effective use of satellite resources.

[0335] It should be noted that the division of units in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0336] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0337] It should be noted here that the above-mentioned device provided in the embodiments of this application can implement all the method steps implemented in the above-mentioned method embodiments and can achieve the same technical effects. Therefore, the same parts and beneficial effects as those in the method embodiments will not be specifically described in this embodiment.

[0338] The embodiments of this application also provide a network device, including a memory 920, a transceiver 910, and a processor 900;

[0339] The memory 920 is used to store computer programs;

[0340] The transceiver 910 is used to receive and send data under the control of the processor 900;

[0341] The processor 900 is used to read the computer program in the memory 920 and perform the following operations:

[0342] Obtain multiple tasks to be executed;

[0343] In a pre-determined network node, determine a first path with the smallest cost increment when executing the first task, where the first path includes at least one of the pre-determined network nodes;

[0344] Determine the first remaining resources of the pre-determined network node according to the resources required to execute the first task on the first path;

[0345] For the i-th task among the multiple tasks, according to the (i - 1)-th remaining resources, in the pre-determined network nodes, determine the i-th path with the smallest cost increment when executing the i-th task, and determine the i-th remaining resources of the pre-determined network nodes according to the resources required to execute the i-th task along the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and i takes each integer from 2 to n, and n represents the number of the multiple tasks.

[0346] Optionally, when the processor 900 determines the i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes according to the (i - 1)-th remaining resources, it is specifically configured to:

[0347] Based on Dijkstra's algorithm, and according to the (i - 1)-th remaining resources, determine the i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes.

[0348] Optionally, when the processor 900 determines the i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes based on Dijkstra's algorithm and according to the (i - 1)-th remaining resources, it is specifically configured to:

[0349] Determine the start network node and the destination network node of the multiple tasks from the pre-determined network nodes;

[0350] Obtain the path from the start network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network nodes include the network nodes other than the start network node among the pre-determined network nodes, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes;

[0351] According to the (i - 1)-th remaining resources, select the first target path with the smallest cost increment when executing the i-th task from the first candidate paths;

[0352] When the destination network node is at the end of the first target path, determine the first target path as the i-th path.

[0353] Optionally, when the processor 900 determines the i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes based on Dijkstra's algorithm and according to the (i - 1)-th remaining resources, it is further configured to:

[0354] When the node at the end of the first target path is not the destination network node, obtain the path formed after adding the j2-th second candidate network node to the end of the first target path, and use it as the j2-th second candidate path, where the second candidate network nodes include the network nodes in the pre-determined network nodes except the network nodes on the first target path, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes;

[0355] According to the i-1 remaining resources, select the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path in the first candidate paths;

[0356] When the end of the second target path is the destination network node, determine the second target path as the i-th path.

[0357] Optionally, when the processor 900 selects the first target path with the smallest cost increment when executing the i-th task from the first candidate paths according to the i-1 remaining resources, it is specifically used for:

[0358] According to the i-1 remaining resources, calculate the cost increment of each first candidate path when executing the i-th task, and store it as the target parameter of the first candidate path in a pre-established first set;

[0359] Obtain the first minimum value in the first set, and determine the first candidate path to which the first minimum value belongs as the first target path.

[0360] Optionally, when the processor 900 selects the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path in the first candidate paths according to the i-1 remaining resources, it is specifically used for:

[0361] According to the i-1 remaining resources, calculate the cost increment of each second candidate path when executing the i-th task, and use it as the target parameter of the second candidate path;

[0362] In the case where the target parameter of the second candidate path is less than the target parameter of the first candidate path associated with the second candidate path, replace the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, where the first candidate path and the second candidate path with the same end node are associated;

[0363] Obtain the second minimum value in the first set, and determine the path to which the second minimum value belongs as the second target path.

[0364] Optionally, before selecting, according to the i-1 remaining resources, the first target path with the smallest cost increment for executing the i-th task from the first candidate paths, the processor 900 is further configured to:

[0365] Store the starting network node in a pre-established second set;

[0366] Store the network nodes other than the starting network node in the pre-determined network nodes in a pre-established third set;

[0367] After selecting the first target path with the smallest cost increment for executing the i-th task, the processor 900 is further configured to:

[0368] Delete the network node at the end of the first target path from the third set, and store the network node at the end of the first target path in the second set;

[0369] After selecting the second target path with the smallest cost increment for executing the i-th task, the processor 900 is further configured to:

[0370] Delete the network node at the end of the second target path from the third set, and store the network node at the end of the second target path in the second set.

[0371] Optionally, when calculating, according to the i-1 remaining resources, the cost increment of each of the first candidate paths for executing the i-th task, the processor 900 is specifically configured to:

[0372] Determine the reconstruction cost when the end node on the j1-th first candidate path executes the i-th task according to the i-1 remaining resources and the resources required for executing the i-th task

[0373] According to the first preset formula Calculate the first parameter of the j1-th first candidate path where T 00 = 0, T (i-1)d represents the delay of the i-1 path when executing the i-1-th task, represents the delay of the j1-th first candidate path when executing the i-th task;

[0374] According to the second preset formula: Calculate the cost increment of the j1-th first candidate path when executing the i-th task where a and b are pre-determined weight values.

[0375] Optionally, when calculating the cost increment of each of the second candidate paths for executing the i-th task according to the (i-1)-th remaining resource, the processor 900 is specifically configured to:

[0376] Calculate a first cost increment of the first target path for executing the i-th task according to the (i-1)-th remaining resource;

[0377] Calculate a cost increment from the node at the end of the first target path to the j2-th second candidate network node for executing the i-th task according to the (i-1)-th remaining resource, and use it as the j2-th second cost increment;

[0378] Determine the sum of the first cost increment and the j2-th second cost increment as the cost increment of the j2-th second candidate path for executing the i-th task.

[0379] Optionally, when calculating the cost increment from the node at the end of the first target path to the j2-th second candidate network node for executing the i-th task according to the (i-1)-th remaining resource, the processor 900 is specifically configured to:

[0380] Determine the reconstruction cost of the j2-th second candidate network node for executing the i-th task according to the (i-1)-th remaining resource and the resources required for executing the i-th task

[0381] According to a third preset formula:

[0382] Calculate a second parameter of the j2-th second candidate path Where T (i-1)d Represents the delay of the (i-1)-th path when executing the (i-1)-th task, Represents the delay of the first target path when executing the i-th task, Represents the delay from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task;

[0383] According to a fourth preset formula: Calculate the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task Where a and b are pre-determined weight values, and j M Represents that the end node of the first target path is the j M -th first candidate network node.

[0384] Optionally, the processor 900 is further configured to:

[0385] Calculate the latency of the j1-th first candidate path when executing the i-th task And store it as the third parameter of the first candidate path in the pre-established fourth set;

[0386] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the processor 900 is further configured to:

[0387] Determine the second candidate path associated with the first candidate path whose target parameter in the first set is replaced, as the first path to be processed;

[0388] Calculate the latency of the first path to be processed when executing the i-th task, as the fourth parameter of the first path to be processed;

[0389] Replace the third parameter of the first candidate path whose target parameter in the first set is replaced in the fourth set with the fourth parameter of the first path to be processed associated with the first candidate path.

[0390] Optionally, the processor 900 is further configured to:

[0391] According to the fourth preset formula Calculate the fifth parameter of the j1-th first candidate path And store it in the pre-established fifth set, where T 00 = 0;

[0392] Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the processor 900 is further configured to:

[0393] Determine the second candidate path associated with the first candidate path whose target parameter in the first set is replaced, as the first path to be processed;

[0394] According to the fifth preset formula Calculate the sixth parameter of the first path to be processed

[0395] Replace the fifth parameter of the first candidate path whose target parameter in the first set is replaced in the fifth set with the sixth parameter of the first path to be processed associated with the first candidate path;

[0396] Where j M Indicates that the end node of the first target path is the j M -th first candidate network node.

[0397] Optionally, before determining, among the pre-determined network nodes, the first path with the smallest cost increment for executing the first task, the processor 900 is further configured to

[0398] sort the multiple tasks.

[0399] Among them, in Figure 9 the bus architecture may include any number of interconnected buses and bridges, specifically various circuits represented by one or more processors represented by the processor 900 and the memory represented by the memory 920 are linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and thus will not be further described herein. The bus interface provides an interface. The transceiver 910 may be multiple elements, that is, including a transmitter and a receiver, and provides a unit for communicating with various other devices on a transmission medium, and these transmission media include wireless channels, wired channels, optical fiber cables, and other transmission media. The processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 may store data used by the processor 900 when executing operations.

[0400] The processor 900 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD), and the processor 900 may also adopt a multi-core architecture.

[0401] It should be noted here that the above device provided by the embodiments of the present application can implement all the method steps implemented by the above method embodiments, and can achieve the same technical effects. The same parts and beneficial effects as those in the method embodiments will not be specifically described in this embodiment.

[0402] The embodiments of the present application further provide a processor-readable storage medium, and the processor-readable storage medium stores a computer program, and the computer program is used to cause the processor to execute the multi-task path determination method described above.

[0403] The processor-readable storage medium can be any available medium or data storage device accessible by the processor, including but not limited to magnetic memories (such as floppy disks, hard disks, magnetic tapes, magneto-optical discs (MO), etc.), optical memories (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor memories (such as ROM, EPROM, EEPROM, non-volatile memories (NANDFLASH), solid state drives (SSD)), etc.

[0404] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories and optical memories, etc.) containing computer-usable program code.

[0405] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0406] These processor-executable instructions can also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the processor-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0407] These processor-executable instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0408] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these modifications and variations.

Claims

1. A multi-task path determination method, characterized in that, The method includes: Obtaining a plurality of tasks to be executed; Determining a first path with the smallest cost increment when executing the first task among pre-determined network nodes, where the first path includes at least one of the pre-determined network nodes; Determining first remaining resources of the pre-determined network nodes according to resources required for executing the first task along the first path; For the i-th task among the plurality of tasks, based on Dijkstra's algorithm and according to the (i - 1)-th remaining resources, determining an i-th path with the smallest cost increment when executing the i-th task among the pre-determined network nodes, and determining i-th remaining resources of the pre-determined network nodes according to resources required for executing the i-th task along the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and i takes each integer from 2 to n, and n represents the number of the plurality of tasks; where the process of determining the i-th path with the smallest cost increment includes: determining a start network node and a destination network node of the plurality of tasks from the pre-determined network nodes; obtaining a path from the start network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network nodes include network nodes among the pre-determined network nodes other than the start network node, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes; according to the (i - 1)-th remaining resources, selecting a first target path with the smallest cost increment when executing the i-th task from the first candidate paths; when the end of the first target path is the destination network node, determining the first target path as the i-th path; when the node at the end of the first target path is not the destination network node, obtaining a path formed by adding the j2-th second candidate network node to the end of the first target path as the j2-th second candidate path, where the second candidate network nodes include network nodes among the pre-determined network nodes other than the network nodes on the first target path, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes; according to the (i - 1)-th remaining resources, selecting a second target path with the smallest cost increment when executing the i-th task from the second candidate paths and paths among the first candidate paths other than the first target path; when the end of the second target path is the destination network node, determining the second target path as the i-th path.

2. The method according to claim 1, wherein The step of selecting a first target path with the smallest cost increment when executing the i-th task from the first candidate paths according to the (i - 1)-th remaining resources includes: Calculating a cost increment of each first candidate path when executing the i-th task according to the (i - 1)-th remaining resources, and storing it as a target parameter of the first candidate path in a pre-established first set; Obtaining a first minimum value in the first set, and determining the first candidate path to which the first minimum value belongs as the first target path.

3. The method according to claim 2, wherein Selecting, according to the (i - 1)-th remaining resource, a second target path with the minimum cost increment when performing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths, includes: Calculating, according to the (i - 1)-th remaining resource, the cost increment of each second candidate path when performing the i-th task as the target parameter of the second candidate path; When the target parameter of the second candidate path is less than the target parameter of the first candidate path associated with the second candidate path, replacing, in the first set, the target parameter of the first candidate path with the target parameter of the second candidate path associated with the first candidate path, where the first candidate path and the second candidate path with the same end node are associated; Obtaining the second minimum value in the first set and determining the path to which the second minimum value belongs as the second target path.

4. The method according to claim 1, characterized in that, Before selecting, according to the (i - 1)-th remaining resource, a first target path with the minimum cost increment when performing the i-th task from the first candidate paths, the method further includes: Storing the starting network node in a pre-established second set; Storing the network nodes other than the starting network node in the pre-determined network nodes in a pre-established third set; After selecting a first target path with the minimum cost increment when performing the i-th task, the method further includes: Deleting the network node at the end of the first target path from the third set and storing the network node at the end of the first target path in the second set; After selecting a second target path with the minimum cost increment when performing the i-th task, the method further includes: Deleting the network node at the end of the second target path from the third set and storing the network node at the end of the second target path in the second set.

5. The method according to claim 2, wherein Calculating, according to the (i - 1)-th remaining resource, the cost increment of each of the first candidate paths when performing the i-th task, includes: Determine the reconstruction cost when the end node on the j1-th first candidate path executes the i-th task according to the (i - 1)-th remaining resource and the resources required to execute the i-th task According to the first preset formula Calculate the first parameter of the j1-th first candidate path where T 00 = 0, T (i-1)d represents the latency of the (i - 1)-th path when executing the (i - 1)-th task, represents the latency of the j1-th first candidate path when executing the i-th task; According to the second preset formula: Calculate the cost increment of the j1-th first candidate path when executing the i-th task where a and b are pre-determined weight values.

6. The method according to claim 3, wherein Calculating, according to the (i - 1)-th remaining resource, the cost increment of each of the second candidate paths when performing the i-th task, includes: Calculating, according to the (i - 1)-th remaining resource, the first cost increment of the first target path when performing the i-th task; Calculating, according to the (i - 1)-th remaining resource, the cost increment from the node at the end of the first target path to the j2-th second candidate network node when performing the i-th task as the j2-th second cost increment; Determining the sum of the first cost increment and the j2-th second cost increment as the cost increment of the j2-th second candidate path when performing the i-th task.

7. The method according to claim 6, wherein Calculating, according to the (i - 1)-th remaining resource, the cost increment from the node at the end of the first target path to the j2-th second candidate network node when performing the i-th task, includes: Determine the reconstruction cost when the j2-th second candidate network node executes the i-th task according to the (i-1)-th remaining resource and the resources required to execute the i-th task According to a third preset formula: Calculate the second parameter of the j2-th second candidate path where T (i-1)d represents the delay of the (i - 1)-th path when executing the (i - 1)-th task represents the delay of the first target path when executing the i-th task represents the delay from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task, j M indicates that the end node of the first target path is the j M -th first candidate network node; According to the fourth preset formula: Calculate the cost increment from the node at the end of the first target path to the j2-th second candidate network node when executing the i-th task where a and b are pre-determined weight values.

8. The method according to claim 3, wherein The method further includes: Calculate the latency of the j1-th first candidate path when executing the i-th task And store it as the third parameter of the first candidate path in a pre-established fourth set; Wherein, after replacing, in the first set, the target parameter of the first candidate path with the target parameter of the second candidate path associated with the first candidate path, the method further includes: Determine a second candidate path associated with the first candidate path in which the target parameter in the first set is replaced, as the first path to be processed; Calculate the latency of the first path to be processed when the i-th task is executed, as the fourth parameter of the first path to be processed; Replace the third parameter of the first candidate path in which the target parameter in the first set is replaced in the fourth set with the fourth parameter of the first path to be processed associated with the first candidate path.

9. The method according to claim 3, wherein The method further includes: According to the fourth preset formula Calculate the fifth parameter of the j1-th first candidate path And store it in the pre-established fifth set, where T 00 = 0; Wherein, after replacing the target parameter of the first candidate path in the first set with the target parameter of the second candidate path associated with the first candidate path, the method further includes: Determine a second candidate path associated with the first candidate path in which the target parameter in the first set is replaced, as the first path to be processed; According to the fifth preset formula Calculate the sixth parameter of the first path to be processed Replace the fifth parameter of the first candidate path in which the target parameter in the first set is replaced in the fifth set with the sixth parameter of the first path to be processed associated with the first candidate path; where j M indicates that the end node of the first target path is the jth M first candidate network node.

10. The method according to claim 1, characterized in that, Before determining, in the pre-determined network nodes, the first path with the smallest cost increment when the first task is executed, the method further includes: Sort the multiple tasks.

11. A multi-task path determination device, characterized in that, The apparatus includes: A task acquisition module, configured to acquire multiple tasks to be executed; A path determination module, configured to determine, in the pre-determined network nodes, the first path with the smallest cost increment when the first task is executed, where the first path includes at least one of the pre-determined network nodes; A remaining resource determination module, configured to determine the first remaining resources of the pre-determined network nodes according to the resources required for the first path to execute the first task; The path determination module is further configured to, for the i-th task among the multiple tasks, based on Dijkstra's algorithm and according to the i-1 remaining resources, determine the i-th path with the smallest cost increment when the i-th task is executed in the pre-determined network nodes; The remaining resource determination module is further configured to determine the i-th remaining resource of the pre-determined network node according to the resources required for the i-th task to be executed on the i-th path, where the i-th path includes at least one of the pre-determined network nodes, and i takes each integer from 2 to n, and n represents the number of the multiple tasks; wherein, the determination process of the i-th path with the smallest cost increment includes: determining the start network node and the destination network node of the multiple tasks from the pre-determined network nodes; obtaining the path from the start network node to the j1-th first candidate network node as the j1-th first candidate path, where the first candidate network node includes the network nodes other than the start network node among the pre-determined network nodes, and j1 takes each integer from 1 to m1, and m1 represents the number of the first candidate network nodes; according to the (i - 1)-th remaining resource, selecting the first target path with the smallest cost increment when executing the i-th task from the first candidate paths; when the end of the first target path is the destination network node, determining the first target path as the i-th path; when the node at the end of the first target path is not the destination network node, obtaining the path formed after adding the j2-th second candidate network node to the end of the first target path as the j2-th second candidate path, where the second candidate network node includes the network nodes other than the network nodes on the first target path among the pre-determined network nodes, and j2 takes each integer from 1 to m2, and m2 represents the number of the second candidate network nodes; according to the (i - 1)-th remaining resource, selecting the second target path with the smallest cost increment when executing the i-th task from the second candidate paths and the paths other than the first target path among the first candidate paths; when the end of the second target path is the destination network node, determining the second target path as the i-th path.

12. A network device, characterized in that, Comprising a memory, a transceiver, and a processor: The memory is used for storing a computer program; the transceiver is used for transceiving data under the control of the processor; the processor is used for executing the method according to any one of claims 1 to 10.

13. A processor-readable storage medium, characterized in that, The processor-readable storage medium stores a computer program, and the computer program is used for causing the processor to execute the method according to any one of claims 1 to 10.

Images