Satellite network deterministic scheduling method based on cross-layer multi-dimensional state information

Abstract

The invention discloses a satellite network deterministic scheduling method based on cross-layer multi-dimensional state information. The satellite network deterministic scheduling method comprises the following steps: S1, a satellite network management center acquires deterministic service requirements of access users; s2, establishing a space-time expansion model in the satellite network by using a time slicing method; s3, constructing a potential forwarding node set based on a spatio-temporal expansion model; s4, traversing and searching reachable paths between all source and destination nodes in the potential forwarding node set; s5, acquiring state parameters of link layers and network layers of the satellite nodes in all reachable paths, and establishing a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information; s6, calculating an error between the time delay of each reachable path and the deterministic time delay of the user task by taking the deterministic scheduling index parameters as input; and S7, taking the path with the minimum time delay error in all the deterministic scheduling paths obtained in the step S6 as the optimal deterministic scheduling path, namely the transmission path finally adopted by the user service.

Description

Technical Field

[0001] The invention relates to a satellite network deterministic scheduling method based on cross-layer multi-dimensional state information, belonging to the technical field of satellite communications. Background Art

[0002] With the continuous development of the integrated space-ground network, various civil and commercial application scenarios are connected to the satellite network, which has led to an increase in the types of services carried by the satellite network (such as 5G / 6G communication services, remote sensing services, computing services, etc.). The QoS services and latency requirements of multi-scenario and multi-type services are not the same. In particular, time-sensitive services require deterministic scheduling strategies to ensure bounded latency and latency jitter. The integrated space-ground network is affected by factors such as the asymmetry of the satellite-to-ground downlinks across terminal types and the stability of the inter-satellite links, which causes the satellite network to produce different channel states. The existing satellite network protocol is based on the TCP / IP protocol and evolved from the ground wired network. The functions of each layer of the protocol stack are relatively independent. The dynamic changes in the satellite network channel state, the dynamic nature of the satellite network topology, and the wireless transmission of the inter-satellite link cause changes in service rate and transmission latency, which makes it difficult to adapt to the guarantee capability requirements of deterministic services. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and meet the deterministic guarantee capability requirements of user services.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A satellite network deterministic scheduling method based on cross-layer multi-dimensional state information, firstly, the satellite network management center obtains the deterministic service demand of the access user, including the satellite source node, the destination node and the deterministic scheduling index parameters such as the deterministic delay and the delay jitter; secondly, the time-space expansion diagram is formed in the satellite network by using the time slicing method, based on which, based on the idea of ​​distance gradient descent in the service transmission direction, the vector distance between the source node and the destination node is used as the gradient direction of descent to screen the satellite nodes to form a potential forwarding node set; and all reachable paths between the source and destination nodes are traversed and searched in the potential forwarding node set; then, the link layer, network layer and other layer state parameters of the satellite nodes in all reachable paths are obtained; then, the state parameters of each layer and the deterministic scheduling index parameters are used as constraints to establish a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information; finally, the model is used for calculation, and the path with the smallest calculation output error is used as the optimal deterministic scheduling path for data transmission. If the optimal deterministic scheduling path is not obtained, it is determined that the deterministic scheduling of the task fails, and the task is re-planned or the range of the deterministic scheduling technical index is relaxed for recalculation. Ultimately, efficient cross-layer resource utilization and performance improvement of satellite networks can be achieved, ensuring deterministic QoS requirements for multiple users and multiple services.

[0006] Specifically include:

[0007] S1. When a user task accesses the satellite network, the satellite network management and control center obtains the deterministic service demand parameters of the user task, including the source satellite node S, the destination satellite node D, the task size M, and the deterministic delay t det , delay jitter error t jit ;

[0008] S2. Establish a satellite network spatiotemporal expansion graph model by:

[0009] S21. Divide the satellite network system operation period T into n consecutive and equal-sized time slots {1, ..., n}, and the time slot duration is defined as Δt = T / n;

[0010] S22. Along the direction of time growth, virtual links are established between different replica nodes of the same satellite in adjacent topologies to represent the process of data caching on the satellite, establish connections between adjacent topologies, and connect them in chronological order to form a steady-state space-time extension graph model.

[0011] S3. Construct a set of potential forwarding satellite nodes. The specific method is as follows:

[0012] S31. Calculate the vector distance between the source satellite node and the destination satellite node. The calculation method is: Based on the two lines of orbital data of the low-orbit satellite in a single time slot, the three-dimensional coordinates (xk (τ),y k (τ),z k (τ)), then the corresponding three-dimensional coordinates of the source node and the destination node are (x S (τ),y S (τ),z S (τ)) and (x D (τ),y D (τ),z D (τ)), thereby determining the vector distance between the source node and the destination node in the service, denoted as

[0013] S32. Use the vector distance as the descending gradient direction to screen which nodes should be included in the potential forwarding node set. Specifically: Combined with the directionality of the service transmission between the source satellite node and the destination satellite node, starting from the source satellite node in the time sequence of the time slot, take the source satellite node as the starting satellite node, search and traverse all adjacent satellite nodes of the starting satellite node, and compare the sum of the distances between the adjacent satellite node and the destination satellite node. If it is less than d(τ), the node is included in the forwarding node set; if it is greater than d(τ), turn to the next adjacent node and repeat the above process;

[0014] S33. Take the satellite node in the potential forwarding node set as the starting satellite node, and repeat step S32 until the node newly included in the potential forwarding node set is the destination satellite node.

[0015] S4. Perform path search in the potential forwarding node set to obtain the set of all reachable paths P(S, D) between the source satellite node and the destination satellite node. The specific method is as follows:

[0016] Starting from the source satellite node S, search each adjacent satellite node of the source node to determine whether the adjacent satellite node is the destination satellite node D. If so, stop searching and record the path at this time. If not, continue searching with the adjacent satellite node as the new starting point until all reachable satellite nodes of the source satellite node have been visited.

[0017] S5. Establish a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information, specifically:

[0018] S51. Obtain the cross-layer multi-dimensional state information of satellite nodes in all reachable paths, including the distance d between satellite nodes po (τ), inter-satellite link transmission rate link po (τ), the rate at which data packets arrive at satellite nodes λ(τ), and the rate at which satellite nodes process data packets μ(τ), where p,o is any adjacent satellite node in the reachable path;

[0019] S52. Based on the above state information, the propagation delay is calculated respectively Transmission delay Queuing delay t queue (τ), calculated as: c is the propagation speed of electromagnetic waves in vacuum; M is the size of the task; t queue The calculation formula of (τ) is a function related to the queuing model queue(p,o) adopted by the satellite node, the rate at which data packets arrive at the satellite node λ(τ), and the rate at which the satellite node processes data packets μ(τ), that is, t queue (τ)~(queue(p,o),λ(τ),μ(τ));

[0020] S53. Therefore, the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information can be expressed as Among them, path(S,D) represents a transmission path in the user task reachable path set P(S,D), S is the source satellite node, and D is the destination satellite node.

[0021] S6. Taking the deterministic scheduling joint optimization algorithm model t(path(S,D)) based on cross-layer multi-dimensional state information as input, calculate the error between the delay of each reachable path and the deterministic delay of the user task, and take the reachable path within the error range as the deterministic scheduling path, specifically:

[0022] S61. Calculate the delay t(path) of each reachable path in the reachable path set P(S,D) according to the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information. i (S,D)),i∈P(S,D);

[0023] S62. Calculate the error error between the delay of each reachable path and the deterministic delay requirement of the user task i (S,D)), where

[0024] S63. Determine the error error (path i (S, D)) is within the allowable range of service delay jitter error. If error(path i (S,D))≤t jit , then record the reachable path as the deterministic scheduling path of the user task; if error(path i (S,D))>t jit , then the reachable path is determined to be a non-deterministic scheduling path; return to step S62 to calculate the delay error of the next reachable path.

[0025] S7. The path with the smallest delay error among all the deterministic scheduling paths obtained in step S6 is taken as the optimal deterministic scheduling path, that is, the transmission path finally adopted by the user service; if the deterministic scheduling path is not obtained after calculation in step S6, it is determined that the deterministic scheduling of the user task has failed, and the user needs to re-plan the task and return to step S1 or relax the range of deterministic scheduling technical indicators and return to step S6 for recalculation.

[0026] A computer program product stored on a non-transitory computer-readable medium includes program code for implementing a satellite network deterministic scheduling method based on cross-layer multi-dimensional state information.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] (1) The present invention proposes a deterministic scheduling joint optimization method based on cross-layer multi-dimensional state information. It takes the parameters of each layer, such as the link layer and the network layer, as input to establish an end-to-end transmission delay model. At the same time, it takes the deterministic scheduling parameters of the user task as constraints to perform cross-layer deterministic scheduling joint optimization. The path with the smallest output error is calculated as the optimal deterministic scheduling path for data transmission, thus meeting the deterministic delay guarantee requirements of user services.

[0029] (2) The present invention proposes a method for constructing a set of potential forwarding satellite nodes. Based on the idea of ​​distance gradient descent, the vector distance between the source node and the destination node is used as the gradient direction of descent to screen satellite nodes and form a set of potential forwarding nodes, thereby reducing the number of satellite nodes in the deterministic path search and shortening the algorithm convergence time.

[0030] (3) The present invention proposes a method for establishing a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information, which comprehensively considers the parameters of each layer such as the link layer and the network layer as well as the state information such as the queue cache in the satellite node, so as to more accurately calculate the end-to-end delay of each path and realize the efficient utilization of satellite network resources.

[0031] (4) The present invention proposes a method for calculating a deterministic scheduling path, based on a deterministic scheduling joint optimization algorithm model of cross-layer multi-dimensional state information, with user deterministic delay parameters as constraints, which can optimally match user QoS requirements and realize deterministic scheduling of satellite network services. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The figure is a flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0034] A satellite network deterministic scheduling method based on cross-layer multi-dimensional state information, firstly, the satellite network management center obtains the deterministic service demand of the access user, including the satellite source node, the destination node and the deterministic scheduling index parameters such as the deterministic delay and the delay jitter; secondly, the time-space expansion diagram is formed in the satellite network by using the time slicing method, based on which, based on the idea of ​​distance gradient descent in the service transmission direction, the vector distance between the source node and the destination node is used as the gradient direction of descent to screen the satellite nodes to form a potential forwarding node set; and all reachable paths between the source and destination nodes are traversed and searched in the potential forwarding node set; then, the link layer, network layer and other layer state parameters of the satellite nodes in all reachable paths are obtained; then, the state parameters of each layer and the deterministic scheduling index parameters are used as constraints to establish a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information; finally, the model is used for calculation, and the path with the smallest calculation output error is used as the optimal deterministic scheduling path for data transmission. If the optimal deterministic scheduling path is not obtained, it is determined that the deterministic scheduling of the task fails, and the task is re-planned or the range of the deterministic scheduling technical index is relaxed for recalculation. Ultimately, efficient cross-layer resource utilization and performance improvement of satellite networks can be achieved, ensuring deterministic QoS requirements for multiple users and multiple services.

[0035] like Figure 1 As shown, specifically including:

[0036] S1. In this example, when a user task accesses the satellite network, the satellite network management and control center obtains the deterministic service demand parameters of the user task, including the source satellite node S, the destination satellite node D, the task size M, and the deterministic delay t det , delay jitter error t jit ;

[0037] S2. Establish a satellite network spatiotemporal expansion graph model by:

[0038] S21. Divide the satellite network system operation period T into n consecutive and equal-sized time slots {1, ..., n}, and the time slot duration is defined as Δt = T / n;

[0039] S22. Along the direction of time growth, virtual links are established between different replica nodes of the same satellite in adjacent topologies to represent the process of data caching on the satellite, establish connections between adjacent topologies, and connect them in chronological order to form a steady-state space-time extension graph model.

[0040] S3. Construct a set of potential forwarding satellite nodes. The specific method is as follows:

[0041] S31. Calculate the vector distance between the source satellite node and the destination satellite node. The calculation method is: Based on the two lines of orbital data of the low-orbit satellite in a single time slot, the three-dimensional coordinates (x k (τ),y k (τ),z k (τ)), then the corresponding three-dimensional coordinates of the source node and the destination node are (x S (τ),y S (τ),z S (τ)) and (x D (τ),y D (τ),z D (τ)), thereby determining the vector distance between the source node and the destination node in the service, denoted as

[0042] S32. Use the vector distance as the descending gradient direction to screen which nodes should be included in the potential forwarding node set. Specifically: Combined with the directionality of the service transmission between the source satellite node and the destination satellite node, starting from the source satellite node in the time sequence of the time slot, take the source satellite node as the starting satellite node, search and traverse all adjacent satellite nodes of the starting satellite node, and compare the sum of the distances between the adjacent satellite node and the destination satellite node. If it is less than d(τ), the node is included in the forwarding node set; if it is greater than d(τ), turn to the next adjacent node and repeat the above process;

[0043] S33. Take the satellite node in the potential forwarding node set as the starting satellite node, and repeat step S32 until the node newly included in the potential forwarding node set is the destination satellite node.

[0044] S4. Perform path search in the potential forwarding node set to obtain the set of all reachable paths P(S, D) between the source satellite node and the destination satellite node. The specific method is as follows:

[0045] Starting from the source satellite node S, search each adjacent satellite node of the source node to determine whether the adjacent satellite node is the destination satellite node D. If so, stop searching and record the path at this time. If not, continue searching with the adjacent satellite node as the new starting point until all reachable satellite nodes of the source satellite node have been visited.

[0046] S5. Establish a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information, specifically:

[0047] S51. Obtain the cross-layer multi-dimensional state information of satellite nodes in all reachable paths, including the distance d between satellite nodes po (τ), inter-satellite link transmission rate linkpo (τ), the rate at which data packets arrive at satellite nodes λ(τ), and the rate at which satellite nodes process data packets μ(τ), where p,o is any adjacent satellite node in the reachable path;

[0048] S52. Based on the above state information, the propagation delay is calculated respectively Transmission delay Queuing delay t queue (τ), calculated as: c is the propagation speed of electromagnetic waves in vacuum; M is the size of the task; the satellite node adopts the M / M / 1 / N queuing theory model, then the queuing delay W s is the sojourn time, the calculation method of which is specifically described in the M / M / 1 / N queueing theory and will not be repeated here;

[0049] S53. Therefore, the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information can be expressed as Among them, path(S,D) represents a transmission path in the user task reachable path set P(S,D), S is the source satellite node, and D is the destination satellite node.

[0050] S6. Taking the deterministic scheduling joint optimization algorithm model t(path(S,D)) based on cross-layer multi-dimensional state information as input, calculate the error between the delay of each reachable path and the deterministic delay of the user task, and take the reachable path within the error range as the deterministic scheduling path, specifically:

[0051] S61. Calculate the delay t(path) of each reachable path in the reachable path set P(S,D) according to the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information. i (S,D)),i∈P(S,D);

[0052] S62. Calculate the error error between the delay of each reachable path and the deterministic delay requirement of the user task i (S,D)), where

[0053] S63. Determine the error error (path i (S, D)) is within the allowable range of service delay jitter error. If error(path i (S,D))≤t jit , then the reachable path is used as the deterministic scheduling path of the user task; if error(path i (S,D))>t jit, then the reachable path is determined to be a non-deterministic scheduling path; return to step S62 to calculate the delay error of the next reachable path.

[0054] S7. The path with the smallest delay error among all the deterministic scheduling paths obtained in step S6 is taken as the optimal deterministic scheduling path, that is, the transmission path finally adopted by the user service; if the deterministic scheduling path is not obtained after calculation in step S6, it is determined that the deterministic scheduling of the user task has failed, and the user needs to re-plan the task and return to step S1 or relax the range of deterministic scheduling technical indicators and return to step S6 for recalculation.

[0055] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

[0056] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A satellite network deterministic scheduling method based on cross-layer multi-dimensional state information, characterized in that: include: S1. The satellite network management center obtains the deterministic service requirements of access users; S2, using the time slicing method to establish a spatiotemporal expansion model in satellite networks; S3, based on the space-time expansion model, based on the idea of ​​distance gradient descent in the service transmission direction, the vector distance between the source node and the destination node is used as the gradient direction of descent to screen satellite nodes and build a set of potential forwarding nodes; S4, traverse and search for all reachable paths between source and destination nodes in the potential forwarding node set; S5. Obtain the link layer and network layer state parameters of satellite nodes in all reachable paths; use the state parameters of each layer as the constraint conditions for delay calculation, and establish a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information; S6, taking the deterministic scheduling index parameter as input, calculating the error between the delay of each reachable path and the deterministic delay of the user task; S7. The path with the smallest delay error among all the deterministic scheduling paths obtained in step S6 is taken as the optimal deterministic scheduling path, that is, the transmission path ultimately adopted by the user service.

2. The satellite network deterministic scheduling method according to claim 1, characterized in that: The deterministic service requirements in S1 include satellite source nodes, destination nodes, deterministic delay and delay jitter.

3. The satellite network deterministic scheduling method according to claim 1, characterized in that: If the optimal deterministic scheduling path is not obtained in S6, it is determined that the deterministic scheduling of the task has failed, and the task is replanned or the deterministic scheduling technical indicator range is relaxed and recalculated.

4. The satellite network deterministic scheduling method according to claim 1, characterized in that: The method for establishing the satellite network spatiotemporal expansion graph model in S2 is: S21. Divide the satellite network system operation period T into n consecutive and equal-sized time slots {1, ..., n}, and the time slot duration is defined as Δt = T / n; S22. Along the direction of time growth, virtual links are established between different replica nodes of the same satellite in adjacent topologies to represent the process of data caching on the satellite, establish connections between adjacent topologies, and connect them in chronological order to form a steady-state space-time extension graph model.

5. The satellite network deterministic scheduling method according to claim 1, characterized in that: The calculation method of the vector distance between the source satellite node and the destination satellite node in S3 is as follows: Based on the two lines of orbital data of the low-orbit satellite in a single time slot, the three-dimensional coordinates (x k (τ),y k (τ),z k (τ)), then the corresponding three-dimensional coordinates of the source node and the destination node are (x S (τ),y S (τ),z S (τ)) and (x D (τ),y D (τ),z D (τ)), thereby determining the vector distance between the source node and the destination node in the service, denoted as 6. The satellite network deterministic scheduling method according to claim 5, characterized in that: In S3, the vector distance is used as the descending gradient direction to screen which nodes should be included in the potential forwarding node set. Specifically, in combination with the directionality of service transmission between the source satellite node and the destination satellite node, starting from the source satellite node in the time order of the time slot, the source satellite node is used as the starting satellite node, and all adjacent satellite nodes of the starting satellite node are searched and traversed, and the sum of the distances between the adjacent satellite node and the destination satellite node is compared. If it is less than d(τ), the node is included in the forwarding node set; if it is greater than d(τ), turn to the next adjacent node and repeat the above process; take the satellite node in the potential forwarding node set as the starting satellite node, and repeat the above steps until the node newly included in the potential forwarding set is the destination satellite node.

7. The satellite network deterministic scheduling method according to claim 1, characterized in that: In S4, the set of all reachable paths P(S, D) between the source satellite node and the destination satellite node is obtained. The specific method is: Starting from the source satellite node S, search each adjacent satellite node of the source node to determine whether the adjacent satellite node is the destination satellite node D. If so, stop searching and record the path at this time. If not, continue searching with the adjacent satellite node as the new starting point until all reachable satellite nodes of the source satellite node have been visited.

8. The satellite network deterministic scheduling method according to claim 1, characterized in that: In S5, a deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information is established as follows: S51. Obtain the cross-layer multi-dimensional state information of satellite nodes in all reachable paths, including the distance d between satellite nodes po (τ), inter-satellite link transmission rate link po (τ), the rate at which data packets arrive at satellite nodes λ(τ), and the rate at which satellite nodes process data packets μ(τ), where p,o is any adjacent satellite node in the reachable path; S52. Based on the above state information, the propagation delay is calculated respectively Transmission delay Queuing delay t queue (τ), calculated as: c is the propagation speed of electromagnetic waves in vacuum; M is the size of the task; the satellite node adopts the M / M / 1 / N queuing theory model, then the queuing delay W s for the duration of stay; S53. Therefore, the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information can be expressed as Among them, path(S,D) represents a transmission path in the user task reachable path set P(S,D), S is the source satellite node, and D is the destination satellite node.

9. The satellite network deterministic scheduling method according to claim 8, characterized in that: In S6, the deterministic scheduling joint optimization algorithm model t(path(S,D)) based on cross-layer multi-dimensional state information is used as input to calculate the error between the delay of each reachable path and the deterministic delay of the user task, and the reachable path within the error range is taken as the deterministic scheduling path, specifically: S61. Calculate the delay t(path) of each reachable path in the reachable path set P(S,D) according to the deterministic scheduling joint optimization algorithm model based on cross-layer multi-dimensional state information. i (S,D)),i∈P(S,D); S62. Calculate the error error between the delay of each reachable path and the deterministic delay requirement of the user task i (S,D)), where S63. Determine the error error (path i (S, D)) is within the allowable range of service delay jitter error. If error(path i (S,D))≤t jit , then record the reachable path as the deterministic scheduling path of the user task; if error(path i (S,D))>t jit , then the reachable path is determined to be a non-deterministic scheduling path; return to step S62 to calculate the delay error of the next reachable path.

10. A computer program product stored on a non-transitory computer readable medium, the computer program product comprising a program code for carrying out the method according to any one of claims 1 to 9.

Images