Task-oriented multi-layer low earth orbit satellite mega-constellation clustering networking method

Abstract

The application relates to a task-oriented multi-layer low-orbit satellite mega-constellation clustering networking method, which comprises the following steps: 1. initializing satellite constellation data, including satellite load resources, satellite geographical position and orbit information and satellite inter-satellite link establishment, and receiving task data by a control center located on the ground or in a high orbit satellite when a task is generated; 2. establishing a task model according to on-satellite resources required by the task and a satellite model according to orbit parameters and on-satellite load capacity data of the satellite; a specific task and satellite model establishment mechanism; 3. determining restriction conditions according to different task types, etc. The superior technical effect of the application lies in that the cooperation relationship between satellites and the matching relationship between the satellites and task requirements are combined through node enhancement graphs by mathematical modeling of the task and the satellites, thereby laying a foundation for task-oriented on-demand clustering.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of mobile communication, and particularly relates to a task-oriented multi-layer low-orbit satellite mega-constellation clustering networking method. BACKGROUND

[0002] Star communication networks have wide coverage and high flexibility, and have become an important part of the sixth generation mobile communication technology (6G). With the construction of satellite constellations, many satellite constellation construction plans such as Starlink and OneWeb have been launched, forming a ubiquitous multi-layer mega-satellite constellation in space. With the maturity of inter-satellite link technology and the enhancement of satellite payload capacity, satellite on-board self-organizing cluster networking has become a key research direction. In the future, the current situation of independent operation of satellite groups isolated from each other in terms of resources, functions and communication systems will be gradually broken. Flexible networking of distributed satellite groups based on on-board self-organization will become a trend. The core concept is that heterogeneous small satellites improve network performance through inter-satellite cooperation to form a large "virtual satellite". When the task demand changes or faces a sudden situation, the satellites already deployed in orbit can quickly form a new network to form new capabilities without launching new satellites. This way will significantly improve the reaction speed and flexibility in response to sudden demand.

[0003] Multi-layer low-orbit satellite mega-constellation network refers to a layered satellite network composed of a large number of low-earth orbit (LEO) satellites. The satellites in the network are usually in multiple different altitude and inclination orbits and have inter-satellite links. The networking forms between satellites are also different, including satellite constellation networking and satellite group formation networking. In addition, there are military and civilian satellites in the multi-layer low-orbit satellite mega-constellation network, and the communication systems and on-board payloads of these satellites are also different. Some satellites may have cross-domain communication capabilities, such as carrying cross-domain routing protocols and multi-beam laser links, and can connect with satellites of different communication systems and protocols. Although the multi-layer low-orbit satellite mega-constellation network has great application prospects, it also faces many challenges. Due to its huge scale, highly complex inter-satellite cooperation relationship, complex management between heterogeneous satellites and dynamically changing network topology, how to efficiently manage the entire multi-layer low-orbit satellite mega-constellation network becomes a difficult problem to solve.

[0004] As a popular strategy to achieve flexible control network in ground network, clustering network is an effective way to further improve the performance of satellite network, which is usually divided into two types: clustering network technology of wireless sensor network (WSN) and clustering network technology of mobile ad hoc network (MANET). In WSN, the main purpose of network clustering technology is to prolong the network lifetime under the condition of limited energy. The representative article of this kind of clustering technology is LEACH clustering algorithm proposed in "Energy-efficient communication protocol for wireless microsensor networks", which randomly selects cluster head and member and balances energy consumption through cluster head rotation. In MANET, mobility between nodes is the main consideration of network clustering technology, and most of such researches are weight-based clustering algorithms, the selection of weight is very flexible, which can be based on mobility, QoS, connectivity, etc. Typical MANET networking strategies are also widely used in vehicle ad hoc network (VANET) and unmanned aerial vehicle ad hoc network (UAV). The article "QMM-VANET: An efficient clustering algorithm based on QoS and monitoring of malicious vehicles in vehicular ad hoc networks" proposes QMM-VANET clustering network algorithm, which realizes low latency and high stability of network clustering in VANET by considering QoS requirements, untrusted value parameters and mobility constraints. Chinese invention patent application No. CN201910197312.9 discloses a cluster unmanned aerial vehicle ad hoc network clustering method, which includes: initializing cluster head, the cluster head broadcasts cluster head invitation information to surrounding nodes; according to the bird flocking-based regional division method, the cluster head divides the surrounding area, and divides the cluster structure in the self-organizing network into: cluster head area, node attraction area and node repulsion area; wherein, the nodes in the cluster head area are cluster head candidate nodes; the node attraction area includes: gateway area and member area, the nodes in the gateway area are gateway nodes in the cluster, and the nodes in the member area are member nodes in the cluster; the nodes in the node repulsion area include: idle nodes and nodes in other clusters; when the idle node receives multiple cluster head broadcast invitation information within the preset first time period, the state of the idle node is marked according to the signal strength of the received invitation information, and the node deployment is completed. The method uses the bird flocking-based regional division method to mark the node state according to the signal strength of the received invitation information, so as to provide a more stable cluster structure.

[0005] There are few studies on clustering networking algorithms in current satellite networks, and academia has been more inclined to study layered satellite network structures in the past. "MLSR: A novel routing algorithm for multilayered satellite IP networks" proposes a satellite clustering architecture for MLSN, which is composed of a MEO / GEO satellite and all LEO satellites within its coverage, effectively reducing the complexity and communication load of on-board routing table calculation. Chinese invention patent application No. CN201811249213.2 discloses a communication method, device and system based on satellite network, the method comprises: a user terminal receives a probe signal sent by a management satellite on a management channel, the management satellite manages one or more service satellites; the user terminal sends a breathing signal to the management satellite, the breathing signal carries the information of the user terminal, and the information of the user terminal is used to determine the service satellite information of the user terminal. Due to the operations of the user terminal receiving the probe signal and sending the breathing signal, the management satellite or the ground station can uniformly schedule the service satellite to serve the user terminal. When the service satellite switches, the step of inter-satellite negotiation can be omitted, saving signaling overhead, and the satellite communication experience of the user terminal is smooth, that is, in this method, a management satellite and multiple service satellites form a satellite "super cell", and this concept of dividing the satellite network into multiple "super cells" is consistent with the clustering networking technology. The concept of clustering networking is also mentioned in satellite networks based on SDN, such as the article "Dynamic SDN controller placement in a LEO constellation satellite network". Chinese invention patent application No. CN202110812489.2 discloses a kind of space-ground integrated adaptive dynamic QoS routing method based on SDN, which includes the following steps: establishing a hierarchical clustering network model based on SDN; establishing network resource mapping; establishing a multi-constraint QOS adaptive routing algorithm SDN-AD, this invention can effectively reduce control overhead and improve transmission efficiency by reducing long-distance transmission of control packets and shortest distance clustering, formulating the multi-constraint QoS problem as an optimization problem with the minimum transmission cost as the target, which can effectively calculate the transmission cost of different types of links, and shield the differences between satellites and ground networks at different levels to better adapt to network changes, thereby providing services that meet different quality of service requirements. This invention solves the optimization problem and realizes adaptive routing.In addition, the article "Reliable and Low-Overhead Clustering in LEO Small Satellite Networks" proposes a distributed online scheme based on coalition game theory, which realizes high reliability and low overhead satellite clustering networking, and is verified in a single-layer satellite network, fully proving that the clustering networking idea has great application prospects in satellite networks.

[0006] Unlike the cluster structure of ground clustering networking, the future multi-layer low-orbit satellite constellation has satellites in multiple different orbits with different altitudes and inclinations, and the networking form between satellites is diverse, and there are cross-layer inter-satellite links between satellite layers. These characteristics make the clustering network structure of the multi-layer low-orbit satellite constellation network usually a three-dimensional cluster structure across layers and domains.

[0007] And the ground clustering networking algorithm is based on a homogeneous network structure, and the nodes between them are often equal. In the wireless ad hoc network scene, mobile ad hoc network and Internet of Vehicles scene, the cluster structure formed by the clustering networking technology is a plane cluster. Because the ground clustering networking algorithm depends on the network topology, it cannot meet the demand of multi-layer low-orbit satellite constellation network for establishing a three-dimensional cluster structure across layers.

[0008] The unmanned aerial vehicle ad hoc network and the multi-layer low-orbit satellite constellation network have similarities in three-dimensional network clusters, but the clustering networking of the unmanned aerial vehicle ad hoc network mostly takes the signal strength between unmanned aerial vehicles as the basis for clustering. However, because satellites in satellite networks use high-power transmitting equipment and often use directional antennas, the control of signal loss in the transmission process is more stable than that of unmanned aerial vehicle ad hoc networks, making the signal strength factor less influential on satellite networks. Therefore, clustering networking based on signal strength cannot be directly used in satellite networks. The unmanned aerial vehicles in the unmanned aerial vehicle ad hoc network can change their positions according to the networking requirements, while the satellites are difficult to do so. In addition, the scale of the unmanned aerial vehicle network is much smaller than that of the satellite constellation network, and its clustering algorithm is designed for small-scale networks, which is usually simplified and difficult to meet the expansion requirements of large-scale satellite clustering networks.

[0009] In summary, the current clustering method in satellite networks is mainly applied to the networking of homogeneous satellites in the same layer, aiming to optimize network management. The clustering method in the prior art ignores the cooperation potential between different satellites and different constellations, and cannot flexibly and efficiently utilize the increasingly rich resource payloads on the satellites to form functional clusters across constellations and resource domains. SUMMARY

[0010] The purpose of the present application is to overcome the problems existing in the prior art, and to provide a task-oriented multi-layer low-orbit satellite constellation clustering networking method.

[0011] The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method comprises the following steps:

[0012] Step 1. Initialize satellite constellation data, including satellite load resources, satellite geographic position and orbit information and satellite inter-satellite link state, when a task is generated, the task data is received by a control center located on the ground or a high-orbit satellite;

[0013] Step 2. The control center establishes a task model according to the on-satellite resources required by the task, and establishes a satellite model according to the orbit parameters and on-satellite load capacity data of the satellite; specific task and satellite model establishment mechanism;

[0014] Step 3. Determine the restriction condition according to different task types;

[0015] Step 4. Model the cooperation between satellites and the matching relationship between satellites and tasks as a Markov transition probability model, and establish a transition probability matrix;

[0016] Step 5. Obtain the candidate satellite members and candidate cluster head satellites through the transition probability matrix obtained in step 4, and calculate the functional cluster matched with the current task based on the cluster head satellite as the center and the restriction condition in step 3;

[0017] Step 6. Upload the networking data to the satellite network, and the satellite network establishes the functional cluster through the inter-satellite link state notification mechanism after receiving the networking data, and saves the intra-cluster routing table.

[0018] Further, in step 1, the ground or high-orbit satellite control center initializes and maintains various data of the satellite constellation through ephemeris information and satellite information, including satellite load resources, satellite geographic position and orbit information and satellite inter-satellite link state, when a task is generated, the task requirement is first transmitted to the ground or high-orbit satellite control center, and the control center performs the subsequent steps to divide the satellite task cluster.

[0019] Further, in step 2, the control center establishes a task model according to the on-satellite resources required by the task, and establishes a satellite model according to the orbit parameters and on-satellite load capacity data of the satellite; the specific task and satellite model establishment mechanism is as follows:

[0020] Step 2.1. Task model establishment mechanism

[0021] The ad hoc task is defined as a set TASK, and its composition is as follows formula (1):

[0022] TASK={task1,task2,…,task TN}...... formula (1),

[0023] Where: taski denotes a subtask in a mission task, TN denotes the number of subtasks, and the definition of a subtask tk is given by the following equations (2) and (3):

[0024]

[0025] wherein type i denotes the type of the subtask, time i denotes the execution time required by the subtask, φ i denotes the set of load resources required by the subtask, wherein, denotes the value of the jth resource requirement of the ith subtask, RN denotes the total number of resource requirements, The greater the value of φ i1 is, the greater the demand of the subtask for the resource, and if the subtask has no demand for the resource, the value of φ i1 is 0.

[0026] Step 2.2 Establishment mechanism of satellite model

[0027] The definition of a satellite is a set s i , which is composed of the following equations (4) and (5):

[0028]

[0029]

[0030] L i = {a, e, θ, Ω, g, v}...... (6),

[0031] N i = {s1, s2, …, s N , N = |N i |}...... (7),

[0032] wherein id i denotes the number of the satellite, denotes the cluster head satellite number of the cluster to which the satellite belongs, tw i denotes the time window in which the satellite covers the mission area, δ i denotes the set of on-board load resources of the satellite, denotes the value of the jth load resource on the ith satellite, and L is in one-to-one correspondence with , L i denotes the six elements of the orbit of the satellite, a denotes the semi-major axis, e denotes the eccentricity, θ denotes the orbit inclination, Ω denotes the ascending node longitude, g denotes the perigee argument, and v denotes the perigee epoch true anomaly, N i denotes the set of neighbor satellites of satellite i, and N denotes the number of neighbor satellites.

[0033] Further, in step 3, the restriction condition is determined according to different task types, specifically:

[0034] For the functional satellite cluster meeting the task requirements, the satellite functional cluster overhead formula is proposed as the restriction condition of the satellite cluster, and a typical restriction mode is given through the cluster size, resource redundancy rate, total hop number and task execution time, and the restriction condition is shown in the following formula (8)-(12):

[0035] S(CL) = a · K + b · R(CL) + g · H(CL) (8),

[0036] CL = {CH, CM1, CM2, …, CM K-1}, K = |CL| (9), | CL | (10),

[0037]

[0038]

[0039] Wherein: S(CL) represents the cluster management cost, CL represents the set of satellite functional cluster members, CH represents the cluster head satellite, CM represents the cluster member satellite, K represents the size of the satellite cluster, R(CL) represents the redundancy of resources within the cluster, H(CL) represents the total hop number of the route from the member satellite to the cluster head satellite, a, b, g are weight parameters for adjusting the influence of various factors on S(CL); formula (11) and formula (12) are resource and time constraints of the satellite functional cluster, for a satellite functional cluster meeting the task requirements, the total of all cluster member satellites for a certain task demand resource is greater than the total demand of the task for the resource, and the time window of any member satellite meets the demand of the satellite assigned to the subtask for the satellite coverage time.

[0040] Further, in step 4, the cooperation between satellites and the matching relationship between satellites and tasks are modeled as a Markov transition probability model, and a transition probability matrix is established, which specifically includes the following steps:

[0041] Step 4.1 Establishing a Markov transition probability model

[0042] The process of the satellite being divided into different satellite clusters is regarded as the transition of the cooperation state between the satellite and different satellites, which is modeled as a Markov process, let S t represent the set of all possible cooperation states in the satellite network, state represent the specific cooperation state of a certain node at time t, and the probability model of the transition from state state i to state state j is shown in the following formula (13)-(16):

[0043] P ij (t) = Pr(S t+1 = state j | S t = state i )... (13),

[0044]

[0045]

[0046] wherein the transition probability satisfies the normalization, boundedness and symmetry conditions, and the value of the transition probability depends on the cooperation possibility between satellites due to the task demand, specifically, by constructing a weighted graph based on the satellite topology structure, the cooperation relationship between satellites and the matching relationship between satellites and task demands are characterized;

[0047] In this step, G(V, E) is used to represent the satellite structure topology, wherein V represents the satellite set, and E represents the inter-satellite link (ISL) set of the satellite, on the basis of the graph G, a node enhanced graph is proposed: wherein V st represents a satellite node, V at represents a sub-task node in TK, the edge connecting the structure points is called a structure edge, and represents the possible ISL set between satellites within the communication range, the edge connecting the structure points and the attribute points is called an attribute edge, and represents the matching relationship between the satellite and the sub-task, the weights of the structure edge and the attribute edge are represented by ω i,j , a weight definition is given based on the task demand resources and the on-board resources as follows formula (17):

[0048]

[0049] For the structure edge, ω i,j represents the similarity of resources between two structure nodes, and is positively related to the number of common resources between the structure nodes, for the attribute edge, when the matching resource number between the structure node and the attribute node is not 0, ω i,j is positive, and ω baseSt and ω baseAt are correction values of the weights of the structure edge and the attribute edge, the size of the correction value is set according to actual needs, and herein ω baseSt and ω baseAt are respectively 10 and 100, in addition, the attribute nodes are not connected to each other;

[0050] In the node enhanced graph, if the multi-hop neighbor With the same subtask, a new neighbor is shared, i.e. an attribute node It becomes two-hop neighbors, shortens their effective distance in the graph, and increases the possibility of being clustered in the same cluster. The transition probability of structure edge and attribute edge is defined as follows:

[0051]

[0052] Where N1 represents the number of neighbor structure points of the node, and N2 represents the number of neighbor attribute points of the node. By combining equations (17) to (18), the transition probability matrix P between all structure points and attribute points is obtained as follows:

[0053]

[0054] Step 4.2: Solve the Markov clustering algorithm (MCL) according to the transition probability matrix P.

[0055] MCL algorithm simulates random walk process through matrix multiplication. MCL algorithm captures the relationship between related nodes in the clustering process, and obtains clusters from graph structure by expanding and shrinking the transition probability matrix P. The MCL algorithm is as follows:

[0056] Step 4.2.1: Perform expansion operation on the transition probability matrix P, i.e. P = P ε , simulate multi-step random walk process, and diffuse transition probability, where ε is the expansion parameter, which is greater than 1, representing the number of steps of random walk. In order to add inertia to the transition probability and ensure that the transition probability matrix P can finally converge, a self-loop is added to each structure point and attribute node, and the weight of the self-loop is fixed as ω baseAt ;

[0057] Step 4.2.2: Perform inflation operation to further enhance transition probability and suppress low transition probability. The inflation operation includes taking the r-th power of each element of the transition probability matrix P, and then normalizing the obtained element:

[0058]

[0059] Where r>1 represents the inflation parameter, which is used to control the granularity of clustering. The larger the inflation parameter, the more cluster structures will be obtained.

[0060] Step 4.2.3: Repeat step 4.2.1 and step 4.2.2 until the matrix converges and a stable probability distribution is obtained. In the converged transition probability matrix, the high value elements of each column will be concentrated in the same row with the high value elements of other columns, indicating that the elements in this row belong to the same cluster.

[0061] Further, step 5, the probability transition matrix obtained by step 4, obtains candidate satellite members and candidate cluster head satellites, and the cluster head satellite is taken as the center and the restriction condition in step 3 is taken as the basis to calculate the functional cluster matched with the current task, and the specific steps include the following steps:

[0062] Step 5.1, according to the ephemeris information, a task TK' for selecting the cluster head satellite node is added, all satellite nodes are taken as structure points, and the task node TK' is taken as the only attribute point to construct a node enhanced graph and the probability transition matrix P'. and P' are input into the MCL algorithm to obtain a candidate set CH cd of potential cluster head satellites.

[0063] Step 5.2, taking each cluster head CH cd in the candidate set CH i as the center, reducing the edge weight ω max of all satellites outside the maximum hop range hop i,j , and the edge weight ω i,j of all satellites whose corresponding sub-task demand has been met.

[0064] Step 5.3, a new probability transition matrix P is constructed, which includes all satellite nodes and all sub-task (except TK') nodes, and the matrix is input into the MCL algorithm to obtain a functional cluster candidate CL i . Steps 3 and 4 are repeated to obtain a functional cluster candidate set CL.

[0065] Step 5.4, the cluster management cost S(CL) is calculated for each functional cluster candidate CL i , and the functional cluster candidate with the minimum S(CL) is returned as the finally generated functional cluster.

[0066] The present application has the following superior technical effects compared with the prior art:

[0067] 1. The existing clustering networking algorithm only considers the similarity between nodes due to the topological structure, and cannot cluster on demand for tasks. The task-oriented multi-layer low-orbit satellite mega-constellation clustering networking method disclosed by the present application lays a foundation for task-oriented on-demand clustering by combining the cooperation relationship between satellites and the matching relationship between satellites and task demands through node enhanced graphs, not only reduces the pressure of multi-layer mega-constellation network management through clustering networking, but also more efficiently and flexibly utilizes the rich on-board load resources of the satellite network.

[0068] 2. The mission-oriented multi-layer low-Earth orbit satellite constellation clustering and networking method described in this invention models the clustering and networking problem as a Markov transition probability process. The state transitions in the Markov process depend only on the current probability distribution, providing greater flexibility for network clustering. This allows the satellite clustering network to dynamically adjust its clustering strategy based on current mission requirements, satellite status, and environmental identifiers. In contrast, traditional clustering algorithms based on fixed rules or predictions lack flexible responses to real-time environmental changes and mission requirements. Furthermore, the clustering method based on the Markov transition probability matrix can effectively assess the cooperation potential between satellites from a global perspective, thus making it easier to achieve globally optimal clustering compared to traditional methods.

[0069] 3. The mission-oriented multi-layer low-Earth orbit satellite constellation cluster networking method described in this invention is not limited to homogeneous and same-layer satellite networks when constructing node enhancement graphs and transition probability matrices, but considers the potential cooperative objects of each satellite, thereby achieving support for heterogeneous and cross-layer satellite networks.

[0070] 4. The mission-oriented multi-layer low-Earth orbit satellite constellation clustering and networking method described in this invention can effectively realize mission-oriented clustering and networking of satellite networks, reduce cluster size, resource occupancy rate, total network hops and network construction time, and improve the on-board resource utilization rate of the multi-layer low-Earth orbit satellite constellation; actual verification results show the adaptability of the method described in this invention to the multi-layer low-Earth orbit satellite constellation. Attached Figure Description

[0071] Figure 1 A schematic diagram of a mission-oriented multi-layer low-Earth orbit satellite cluster networking scenario in the constellation Giants.

[0072] Figure 2 Flowchart of a mission-oriented multi-layer low-Earth orbit satellite clustering and networking method in the constellation Gigantote;

[0073] Figure 3 A schematic diagram of the mission-oriented satellite functional cluster generation algorithm;

[0074] Figures 4a-4d This diagram illustrates the impact of the number of different mission resource requirements on satellite clusters. Detailed Implementation

[0075] To better understand the above-mentioned objectives, features and advantages of the present invention, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0076] Example

[0077] like Figure 1 As shown in Figure 4, the mission-oriented multi-layer low-Earth orbit satellite constellation cluster networking method of the present invention includes the following steps:

[0078] Step 1. Initialize satellite constellation data, including satellite payload resources, satellite geographic position and orbit information and satellite inter-satellite link conditions. When a task is generated, the task data is received by a control center located on the ground or a high-orbit satellite;

[0079] Step 2. The control center establishes a task model according to the on-satellite resources required by the task, and establishes a satellite model according to the orbit parameters and on-satellite load capacity data of the satellite; specific task and satellite model establishment mechanism;

[0080] Step 3. Determine the restriction condition according to different task types;

[0081] Step 4. Model the cooperation between satellites and the matching relationship between satellites and tasks as a Markov transition probability model, and establish a transition probability matrix;

[0082] Step 5. Obtain the candidate satellite members and candidate cluster head satellites through the transition probability matrix obtained in step 4, and calculate the functional cluster matching the current task based on the cluster head satellite as the center and the restriction condition in step 3;

[0083] Step 6. Upload the networking data to the satellite network, and the satellite network establishes a functional cluster through the inter-satellite link state notification mechanism after receiving the networking data, and saves the intra-cluster routing table.

[0084] As a specific embodiment of the present application, in step 1, the ground or high-orbit satellite control center initializes and maintains various data of the satellite constellation through ephemeris information and satellite information, including satellite payload resources, satellite geographic position and orbit information and satellite inter-satellite link conditions. When a task is generated, the task requirements are first transmitted to the ground or high-orbit satellite control center, and the control center performs the subsequent steps to divide the satellite task cluster.

[0085] As a specific embodiment of the present application, in step 2, the control center establishes a task model according to the on-satellite resources required by the task, and establishes a satellite model according to the orbit parameters and on-satellite load capacity data of the satellite; the specific task and satellite model establishment mechanism is as follows:

[0086] Step 2.1. Task model establishment mechanism

[0087] The temporary task is defined as a set TASK, which is composed of the following formula (1):

[0088] TASK={task1,task2,…,task TN}......(1),

[0089] Where: task iLet TN represent the number of subtasks in an ad hoc task, and let TN represent the number of subtasks. The definition of subtask tk is as follows (2)-(3):

[0090] task i ={type i ,φ i ,time i}......(2),

[0091]

[0092] Where: type i Indicates the type of subtask, time i φ represents the execution time required for the subtask. i This represents the set of payload resources required by the subtask, where... Let represent the numerical value of the j-th type of resource requirement for the i-th subtask, and RN represent the total number of resource requirements. The larger the value ≥ 0, the greater the subtask's demand for this type of resource. If the subtask has no demand for this type of resource, then... The value is 0;

[0093] Step 2.2 Mechanism for Establishing Satellite Models

[0094] A satellite is defined as a set s i Its composition is as follows (4)-(5):

[0095]

[0096]

[0097] L i ={a,e,θ,Ω,g,v}......(6),

[0098] N i ={s1,s2,…,s N ,N=|N i |}......(7),

[0099] Where: id i Indicates the satellite's number, This indicates the cluster head satellite number of the cluster to which the satellite belongs, tw i The time window representing the satellite's mission coverage area, δ i This represents the set of onboard payload resources of a satellite. This represents the value of the j-th type of payload resource on the i-th satellite, and... One-to-one correspondence, L idenotes the six elements of the orbit of satellite i, a denotes semi-major axis, e denotes eccentricity, denotes inclination, denotes longitude of ascending node, g denotes argument of perigee, denotes mean anomaly at epoch, N i denotes the neighbor satellite set of satellite i, N denotes the number of neighbor satellites.

[0100] As a specific embodiment of the present application, in step 3, the restriction condition is determined according to different task types, specifically:

[0101] For the functional satellite cluster satisfying the task requirements, the cost formula of the satellite functional cluster is proposed as the restriction condition of the satellite cluster, and a typical restriction mode is given through the cluster size, resource redundancy rate, total hop number and task execution time, and the restriction condition is shown in the following formula (8)-(12):

[0102] S(CL)=α·K+β·R(CL)+γ·H(CL)......(8),

[0103] CL={CH,CM1,CM2,…,CM K-1},K= | CL | ......(9),

[0104]

[0105]

[0106] Wherein: S(CL) denotes the cluster management cost, CL represents the set of satellite functional cluster members, CH represents the cluster head satellite, CM represents the cluster member satellite, K represents the size of the satellite cluster, R(CL) represents the redundancy of resources within the cluster, H(CL) represents the total hop number of routing from the member satellite to the cluster head satellite, and a, b, g are weight parameters for adjusting the influence of various factors on S(CL); formula (11) and formula (12) are resource and time constraints of the satellite functional cluster, for a satellite functional cluster satisfying the task requirements, the total sum of all cluster member satellites for a certain task demand resource is greater than the total demand of the task for the resource, and the time window of any member satellite satisfies the demand of the satellite assigned to the subtask for the satellite coverage time.

[0107] As a specific embodiment of the present application, in step 4, the cooperation between satellites and the matching relationship between satellites and tasks are modeled as a Markov transition probability model, and a transition probability matrix is established, which specifically includes the following steps:

[0108] Step 4.1 Establishing a Markov transition probability model

[0109] The process of a satellite being assigned to different satellite clusters can be viewed as a transition of cooperative states between that satellite and different satellites, and this can be modeled as a Markov process, where S... t Let state represent the set of all possible cooperative states in a satellite network, and let state represent a specific cooperative state of a node at time t. i Transition to state j The probability model is shown in equations (13)-(16) below:

[0110]

[0111]

[0112] Among them, the transition probability satisfies the conditions of normalization, boundedness and symmetry. The value of the transition probability depends on the possibility of cooperation between satellites due to mission requirements. Specifically, by constructing a weighted map based on satellite topology, the cooperation relationship between satellites and the matching relationship between satellites and mission requirements are characterized.

[0113] In this step, G(V,E) is used to represent the satellite structure topology, where V represents the set of satellites and E represents the set of inter-satellite links (ISLs). Based on graph G, a node-enhanced graph is proposed: Among them, V st V represents a satellite node. at Represents the subtask nodes in TK, and the edges connecting structure points. Called a structural edge, it represents the set of possible ISLs between satellites within the communication range, and is an edge connecting structural points and attribute points. These are called attribute edges, representing the matching relationship between satellites and subtasks. The weights of structural edges and attribute edges are denoted by ω. i,j Based on the mission requirements and onboard resources, a weighting definition is given as follows (17):

[0114]

[0115] For structural edges, ω i,j ω represents the similarity of resources between two structural nodes and is positively correlated with the number of shared resources between them. For attribute edges, ω is positively correlated when the number of matching resources between the structural node and the attribute node is not zero. i,j The value is positive, ω baseSt and ω baseAt This is a correction value for the weights of structural edges and attribute edges. The magnitude of the correction value is set according to actual needs. Here, ω... baseSt and ω baseAt The values ​​are 10 and 100 respectively; in addition, the attribute nodes are not connected to each other.

[0116] In the node-enhanced graph, if multi-hop neighbors match the same subtask, share a new neighbor, i.e., an attribute node become two-hop neighbors, shorten their effective distance in the graph, and increase their likelihood of being clustered together, the transition probabilities of the structural and attribute edges are defined as follows:

[0117]

[0118] where N1 represents the number of neighbor structural points of a node, and N2 represents the number of neighbor attribute points of a node. By combining Equations (17) to (18), the transition probability matrix P between all structural and attribute points is obtained as follows:

[0119]

[0120] Step 4.2. Solving the Markov clustering algorithm (MCL) based on the transition probability matrix P

[0121] The MCL algorithm simulates a random walk process through matrix multiplication. The MCL algorithm captures the relationships between related nodes in the clustering process and obtains clusters from the graph structure by expanding and contracting the transition probability matrix P. The MCL algorithm is as follows:

[0122] Step 4.2.1. Perform an expansion operation on the transition probability matrix P, i.e., P = P ε , to simulate a multi-step random walk process and spread out the transition probabilities. Here, ε is an expansion parameter that is greater than 1 and represents the number of steps of the random walk. To add inertia to the transition probabilities and ensure that the transition probability matrix P eventually converges, a self-loop is added to each structural point and attribute node, and the weight of the self-loop is fixed at ω baseAt ;

[0123] Step 4.2.2. Perform an inflation operation to further increase the transition probabilities and suppress low transition probabilities. The inflation operation includes raising each element of the transition probability matrix P to the power of r and then normalizing the resulting elements:

[0124]

[0125] where r > 1 represents an inflation parameter that is used to control the granularity of the clusters. A larger inflation parameter will result in more cluster structures.

[0126] Step 4.2.3. Repeat steps 4.2.1 and 4.2.2 until the matrix converges and a stable probability distribution is obtained. In the converged transition probability matrix, the high-value elements of each column will be concentrated in the same row as the high-value elements of other columns, indicating that the elements in this row belong to the same cluster.

[0127] Further, in step 5, the probability transition matrix obtained in step 4 is used to obtain candidate satellite members and candidate cluster head satellites, and a functional cluster matching the current task is calculated based on the cluster head satellite as the center and the restriction condition in step 3, as shown in the following formula: Figure 3 Specifically, the method comprises the following steps:

[0128] Step 5.1, according to the ephemeris information, a task TK' for selecting a cluster head satellite node is added, and a node enhanced graph is constructed by taking all satellite nodes as structure points and the task node TK' as a unique attribute point and the probability transition matrix P'. and P' are input into the MCL algorithm to obtain a candidate set CH cd of potential cluster head satellites.

[0129] Step 5.2, taking each cluster head CH cd in the candidate set CH i as the center, reducing the edge weight ω max of all satellites beyond the maximum hop range hop i,j , and the edge weight ω i,j of all satellites whose corresponding sub-task demand has been met.

[0130] Step 5.3, a new probability transition matrix P is constructed, which includes all satellite nodes and all sub-task nodes (except TK'), and the matrix is input into the MCL algorithm to obtain a functional cluster candidate CL i . Steps 3 and 4 are repeated to obtain a functional cluster candidate set CL.

[0131] Step 5.4, the cluster management cost S(CL) is calculated for each functional cluster candidate CL i , and the functional cluster candidate with the minimum S(CL) is returned as the final generated functional cluster.

[0132] Verification example

[0133] In the verification of the method, a multi-layer low earth orbit satellite network is used, which contains 2825 LEO satellites, and the satellite constellation constitutes various parameters as shown in Table 1.

[0134] Table 1 Satellite constellation composition

[0135] Satellite tier Tier 1 Tier 2 Tier 3 Tier 4 Number of orbits 6 5 8 32 Number of satellites / orbit 75 75 50 50 Orbit altitude (km) 1325 1275 1130 1110 Orbit inclination (°) 70 81 74 53.8

[0136] The simulation time is 86400s, the time of task generation is 720s, the snapshot is taken every 60s, the topology of satellite network remains unchanged during the snapshot by default, the number of tasks is 5, the number of subtasks of each task is 5, corresponding to five different on-board payloads. The processing rate of each satellite to the data packet is the same, which is 5ms, the maximum number of inter-satellite link established is 4, in addition, the control information of the clustering network will be transmitted preferentially in the satellite network, therefore, the link congestion and link breakage are not considered. The simulation indexes are: average satellite cluster size, resource waste rate, total hop number and network construction time.

[0137] The average satellite cluster size refers to the number of all satellites in the satellite cluster, the mathematical expression is |CL|; the resource redundancy rate refers to the part of the total resources in the satellite cluster that exceeds the demand of the task, the mathematical expression is shown by formula (9); the total hop number refers to the total sum of the routing hop numbers of all cluster member satellites to the cluster head satellite in the satellite cluster structure, the calculation formula is as follows:

[0138]

[0139] The network construction time is divided into two parts: 1) clustering time T clu , which represents the time required for establishing communication between the cluster head and the cluster member with the maximum hop number, 2) sending time T del , which represents the time required for sending the control information from the control center to the cluster head satellite, specifically, the time required for the control information of the control plane to be uploaded to the possible access satellite in the cluster and then transmitted to the cluster head, which is defined as follows:

[0140] T net =T clu +T del ......(22),

[0141]

[0142] Figures 4a-4cThe influence of the number of different task resource requirements on the satellite cluster is shown, and the algorithm proposed by the method is marked as TOC. With the increase of the number of task requirements, all algorithms show an upward trend, and the performance of the TOC algorithm is always better than that of other algorithms in the technical field. In terms of average cluster size, the TOC algorithm reaches 29.6, which is about 40.1% to 43.9% lower than other algorithms; in terms of resource redundancy, the performance of the TOC algorithm is lower than 12.1%, which is about 69.7% lower than other algorithms in the technical field; in terms of total hop count, the TOC algorithm is about 92 hops, while other algorithms in the technical field reach a maximum of 426 hops. This is because other algorithms in the technical field often rely on expanding the cluster size to absorb more nodes to meet the task requirements. This expansion of the cluster size is blind, leading to inevitable resource redundancy. A satellite that serves a task requirement that has been met may join the satellite cluster due to the expansion of the cluster size. However, the TOC algorithm solves this problem by actively reducing the weight of the attribute edge of the sub-task that has met the requirements.

[0143] Figure 4d The influence of different satellite layer numbers on the network construction time of the satellite cluster is shown. As shown in the figure, LEACH and LID algorithms show an upward trend, while QMM and TOC algorithms show a downward trend. In a single-layer constellation, the network construction time of all algorithms is similar. With the increase of the number of layers, the network construction time of the TOC algorithm is continuously reduced to a minimum value of 36.6 ms. This value is about 82.8% lower than LEACH and LID algorithms, and 38.2% lower than QMM algorithm. This is because when there is only one layer of satellites in space, the number of satellites is small and the spatial distribution is sparse. In this case, all clustering algorithms almost cluster all satellites in the task area to meet the task requirements, so there is no significant difference between them. However, with the increase of the number of satellite layers and the density of satellites in space, the LID and LEACH algorithms that rely on structure and probability tend to form satellite clusters of excessive size, thereby increasing the network construction time. On the contrary, the TOC algorithm can find more possible cluster structures by constructing clusters across layers, and reduce the network construction time by controlling the total hop count.

[0144] In summary, the method can effectively realize task demand-oriented clustering of satellite networks, reduce cluster size, resource occupancy, total network hop count, and network construction time, and improve the utilization of on-board resources in a multi-layer low-orbit satellite mega-constellation. Finally, through examples, the method proves its adaptability to multi-layer low-orbit satellite mega-constellations.

[0145] The present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the present application, and various changes and improvements can be made to the present application without departing from the spirit and scope of the present application, and such changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims.

Claims

1. A task-oriented multi-layer low-orbit satellite mega constellation clustering networking method, comprising the following steps: Step 1. Initialize satellite constellation data, including satellite load resources, satellite geographic position and orbit information, and satellite inter-satellite link status, when a task is generated, the task data is received by the control center located on the ground or high orbit satellite; Step 2. The control center establishes a task model according to the on-board resources required by the task, and establishes a satellite model according to the orbit parameters and on-board load capacity data of the satellite; Step 3. Determine the restriction condition according to different task types; Step 4. The cooperation between satellites and the matching relationship between satellites and tasks are modeled as a Markov transition probability model, and a transition probability matrix is established; Step 5. Through the transition probability matrix obtained in step 4, the candidate satellite members and candidate cluster head satellites are obtained, and the functional cluster matching the current task is calculated based on the cluster head satellite as the center and the restriction condition in step 3; Step 6. Upload the networking data to the satellite network, and the satellite network establishes the functional cluster through the inter-satellite link state notification mechanism after receiving the networking data, and saves the intra-cluster routing table.

2. The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method according to claim 1, in step 1, the ground or high-orbit satellite control center initializes and maintains various data of the satellite constellation through ephemeris information and satellite information, including satellite load resources, satellite geographic position and orbit information, and satellite inter-satellite link status, when a task is generated, the task demand is first transmitted to the ground or high-orbit satellite control center, and the control center performs the subsequent steps to divide the satellite task cluster.

3. The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method according to claim 1, in step 2, the control center establishes a task model according to the on-board resources required by the task, and establishes a satellite model according to the orbit parameters and on-board load capacity data of the satellite; The specific task and satellite model establishment mechanism is as follows: Step 2.

1. Task model establishment mechanism The temporary task is defined as a set TASK, which is composed of the following formula (1): TASK = {task1, task2,..., task TN}... Equation (1), where: task i denotes a subtask in a machine task, TN denotes the number of subtasks, and the definition of a subtask tk is given by the following equations (2) and (3): task i = {type i , φ i , time i}... (2), ......(3), wherein: type i denotes the type of the subtask, time i denotes the execution time required by the subtask, φ i denotes the set of load resources required by the subtask, wherein, denotes the value of the jth resource requirement of the ith subtask, RNdenotes the total number of resource requirements, the greater the value of φ, the greater the demand of the subtask for this type of resource, if the subtask has no demand for this type of resource, then the value of φ is 0; Step 2.

2. Satellite model establishment mechanism A satellite is defined as a set s i consisting of the following equations (4)-(5): ......(4), ......(5), L i = {a, e, 0, W, g, v}... (6), N i = {s1, s2,..., s N N = |N i |}... (7), where id i denotes the number of the satellite, denotes the cluster head satellite number of the cluster to which the satellite belongs, tw i denotes the time window in which the satellite covers the mission area, δ i denotes the on-board payload resource set of the satellite, denotes the value of the jth payload resource on the ith satellite, and corresponds one-to-one, L i denotes the orbital elements of the satellite, a denotes the semi-major axis, e denotes the eccentricity, θ denotes the inclination, Ω denotes the longitude of the ascending node, g denotes the argument of perigee, v denotes the mean anomaly at perigee epoch, N i denotes the neighbor satellite set of satellite i, N denotes the number of neighbor satellites.

4. The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method according to claim 1, in step 3, the restriction condition is determined according to different task types, specifically: For the functional satellite cluster that meets the task requirements, the satellite functional cluster cost formula is proposed as the restriction condition of the satellite cluster, and a typical restriction method through cluster size, resource redundancy rate, total hop number and task execution time is given, the restriction condition is shown in the following formula (8)-(12): S(CL)=α·K+β·R(CL)+γ·H(CL)......(8), CL = {CH, CM1, CM2,..., CM K-1}, K = |CL|... (9), ......(10), ......(11), ......(12), Wherein: S(CL) represents the cluster management cost, CL represents the set of satellite functional cluster members, CH represents the cluster head satellite, CM represents the cluster member satellite, K represents the scale of the satellite cluster, R(CL) represents the redundancy of the resources within the cluster, H(CL) represents the total number of hops of the route from the member satellite to the cluster head satellite, and a, b, g are weight parameters for adjusting the influence of various factors on S(CL); formula (11) and formula (12) are the resource and time constraints of the satellite functional cluster, and for a satellite functional cluster that meets the task demand, the total sum of the resources of all cluster member satellites for a certain task demand is greater than the total demand of the task for the resource, and the time window of any member satellite meets the demand of the satellite coverage time of the subtask to which the satellite is assigned.

5. The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method according to claim 1, in step 4, the cooperation between satellites and the matching relationship between satellites and tasks are modeled as a Markov transition probability model, and a transition probability matrix is established, which specifically includes the following steps: Step 4.1 Establishing a Markov transition probability model The process of classifying a satellite into different satellite clusters is regarded as the transition of the cooperation state between the satellite and different satellites, which is modeled as a Markov process, let S t denote the set of all cooperation states in the satellite network, state denote the specific cooperation state of a certain node at time t, and the probability model of the transition from state state i to state state j is shown in the following equations (13)-(16): P ij (t) = Pr(S t+1 = state j | S t = state i )... (13), ......(14), ......(15), ......(16), Wherein: the transition probability satisfies the normalization, boundedness and symmetry conditions, and the value of the transition probability depends on the cooperation possibility between satellites due to task demand, and specifically, a weighted graph based on the satellite topology structure is constructed to represent the cooperation relationship between satellites and the matching relationship between satellites and task demands; In this step, G(V, E) is used to represent the satellite structure topology, where V represents the satellite set, E represents the inter-satellite link (ISL) set of the satellite, and on the basis of the graph G, the node enhanced graph is proposed: wherein V st represents the satellite node, V at represents the sub-task node in TK, the edge connecting the structure point and the structure point is called a structure edge, represents the ISL set between the satellites within the communication range, and the edge connecting the structure point and the attribute point is called an attribute edge, represents the matching relationship between the satellite and the sub-task, and the weight of the structure edge and the attribute edge is represented by ω i,j , a weight value is given based on the task demand resource and the on-board resource, and the definition is as follows in equation (17): ......(17), For structural edges, ω i,j represents the similarity of resources between two structural nodes, and is positively related to the number of common resources between the structural nodes. For attribute edges, ω i,j is positive when the number of matching resources between the structural node and the attribute node is not 0. ω baseSt and ω baseAt are the modified values of the weights of the structural edges and the attribute edges, respectively. The values of ω baseSt and ω baseAt are 10 and 100, respectively. The attribute nodes are not connected to each other. In the node-enhanced graph, if multi-hop neighbors and match the same subtask, share a new neighbor, the attribute node becomes a two-hop neighbor, shortens their effective distance in the graph, and increases the possibility of being divided into the same cluster. The transition probabilities of the structure edge and the attribute edge are defined as follows: ......(18), Wherein, N1 represents the number of neighbor structure points of the node, and N2 represents the number of neighbor attribute points of the node, and by combining formula (17) to formula (18), the transition probability matrix (19) between all structure points and attribute points is obtained: ......(19), Step 4.2 Solving the Markov clustering algorithm (MCL) according to the transition probability matrix P The MCL algorithm simulates the random walk process through matrix multiplication, and the MCL algorithm captures the relationship between related nodes in the clustering process, and obtains clusters from the graph structure by expanding and contracting the transition probability matrix P, and the MCL algorithm is as follows: Step 4.2.1 The transition probability matrix P is extended, i.e. P = P ε , to simulate a multi-step random walk process to diffuse the transition probabilities, where ε is an extension parameter greater than 1 representing the number of steps of the random walk, and to add inertia to the transition probabilities to ensure that the transition probability matrix P converges eventually, a self-loop is added to each structure point and attribute node, and the weight of the self-loop is fixed as ω baseAt ; Step 4.2.2 Perform inflation operation to further improve the transition probability and suppress low transition probability, and the inflation operation includes performing r power operation on each element of the transition probability matrix P, and then normalizing the obtained element: ......(20), Wherein, r>1 represents the inflation parameter for controlling the granularity of the cluster, and the larger the inflation parameter, the more cluster structures will be obtained; Step 4.2.3 Repeat step 4.2.1 and step 4.2.2 until the matrix converges and a stable probability distribution is obtained, and in the converged transition probability matrix, the high value elements of each column will be concentrated in the same row with the high value elements of other columns, indicating that the elements in this row belong to the same cluster.

6. The task-oriented multi-layer low-orbit satellite mega constellation clustering networking method according to claim 1, in step 5, the probability transition matrix obtained in step 4 is used to obtain candidate satellite members and candidate cluster head satellites, and based on the cluster head satellite as the center and the restriction condition in step 3, the functional cluster matching the current task is calculated, which specifically includes the following steps: Step 5.1 adds the task TK' of the selected cluster head satellite node according to the ephemeris information, constructs a node enhanced graph with all satellite nodes as structure points and the task node TK' as the only attribute point and the probability conversion matrix P'; input and P' into the MCL algorithm to obtain the candidate set CH of potential cluster head satellites cd ; Step 5.2 Take each cluster head CH in the candidate set CH cd as the center, reduce the edge weight ω i of all satellites outside the maximum hop range hop max , and the edge weight ω i,j of all satellites whose corresponding subtask requirements have been met; and i,j ​ Step 5.3 Construct a new probability transition matrix P, which includes all satellites and all subtask nodes except TK', input the matrix into MCL algorithm, and a functional cluster candidate CL can be obtained i ; repeat step 3 and step 4 to obtain a functional cluster candidate set CL; Step 5.4 is for each functional cluster candidate CL i Calculate the cluster management cost S(CL), return the functional cluster candidate with the minimum S(CL) as the final generated functional cluster.

Images