Dynamic ultra-low orbit satellite feed link routing method based on clusters

Abstract

The invention discloses a cluster-based dynamic ultra-low orbit satellite feed link routing method. Relates to the technical field of ultra-low earth orbit satellite routing, in particular to the technical field of cluster-based dynamic ultra-low earth orbit satellite feed link routing methods. The problems that a satellite is poor in adaptability, low in resource utilization rate, prominent in delay and reliability problems, limited in expansibility and the like in a high-dynamic environment are effectively solved. The method comprises the following steps: 1, carrying out region division according to the coverage range of each gateway station in a very low orbit; 2, starting from the first time slice, calculating the distance between each satellite and a ground gateway station and the energy consumption of the satellite in unit time; 3, inputting the distance between the satellite and the ground gateway station, the energy consumption in unit time and the neighborhood value into fuzzy logic for processing to obtain the probability that the satellite becomes a cluster head and the communication competition radius; 4, performing cluster head election according to the probability that the satellite becomes the cluster head and the communication competition radius; and 5, a cluster forming process is completed, and routing distribution is carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of very low earth orbit satellite routing, and in particular to the technical field of cluster-based very low earth orbit satellite routing. Background Art

[0002] In the field of satellite communications, especially for very low earth orbit (VLEO) satellite networks, efficient feeder link routing methods are the key to ensuring data transmission stability, reducing latency and optimizing resource utilization. With the rapid development of space technology, VLEO satellites can provide lower communication latency and stronger signal strength due to their low orbit characteristics, but they also face challenges such as rapid satellite movement and frequent changes in network topology, which puts higher requirements on feeder link routing strategies. At present, the mainstream solutions for feeder link routing of VLEO satellite networks are mainly based on static or semi-dynamic routing algorithms. These algorithms usually rely on pre-set satellite orbit prediction models and combine ground stations and network control centers for route planning.

[0003] The prior art has the following problems:

[0004] 1. Poor adaptability: Static or semi-dynamic routing algorithms are difficult to adapt to the high dynamics of VLEO satellite networks. The rapid changes in satellite positions cause predefined routes to quickly become outdated. The algorithms do not fully consider the rapid changes in satellite orbits and the dynamics of network topology, and lack a real-time routing adjustment mechanism.

[0005] 2. Low resource utilization: Since routing decisions are based on global optimization rather than local optimization, some links may be overloaded while other links are idle. Distributed computing capabilities are not fully utilized, making it difficult to achieve fine-grained resource allocation and load balancing.

[0006] 3. Latency and reliability issues: In a highly dynamic environment, untimely updates of routing information may lead to increased data transmission delays or even link interruptions. The propagation delay of routing information from the network control center to the satellite and the insufficient response speed of routing adjustments.

[0007] 4. Scalability limitations: As the number of satellites increases, the computing burden of the network control center rises sharply, affecting the scalability of the overall system. Centralized routing management methods are difficult to effectively handle complex routing decisions in large-scale networks.

[0008] In summary, the existing technology has significant deficiencies in dealing with the high dynamics and resource optimization of VLEO satellite networks. There is an urgent need for a new feeder link routing method that can adapt to the rapidly changing network environment, efficiently utilize resources and ensure data transmission reliability. Summary of the invention

[0009] In view of the above problems, the present invention discloses a cluster-based dynamic very low orbit satellite feeder link routing method, which relates to the field of very low orbit satellite technology and effectively solves the technical difficulties existing in the feeder link routing method in a very low orbit (VLEO) satellite network, especially the problems of poor adaptability, low resource utilization, prominent delay and reliability problems and limited scalability of the prior art in a highly dynamic environment.

[0010] The method comprises the following steps:

[0011] S1. According to the coverage range of each gateway station in the very low orbit, the region is divided. K satellites and 1 gateway station in the region form a very low orbit satellite communication system. The coverage range of the gateway station is abstracted into a coordinate system.

[0012] S2, define S time slices {T1, T2..., T N ,...,T S}, S represents the total number of time slices;

[0013] S3. Starting from the first time slice T1, the gateway calculates the distance d of each satellite from the ground gateway and estimates the energy consumption E of the satellite per unit time. tota1 , the number of neighboring communicable satellites of the satellite is recorded as the neighborhood value m;

[0014] S4, the distance d between the satellite and the ground gateway, the energy consumption E per unit time of the satellite tota1 The three parameters of the satellite's neighborhood value m are combined with fuzzy logic for pre-clustering. After screening the rule base, the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node are obtained.

[0015] S5, according to the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node, a cluster head election is performed to obtain the formal cluster head;

[0016] S6, completing the cluster formation process and performing routing allocation;

[0017] S7. If the current time slice number N≠S, re-enter step S3 to continue collecting information, otherwise terminate data transmission.

[0018] Furthermore, the area is divided into arc circles, the radius of the arc circle is r area By formula: It is obtained that d0 represents the distance between the center of the arc circle and the ground, and θ represents the maximum elevation angle at which the gateway can establish a connection with the satellite.

[0019] Furthermore, the coordinate system takes the gateway station as the coordinate origin, and the horizontal and vertical coordinates simulate the constellation orbit, and the values ​​are discrete values ​​that conform to the orbit characteristics.

[0020] Furthermore, the length of the time slice T seg By formula Calculation, R is the radius of the earth, G is the gravitational constant, h is the orbital height of a single star, M is the mass of the earth, and by default the topology does not change within the same time slice.

[0021] Furthermore, the calculation formula of the distance d between the satellite and the ground gateway is: where x i ,y i Satellite i The value of the horizontal and vertical coordinates of the satellite. The energy consumption per unit time of the satellite is estimated as follows: total =E TX +E RX +E agg , where E TX is the energy consumption of satellite data transmission, E RX is the energy consumption of data reception, E agg Aggregate energy consumption for data; the number of neighboring communicable satellites of the satellite represents the number of all communicable satellites within the maximum communication radius of the satellite.

[0022] Furthermore, the distance d between the satellite and the ground gateway, the energy consumption E per unit time of the satellite total The three parameters of the satellite's neighborhood value m are combined with fuzzy logic for pre-clustering: the distance d between the satellite and the ground gateway, the satellite's energy consumption per unit time E total The neighborhood value m of the satellite is input into the existing fuzzy logic system, and the rule base is a dedicated component of the fuzzy logic system; the rule base satisfies the rule that the higher the neighborhood value, the greater the probability p of the satellite being selected as the cluster head, and the communication contention radius r is negatively correlated with the neighborhood value m.

[0023] Furthermore, the cluster head election means that each satellite decides whether to become a candidate cluster head according to the probability p of the satellite being selected as the cluster head. For all satellites among the candidate cluster heads, if the probability p of the satellite being selected as the cluster head within its communication competition radius and the communication competition radius r of the satellite as a cluster head node are better than other satellites, then it will be selected as the cluster head satellite, otherwise it will become a discrete satellite.

[0024] Furthermore, the cluster formation process is specifically as follows: if the energy consumption E of the cluster head per unit time is total The priorities of the distance from the discrete satellite to the cluster head and the distance from the cluster head to the ground station satisfy priority1>priority2>priority3, that is, the cluster formation process is completed; priority1 represents the energy consumption E of the cluster head per unit time. totalPrioriyt2 represents the priority of the distance from the discrete satellite to the cluster head, and prioriyt3 represents the priority of the distance d from the cluster head to the ground station;

[0025] The routing allocation is specifically as follows: if the satellite i If it is a non-cluster head satellite, then Satellite i The cluster head satellite is used as the Satellite i A relay for communicating with a gateway. If the Satellite i The cluster head satellite is not on the same orbital plane, and the distance threshold d is set. th , distance threshold d th Indicates the maximum communication radius of the satellite. i The distance between the cluster head satellite and the cluster head satellite is greater than d th , then the nearest adjacent cluster member satellite is selected as the next hop route, and the two communicate directly through the intersatellite link; if the satellite i If it is the cluster head satellite, its next hop is the ground gateway, and the routing allocation is completed.

[0026] Furthermore, the current time slice number N≠S, that is, the next time slice T N+1 The satellite has not reached the average maximum operating time threshold, which indicates the maximum value of the time slices that can be divided.

[0027] The beneficial effects of the method of the present invention are:

[0028] (1) The present invention significantly enhances routing adaptability and accuracy by deeply discussing the mobility of the VLEO satellite communication system and establishing a dynamic model close to reality.

[0029] (2) The present invention proposes a feeder link routing strategy based on a cluster model, which takes into account both the energy consumption and dynamic coverage of VLEO satellites, effectively improves resource utilization, enhances network performance and user experience, and prolongs the working life of network nodes and the overall network.

[0030] (3) The present invention adopts a cluster-based hierarchical management structure, which not only reduces the computational burden of the network control center, but also improves the scalability and maintainability of the system. As the number of satellites increases, the present invention can easily cope with the expansion of the network scale without large-scale transformation of the original system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of the method described in an embodiment of the present invention;

[0032] Figure 2 Schematic diagram of communication between a gateway and a satellite in an embodiment of the present invention;

[0033] Figure 3 In FIG. 1 , (a) is a schematic diagram of the membership function relationship of the probability p of the satellite becoming the head node of its neighborhood satellite cluster in an embodiment of the present invention, the horizontal axis is the value of the probability p of the satellite becoming the head node of its neighborhood satellite cluster, and the vertical axis is the membership of the probability p of the satellite becoming the head node of its neighborhood satellite cluster to the fuzzy level.

[0034] (b) is a schematic diagram of the membership function relationship of the communication competition radius r of the satellite as the cluster head node in an embodiment of the present invention, wherein the horizontal axis is the value of the communication competition radius r of the satellite as the cluster head node, and the vertical axis is the membership of the communication competition radius r of the satellite as the cluster head node to the fuzzy level;

[0035] Figure 4 (a) is a graph showing the variation of the output value of the communication contention radius r of the satellite as the cluster head node with the input value of the neighborhood value m and the distance d between the satellite and the ground gateway in an embodiment of the present invention.

[0036] (b) is a schematic diagram of the relationship between the fuzzy rule base of the fuzzy logic system in an embodiment of the present invention, showing the input distance d between the satellite and the ground gateway, the satellite energy consumption per unit time E total When the fuzzy level of the satellite's neighborhood value m is combined in different ways, the output of the mapping is the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the fuzzy level combination of the communication contention radius r of the satellite as the cluster head node. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] This embodiment provides a cluster-based dynamic VLO satellite feeder link routing method, the flow chart of the method is as follows: Figure 1 , the method comprises the following steps:

[0039] S1. The regions are divided according to the coverage of each gateway station in the very low orbit. K satellites and 1 gateway station in the region constitute a very low orbit satellite communication system, and the coverage of the gateway station is abstracted into a coordinate system.

[0040] The relevant operations of step S1 are introduced with a specific example:

[0041] For the entire satellite network, regional division is carried out according to the distribution of gateway stations. Regional division means dividing the very low orbit space covered by the satellite network into multiple small areas according to the distribution of gateway stations on the ground. There is only one gateway station in each divided area, so that the satellites in each small area can communicate with the gateway station. Figure 2 shown.

[0042] The divided area is a curved circle. Since the curved circle is similar to the tangent circle, the radius r of the curved circle is area The radius of the tangent circle is obtained by solving the formula, which is: Where d0 represents the distance from the center of the arc circle to the ground, and θ represents the maximum elevation angle at which the gateway can establish a connection with the satellite.

[0043] A very low orbit satellite communication system is established, which includes a gateway station and k VLEO satellites within the coverage area of ​​the very low orbit space of the gateway station, such as Figure 2 shown.

[0044] Let K VLEO satellites be {Satellite1,Satellite2,..,Satellite i ,...Satellite k The coverage of the gateway station is abstracted into a coordinate system at a height h from the ground. k VLEO satellites are randomly distributed in the orbit. Each satellite has different horizontal and vertical coordinates within the range. The horizontal and vertical coordinates simulate the constellation orbit, and their values ​​are discrete values ​​that conform to the orbit characteristics.

[0045] S2, define S time slices {T1, T2..., T N ,...,T S}.

[0046] The relevant operations of step S2 are introduced with a specific example:

[0047] The total duration of the time slices to be divided is obtained by min{total duration of ground gateway and satellite communication, average maximum continuous working duration of satellite}. The total duration of ground gateway and satellite communication represents the average overhead time of the satellite in the VLO coverage area. The average maximum continuous working duration of the satellite Δt is given by E tota1 Calculation, E total Represents the energy consumption of the satellite per unit time. Divide into S time slices {T1, T2..., T N ,...,T S}, T0 is defined as the initial state, until the next time slice, the satellite runs at a distance v·T in its orbit seg , where T seg is the time slice length, according to the formula It is calculated that v is the speed of the satellite, and the formula It is calculated that R is the radius of the earth, G is the gravitational constant, h is the orbital height of a single star, and M is the mass of the earth. By default, the topology does not change within the same time slice.

[0048] The distribution coordinates of the satellites in the next time slice need to be determined in combination with the moving direction and moving distance of the satellites. The specific parameter values ​​of this embodiment are: orbit height h = 400km, number of satellites k = 100, coverage radius r area =200km.

[0049] S3. Starting from the first time slice T1, the gateway calculates the distance d of each satellite from the ground gateway and estimates the energy consumption E of the satellite per unit time. total , the number of nearby communicable satellites of the satellite is recorded as the neighborhood value m.

[0050] The relevant operations of step S3 are introduced with a specific example:

[0051] According to the distribution of satellites in step S2 and the VLO coverage area of ​​the ground station, the ground gateway collects information such as the data transmission requirements of all satellites in the coverage area. The data transmission requirements of the satellite are specifically the total amount of data requested or received from the satellite by all users or ground stations within the satellite coverage area within Δt time. Consider the total service demand D of a single satellite uplink, and then according to Calculate the distance d between the satellite and the ground gateway, where d0 is approximately the orbital height of the satellite, and x i is the horizontal coordinate of the satellite in the VLO coverage area, y i is the vertical coordinate of the satellite in the VLE coverage area. Since all satellites are in VLE satellite orbits, their orbital heights are consistent. Calculate the energy consumption per unit time E total , the energy consumption of a single satellite is E total =E TX +E RX +E agg , where E TX is the energy consumption of satellite data transmission, E RX is the energy consumption of data reception, E agg is the energy consumption of data aggregation; the energy consumption of satellite data transmission is Its P send is the transmission power, G t is the transmitting antenna gain, G r is the receiving antenna gain, Δt is the data transmission duration, which can be approximated to the length of a time slice under full load working conditions, that is, the average maximum continuous working duration of the satellite. L is the free space path loss, which is specifically expressed as: Where d is the distance between the sender and the receiver, which generally refers to the information sender and the receiver, which can be between satellites or between satellites and gateways, f is the signal frequency, and c is the speed of light. The energy consumption of data reception is expressed as: Where P recv is the receiver power, D recv is the total amount of data received, R recv is the data receiving rate. Data aggregation energy consumption E agg =(P proc +P storage )×T proc , where P proc is the satellite processor power, P storage is the memory power, T proc The total time of data processing.

[0052] The unit time is the length of S time slices. The unit time length needs to take into account the average maximum continuous working time of the satellite. Specifically, the average maximum continuous working time of the satellite is limited by the battery capacity and load threshold, and the average maximum continuous working time of the satellite is less than the total time of the ground gateway and satellite communication.

[0053] The number of neighboring communicable satellites of each single satellite is recorded, and the number is recorded as the neighborhood value m of the satellite, wherein the neighboring communicable satellites can be determined according to the maximum communication radius of the single satellite in the same orbit and in different orbits. Specifically, the neighboring communicable satellites are all satellites included in the maximum communication radius of the single satellite in the same orbit and in different orbits. In this embodiment, the total service demand D=4Gbps, S=1500, and the maximum communication radius R MAXeq =50km, maximum communication radius R MAXdiff =25km.

[0054] S4, the distance d between the satellite and the ground gateway, the energy consumption E per unit time of the satellite total The three parameters of the satellite's neighborhood value m are combined with fuzzy logic for pre-clustering. After screening the rule base, the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node are obtained.

[0055] The relevant operations of step S4 are introduced with a specific example:

[0056] According to the distance d between the satellite and the ground gateway in step S3, the energy consumption E per unit time of the satellite total , the neighborhood value m of the satellite, and use the fuzzy logic tool to perform the pre-clustering preprocessing process, setting the input and output of the fuzzy logic system, because the distance d from the satellite to the ground gateway is determined by The d0 parameter is the orbital height, and the orbital heights of each satellite are consistent. Therefore, in this embodiment, the distance d from the satellite to the ground gateway is simplified to the distance d from the satellite to the origin of the coordinate system, and the calculation formula is: It can also indirectly indicate the distance between the satellite and the ground gateway. The fuzzy logic input is the distance d, energy consumption E total , neighborhood value m, the output is the probability p of the satellite becoming the head node of its neighborhood satellite cluster, and the fuzzy level of the communication competition radius r of the satellite as the head node, and obtains the membership function corresponding to its fuzzy level. The internal process of fuzzy logic is specifically divided into four parts. First, the three input variable sets are fuzzified in turn. For the distance d, three fuzzy levels of far, medium and near are set, representing the degree of distance from the satellite to the ground gateway; for the satellite energy consumption per unit time E total , set three fuzzy levels of low, medium and high, representing the energy consumption of satellites for communication within a unit time; for the neighborhood value m, set three fuzzy levels of more, medium and less, representing the level of the number of neighboring satellites that can establish a communication link within the maximum communication radius. After setting the fuzzy level of the input parameter, establish a functional relationship between the input parameter value and the fuzzy level. The corresponding relationship between the input and the input fuzzy level is the attribution function of the input corresponding to the input level, that is, the membership function. The membership functions of the distance d corresponding to the fuzzy levels of "near", "middle" and "far" are:

[0057]

[0058] Among them, d max It is the maximum distance between the satellite and the origin of the coordinate system within the VLE coverage area.

[0059] The membership functions of the neighborhood value m corresponding to the fuzzy levels of "few", "medium", "average", and "enormous" are:

[0060]

[0061] Among them, m max It is the maximum number of satellites that can communicate with a single satellite within the coverage of the VLO. The membership functions of energy consumption E corresponding to the fuzzy levels of "low", "medium", and "middle" are:

[0062]

[0063]

[0064] Among them, e max It is the maximum energy consumption of a single satellite within the VLE coverage area during the maximum connection establishment time.

[0065] The distance d, energy consumption E, and neighborhood value m are input into the fuzzy system, and the actual value is fuzzified into an appropriate language variable through the corresponding membership function. The specific process is as follows: first, the distance, energy consumption, and neighborhood values ​​are preprocessed, d and E are normalized, and the processed actual values ​​are brought into the corresponding membership function as variables to obtain the probability that the actual value belongs to the corresponding degree level. In this example, when the neighborhood value is 60, according to the membership function formula, the probability of belonging to "few" / "less" is 0, the probability of belonging to "medium" / "average" is 0.5, and the probability of belonging to "many" / "enormous" is 0.2. Therefore, when the domain value is 60, it is possible to belong to the "medium" and "many" degree levels. After normalizing the probability, the probability of belonging to "medium" is The probability of belonging to "many" is The degree of membership is determined according to the probability distribution value. The specific implementation method in this embodiment is to generate a random number within 1. If it is less than If it is too low, it belongs to “middle”; otherwise, it belongs to “more”.

[0066] The above steps fuzzify the clear input values ​​into appropriate language variables. Then set different weights ω1, ω2, and ω3 for the three input variables. ω1 represents the weight of the distance d between the satellite and the ground gateway, and ω2 represents the energy consumption E per unit time of the satellite. total The weight of ω3 represents the weight of the satellite's neighborhood value m, where ω1>ω2>ω3. The membership function of the output parameter is the probability p of the satellite becoming the head node of its neighborhood satellite cluster. Figure 3 (a), the horizontal axis is the probability p of the satellite becoming the head node of its neighborhood satellite cluster, and the vertical axis is the membership of the probability p of the satellite becoming the head node of its neighborhood satellite cluster to the fuzzy level. The membership function corresponding to the communication competition radius r of the satellite as the cluster head node is as follows: Figure 3 (b), the horizontal axis is the value of the communication competition radius r of the satellite as the cluster head node, in km, and the vertical axis is the membership of the communication competition radius r of the satellite as the cluster head node to the fuzzy level. The probability p of the satellite becoming the head node of its neighborhood satellite cluster is divided into nine fuzzy levels, and the communication competition radius r of the satellite as the cluster head node is divided into seven fuzzy levels. A corresponding rule base of input to output is established, and the logical relationship between input and output satisfies that the higher the neighborhood value, the greater the probability p of the satellite becoming the head node of its neighborhood satellite cluster, and the communication competition radius r of the satellite as the cluster head node is negatively correlated with the neighborhood value m. Figure 4(a) shows the output value of the communication contention radius r of the satellite as the cluster head node as the input value of the neighborhood value m and the distance d of the satellite from the ground gateway. r represents the communication contention radius of the satellite as the cluster head node, in km. m represents the neighborhood value of the satellite, which is a normalized value. d represents the distance of the satellite from the ground gateway, which is a normalized value. Figure 4 (b) A schematic diagram of the relationship between the fuzzy rule base of the fuzzy logic system, showing that when the fuzzy levels of the input satellite distance d from the ground gateway, the satellite energy consumption per unit time E_re and the satellite's neighborhood value m are taken in different combinations, the mapped output is the probability p of the satellite becoming the head node of its neighborhood satellite cluster and the fuzzy level combination of the communication competition radius r of the satellite as the cluster head node, and the membership function corresponding to the fuzzy level is obtained. The energy consumption input E_re in the rule base is the remaining available energy, and its value is equal to the energy consumption used for the average maximum working time of the satellite minus the energy consumption during the actual working time.

[0067] The specific operation is as follows: after the input is fuzzified into appropriate linguistic variables, the fuzzy level of the output parameter is obtained by referring to the rule base of fuzzy logic. After obtaining the fuzzy levels of the two parameters, the probability p of the output satellite becoming the cluster head node of its neighborhood satellite and the communication competition radius r of the satellite as the cluster head node, the defuzzification method is used to calculate the expected value of all possible outputs corresponding to each input.

[0068] The area center method is used to defuzzify the probability p of the satellite becoming the cluster head node of its neighborhood satellite and the fuzzy level of the communication competition radius r of the satellite as the cluster head node. The area center method is expressed as: Among them, μ(x) is the membership function, z is the defuzzified value of the corresponding membership function, and the center of gravity (center of mass) of the fuzzy set is calculated according to the area center method to determine the probability p of the satellite becoming the head node of its neighborhood satellite cluster and the clear value of the communication competition radius r of the satellite as the cluster head node after defuzzification, that is, the communication competition radius r of the satellite as the cluster head node.

[0069] S5. According to the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node, a cluster head election is performed to obtain a formal cluster head.

[0070] The following is an introduction to the VLO satellite routing operations using specific examples:

[0071] According to step S4, the probability p of each satellite becoming the cluster head node of its neighboring satellites and the communication competition radius r of the satellite as the cluster head node are calculated, and then the cluster head election is carried out. i The probability of becoming the head node of its neighborhood satellite cluster is p i , the communication competition radius of the cluster head node is r iEach satellite in the entire network has the probability p of becoming the cluster head node of its neighboring satellites. i Decide whether to become a candidate cluster head. If the satellite i Becomes a candidate cluster head, within the communication competition radius r of the cluster head node i There are other candidate cluster head satellites j , then compare p i 、p j and r i 、r j , if there is p i <p j And r i <r j , namely Satellite j The probability value and competition radius are better than Satellite i , then cancel the satellite i The candidate identities of the cluster heads are used as discrete satellites to wait for the selection of cluster heads to enter the cluster stage. After the election is completed, all candidate cluster heads become formal cluster heads, completing the star cluster formation stage.

[0072] S6. Complete the cluster formation process and perform route allocation.

[0073] The following is an introduction to the VLO satellite routing operations using specific examples:

[0074] According to the cluster head distribution in step S5, route allocation and task planning are performed. After the satellite as the cluster head is determined in the entire network, the remaining discrete satellites are i There may be more than one selectable cluster head satellite within its communication radius r. For multiple cluster heads within the range, discrete satellites i According to the different parameters of the neighboring cluster head satellites, the optimal cluster joining is determined within the range, and the parameter is considered in units of T seg The energy consumption E of the cluster head in time total , the distance from the discrete satellite to the cluster head, the distance d from the cluster head to the ground station, etc., energy consumption E total The priority of the satellite is priority 1, the priority of the satellite distance is priority 2, and the priority of the distance d from the cluster head to the ground station is priority 3. The three satisfy priority 1>priority 2>priority 3. The cluster head with the lowest energy consumption per unit time within the competition radius is selected as the cluster head, and the distance from the satellite to the cluster head is considered. The cluster formation process is completed and the route allocation is performed. If the satellite i Non-cluster head satellite, Satellite i Cluster Head Satellitej As Satellite i The relay communicating with the gateway, in this embodiment, the distance threshold d th Set to the maximum communication radius of the different track, that is, d th =R MAXdiff =25km, if Satellite j With Satellite i Not on the same orbital plane and the distance between them is greater than the distance threshold d th , then the nearest adjacent cluster member satellite is selected as the next hop route, and the two communicate directly through the intersatellite link. i To Satellite j The longer the distance, the more it will affect the number of hops between the two. i The cluster head satellite is the next hop to the ground gateway. After the route allocation is completed, the mission planning is carried out. i When serving as a cluster head, it waits for the service data of cluster members. At this stage, it is also necessary to consider that the satellite cache space is limited and the space utilization needs to be improved. i When receiving business data from cluster members, it should compare the existing cache and aggregate duplicate data before saving. i When it is a non-cluster head, after receiving service data from the ground gateway or from other cluster member satellites, it performs data aggregation, discards redundant data, and then sends the aggregated data to the next hop through the inter-satellite link until its service data is transmitted to the cluster head satellite. j , Satellite j Wait for the service data of all cluster members in the cluster to be collected within the specified time, aggregate the data after receiving it, and send the aggregated data to the ground gateway through the feeder link. i Routing allocation and data transmission within.

[0075] S7, if the current time slice number N≠S, then the next time slice T N+1 If the satellite does not reach the average maximum working time threshold, which indicates the maximum value of the time slice that can be divided, then the process reenters step S3, otherwise the data transmission ends.

Claims

1. A cluster-based dynamic very low earth orbit satellite feeder link routing method, characterized in that: The method comprises the following steps: S1. According to the coverage range of each gateway station in the very low orbit, the region is divided. K satellites and 1 gateway station in the region form a very low orbit satellite communication system. The coverage range of the gateway station is abstracted into a coordinate system. S2, define S time slices {T1, T2..., T N ,...,T S }, S represents the total number of time slices; S3. Starting from the first time slice T1, the gateway calculates the distance d of each satellite from the ground gateway and estimates the energy consumption E of the satellite per unit time. total , the number of neighboring communicable satellites of the satellite is recorded as the neighborhood value m; S4, the distance d between the satellite and the ground gateway, the energy consumption E per unit time of the satellite total The three parameters of the satellite's neighborhood value m are combined with fuzzy logic for pre-clustering. After screening the rule base, the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node are obtained. S5, according to the probability p of the satellite becoming the cluster head node of its neighborhood satellites and the communication competition radius r of the satellite as the cluster head node, a cluster head election is performed to obtain the formal cluster head; S6, completing the cluster formation process and performing routing allocation; S7. If the current time slice number N≠S, re-enter step S3 to continue collecting information, otherwise terminate data transmission.

2. The cluster-based dynamic VLO satellite feeder link routing method according to claim 1, characterized in that: The area is divided into arc circles, the radius of the arc circle is r area By formula: It is obtained that d0 represents the distance between the center of the arc circle and the ground, and θ represents the maximum elevation angle at which the gateway can establish a connection with the satellite.

3. The cluster-based dynamic VLO satellite feeder link routing method according to claim 1, characterized in that: The coordinate system takes the gateway station as the coordinate origin, and the horizontal and vertical coordinates simulate the constellation orbit, and the values ​​are discrete values ​​that conform to the orbit characteristics.

4. The cluster-based dynamic VLO satellite feeder link routing method according to claim 2, characterized in that: The length of the time slice T seg By formula Calculation, R is the radius of the earth, G is the gravitational constant, h is the orbital height of a single star, M is the mass of the earth, and by default the topology does not change within the same time slice.

5. The cluster-based dynamic VLO satellite feeder link routing method according to claim 2, characterized in that: The calculation formula of the distance d between the satellite and the ground gateway is: where x i ,y i Satellite i The satellite energy consumption per unit time is estimated as follows: total =E TX +E RX +E agg , where E TX is the energy consumption of satellite data transmission, E RX is the energy consumption of data reception, E agg Aggregate energy consumption for data; the number of neighboring communicative satellites of the satellite represents the number of all communicative satellites within the maximum communication radius of the satellite.

6. The cluster-based dynamic VLO satellite feeder link routing method according to claim 5, characterized in that: The distance d between the satellite and the ground gateway, the energy consumption E per unit time of the satellite total The three parameters of the satellite's neighborhood value m are combined with fuzzy logic for pre-clustering: the distance d between the satellite and the ground gateway, the satellite's energy consumption per unit time E total The neighborhood value m of the satellite is input into the existing fuzzy logic system, and the rule base is a dedicated component of the fuzzy logic system; the rule base satisfies the rule that the higher the neighborhood value, the greater the probability p of the satellite being selected as the cluster head, and the communication contention radius r is negatively correlated with the neighborhood value m.

7. The cluster-based dynamic VLO satellite feeder link routing method according to claim 6, characterized in that: The cluster head election means that each satellite decides whether to become a candidate cluster head according to the probability p of the satellite being selected as the cluster head. For all satellites among the candidate cluster heads, if the probability p of the satellite being selected as the cluster head and the communication competition radius r of the satellite as the cluster head node are better than other satellites within its communication competition radius, it will be selected as the cluster head satellite, otherwise it will become a discrete satellite.

8. The cluster-based dynamic VLO satellite feeder link routing method according to claim 7, characterized in that: The cluster formation process is specifically as follows: if the energy consumption of the cluster head per unit time is E total The priorities of the distance from the discrete satellite to the cluster head and the distance from the cluster head to the ground station satisfy priority1>priority2>priority3, that is, the cluster formation process is completed; priority1 represents the energy consumption E of the cluster head per unit time. total priority, priority2 represents the priority of the distance from the discrete satellite to the cluster head, and priority3 represents the priority of the distance d from the cluster head to the ground station; The routing allocation is specifically as follows: if the satellite i If it is a non-cluster head satellite, then Satellite i The cluster head satellite is used as the Satellite i A relay for communicating with a gateway. If the Satellite i The cluster head satellite is not on the same orbital plane, and the distance threshold d is set. th , distance threshold d th Indicates the maximum communication radius of the satellite. i The distance between the cluster head satellite and the cluster head satellite is greater than d th , then the nearest adjacent cluster member satellite is selected as the next hop route, and the two communicate directly through the intersatellite link; if the satellite i If it is the cluster head satellite, its next hop is the ground gateway, and the routing allocation is completed.

9. The cluster-based dynamic VLO satellite feeder link routing method according to claim 1, characterized in that: The current time slice number N≠S, that is, the next time slice T N+1 The satellite has not reached the average maximum operating time threshold, which indicates the maximum value of the time slices that can be divided.

Images