A method for selecting access nodes in large-scale low-orbit satellite constellation networks

Abstract

The present invention relates to a method for selecting network access nodes for a large-scale low-orbit satellite constellation, belonging to the field of satellite communication technology. The method first calculates the position and beam coverage of high- and low-orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris, platform, and payload information. Then, a fitness function is designed based on the inter-satellite positional relationship, beam coverage, communication link channel quality, and service requirements. Finally, a genetic algorithm is used to select satellites suitable as network access nodes within the large-scale low-orbit satellite constellation. The present invention can be used to achieve efficient and dynamic connection node selection in a joint high- and low-orbit satellite network, thereby realizing functions such as inter-satellite resource coordination and load balancing, and improving the deployment efficiency of large-scale low-orbit satellite constellations.

Description

Technical Field

[0001] The present invention belongs to the field of satellite communication technology, and in particular to a method for selecting network access nodes in a large-scale low-orbit satellite constellation. The method is suitable for selecting network access nodes in a large-scale low-orbit satellite constellation. Background Art

[0002] The Starlink project proposed by the US company SpaceX is the largest low-orbit satellite constellation with the largest number of satellites to date. The entire plan will deploy about 42,000 small satellites in low-Earth orbit, including satellite constellations at different orbital altitudes and several ground stations. There was no intersatellite link between early Starlink satellites, and communications between satellites had to be relayed through earth stations, which resulted in Starlink services being unable to cover areas where there was no way to establish earth stations. The biggest improvement in the 1.5 version of the Starlink satellite is the ability to provide inter-satellite laser communication. Since the first batch of improved satellites were launched into orbit in September 2021, subsequent satellites launched have been equipped with laser intersatellite link components, which enable information transmission and exchange between Starlink satellites in the air, and data can be transmitted directly between satellites. Direct communication between Starlink satellites in orbit greatly reduces dependence on earth stations and improves the low-latency service guarantee capability of the Starlink system.

[0003] Large-scale low-orbit satellite constellations with inter-satellite links are the future development direction of low-orbit satellite communications. Furthermore, unified networking of low-orbit and high-orbit satellites can further reduce the satellite constellation's dependence on ground stations and improve system reliability. In this unified networking of low-orbit and high-orbit satellites, a subset of low-orbit satellites must be selected as network nodes for the large-scale low-orbit constellation, serving as service aggregation. However, no such solution exists in existing technologies. Summary of the Invention

[0004] This paper addresses the problem of selecting network access nodes in large-scale low-orbit satellite constellations and proposes a method for selecting network access nodes in large-scale low-orbit satellite constellations. The method uses a genetic algorithm to select satellites suitable as network access nodes in a large-scale low-orbit satellite constellation.

[0005] The technical solution adopted by the present invention is:

[0006] A method for selecting access nodes in a large-scale low-orbit satellite constellation network comprises the following steps:

[0007] Step 1: Calculate the position and beam coverage of high and low orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris, platform, and payload information.

[0008] Step 2: Design a fitness function based on the inter-satellite location relationship, beam coverage, communication link channel quality, and service demand.

[0009] Step 3: Based on the fitness function designed in step 2, a genetic algorithm is used to select satellites suitable as network access nodes in a large-scale low-orbit satellite constellation.

[0010] Furthermore, the specific method of step 1 is:

[0011] Step 1-1: Calculate the position of the high-orbit and low-orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris information. The high-orbit satellite's transponder is transparent forwarding or processed forwarding, while the low-orbit satellite's transponder is processed forwarding.

[0012] Step 1-2: Calculate the satellite beam coverage of high-orbit and low-orbit satellites at each moment through the satellite platform and payload information, including the coverage of high-orbit satellites over low-orbit satellites and the ground, as well as the coverage of low-orbit satellites over the ground, and establish a communication link between high-orbit satellites and low-orbit satellites.

[0013] Furthermore, the specific method of step 2 is:

[0014] In step 2-1, based on the positional relationship between satellites, the following fitness function is designed to ensure that the network access node has as many adjacent low-orbit satellites as possible:

[0015] f1=n / N

[0016] Where n is the number of adjacent low-orbit satellites of the candidate network access node, and N is the total number of low-orbit satellites;

[0017] In step 2-2, based on the positional relationship between satellites and the coverage of high-orbit satellite beams, the following fitness function is designed to ensure that network access nodes have more connections with high-orbit satellites and are as close to them as possible:

[0018] f2=l / L+10 / min(d1 d2 d3…d l )

[0019] Where l is the number of high-orbit satellites that the alternative network access node can connect to, L is the total number of high-orbit satellites, d1d2d3…d l The distance between the candidate network access node and each high-orbit satellite that can be connected, in km, min(*) indicates the minimum value;

[0020] Step 2-3: Based on the payloads of high-orbit satellites and low-orbit satellites, and considering the signal-to-noise ratio and signal bandwidth of the communication link between the high-orbit satellite and the network access node, the following fitness function is designed:

[0021] f3=SNR1+0.1*BW1+SNR2+0.1*BW2+…SNR l +0.1*BWl

[0022] Where SNR is the signal-to-noise ratio of the communication link between the high-orbit satellite and the alternative network access node, measured in dB; BW is the signal bandwidth of the communication link between the high-orbit satellite and the alternative network access node, measured in kHz; and l is the number of high-orbit satellites that the alternative network access node can connect to.

[0023] In steps 2-4, consider the service traffic of the network access node and design the following fitness function:

[0024] f4=S0+S1+S2+…S n

[0025] Among them, S0 is the expected traffic volume of the candidate network access node, n is the number of adjacent low-orbit satellites of the candidate network access node, S1, S2, ..., S n The traffic volume between the candidate network access node and each adjacent low-orbit satellite, in Kbyte;

[0026] Steps 2-5, combine various fitness functions to obtain the final fitness function:

[0027] f=af1+bf2+cf3+df4

[0028] Among them: a, b, c, d are weights.

[0029] Furthermore, the specific method of step 3 is:

[0030] Step 3-1: Set the number of low-orbit satellites N, the number of high-orbit satellites L, the population size M of the genetic algorithm, as well as the initial network node probability P, mutation probability Pm, crossover probability Pc, and number of iterations G of the genetic algorithm. The chromosome length is equal to the number of low-orbit satellites N.

[0031] Step 3-2: Binary encode the population, set the satellites that serve as network nodes to "1", and set the other satellites to "0". Initially, the position of "1" is randomly set according to probability P;

[0032] Step 3-3: Calculate the satellite's position at a certain moment based on the satellite's ephemeris, configure the signal-to-noise ratio and signal bandwidth of the inter-satellite communication link, and the traffic volume of each low-orbit satellite, calculate the fitness f of the node in the individual, and then obtain the current fitness of each individual and calculate the selection probability;

[0033] Step 3-4, crossover is performed on the population with probability Pc and mutation is performed with probability Pm;

[0034] Steps 3-5: Update the population and select the individuals with the highest fitness to replace the individuals with the lowest fitness;

[0035] Step 3-6, determine whether the number of iterations has been reached. If so, the algorithm ends and the "1" in the individual represents the location of the network access node. Otherwise, return to step 3-3 and continue iteration.

[0036] Compared with the background technology, the present invention has the following advantages:

[0037] 1. The suitability function designed in this invention takes into account the number of adjacent satellites, the positional relationship between satellites, beam coverage, communication link channel quality, and the traffic volume between the local node and adjacent satellites, thereby better selecting satellites suitable as network nodes.

[0038] 2. The large-scale low-orbit satellite constellation network access nodes selected by the present invention can realize efficient dynamic connection between high-orbit and low-orbit satellites, achieve resource coordination and load balancing among satellites, and improve the deployment efficiency of large-scale low-orbit satellite constellations. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of an embodiment of the present invention.

[0040] Figure 2 It is a schematic diagram of an application scenario of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be described in detail below with reference to the accompanying drawings.

[0042] A method for selecting network nodes in a large-scale low-orbit satellite constellation network is proposed. The method first calculates the position and beam coverage of high- and low-orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's orbit, platform, and payload information. Then, a fitness function is designed based on the positional relationship between satellites, beam coverage, communication link channel quality, and service demand. Finally, a genetic algorithm is used to select satellites suitable as network access nodes in the large-scale low-orbit satellite constellation.

[0043] like Figure 1 As shown, the method specifically includes the following steps:

[0044] Step 1: Calculate the position and beam coverage of high and low orbit satellites in the ground-fixed coordinate system at each moment through the satellite's ephemeris, platform and payload information. Application scenarios include Figure 2 shown.

[0045] The specific steps for step 1 are as follows:

[0046] Step 1-1: Calculate the position of the high-orbit and low-orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris information. The high-orbit satellite's transponder is transparent forwarding or processed forwarding, while the low-orbit satellite's transponder is processed forwarding.

[0047] Step 1-2: Calculate the satellite beam coverage of high-orbit and low-orbit satellites at each moment through the satellite platform and payload information, including the coverage of high-orbit satellites over low-orbit satellites and the ground, as well as the coverage of low-orbit satellites over the ground, and establish a communication link between high-orbit satellites and low-orbit satellites.

[0048] Step 2: Design a fitness function based on the positional relationship between satellites, beam coverage, communication link channel quality, and service demand.

[0049] The specific steps for step 2 are as follows:

[0050] In step 2-1, based on the positional relationship between satellites, the following fitness function is designed to ensure that the network access node has as many adjacent low-orbit satellites as possible:

[0051] f1=n / N

[0052] Where n is the number of adjacent low-orbit satellites of the candidate network access node, and N is the total number of low-orbit satellites;

[0053] In step 2-2, based on the positional relationship between satellites and the coverage of high-orbit satellite beams, the following fitness function is designed to ensure that network access nodes have more connections with high-orbit satellites and are as close to them as possible:

[0054] f2=l / L+10 / min(d1 d2 d3…d l )

[0055] Where l is the number of high-orbit satellites that the alternative network access node can connect to, L is the total number of high-orbit satellites, d1d2d3…d l The distance between the candidate network access node and each high-orbit satellite that can be connected, in km, min(*) indicates the minimum value;

[0056] Step 2-3: Based on the payloads of high-orbit satellites and low-orbit satellites, and considering the signal-to-noise ratio and signal bandwidth of the communication link between the high-orbit satellite and the network access node, the following fitness function is designed:

[0057] f3=SNR1+0.1*BW1+SNR2+0.1*BW2+…SNR l +0.1*BW l

[0058] Where SNR is the signal-to-noise ratio of the communication link between the high-orbit satellite and the alternative network access node, measured in dB; BW is the signal bandwidth of the communication link between the high-orbit satellite and the alternative network access node, measured in kHz; and l is the number of high-orbit satellites that the alternative network access node can connect to.

[0059] In steps 2-4, consider the service traffic of the network access node and design the following fitness function:

[0060] f4=S0+S1+S2+…S n

[0061] Among them, S0 is the expected traffic volume of the candidate network access node, n is the number of adjacent low-orbit satellites of the candidate network access node, S1, S2, ..., S n The traffic volume between the candidate network access node and each adjacent low-orbit satellite is in Kbyte;

[0062] Steps 2-5, combine various fitness functions to obtain the final fitness function:

[0063] f=af1+bf2+cf3+df4

[0064] Among them, a, b, c, and d are weights. The purpose of weighting is to make the values ​​of af1, bf2, cf3, and df4 as close as possible. In this example, a = 0.8, b = 0.6, c = 0.001, and d = 0.001.

[0065] Step 3: Based on the fitness function designed in step 2, a genetic algorithm is used to select satellites suitable as network nodes in a large-scale low-orbit satellite constellation.

[0066] The specific steps for step 3 are as follows:

[0067] Step 3-1: Assume that the number of low-orbit satellites N = 1000, the number of high-orbit satellites L = 10, the population size of the genetic algorithm M = 50, the chromosome length is equal to the number of low-orbit satellites N, the mutation probability of the genetic algorithm Pm = 0.05, the crossover probability Pc = 0.8, and the number of iterations G = 200;

[0068] Step 3-2: Binary code the population, set the satellites that are network nodes to "1", otherwise set to "0", and randomly set the positions of "1". The probability of setting "1" in an individual is P = 0.1;

[0069] Step 3-3: Calculate the satellite's position at a certain moment based on the satellite's ephemeris, configure the signal-to-noise ratio and signal bandwidth of the connectable inter-satellite communication link, and the traffic volume of each low-orbit satellite, calculate the fitness f of the node in the individual, and then obtain the current fitness of each individual and calculate the selection probability;

[0070] Step 3-4: crossover is performed on the population with a probability of Pc = 0.8 and mutation is performed with a probability of Pm = 0.05;

[0071] Steps 3-5: Update the population and select individuals with high fitness to replace individuals with low fitness;

[0072] Step 3-6, determine whether the number of iterations has been reached. If so, the algorithm ends and the "1" in the individual represents the location of the network access node. Otherwise, return to step 3-3 and continue iteration.

[0073] The present invention can be used to realize the efficient dynamic connection node selection in the joint networking of high- and low-orbit satellites, thereby realizing functions such as resource coordination and load balancing among satellites, and improving the deployment efficiency of large-scale low-orbit satellite constellations.

Claims

1. A method for selecting access nodes in a large-scale low-orbit satellite constellation network, characterized in that: The steps include: Step 1: Calculate the position and beam coverage of high and low orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris, platform, and payload information. Step 2: Design a fitness function based on the inter-satellite location relationship, beam coverage, communication link channel quality, and service requirements. The specific method is as follows: In step 2-1, based on the positional relationship between satellites, the following fitness function is designed to ensure that the network access node has as many adjacent low-orbit satellites as possible: in, is the number of adjacent low-orbit satellites of the alternative network access node, is the total number of low-orbit satellites; In step 2-2, based on the positional relationship between satellites and the coverage of high-orbit satellite beams, the following fitness function is designed to ensure that network access nodes have more connections with high-orbit satellites and are as close to them as possible: in, is the number of high-orbit satellites that the alternative network access node can connect to, is the total number of high-orbit satellites, The distance from the alternative network access node to each high-orbit satellite that can be connected, in km, Indicates taking the minimum value; Step 2-3: Based on the payloads of high-orbit satellites and low-orbit satellites, and considering the signal-to-noise ratio and signal bandwidth of the communication link between the high-orbit satellite and the network access node, the following fitness function is designed: Among them, SNR is the signal-to-noise ratio of the communication link between the high-orbit satellite and the alternative network access node, the unit is dB, BW is the signal bandwidth of the communication link between the high-orbit satellite and the alternative network access node, the unit is kHz, The number of high-orbit satellites that the alternative network access nodes can connect to; In steps 2-4, consider the service traffic of the network access node and design the following fitness function: in, is the expected traffic volume of the candidate network access node, n is the number of adjacent low-orbit satellites of the candidate network access node, The traffic volume between the candidate network access node and each adjacent low-orbit satellite, in Kbyte; Steps 2-5, combine various fitness functions to obtain the final fitness function: Among them, a, b, c, and d are weights; Step 3: Based on the fitness function designed in step 2, a genetic algorithm is used to select satellites suitable as network access nodes in the large-scale low-orbit satellite constellation.

2. The method for selecting access nodes in a large-scale low-orbit satellite constellation network according to claim 1, wherein: The specific method of step 1 is: Step 1-1: Calculate the position of the high-orbit and low-orbit satellites in the Earth-fixed coordinate system at each moment using the satellite's ephemeris information. The high-orbit satellite's transponder is transparent forwarding or processed forwarding, while the low-orbit satellite's transponder is processed forwarding. Step 1-2: Calculate the satellite beam coverage of high-orbit and low-orbit satellites at each moment through the satellite platform and payload information, including the coverage of high-orbit satellites over low-orbit satellites and the ground, as well as the coverage of low-orbit satellites over the ground, and establish a communication link between high-orbit satellites and low-orbit satellites.

3. The method for selecting access nodes of a large-scale low-orbit satellite constellation network according to claim 1, characterized in that: The specific method of step 3 is: Step 3-1: Set the number of low-orbit satellites N, the number of high-orbit satellites L, the population size M of the genetic algorithm, as well as the initial network node probability P, mutation probability Pm, crossover probability Pc, and number of iterations G of the genetic algorithm. The chromosome length is equal to the number of low-orbit satellites N. Step 3-2: Binary encode the population, set the satellites that serve as network nodes to "1", and set the other satellites to "0". Initially, the position of "1" is randomly set according to probability P; Step 3-3: Calculate the satellite's position at a certain moment based on the satellite's ephemeris, configure the signal-to-noise ratio and signal bandwidth of the inter-satellite communication link, and the traffic volume of each low-orbit satellite, calculate the fitness f of the node in the individual, and then obtain the current fitness of each individual and calculate the selection probability; Step 3-4, crossover is performed on the population with probability Pc and mutation is performed with probability Pm; Steps 3-5: Update the population and select the individuals with the highest fitness to replace the individuals with the lowest fitness; Step 3-6: Determine whether the number of iterations has been reached. If so, the algorithm ends and the "1" in the individual represents the location of the network access node. Otherwise, return to step 3-3 and continue iterating.

Images