Low-cost communication path real-time calculation and optimization method for multi-user dynamic inter-satellite network

Abstract

The invention discloses a low-cost communication path real-time calculation and optimization method for a multi-user dynamic inter-satellite network, and the method comprises the steps: 1, modeling an inter-satellite network in a time period to an undirected graph, and distributing a time-related communication cost function for each link; step 2, calculating an intersection point of a link communication cost function, dividing a stable link period according to potential path change time corresponding to the intersection point, and calculating a double-connected component and an initial optimal communication path of a satellite topological graph; 3, judging links related to each cross point by using a filtering mechanism, and updating a path by using a link exchange strategy; and step 4, outputting a stable communication period and dynamic update of the optimal communication path. According to the method, the communication path is calculated and updated in real time by using the thought of stable communication period and link exchange, and the communication cost is ensured to be minimized. According to the method, the communication cost and response time of the inter-satellite network with the link cost changing along with time can be obviously reduced, and repeated calculation of the optimal communication path is avoided.

Description

Technical Field

[0001] The invention relates to a real-time calculation and optimization method for a low-cost communication path of a dynamic inter-satellite network for multi-user applications, and belongs to the technical field of satellite communications. Background Art

[0002] With the acceleration of globalization and the rapid development of information technology, the global communication demand continues to grow. Especially in the context of low earth orbit (LEO) satellite networks, they have become a key infrastructure to support international communications, distance education, and global emergency response initiatives with their low communication latency and wide coverage capabilities. A key issue of traditional terrestrial networks is accessibility in rural and remote areas, such as deserts and mountainous areas. In addition, with the increase in Internet of Things (IoT) devices and the surge in mobile Internet users, the importance of satellite networks has increased significantly. They not only connect remote areas, but also support global navigation and communication for marine, air, and land vehicles. In recent years, more and more applications have put forward higher requirements on communication costs, bringing challenges to the design of communication satellite networks. To meet the diverse needs of network applications, LEO satellite constellations are developing towards large-scale, high-density, and multi-layer configurations. With the expansion of satellite constellation size and the enhancement of satellite dynamic characteristics, the satellite communication environment has become increasingly complex. These networks require dynamic topology control algorithms to optimize communication paths, minimize energy consumption, and improve network performance. LEO satellite constellations consist of multiple satellites distributed in their respective orbits. Compared with terrestrial network communications, factors such as satellite mobility and link fluctuations may lead to significant differences between LEO satellite networks and terrestrial networks. Specifically: 1) Predictability: Orbital dynamics can be used to accurately predict the trajectory of each satellite node, allowing anticipated adjustments to the network topology. This enables network planners to pre-compute optimal routes and adapt to changes in connectivity, ensuring continuous and reliable communication. 2) Link dynamics: The rapid movement of satellites causes the relative distance between them to continuously change. This requires real-time adaptation to maintain the optimal communication path. The dynamic nature of satellite movement poses a significant challenge to maintaining a stable and efficient network topology, as the optimal route for data transmission may change frequently. 3) Resource limitations: Each satellite has a limited number of optical transceivers, and these resources need to be used efficiently to maintain network connectivity and performance. The limited availability of transceivers means that the network must be carefully managed to ensure that all users have access to the necessary communication channels without overloading any single satellite node.

[0003] Currently, most studies on implementing real-time optimization of dynamic topology in satellite communication networks rely on static or predefined time periods to manage the network topology. Satellite algorithms can be divided into static and dynamic algorithms depending on whether satellite dynamics are considered. Static algorithms exploit the predictability and periodicity of satellite networks to cope with the dynamic changes of satellite topology, assuming that the network topology remains unchanged over a certain period of time. Therefore, these static algorithms can simplify the topology by taking advantage of the periodic characteristics of satellite operation and generally show lower complexity. However, as satellites move and the network size increases, the connection cost of inter-satellite links (ISLs) keeps changing, leading to an increase in the overall communication cost. Given the dynamic nature of satellite topology, dynamic satellite algorithms have been proposed to enhance the reliability of paths. These algorithms exploit the information acquisition and processing capabilities of satellites to update the status information of topology and ISLs. Subsequently, reliable network routes are calculated based on the collected status information. Dynamic algorithms allow path optimization based on real-time satellite status information and better adapt to the dynamic changes of satellite network environment. However, most algorithms optimize for a single point in time and need to be recalculated frequently as conditions change, aggregating the minimum spanning tree as a static discrete snapshot. This not only increases the computational burden, but may also affect the quality and reliability of communication due to untimely responses. Summary of the invention

[0004] In view of the limitations of the existing technology, the present invention proposes a low-cost real-time calculation and optimization method for communication paths of multi-user dynamic intersatellite networks, aiming to optimize satellite multi-user communications in real time, utilizing the predictive adjustment of orbital dynamics and the processing capabilities of continuous network topology changes, and performing dynamic local updates through filtering mechanisms and link exchange strategies to more accurately maintain the optimal communication path, thereby improving network performance and reducing energy consumption.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] The present invention proposes a method for real-time calculation and optimization of low-cost communication paths of a multi-user dynamic intersatellite network, the method comprising:

[0007] Step 1: Given a time period [1, K], the intersatellite network of the entire time period is modeled to obtain an undirected graph G = (V, E), where V represents the set of satellite nodes and E represents the set of links between satellites in the time period. The communication cost function of each intersatellite link is obtained by taking the intersatellite link communication cost into account with the relative distance between satellites. This is usually a function of time and is expressed as where e ij represents the link between two satellites i and j, t is a time variable, and t∈[1,K], assigns edge weight function to each edge of G=(V,E) in time period [1,K];

[0008] Step 2: Calculate the intersection point set Set on the time axis based on the predicted communication cost functions of multiple intersatellite links point , mathematically, this can be expressed as finding the value of time t such that Where e1 and e2 represent any two edges in the satellite network. and It is the communication cost function of two different links e1 and e2. The value of time t represents the potential path change time PPCT ε According to the potential path change time, the time period [1, K] is divided into multiple stable link periods SIP α =[α s , α e ], where ε is the time-ordered index of potential path change times, α is the index of stable link periods, and α s and α e Respectively represent the start time and end time of the αth stable link period. Calculate the dual-connected components of the satellite topology graph, and select the communication path at time t = 1 through the classic minimum spanning tree greedy algorithm, which is represented by Path opt (1).

[0009] Step 3: Use a filtering mechanism to determine whether to update the communication path at each intersection in the potential path change time. The current communication path will be updated if and only if the following conditions are met: the intersection involves a selected intersatellite link and an unselected link in the current optimal communication path, both of which belong to the same dual-connected component, and the communication cost value of the unselected link after the intersection is less than the communication cost value of the selected intersatellite link.

[0010] Step 4: loop step 3, and output each stable communication cycle SCP based on the updated result in step 3 β =[β s , β e ], where β is the index of the stable communication period, β s and β e They represent the start time and end time of the stable communication cycle respectively, so as to obtain the optimal communication path with the lowest communication cost in the complete time period [1, K] and its dynamic update.

[0011] In the step 1:

[0012] Step 1.1) Given a time period [1, K], this time period can be defined as the inter-satellite communication time period, or it can be determined based on the periodic characteristics of satellite motion, defined as the length of time it takes for a satellite to complete a motion cycle.

[0013] Step 1.2) The communication cost of the intersatellite network link in this time period varies with time. The intersatellite network in this time period is modeled as an undirected weighted graph G = (V, E). Where V = {1, 2, ..., n} represents the vertex set corresponding to all satellite nodes, E = {e ij |i, j∈V, i≠j} represents a set of edges indicating interconnectedness. Assume that there are n satellites and m intersatellite links in this satellite network, and the m intersatellite links can ensure the connectivity of n satellites.

[0014] Step 1.3) In order to establish the initial topology map, it is necessary to determine whether a connection can be established between satellites. In a satellite network, the line of sight (LOS) condition must be met between two satellites, that is, the signal transmission path between the two satellites is not blocked by the earth, atmosphere or other physical obstacles.

[0015] In order to have a LOS connection between satellites i and j in a satellite network, the central angle and Certain conditions must be met to ensure that the intersatellite link is not obstructed by the Earth, the atmosphere, or other obstacles, among which e ij represents the link between satellites i and j, st i 、st j represent the straight line from satellite i to the center of the earth and the straight line from satellite j to the center of the earth, respectively, and the central angle and Respectively represent e ij With st i The angle formed, e ij With st j The angle formed. Center angle and The calculation is as follows (1):

[0016]

[0017] Among them, L ij (t) is the relative distance between satellites i and j at time t, d i and d j They represent the orbital heights of satellites i and j from the center of the earth respectively.

[0018] When the link e between satellites i and j ij When tangent to the atmosphere, the two satellites are geometrically visible to each other, so to ensure that the two satellites are visible to each other, the central angle and Certain conditions must be met, taking into account the Earth's radius R e and the thickness of the atmosphere h a , these conditions are as follows:

[0019]

[0020] Although the central angle and The existence of a direct path is confirmed, but communication must be done by the actual distance L ij (t) occurs, i.e., the relative distance between satellites i and j at time t. This distance is crucial for calculating transmission power, signal attenuation, and the overall intersatellite link budget. It varies predictably over time due to the motion of the satellites and must satisfy the following condition (3):

[0021]

[0022] Among them, when the intersatellite link e ij When tangent to the atmosphere, and Respectively represent the relative distances from the tangent point to satellites i and j.

[0023] The visibility condition formula (3) ensures that the intersatellite link is not blocked by the earth or the atmosphere. Considering the characteristics of laser transmission in LEOSN, the Euclidean distance is used to estimate the relative distance between satellites. The Euclidean distance L between two satellites at any time t ij (t) is calculated as follows:

[0024]

[0025] ζ=1-cosψ i cosψ j cos(λ i -λ j )-sinψ i sinψ j (5)

[0026] ψ=arcsin(sinτsinμ)(6)

[0027] λ=Ω+arctan(cosτtanμ)-ω e t(7)

[0028] μ=ω s t+γ(8)

[0029] Where r is the sum of the earth's radius and the satellite's orbital height, τ is the satellite's orbital inclination, and Ω is the position of the ascending intersection point in degrees. s It represents the angular velocity of the satellite orbiting the earth, in degrees per second. γ represents the initial phase angle of the satellite, ω e Represents the angular velocity of the Earth's rotation in degrees per second.

[0030] Step 1.4) In this network, assume that each edge e ijis assigned a time-varying non-negative edge weight that corresponds to the communication cost function of the intersatellite link. The cost is symmetric, meaning that the communication cost from satellite i to j is the same as the communication cost from satellite j to i, i.e. For existing edges, the intersatellite link communication cost is related to power consumption. In order to maintain communication, a certain amount of power consumption is required, and its value is related to the Euclidean distance L between satellites i and j. ij (t) is proportional to the square. Therefore, the intersatellite link communication cost Where k is a proportionality constant representing the effect of distance on communication cost.

[0031] In the step 2:

[0032] Step 2.1) Calculate the biconnected components of the topological structure of the undirected graph G, that is, the communication subnet to which each link in the satellite network belongs, and create an intersatellite link table to store the detailed information of each link. This table includes the unique ID of each intersatellite link, the two connected satellites and the communication subnet to which they belong.

[0033] Step 2.2) Calculate the predicted communication cost function for each intersatellite link at the intersection of the time period [1, K]. Mathematically, this can be expressed as finding the value of t such that Based on the time t of these intersections, the time point at which the optimal communication path may change can be identified. This time point is defined as the potential path change time PPCT ε , which divides the time period [1, K] into multiple non-overlapping stable link periods SIP α =[α s , α e ], all intersections are arranged in time order, and each intersection is stored as an intersection value PPCT ε represents the εth potential path change time, η represents the potential path change time PPCT ε The nth intersection point, e1, e2, ..., e k represents the links involved in this intersection, k represents the number of intersatellite links involved in this intersection, and its value is less than or equal to the number of links m in the entire network.

[0034] Step 2.3) At time point t = 1, use the classic minimum spanning tree greedy algorithm (such as Prim, Kruskal, etc.) to calculate the communication path with the minimum communication cost, and create a bit vector table to represent the state of each link, where each column corresponds to a link in the path and each row represents an optimal communication path. If the link exists in the optimal path, the value corresponding to the link is 1, otherwise it is 0. The initial optimal communication path is the optimal communication path at time t = 1, marked as Pathopt (1). For each subsequent optimal communication path update at each path switching time, a new row needs to be added to the bit vector table accordingly.

[0035] In the step three:

[0036] When processing the intersection set Set obtained in step 2.2) point , apply filters to filter out those intersections that will not change the optimal communication path: if the links involved in the intersection already exist in the current optimal communication path, or do not exist in the current optimal communication path, these intersections will not cause changes in the optimal path and can be filtered directly; if the links involved in the intersection belong to different communication subnets (i.e., biconnected components), these intersections can also be filtered because they will not affect the selection of communication paths across subnets. After applying these filters, they are grouped by biconnected components, and the previous two layers of filtering are performed again in each group. After these filters, check the relative order of the communication costs of the links involved in the intersection before and after the corresponding potential path change time. If the relative order before and after is the same, the intersection is filtered out. If no intersection is filtered after applying all filters, the current optimal communication path needs to be updated. The updating method is as follows: if the intersection involves only two links and they belong to the same base ring, the link with higher communication cost can be removed from the current optimal path through the "link exchange" strategy and replaced with an intersatellite link with lower communication cost; if the intersection involves more than two intersatellite links, the selected intersatellite link with higher communication cost can be replaced with the unselected intersatellite link with lower communication cost.

[0037] In the step 4:

[0038] If all the intersections in the potential link change time do not change the optimal communication path, the two stable link cycles adjacent to the potential link change time will be merged to form a new stable communication cycle. In this new stable communication cycle, the communication cost of the optimal communication path of the original stable link cycle remains the lowest; if the intersatellite link involved in the current intersection changes the optimal communication path, the potential path change time of the intersection is the path switching time. After the new optimal path is generated, its effective time period starts from the potential path change time of the intersection, corresponding to the beginning of the new stable communication cycle, and the effective time period of the previous optimal path ends at this time, corresponding to the end of the previous stable communication cycle. The previous optimal path and its stable communication cycle will be output to form a record of the optimal communication path in this time period. Therefore, the dynamic update of the optimal communication path with the lowest communication cost in the entire time period t∈[1, K] is obtained.

[0039] Beneficial effects of the present invention:

[0040] 1. The present invention aims at optimizing the minimum communication cost of the time-varying low-orbit satellite network. By identifying the stable link period and the stable communication period, the effective time and change frequency of the optimal path can be determined, thereby effectively solving the communication cost optimization problem in the time-varying LEOSN without having to recalculate from the beginning, aiming to always maintain a communication path with the lowest communication cost during the network communication process. The application of this concept converts discrete changes in the satellite into continuous changes, so that the communication path at each moment is the optimal path.

[0041] 2. The present invention locally adjusts the optimal path in a time-varying satellite network through a filtering mechanism and a link exchange strategy. Compared with the prior art, the algorithm significantly reduces the response time and resource consumption.

[0042] 2. The present invention uses lightweight data structures (such as bit vectors and hash tables) to store intersatellite link information and communication path status, so that the algorithm can quickly search and update spanning tree information under the limited computing and storage resources of satellite nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of a method for real-time calculation and optimization of low-cost communication paths of a multi-user dynamic intersatellite network in an embodiment of the present invention.

[0044] Figure 2 It is a structural schematic diagram of a model of a low-cost communication path real-time calculation and optimization method for a multi-user dynamic intersatellite network in an embodiment of the present invention.

[0045] Figure 3 It is a schematic diagram of a function of link communication cost varying with time in an embodiment of the present invention.

[0046] Figure 4 It is a schematic diagram of the dynamic update of the optimal communication path in the embodiment of the invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings:

[0048] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0049] The present invention aims at how to maintain the optimal communication path in a dynamically changing satellite network in real time, utilizes the predictive adjustment of orbital dynamics and the processing ability of continuous network topology changes, defines a stable link period and a stable communication period to ensure real-time optimization of the optimal communication path, takes filtering mechanism and link exchange strategy as the core, avoids unnecessary repeated calculations, improves the response rate of the entire network, and reduces power consumption.

[0050] The present invention proposes a real-time calculation and optimization method for low-cost communication paths of a multi-user dynamic intersatellite network, which stipulates the maximum power consumption of the link, the communication cost that the satellite can calculate, and no network congestion.

[0051] Under the above conditions, the real-time optimization steps of low-cost communication paths between multiple users are as follows: Figure 1 As shown. Specifically including the following steps:

[0052] Step 1: Given a time period [1, K], the intersatellite network of the entire time period is modeled to obtain an undirected graph G = (V, E), where V represents the set of satellite nodes and E represents the set of links between satellites in the time period. The communication cost function of each intersatellite link is obtained by taking the intersatellite link communication cost into account with the relative distance between satellites. This is usually a function of time and is expressed as where e ij Denotes the link between two satellites i and j, t is a time variable, and t∈[1, K], assigns edge weight function to each edge of G=(V, E) in time period [1, K], the specific steps are as follows:

[0053] Step 1.1) Given a time period [1, K], this time period can be defined as the inter-satellite communication time period, or it can be determined based on the periodic characteristics of satellite motion, defined as the length of time it takes for a satellite to complete a motion cycle.

[0054] Step 1.2) The communication cost of the intersatellite network link in this time period varies with time. The intersatellite network in this time period is modeled as an undirected weighted graph G = (V, E). Where V = {1, 2, ..., n} represents the vertex set corresponding to all satellite nodes, E = {e ij |i, j∈V, i≠j} represents a set of edges indicating interconnectedness. Assume that there are n satellites and m intersatellite links in this satellite network, and the m intersatellite links can ensure the connectivity of n satellites.

[0055] Step 1.3) In order to establish the initial topology map, it is necessary to determine whether a connection can be established between satellites. In a satellite network, the line of sight (LOS) condition must be met between two satellites, that is, the signal transmission path between the two satellites is not blocked by the earth, atmosphere or other physical obstacles.

[0056] In order to have a LOS connection between satellites i and j in a satellite network, the central angle and Certain conditions must be met to ensure that the intersatellite link is not obstructed by the Earth, the atmosphere, or other obstacles, among which e ij represents the link between satellites i and j, st i 、st j represent the straight line from satellite i to the center of the earth c and the straight line from satellite j to the center of the earth c, respectively, and the central angle and Respectively represent e ij With st i The angle formed, e ij With st j The angle formed. Center angle and The calculation is as follows (1):

[0057]

[0058] Among them, L ij (t) is the relative distance between satellites i and j at time t, d i and d j They represent the orbital heights of satellites i and j from the center of the earth respectively.

[0059] When the link e between satellites i and j ij When tangent to the atmosphere, the two satellites are geometrically visible to each other, so to ensure that the two satellites are visible to each other, the central angle and Certain conditions must be met, taking into account the Earth's radius R e and the thickness of the atmosphere h a , these conditions are as follows:

[0060]

[0061] Although the central angle and The existence of a direct path is confirmed, but communication must be done by the actual distance L ij (t) occurs, i.e., the relative distance between satellites i and j at time t. This distance is crucial for calculating transmission power, signal attenuation, and the overall intersatellite link budget. It varies predictably over time due to the motion of the satellites and must satisfy the following condition (3):

[0062]

[0063] Among them, when the intersatellite link e ij When tangent to the atmosphere, and Respectively represent the relative distances from the tangent point to satellites i and j.

[0064] The visibility condition formula (3) ensures that the intersatellite link is not blocked by the earth or the atmosphere. Considering the characteristics of laser transmission in LEOSN, the Euclidean distance is used to estimate the relative distance between satellites. The Euclidean distance L between two satellites at any time t ij (t) is calculated as follows:

[0065]

[0066] ζ=1-cosψ i cosψ j cos(λ i -λ j )-sinψ i sinψ j (5)

[0067] ψ=arcsin(sinτsinμ)(6)

[0068] λ=Ω+arctan(cosτtanμ)-ω e t(7)

[0069] μ=ω s t+γ(8)

[0070] Where r is the sum of the earth's radius and the satellite's orbital height, τ is the satellite's orbital inclination, and Ω is the position of the ascending intersection point in degrees. s It represents the angular velocity of the satellite orbiting the earth, in degrees per second. γ represents the initial phase angle of the satellite, ω e Represents the angular velocity of the Earth's rotation in degrees per second.

[0071] Step 1.4) In this network, assume that each edge e ij is assigned a time-varying non-negative edge weight that corresponds to the communication cost function of the intersatellite link. The cost is symmetric, meaning that the communication cost from satellite i to j is the same as the communication cost from satellite j to i, i.e. For existing edges, the intersatellite link communication cost is related to power consumption. In order to maintain communication, a certain amount of power consumption is required, and its value is related to the Euclidean distance L between satellites i and j. ij (t) is proportional to the square. Therefore, the intersatellite link communication cost Where k is a proportionality constant representing the effect of distance on communication cost.

[0072] Step 2: Calculate the intersection point set Set on the time axis based on the predicted communication cost functions of multiple intersatellite links point, mathematically, this can be expressed as finding the value of time t such that Where e1 and e2 represent any two edges in the satellite network. and It is the communication cost function of two different links e1 and e2. The value of time t represents the potential path change time PPCT ε According to the potential path change time, the time period [1, K] is divided into multiple stable link periods SIP α =[α s , α e ], where ε is the time-ordered index of potential path change times, α is the index of stable link periods, and α s and α e Respectively represent the start time and end time of the αth stable link period. Calculate the dual-connected components of the satellite topology graph, and select the communication path at time t = 1 through the classic minimum spanning tree greedy algorithm, which is represented by Path opt (1), the specific steps are as follows:

[0073] Step 2.1) Calculate the biconnected components of the topological structure of the undirected graph G, that is, the communication subnet to which each link in the satellite network belongs, and create an intersatellite link table to store the detailed information of each link. This table includes the unique ID of each intersatellite link, the two connected satellites and the communication subnet to which they belong.

[0074] Step 2.2) Calculate the predicted intersatellite link communication cost function The intersection set Set in the time period [1, K] point , these intersections Potential Path Change Time PPCT ε , which means that at this moment, the optimal communication path in the satellite network may change. According to the potential path change time, the time period [1, K] is divided into multiple non-overlapping stable link periods, which are defined as follows:

[0075] Stable link period, expressed as SIP α =[α s , α e ], defined as non-overlapping subsequences of time periods [1, K] during which the communication costs of all ISLs remain in a consistent order. During a stable link period, the ordering of ISL communication costs from lowest to highest does not change, thus ensuring a predictable and stable optimal communication path. This period is critical to identifying periods where the network topology can be effectively managed without recalculation.

[0076] Assuming there are m intersatellite links, when the communication cost of these m intersatellite links remains stable, the following formula needs to be satisfied:

[0077]

[0078] Among them, e1e2e3…e m They are m intersatellite links in the satellite network, SIP α Indicates the αth stable link period in SIP.

[0079] All intersections are arranged in order of their potential path change time. There will be one or more intersections in each potential path change time, and each intersection is stored as an intersection value. Through this step of calculation, the time point when the optimal communication path changes can be predicted in advance, thereby improving network performance.

[0080] Step 2.3) At time point t = 1, use the classic minimum spanning tree greedy algorithm (such as Prim, Kruskal, etc.) to calculate the communication path with the minimum communication cost, and create a bit vector table to represent the state of each link, where each column corresponds to a link in the path and each row represents an optimal communication path. If the link exists in the optimal path, the value corresponding to the link is 1, otherwise it is 0. The initial optimal communication path is the optimal communication path at time t = 1, marked as Path opt (1). For each subsequent optimal communication path update at each path switching time, a new row needs to be added to the bit vector table accordingly.

[0081] Step 3: Use a filtering mechanism to determine whether to update the communication path at each intersection in the potential path change time. The current communication path will be updated if and only if the following conditions are met: the intersection involves a selected intersatellite link and an unselected link in the current optimal communication path, both of which belong to the same dual-connected component, and the communication cost value of the unselected link after the intersection is less than the communication cost value of the selected intersatellite link. The specific steps are as follows:

[0082] For the set of intersection points Set obtained in step 2.2 point Traverse, assuming that the current intersection

[0083] Step 3.1) If e1, e2, ..., e k If all of them are selected intersatellite links in the current optimal communication path or all of them are unselected links, then these intersections will not cause the optimal path to change, and the next intersection will be directly determined. If the current intersection is PPCT ε The last intersection of , then the next intersection belongs to That is the next potential path change time PPCT ε+1 If the current intersection is not PPCTε The last intersection of

[0084] Step 3.2) If e1, e2, ..., e k Belonging to different dual-connected components, which means belonging to different communication subnets, these intersections will not lead to changes in the optimal path, because they will not affect the selection of communication paths across subnets, and the next intersection will be judged directly.

[0085] Step 3.3) After the filtering mechanism of step 3.1) and step 3.2), if the intersection is still not filtered, then e1, e2, ..., e k Group by biconnected components B q , where q represents the qth communication subnet. Step 3.1) and step 3.2) are filtered again in each group. After these filters, check the links in each group. First, sort the communication costs of the links before and after the corresponding potential path change time, and then compare the two sorting results O1 and O2. If the relative order of the two sortings remains consistent, it means that the cost fluctuations of these links have no effect on the selection of the optimal communication path, so check the next group B. q+1 Links in .

[0086] Step 3.4) After the filtering mechanism of step 3.1), step 3.2) and step 3.3), it indicates that the current optimal communication path needs to be dynamically updated. q Are there only two links in the network and do they belong to the same public ring? If both conditions are met, the link exchange strategy is used to remove the link with higher communication cost from the current optimal path and replace it with a link with lower communication cost to form a new optimal communication path. opt (Ф), where Φ represents the optimal communication path updated for the Φth time. If there are more than two links in the group, traverse the unselected links in O2, find the base ring to which they belong, remove the link with higher communication cost in the base ring from the current optimal path, and add this link to the optimal path to form a new optimal communication path Path opt (Φ). Step 3.5) If all the conditions for updating the optimal path in steps 3.3) and 3.4) are not met, it means that the intersection will not affect the selection of the optimal path, and the next intersection or Make a judgment.

[0087] Step 4: loop step 3, and output each stable communication cycle SCP based on the updated result in step 3 β =[β s , βe ], where β is the index of the stable communication period, β s and β e Respectively represent the start time and end time of the stable communication cycle, so as to obtain the optimal communication path with the lowest communication cost in the complete time period [1, K] and its dynamic update. The specific steps are as follows:

[0088] According to whether a new optimal communication path Path is formed in step 3.4) opt The result of (Φ) updates the stable communication period, and finally the entire time period is divided into multiple non-overlapping stable communication periods, which are defined as follows:

[0089] Stable communication period, denoted as SCP β =[β s , β e ] is the maximum continuous cycle in the time period [1, K], and the optimal communication path in this cycle remains unchanged. It ensures the continuous optimal path while ensuring the stability of the network.

[0090] If no new optimal communication path is formed, it means that the potential path change time PPCT corresponding to this intersection ε Although there is a change in the intersatellite link communication cost, it does not affect the selection of the optimal path and will be consistent with PPCT ε The two adjacent stable link periods are merged. The original stable link periods are SIP α =[α s , α e ]=[PPCT ε-1 , PPCT ε ] and SIP α+1 =[(α+1) s , (α+1) e ]=[PPCT ε , PPCT ε+1 ] Merged into SCP β =[PPCT ε-1 , PPCT ε+1 ], and this stable communication period SCP β The optimal communication path remains unchanged as long as no new optimal communication path is formed. opt (Φ), the stable communication cycle will continue to merge and update until all time periods [1, K] are covered.

[0091] If a new optimal communication path is formed, it means that the potential path change time PPCT of this intersection ε It is also the path switching time PST ρ, where ρ represents the ρth path switching time. Each path switching time requires adding a new row to the bit vector table, thereby recording the status of all intersatellite links in the satellite network. The original stable link period SIP α =[α s , α e ]=[PPCT ε-1 , PPCT ε ] is defined as a new stable communication period SCP β =[PPCT ε-1 , PPCT ε ]=[PPCT ε-1 , PST ρ ], the optimal communication path in this cycle is Path opt (Ф-1), path switching time PST ρ is the start time of the new optimal communication path, the original stable link period SIP α+1 =[(α+1) s , (ɑ+1) e ]=[PPCT ε , PPCT ε+1 ] is defined as a new stable communication period SCP β+1 =[PPCT ε , β e ]=[PST ρ , β e ], the optimal communication path in this stable communication cycle is Path opt (Φ), where β e is unknown, β e The value needs to be changed by the next potential path change time PPCT ε+1 The decision is made using the filtering mechanism in step three.

[0092] Steps 3 and 4 are repeated to finally determine a set of dynamic optimal communication paths {Path opt (SCP β )), so that each Path opt (SCP β ) minimizes the total cost within the corresponding stable communication period, ensuring that the communication path always maintains the lowest cost within the time period [1, K].

Claims

1. A method for real-time calculation and optimization of low-cost communication paths of a multi-user dynamic intersatellite network, comprising: Step 1: Given a time period [1, K], the intersatellite network of the entire time period is modeled to obtain an undirected graph G = (V, E), where V represents the set of satellite nodes and E represents the set of links between satellites in the time period. The communication cost function of each intersatellite link is obtained by taking the intersatellite link communication cost into account with the relative distance between satellites. This is usually a function of time and is expressed as where e ij represents the link between two satellites i and j, t is a time variable, and t∈[1,K], assigns edge weight function to each edge of G=(V,E) in time period [1,K]; Step 2: Calculate the intersection point set Set on the time axis based on the predicted communication cost functions of multiple intersatellite links point , mathematically, this can be expressed as finding the value of time t such that Where e1 and e2 represent any two edges in the satellite network. and It is the communication cost function of two different links e1 and e2. The value of time t represents the potential path change time PPCT ε . The time period [1, K] is divided into multiple stable link periods according to the potential path change time where ε is the time-ordered index of potential path change times, α is the index of stable link periods, and α s and α e Respectively represent the start time and end time of the αth stable link period. Calculate the dual-connected components of the satellite topology graph, and select the communication path at time t = 1 through the classic minimum spanning tree greedy algorithm, which is represented by Path opt (1). Step 3: Use a filtering mechanism to determine whether to update the communication path at each intersection in the potential path change time. The current communication path will be updated if and only if the following conditions are met: the intersection involves a selected intersatellite link and an unselected link in the current optimal communication path, both of which belong to the same dual-connected component, and the communication cost value of the unselected link after the intersection is less than the communication cost value of the selected intersatellite link. Step 4: Loop through step 3 and output each stable communication cycle based on the updated result in step 3 The optimal communication path within , where β is the index of the stable communication period, β s and They represent the start time and end time of the stable communication cycle respectively, so as to obtain the optimal communication path with the lowest communication cost in the complete time period [1, K] and its dynamic update. In the step 1: Step 1.1) Given a time period [1, K], this time period can be defined as the inter-satellite communication time period, or it can be determined based on the periodic characteristics of satellite motion, defined as the length of time it takes for a satellite to complete a motion cycle. Step 1.2) The communication cost of the intersatellite network link in this time period varies with time. The intersatellite network in this time period is modeled as an undirected weighted graph G = (V, E). Where V = {1, 2, ..., n} represents the vertex set corresponding to all satellite nodes, and E = {e ij |i,j∈V, i≠j} represents a set of edges indicating interconnections. Assume that there are n satellites and m intersatellite links in this satellite network, and the m intersatellite links can ensure the connectivity of n satellites. Step 1.3) In order to establish the initial topology map, it is necessary to determine whether a connection can be established between satellites. In a satellite network, the line of sight (LOS) condition must be met between two satellites, that is, the signal transmission path between the two satellites is not blocked by the earth, atmosphere or other physical obstacles. In order to have a LOS connection between satellite i and , in the satellite network, the central angle and Certain conditions must be met to ensure that the intersatellite link is not obstructed by the Earth, the atmosphere, or other obstacles, among which e ij represents the link between satellites i and j, st i 、st j They represent the straight line from satellite i to the center of the earth, the straight line from satellite j to the center of the earth, and the central angle and Respectively represent e ij With st i The angle formed, e ij With st j The angle formed. Center angle and The calculation is as follows (1): Among them, L ij (t) is the relative distance between satellites i and j at time t, d i and d j They represent the orbital heights of satellites i and j from the center of the earth respectively. When the link e between satellites i and j ij When tangent to the atmosphere, the two satellites are geometrically visible to each other, so to ensure that the two satellites are visible to each other, the central angle and Certain conditions must be met, taking into account the radius of the Earth and the thickness of the atmosphere h a , these conditions are as follows: Although the central angle and The existence of a direct path is confirmed, but communication must be done by the actual distance L ij (t) occurs, i.e., the relative distance between satellites i and j at time t. This distance is crucial for calculating transmission power, signal attenuation, and the overall intersatellite link budget. It varies predictably over time due to the motion of the satellites and must satisfy the following condition (3): Among them, when the intersatellite link e ij When tangent to the atmosphere, and Respectively represent the relative distances from the tangent point to satellites i and j. The visibility condition formula (3) ensures that the intersatellite link is not blocked by the earth or the atmosphere. Considering the characteristics of laser transmission in LEOSN, the Euclidean distance is used to estimate the relative distance between satellites. The Euclidean distance L between two satellites at any time t ij (t) is calculated as follows: ξ=1-cosψ i cosψ j cos(λ i -l j )-sinψ i sinψ j (5) m = oh s t+y(8) Where r is the sum of the earth's radius and the satellite's orbital height, τ is the satellite's orbital inclination, and Ω is the position of the ascending intersection point in degrees. s It represents the angular velocity of the satellite orbiting the earth, in degrees per second. γ represents the initial phase angle of the satellite, ω e Represents the angular velocity of the Earth's rotation in degrees per second. Step 1.4) In this network, assume that each edge e ij is assigned a time-varying non-negative edge weight that corresponds to the communication cost function of the intersatellite link. The cost is symmetric, meaning that the communication cost from satellite i to j is the same as the communication cost from satellite j to i, i.e. For existing edges, the intersatellite link communication cost is related to power consumption. In order to maintain communication, a certain amount of power consumption is required, and its value is related to the Euclidean distance L between satellites i and j. ij The intersatellite link communication cost is proportional to the square of (t). Where k is a proportionality constant representing the effect of distance on communication cost. In the step 2: Step 2.1) Calculate the biconnected components of the topological structure of the undirected graph G, that is, the communication subnet to which each link in the satellite network belongs, and create an intersatellite link table to store the detailed information of each link. This table includes the unique ID of each intersatellite link, the two connected satellites and the communication subnet to which they belong. Step 2.2) Calculate the predicted communication cost function for each intersatellite link at the intersection of the time period [1, K]. Mathematically, this can be expressed as finding the value of t such that Based on the time t of these intersections, the time point at which the optimal communication path may change can be identified. This time point is defined as the potential path change time PPCT ε , which divides the time period [1, K] into multiple non-overlapping stable link periods All intersections are arranged in their time order, and each intersection is stored as an intersection value PPCT ε represents the εth potential path change time, η represents the potential path change time PPCT ε The nth intersection point, e1, e2, ..., e k represents the links involved in this intersection, k represents the number of intersatellite links involved in this intersection, and its value is less than or equal to the number of links m in the entire network. Step 2.3) At time point t = 1, use the classic minimum spanning tree greedy algorithm (such as Prim, Kruskal, etc.) to calculate the communication path with the minimum communication cost, and create a bit vector table to represent the state of each link, where each column corresponds to a link in the path and each row represents an optimal communication path. If the link exists in the optimal path, the value corresponding to the link is 1, otherwise it is 0. The initial optimal communication path is the optimal communication path at time t = 1, marked as Path opt (1). For each subsequent optimal communication path update at each path switching time, a new row needs to be added to the bit vector table accordingly. In the step three: When processing the intersection set Set obtained in step 2.2) point , apply filters to filter out those intersections that will not change the optimal communication path: if the links involved in the intersection already exist in the current optimal communication path, or do not exist in the current optimal communication path, these intersections will not cause changes in the optimal path and can be filtered directly; if the links involved in the intersection belong to different communication subnets (i.e., biconnected components), these intersections can also be filtered because they will not affect the selection of communication paths across subnets. After applying these filters, they are grouped by biconnected components, and the previous two layers of filtering are performed again in each group. After these filters, check the relative order of the communication costs of the links involved in the intersection before and after the corresponding potential path change time. If the relative order before and after is the same, the intersection is filtered out. If no intersection is filtered after applying all filters, the current optimal communication path needs to be updated. The updating method is as follows: if the intersection involves only two links and they belong to the same base ring, the link with higher communication cost can be removed from the current optimal path through the "link exchange" strategy and replaced with an intersatellite link with lower communication cost; if the intersection involves more than two intersatellite links, the selected intersatellite link with higher communication cost can be replaced with the unselected intersatellite link with lower communication cost. In the step 4: If all the intersections in the potential link change time do not change the optimal communication path, the two stable link cycles adjacent to the potential link change time will be merged to form a new stable communication cycle. In this new stable communication cycle, the communication cost of the optimal communication path of the original stable link cycle remains the lowest; if the intersatellite link involved in the current intersection changes the optimal communication path, the potential path change time of the intersection is the path switching time. After the new optimal path is generated, its effective time period starts from the potential path change time of the intersection, corresponding to the beginning of the new stable communication cycle, and the effective time period of the previous optimal path ends at this time, corresponding to the end of the previous stable communication cycle. The previous optimal path and its stable communication cycle will be output to form a record of the optimal communication path in this time period. Therefore, the dynamic update of the optimal communication path with the lowest communication cost in the entire time period t∈[1, K] is obtained.

2. The method for real-time calculation and optimization of low-cost communication paths in a multi-user dynamic intersatellite network according to claim 1, characterized in that The steps of step 2.2) include: Calculate the predicted intersatellite link communication cost function The intersection set Set in the time period [1, K] point , these intersections Potential Path Change Time PPCT ε This means that at this moment, the optimal communication path in the satellite network may change. According to the potential path change time, the time period [1, K] is divided into multiple non-overlapping stable link periods, which are defined as follows: Stable link period, expressed as SIP α =[α s , α e ], defined as non-overlapping subsequences of time periods [1, K] during which the communication costs of all ISLs remain in a consistent order. During a stable link period, the ordering of ISL communication costs from lowest to highest does not change, thus ensuring a predictable and stable optimal communication path. This period is critical to identifying periods where the network topology can be effectively managed without recalculation. Assuming there are m intersatellite links, when the communication cost of these m intersatellite links remains stable, the following formula needs to be satisfied: Among them, e1e2e3…e m They are m intersatellite links in the satellite network, SLP α Indicates the αth stable link period in SIP. All intersections are arranged in order of their potential path change time. There will be one or more intersections in each potential path change time, and each intersection is stored as an intersection value. Through this step of calculation, the time point when the optimal communication path changes can be predicted in advance, thereby improving network performance.

3. The method for real-time calculation and optimization of low-cost communication paths of a multi-user dynamic intersatellite network according to claim 1, characterized in that The steps of step three include: For the set of intersection points Set obtained in step 2.2 point Each intersection point in To traverse. Step 3.1) If e1, e2, ..., e k If all of them are selected intersatellite links in the current optimal communication path or all of them are unselected intersatellite links, then these intersections will not lead to changes in the optimal path, and the next intersection will be directly determined. If the current intersection is the last intersection of PPCTε, then the next intersection belongs to That is the next potential path change time PPCT ε+1 If the current intersection is not PPCT ε The last intersection of Step 3.2) If e1, e2, ..., e k Belonging to different dual-connected components, which means belonging to different communication subnets, these intersections will not lead to changes in the optimal path, because they will not affect the selection of communication paths across subnets, and the next intersection will be judged directly. Step 3.3) After the filtering mechanism of step 3.1) and step 3.2), if the intersection is still not filtered, then e1, e2, ..., e k Group by biconnected components B q , where q represents the qth communication subnet. Step 3.1) and step 3.2) are filtered again in each group. After these filters, the intersatellite links in each group are checked. The communication costs of the intersatellite links are sorted before and after the corresponding potential path change time, and then the two sorting results O1 and O2 are compared. If the relative order of the two sortings is consistent, it means that the cost fluctuations of these intersatellite links have no effect on the selection of the optimal communication path, so the next group B is checked. q+1 intersatellite links. Step 3.4) After the filtering mechanism of step 3.1), step 3.2) and step 3.3), it indicates that the current optimal communication path needs to be dynamically updated. q Whether there are only two intersatellite links in the network and whether they belong to the same public ring. If both conditions are met, the link exchange strategy is used to remove the intersatellite link with higher communication cost from the current optimal path and replace it with the intersatellite link with lower communication cost to form a new optimal communication path. opt (Ф), where Φ represents the optimal communication path updated for the Φth time. If there are more than two intersatellite links in the group, the intersatellite links that are not selected in o2 are traversed to find the base ring to which they belong, and the intersatellite links with higher communication costs in the base ring are removed from the current optimal path, and this intersatellite link is added to the optimal path to form a new optimal communication path Path opt (Ф). Step 3.5) If all the conditions for updating the optimal path in steps 3.3) and 3.4) are not met, it means that the intersection will not affect the selection of the optimal path, and the next intersection or Make a judgment.

4. The method for real-time calculation and optimization of low-cost communication paths in a multi-user dynamic intersatellite network according to claim 1, It is characterized in that in the step 4 ： According to whether a new optimal communication path Path is formed in step 3.4) opt The result of (Ф) updates the stable communication period, and finally the entire time period is divided into multiple non-overlapping stable communication periods, which are defined as follows: Stable communication period, expressed as It is the maximum continuous cycle in the time period [1, K], and the optimal communication path in this cycle remains unchanged. It ensures the stability of the network while ensuring the continuous optimal path. If no new optimal communication path is formed, it means that the potential path change time PPCT corresponding to this intersection ε Although there is a change in the intersatellite link communication cost, it does not affect the selection of the optimal path and will be consistent with PPCT ε The two adjacent stable link periods are merged. The original stable link periods are SIP α =[α s , α e ]=[PPCT ε-1 , PPCT ε ] and SIP α+1 =[(α+1) s , (α+1) e ]=[PPCT ε ,PPCT ε+1 ] Merged into SCP β =[PPCT ε-1 ,PPCT ε+1 ], and this stable communication period SCP β The optimal communication path remains unchanged as long as no new optimal communication path is formed. opt (Ф), the stable communication cycle will continue to merge updates until all time periods [1, K] are covered. If a new optimal communication path is formed, it means that the potential path change time PPCT of this intersection ε It is also the path switching time PST ρ , where ρ represents the ρth path switching time. Each path switching time requires adding a new row to the bit vector table, thereby recording the status of all intersatellite links in the satellite network. The original stable link period Defined as a new stable communication period SCP β =[PPCT ε-1 , PPCT ε ]=[PPCT ε-1 , PST ρ ], the optimal communication path in this cycle is Path opt (Ф-1), path switching time PST ρ is the start time of the new optimal communication path, the original stable link period Defined as a new stable communication period SCP β+1 =[PPCT ε , β e ]=[PST ρ ,β e ], the optimal communication path in this stable communication cycle is Path opt (Φ), where is unknown, The value needs to be changed by the next potential path change time PPCT ε+1 The decision is made using the filtering mechanism in step three. Steps 3 and 4 are repeated to finally determine a set of dynamic optimal communication paths {Path opt (SCP β )}, so that each Path opt (SCP β ) minimizes the total cost within the corresponding stable communication period, ensuring that the communication path always maintains the lowest cost within the time period [1, K].

Images