Method and device for judging dynamic switching of satellite clusters and storage medium

Abstract

The invention discloses a method and device for judging dynamic switching of satellite clusters and a storage medium, and the method comprises the steps: determining a clustering condition in a communication coverage range of a gateway station at a current moment, and enabling the clustering condition to represent the satellite distribution condition of each first cluster in the communication coverage range; calculating the data transmission quantity between the cluster head of each first cluster and the gateway station at the current moment, and determining a corresponding first mean value and a first variance based on the data transmission quantity; acquiring a second mean value and a second variance at each historical moment; the first mean value, the first variance, the multiple second mean values and the multiple second variances are input into a pre-trained probability prediction model, the probability prediction model is used for outputting a first probability of re-clustering at the current moment, and the probability prediction model comprises an RNN model, a full connection layer and a sigmoid function; and based on the first probability and a preset first probability threshold, judging whether re-clustering is needed at the next moment.

Description

Technical Field

[0001] This application relates to the field of satellite communication technology, and in particular to a method, apparatus, and storage medium for determining dynamically switching satellite clusters. Background Technology

[0002] A gateway station can establish communication connections with multiple satellites within its coverage area to transmit data. Within this coverage area, these satellites are divided into multiple clusters, each containing multiple satellites. Within each cluster, a unique cluster head needs to be designated. The cluster head is responsible for managing all nodes within the cluster and coordinating communication between them. Data generated by all nodes within a cluster is typically first sent to the cluster head, which then aggregates the data and sends the aggregated data to the gateway station.

[0003] However, since satellites are constantly in motion, satellites will move out of the gateway station's communication coverage area at any time, and satellites will also enter the gateway station's communication coverage area. Therefore, it is necessary to remove satellites that have been removed from the gateway station's communication coverage area from the cluster based on the actual situation, and to reassign satellites that have entered the gateway station's communication coverage area into the cluster.

[0004] In other words, if the number of satellites within the communication coverage area changes or a cluster head moves out of the gateway station's coverage area, re-clustering is generally required. To reduce unnecessary overhead, existing technologies propose determining whether re-clustering is necessary before re-clustering, and then deciding whether to re-cluster based on the assessment. However, since existing assessment methods only consider the instantaneous situation at a certain moment and ignore the historical trend of cluster changes, when a temporary or local anomaly occurs in the cluster head within a cluster, existing assessment methods may generate incorrect assessment results (e.g., requiring re-clustering). Incorrect assessment results will inevitably cause network oscillations and additional overhead.

[0005] The existing technologies mentioned above, which determine whether clustering is necessary by considering only the instantaneous situation at a certain moment and ignoring the historical trend of clustering changes, have not yet proposed an effective solution to the technical problem that when a cluster head in a cluster experiences a brief anomaly, the incorrect judgment result will cause network oscillation and additional overhead. Summary of the Invention

[0006] The embodiments of this disclosure provide a method, apparatus, and storage medium for determining dynamic switching of satellite clusters, in order to at least solve the technical problem that the current method of determining whether clustering is needed only considers the instantaneous situation at a certain moment and ignores the historical trend of clustering changes. As a result, when a brief anomaly occurs in the cluster head within a cluster, the incorrect judgment result will cause network oscillation and additional overhead.

[0007] According to one aspect of the present disclosure, a method for determining dynamic switching of satellite clusters is provided, comprising: determining the clustering situation within the communication coverage area of ​​a gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; calculating the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determining a corresponding first mean and a first variance based on the data transmission volume; obtaining a second mean and a second variance at each historical time, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at a historical time; inputting the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and using the probability prediction model to output a first probability of re-clustering at the current time, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and determining whether re-clustering is needed at the next time based on the first probability and a pre-set first probability threshold.

[0008] According to another aspect of the present disclosure, a storage medium is also provided, the storage medium including a stored program, wherein, when the program is executed, a processor performs any of the methods described above.

[0009] According to another aspect of the present disclosure, an apparatus for determining dynamic switching of satellite clusters is also provided, comprising: a clustering situation determination module, configured to determine the clustering situation within the communication coverage area of ​​a gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; a first calculation module, configured to calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determine the corresponding first mean and first variance based on the data transmission volume; a first data acquisition module, configured to acquire the second mean and second variance at each historical time, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at a historical time; a probability output module, configured to input the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and output a first probability of re-clustering at the current time using the probability prediction model, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and a first judgment module, configured to determine whether re-clustering is required at the next time based on the first probability and a pre-set first probability threshold.

[0010] According to another aspect of the present disclosure, an apparatus for determining dynamic switching of satellite clusters is also provided, comprising: a processor; and a memory connected to the processor, configured to provide the processor with instructions for processing the following steps: determining the clustering situation within the communication coverage area of ​​a gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; calculating the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determining a corresponding first mean and a first variance based on the data transmission volume; obtaining a second mean and a second variance at each historical time, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at a historical time; inputting the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and using the probability prediction model to output a first probability of re-clustering at the current time, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and determining whether re-clustering is needed at the next time based on the first probability and a pre-set first probability threshold.

[0011] This application provides a method for determining dynamic satellite cluster switching. First, the gateway station determines the clustering status within its communication coverage area at the current moment. Then, the gateway station calculates the data transmission volume between the current moment and the cluster heads of each of the next first clusters, and determines the corresponding first mean and first variance based on the data transmission volume. Further, the gateway station obtains the second mean and second variance at each historical moment. Then, the gateway station inputs the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and uses the probability prediction model to output the first probability of re-clustering at the current moment. Finally, based on the first probability and a pre-set first probability threshold, the gateway station determines whether re-clustering is needed at the next moment.

[0012] As described above, this application not only obtains the load balancing degree among the cluster heads of each first cluster at the current moment, but also obtains the load balancing degree among the cluster heads of each second cluster at multiple historical moments. Therefore, by inputting the first mean and first variance of the cluster heads of each first cluster at the current moment, and multiple second means and multiple second variances of the cluster heads of each second cluster at various historical moments into the probabilistic prediction model, the probabilistic prediction model can determine whether re-clustering is necessary based on the instantaneous data at the current moment and the historical trends at multiple historical moments. The probabilistic prediction model includes an RNN model, a fully connected layer, and a sigmoid function.

[0013] That is, unlike the existing technology that only considers the instantaneous situation at the current moment to determine whether re-clustering is needed, this application combines historical trends with instantaneous situations, considers the clustering situation over a continuous period of time, improves the accuracy of the judgment results, and avoids network oscillations or additional overhead between the gateway station and the satellite.

[0014] This solves the technical problem that the current method of determining whether clustering is needed only considers the instantaneous situation at a certain moment and ignores the historical trend of clustering changes. As a result, when a cluster head in a cluster experiences a brief anomaly, the incorrect judgment result will cause network oscillation and additional overhead. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of this disclosure and form part of this application, illustrate exemplary embodiments of this disclosure and are used to explain this disclosure, but do not constitute an undue limitation of this disclosure. In the drawings: Figure 1 This is a schematic diagram illustrating the communication connection relationship between the gateway station and multiple satellites as described in Embodiment 1 of this application; Figure 2A This is a schematic diagram of the hardware architecture of a plurality of first satellites according to Embodiment 1 of this application; Figure 2B This is a schematic diagram of the hardware architecture of the gateway station according to Embodiment 1 of this application; Figure 3 This is a flowchart of the method for determining dynamic switching of satellite clusters according to Embodiment 1 of this application; Figure 4 This is a schematic diagram of the architecture of the probability prediction model according to Embodiment 1 of this application; Figure 5 This is a schematic diagram of the cluster head determination model according to Embodiment 1 of this application; Figure 6 This is a schematic diagram illustrating the training of the conversion model according to Embodiment 1 of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] Example 1 According to this embodiment, a method embodiment for determining a dynamic switching satellite cluster is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] Figure 1 A schematic diagram illustrating the communication connection between the gateway station and multiple satellites according to this embodiment is shown. (Reference) Figure 1 As shown, the communication coverage area of ​​gateway station 100 includes multiple first satellites 201-20n, and these multiple first satellites 201-20n can be divided into multiple clusters. For example, cluster 301 includes satellites 201-203, cluster 302 includes satellites 204-205, and so on.

[0020] For example, the cluster head in cluster 301 is the first satellite 201, and the first satellites 202 and 203 transmit data information to the gateway station 100 through the first satellite 201. As another example, the cluster head in cluster 302 is the first satellite 204, and the first satellite 205 transmits data information to the gateway station 100 through the first satellite 204.

[0021] In addition, since the satellite is constantly moving, the gateway station 100 needs to determine whether re-clustering is necessary at predetermined time intervals.

[0022] Figure 2A Further shown Figure 1 A schematic diagram of the hardware architecture of several first-generation satellites (201-20n). (Reference) Figure 2AAs shown, multiple first satellites 201-20n include an integrated electronic system, which includes a processor, a memory, a bus management module, and a communication interface. The memory is connected to the processor, allowing the processor to access the memory, read program instructions stored in the memory, read data from the memory, or write data to the memory. The bus management module is connected to the processor and also to a bus such as a CAN bus. Thus, the processor can communicate with onboard peripherals connected to the bus through the bus managed by the bus management module. Furthermore, the processor also communicates with devices such as cameras, star sensors, telemetry and control transponders, and data transmission equipment via the communication interface. Those skilled in the art will understand that... Figure 2A The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, a satellite system may also include... Figure 2A The more or fewer components shown, or having the same Figure 2A The different configurations shown.

[0023] Figure 2B Further shown Figure 1 A schematic diagram of the hardware architecture of CITIC Gateway 100. (Reference) Figure 2B As shown, the gateway station 100 may include one or more processors (processors may include, but are not limited to, microprocessors such as MCUs or programmable logic devices such as FPGAs), a memory for storing data, a transmission device for communication functions, and an input / output interface. The memory, transmission device, and input / output interface are connected to the processor via a bus. In addition, it may also include a display, keyboard, and cursor control device connected to the input / output interface. Those skilled in the art will understand that... Figure 2B The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, the ground system may also include... Figure 2B The more or fewer components shown, or having the same Figure 2B The different configurations shown.

[0024] It should be noted that, Figure 2A and Figure 2B One or more processors and / or other data processing circuits shown herein may generally be referred to as "data processing circuitry". This data processing circuitry may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be integrated, in whole or in part, into any other element in a computing device. As involved in embodiments of this disclosure, the data processing circuitry serves as processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0025] Figure 2A and Figure 2BThe memory shown can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining dynamic switching of satellite clusters in the embodiments of this disclosure. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the method for determining dynamic switching of satellite clusters in the application described above. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.

[0026] It should be noted here that, in some optional embodiments, the above... Figure 2A and Figure 2B The device shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 2A and Figure 2B This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned devices.

[0027] Under the aforementioned operating environment, according to the first aspect of this embodiment, a method for determining dynamically switching satellite clusters is provided, the method comprising: Figure 2B The gateway station 100 shown is implemented. Figure 3 A flowchart illustrating the method is shown below. (Refer to...) Figure 3 As shown, the method includes: S302: Determine the clustering situation within the communication coverage area of ​​the gateway station at the current time, where the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; S304: Calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determine the corresponding first mean and first variance based on the data transmission volume, where the first variance represents the load balancing degree between the cluster heads of each first cluster at the current time. S306: Obtain the second mean and second variance at each historical time point, where the second variance represents the load balancing degree among the cluster heads of each second cluster at each historical time point; S308: Input the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probabilistic prediction model, and use the probabilistic prediction model to output the first probability of re-clustering at the current time step. The probabilistic prediction model includes an RNN model, fully connected layers, and a sigmoid function. S310: Based on the first probability and the pre-set first probability threshold, determine whether re-clustering is needed in the next moment.

[0028] Specifically, first, the gateway station 100 determines the clustering situation within the communication coverage area at the current moment (S302). The clustering situation at the current moment refers to the clustering situation within the communication coverage area of ​​the gateway station 100. For example, at the current moment... Below, the first cluster Includes the first satellite ~ The second cluster Includes the first satellite ~ ..., the first cluster Includes the first satellite ~ And the first cluster Cluster head inside For the first satellite The first cluster Cluster head inside For the first satellite ..., the first cluster Cluster head inside For the first satellite .

[0029] Then, the gateway station 100 calculates the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determines the corresponding first mean and first variance based on the data transmission volume (S304). For example, the gateway station 100 can perform a step of determining whether re-clustering is needed at preset time intervals. Therefore, firstly, the gateway station 100 obtains the current time... Below, each of the first clusters ~ cluster head ~ The amount of data transmitted between the gateway station 100 and the gateway station 100. Of particular note are the data transmission volumes of each first cluster. ~ cluster head ~ Data transmission volume between gateway station 100 ~ Including each first cluster ~ cluster head ~ Data transmission volume between itself and gateway station 100 and corresponding clusters ~ Other nodes within the cluster are accessed through the cluster head. ~ The amount of data transmitted will not be elaborated here.

[0030] Then, the gateway station 100, based on the determined connections with each... ~ cluster head ~ Corresponding data transfer volume ~ The first mean was calculated. and first variance The specific calculation formula is as follows: in, Indicates the current time Below, each of the first clusters ~ cluster head ~ The first average amount of data transmission. This represents the amount of data transmitted to the j-th cluster head. Indicates the current time Below, each of the first clusters ~ cluster head ~ The first variance of the data transmission volume. j = 1~m. And where the first variance represents the load balancing degree among the cluster heads of each first cluster at the current moment.

[0031] Furthermore, the gateway station 100 acquires various historical moments. ~ Second mean ~ Second variance ~ (S306). Here, the second variance represents the load balancing degree among the cluster heads of each second cluster at the corresponding historical time. For example, at the historical time... The second mean of the cluster heads of each second cluster Second variance Historical moment The second mean of the cluster heads of each second cluster Second variance And so on; historical moments The second mean of the cluster heads of each second cluster Second variance .

[0032] Furthermore, the first average value of the 100th station will be... First variance Multiple second means ~ and multiple second variances ~ Input into a pre-trained probabilistic prediction model. Figure 4 This is a schematic diagram of the architecture of the probability prediction model according to an embodiment of this application. (Reference) Figure 4 As shown, the probabilistic prediction model includes an RNN model, fully connected layers, and a sigmoid function. The RNN model is a neural network model specifically designed for processing sequential data.

[0033] Therefore, the probabilistic prediction model in this application uses an RNN model, which can capture temporal dependencies in sequential data by transmitting information over time. Specifically, the RNN model uses hidden states to store information from the previous time step and pass it to the next time step, thus achieving information transfer and memorization. In the RNN model, each time step has an input and an output, as well as a hidden state. The hidden state is determined by the hidden state of the previous time step and the input of the current time step, containing information from all inputs up to this point. Therefore, the RNN model can handle sequential data of arbitrary length and can capture long-term dependencies in the sequence through hidden states. In other words, given the data transmission volume at the current moment and the data transmission volume at each historical moment, this application can use the RNN model to determine the temporal dependencies between the current moment and each historical moment. This allows for the combination of historical trends and instantaneous conditions, resulting in a more accurate prediction output from the probabilistic prediction model.

[0034] That is, the probability prediction model receives the first mean of the above data. First variance Multiple second means ~ and multiple second variances ~ In this case, based on the above information, the first probability of needing to re-cluster at the current moment can be output. (S308).

[0035] Furthermore, the gateway station 100 uses a probabilistic prediction model to output the first probability that re-clustering is required. In this case, it can be based on the first probability. Compared with a pre-set first probability threshold The relationship between their magnitudes determines the next moment. Does re-clustering need to be performed (S310)?

[0036] As described in the background section, since satellites are constantly in motion, satellites will move out of the communication coverage area of ​​the gateway station at any given moment, and satellites will also enter the communication coverage area of ​​the gateway station. Therefore, it is necessary to remove satellites that have been removed from the communication coverage area of ​​the gateway station from the cluster based on the actual situation, and to reassign satellites that have entered the communication coverage area of ​​the gateway station into the cluster.

[0037] In other words, if the number of satellites within the communication coverage area changes or a cluster head moves out of the gateway station's coverage area, re-clustering is generally required. To reduce unnecessary overhead, existing technologies propose determining whether re-clustering is necessary before re-clustering, and then deciding whether to re-cluster based on the assessment. However, since existing assessment methods only consider the instantaneous situation at a certain moment and ignore the historical trend of cluster changes, when a temporary or local anomaly occurs in the cluster head within a cluster, existing assessment methods may generate incorrect assessment results (e.g., requiring re-clustering). Incorrect assessment results will inevitably cause network oscillations and additional overhead.

[0038] In view of this, this application provides a method for determining dynamic switching of satellite clusters. Furthermore, as described above, this application not only obtains the load balancing degree among the cluster heads of each first cluster at the current moment, but also obtains the load balancing degree among the cluster heads of each second cluster at multiple historical moments. Therefore, by inputting the first mean and first variance of the cluster heads of each first cluster at the current moment, and multiple second means and multiple second variances of the cluster heads of each second cluster at various historical moments into a probabilistic prediction model, the probabilistic prediction model can determine whether re-clustering is necessary based on the instantaneous data at the current moment and the historical trends at multiple historical moments. The probabilistic prediction model includes an RNN model, a fully connected layer, and a sigmoid function.

[0039] That is, unlike the existing technology that only considers the instantaneous situation at the current moment to determine whether re-clustering is needed, this application combines historical trends with instantaneous situations, considers the clustering situation over a continuous period of time, improves the accuracy of the judgment results, and avoids network oscillations or additional overhead between the gateway station and the satellite.

[0040] This solves the technical problem that the current method of determining whether clustering is needed only considers the instantaneous situation at a certain moment and ignores the historical trend of clustering changes. As a result, when a cluster head in a cluster experiences a brief anomaly, the incorrect judgment result will cause network oscillation and additional overhead.

[0041] Optionally, the operation of determining whether re-clustering is needed at the next moment based on the first probability and a preset first probability threshold includes: indicating that re-clustering is needed at the next moment if the first probability is greater than the first probability threshold; and indicating that re-clustering is not needed at the next moment if the first probability is less than or equal to the first probability threshold.

[0042] Specifically, in the first probability Greater than the first probability threshold In this case, it indicates the next moment. Re-clustering is required. And in the first probability Less than or equal to the first probability threshold In this case, it indicates the next moment. No need for re-clustering.

[0043] Optionally, if the first probability is greater than the first probability threshold, the re-clustering operation is required, including: identifying multiple first satellites located within the communication coverage area of ​​the gateway station; obtaining multiple first satellite information corresponding to each first satellite using ephemeris information, and determining a first vector corresponding to each first satellite based on the multiple first satellite information; inputting the multiple first vectors into a pre-trained conversion model, and outputting a second vector corresponding to each first satellite based on the conversion model; and determining the clustering situation at the next time step based on the multiple second vectors.

[0044] Specifically, firstly, in the first probability Greater than the first probability threshold In this case, gateway station 100 identifies multiple first satellites located within the communication coverage area of ​​gateway station 100. ~ Then, gateway station 100 uses ephemeris information to obtain information from each of the first satellites. ~ Corresponding multiple first satellite information ~ For example, with the first satellite Corresponding first satellite information Including the x-coordinate of the first satellite The ordinate of the first satellite The vertical coordinates of the first satellite Average signal-to-noise ratio of the input link (Represents the average signal-to-noise ratio of data received from other first satellites), average signal-to-noise ratio of the output link. (Represents the average signal-to-noise ratio of data transmitted to other first satellites), channel bandwidth Average connectivity Cache usage And computing resource utilization Etc. (l parameter information)

[0045] Similarly, gateway station 100 can also acquire the first satellite. ~ The above-mentioned multiple first satellite information ~ .

[0046] Then, based on information from multiple first satellites, the gateway station 100 determines the first vector corresponding to each first satellite. ~ For example, X1=[x 1,1 ,x 2,1 ,...,x l,1 ] T X2=[x 1,2 ,x 2,2 ,...,x l,2 ] T ;...;X n =[x 1,n ,x 2,n ,...,x l,n ] T .

[0047] Furthermore, gateway station 100 will use multiple first vectors ~ The input is fed into a pre-trained transformation model, and the first vector is processed based on the transformation model. ~ Processing is performed to output the results to each of the first satellites. ~ The corresponding second vector ~ The transformation model is used to generate precise parameter information that allows for accurate clustering. For example, with the first satellite... The corresponding second vector Y1=[y 1,1 ,y 2,1 ,...,y l,1 ] T ; with the first satellite The corresponding second vector Y2=[y 1,2 ,y 2,2 ,...,y l,2 ] T And so on; with the first satellite The corresponding second vector Y1=[y 1,n ,y 2,n ,...,y l,n ] T .

[0048] Optionally, the operation of determining the clustering situation at the next time step based on multiple second vectors includes: constructing a matrix based on multiple second vectors, inputting the matrix into a pre-trained cluster head determination model, and determining a second probability and a third probability of each first satellite as a cluster head based on the cluster head determination model, wherein the second probability represents the probability of the first satellite being a cluster head and the third probability represents the probability of the first satellite not being a cluster head; if the second probability is greater than a pre-set second probability threshold, determining multiple second satellites among the multiple first satellites that will be cluster heads; if multiple second satellites are determined, determining multiple third satellites among the multiple first satellites that are not the multiple second satellites; and clustering the multiple second satellites and third satellites.

[0049] Specifically, Figure 5 This is a schematic diagram of the cluster head determination model according to an embodiment of this application. (Reference) Figure 5 As shown, at gateway station 100, a transformation model is used to determine the relationship with each of the first satellites. ~ The corresponding second vector ~ In the case of each second vector ~ Construct the corresponding matrix MY.

[0050] Then, gateway station 100 inputs matrix MY into the pre-trained cluster head determination model, and uses the cluster head determination model to output the first satellites. ~ The second and third probabilities are for the first satellite to be the cluster head. The second probability represents the probability that the first satellite will be the cluster head, and the third probability represents the probability that the first satellite will not be the cluster head. (And refer to...) Figure 5 As shown, the cluster head determination model includes a deep learning module, a fully connected layer, and multiple softmax classifiers 1 to n. The number of softmax classifiers is the same as the number of the first satellite.

[0051] For example, the first satellite Corresponding to the second probability z 1,1 and the third probability z 1,2 First satellite Corresponding to the second probability z 2,1 and the third probability z 2,2 And so on; First satellite Corresponding to the second probability z n,1 and the third probability z n,2 .

[0052] Furthermore, the gateway station 100 determined that it was related to each of the first satellites. ~ The corresponding second probability z 1,1 ~zn,1 and a pre-set second probability threshold z g,1 The size relationship between them, and the identification of multiple first satellites. ~ Multiple second satellites serve as cluster heads. For example, multiple first satellites... ~ China, the first satellite The second probability z 1,1 First satellite The second probability z 4,1 And the first satellite The second probability z n-1,1 Greater than the pre-set second probability threshold z g,1 Thus, the first satellite First satellite And the first satellite As a second satellite that can serve as a cluster head .

[0053] Subsequently, at gateway station 100, a second satellite was identified as a potential cluster head. In this case, multiple first satellites can be further identified. ~ Except for the second satellite Multiple third satellites in addition to For example, in the second satellite Including the first satellite First satellite And the first satellite In the case of the third satellite Including the first satellite ~ and the first satellite .

[0054] Thus, multiple second satellites that could serve as cluster heads were identified at gateway station 100. In this case, only multiple third satellites are needed. Perform clustering.

[0055] Optionally, the clustering operation for multiple second and third satellites includes: determining the channel bandwidth and signal-to-noise ratio between each third satellite and each second satellite, and calculating the corresponding channel capacity based on the channel bandwidth and signal-to-noise ratio; determining the distance between each third satellite and each second satellite; calculating the matching degree between each third satellite and each second satellite based on the distance and channel capacity; and clustering the multiple second and third satellites based on the matching degree.

[0056] Specifically, first, the gateway station 100 determines each third satellite. Each with its own second satellite Channel bandwidth between and signal-to-noise ratio Where a = 1 to n-2, b = 1 to 3. For example, Indicates the third satellite With the second satellite Channel bandwidth between Indicates the third satellite With the second satellite Channel bandwidth between Indicates the third satellite With the second satellite Channel bandwidth between Indicates the third satellite With the second satellite The channel bandwidth between, ... Indicates the third satellite With the second satellite The channel bandwidth between.

[0057] For example, Indicates the third satellite With the second satellite The signal-to-noise ratio between them Indicates the third satellite With the second satellite The signal-to-noise ratio between them Indicates the third satellite With the second satellite The signal-to-noise ratio between them Indicates the third satellite With the second satellite The signal-to-noise ratio between them, ... Indicates the third satellite With the second satellite The signal-to-noise ratio between them.

[0058] Then, Gateway Station 100 calculated the values ​​of each third satellite according to Shannon's formula. With each second satellite The channel capacity. The specific calculation formula is as follows: in, Indicates the third satellite With the second satellite Channel capacity between Indicates the third satellite With the second satellite Channel bandwidth between Indicates the third satellite With the second satellite The signal-to-noise ratio between them.

[0059] Furthermore, the gateway station 100 identifies each third satellite. Each with its own second satellite Distance between .

[0060] Then, gateway station 100 is based on various third satellites Each with its own second satellite Calculate the distance and channel capacity between each third satellite. Each with its own second satellite The matching degree between them. The specific calculation formula is as follows: in, Indicates each third satellite Each with its own second satellite The degree of matching between them. This represents the first weight value corresponding to the signal-to-noise ratio. Indicates each third satellite Each with its own second satellite The signal-to-noise ratio between them. This represents the second weight value corresponding to the distance. And within this, the first weight value... Second weight value For example, it could be a value predetermined by the gateway station 100 based on historical data.

[0061] Therefore when The larger the value, the more likely it is to be a third satellite. Each with its own second satellite The better the match; when The smaller the value, the more likely it is to be a third satellite. Each with its own second satellite The more mismatched they are.

[0062] Therefore, based on the matching degree calculated above, the gateway station 100 can group each satellite to be clustered (i.e., each third satellite) Join the cluster head with the highest matching degree (i.e., the corresponding second satellite). In the second cluster corresponding to ), it is called a cluster member of the second cluster.

[0063] Optionally, it also includes: when the satellite distribution within the communication coverage area of ​​the gateway station changes, calculating the matching degree between the fourth satellite that caused the satellite distribution change and the cluster heads of each second cluster; and adding the fourth satellite to the second cluster with the highest matching degree.

[0064] Specifically, if the gateway station 100 determines that the satellite distribution within its communication coverage area has changed (for example, a satellite may enter or leave the communication coverage area of ​​the gateway station 100), then the matching degree between the fourth satellite that caused the change in satellite distribution and the cluster heads of each second cluster is calculated. The calculation method for the matching degree is the same as described above and will not be repeated here.

[0065] Finally, gateway station 100 added the fourth satellite to the second cluster with the highest matching degree.

[0066] In addition, before the gateway station 100 uses the conversion model and determines multiple second satellite information based on multiple first satellite information, the conversion model needs to be trained in advance. Figure 6 This is a schematic diagram illustrating the training of the conversion model according to an embodiment of this application. (Reference) Figure 6 As shown, the specific steps are as follows: First, the gateway station 100 obtains sample cluster members. The first satellite information sample is obtained, and the corresponding first vector sample is constructed. The gateway station 100 acquired two sample cluster heads. and The first satellite information sample is obtained, and the corresponding first vector sample is constructed. ~ .

[0067] Furthermore, the gateway station 100 will send the first vector sample First vector sample and the first vector sample The inputs are respectively fed into the initial transformation models 1-3, and the second vector samples are output using transformation models 1-3. Second vector sample Second vector sample .

[0068] Then, gateway station 100 will send the second vector sample Second vector sample Second vector sample The data are input into the MLP model and fused using the MLP model to generate fused information.

[0069] Finally, the gateway station 100 inputs the fused information into the binary classifier and uses the output of the binary classifier to add it to the cluster head. The probability of being in the sample cluster and adding cluster head The probability of being in the sample cluster .

[0070] After that, the probability of the message gate being 100% will be... With sample cluster head The results of the sample clusters are compared (i.e., the loss function), and the probabilities are... With sample cluster head The results of the sample clusters are compared (i.e., the loss function) to train the transformation model.

[0071] In addition, refer to Figure 1 As shown, according to a second aspect of this embodiment, a storage medium is provided. The storage medium includes a stored program, wherein, when the program is executed, a processor performs any of the methods described above.

[0072] Therefore, according to this embodiment, the accuracy of determining whether re-clustering is needed can be improved, and the technical effect of avoiding network oscillation or additional overhead between the gateway station and the satellite can be achieved.

[0073] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0075] Example 2 The device for determining dynamic switching of satellite clusters in this application corresponds to the method described in Embodiment 1. The device includes: a clustering situation determination module, used to determine the clustering situation within the communication coverage area of ​​the gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; a first calculation module, used to calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determine the corresponding first mean and first variance based on the data transmission volume; a first data acquisition module, used to acquire the second mean and second variance at each historical time, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at a historical time; a probability output module, used to input the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and use the probability prediction model to output the first probability of re-clustering at the current time, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and a first judgment module, used to determine whether re-clustering is needed at the next time based on the first probability and a pre-set first probability threshold.

[0076] Optionally, the first judgment module includes: a first clustering module, used to indicate that re-clustering is required at the next time step when the first probability is greater than the first probability threshold; and a second clustering module, used to indicate that re-clustering is not required at the next time step when the first probability is less than or equal to the first probability threshold.

[0077] Optionally, the second clustering module includes: a first satellite determination module, used to determine multiple first satellites located within the communication coverage area of ​​the gateway station; a first vector determination module, used to obtain information about multiple first satellites corresponding to each first satellite using ephemeris information, and to determine a first vector corresponding to each first satellite based on the information about multiple first satellites; a second vector determination module, used to input multiple first vectors into a pre-trained conversion model, and to output a second vector corresponding to each first satellite based on the conversion model; and a clustering status determination module, used to determine the clustering status at the next time step based on multiple second vectors.

[0078] Optionally, the clustering determination module includes: a probability determination module, used to construct a matrix based on multiple second vectors, input the matrix into a pre-trained cluster head determination model, and determine a second probability and a third probability of each first satellite as a cluster head based on the cluster head determination model, wherein the second probability represents the probability of the first satellite being a cluster head and the third probability represents the probability of the first satellite not being a cluster head; a second satellite determination module, used to determine multiple second satellites as cluster heads among the multiple first satellites when the second probability is greater than a pre-set second probability threshold; a third satellite determination module, used to determine multiple third satellites among the multiple first satellites other than the multiple second satellites when multiple second satellites are determined; and a clustering module, used to cluster the multiple second satellites and third satellites.

[0079] Optionally, the clustering module includes: a channel capacity calculation module, used to determine the channel bandwidth and signal-to-noise ratio between each third satellite and each second satellite, and to calculate the corresponding channel capacity based on the channel bandwidth and signal-to-noise ratio; a distance determination module, used to determine the distance between each third satellite and each second satellite; a matching degree calculation module, used to calculate the matching degree between each third satellite and each second satellite based on the distance and channel capacity; and a clustering submodule, used to cluster multiple second and third satellites based on the matching degree.

[0080] Optionally, the apparatus further includes: a matching degree calculation submodule, used to calculate the matching degree between the fourth satellite that caused the satellite distribution change and the cluster heads of each second cluster when the satellite distribution within the communication coverage area of ​​the gateway station changes; and a matching module, used to add the fourth satellite to the second cluster with the highest matching degree.

[0081] Therefore, according to this embodiment, the accuracy of determining whether re-clustering is needed can be improved, and the technical effect of avoiding network oscillation or additional overhead between the gateway station and the satellite can be achieved.

[0082] Example 3 The apparatus for determining dynamic switching of satellite clusters in this application corresponds to the method described in Embodiment 1. The apparatus includes: a processor; and a memory connected to the processor, used to provide the processor with instructions to process the following steps: determining the clustering situation within the communication coverage area of ​​the gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; calculating the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determining the corresponding first mean and first variance based on the data transmission volume; obtaining the second mean and second variance at each historical time, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at a historical time; inputting the first mean, first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and using the probability prediction model to output the first probability of re-clustering at the current time, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and determining whether re-clustering is needed at the next time based on the first probability and a pre-set first probability threshold.

[0083] Therefore, according to this embodiment, the accuracy of determining whether re-clustering is needed can be improved, and the technical effect of avoiding network oscillation or additional overhead between the gateway station and the satellite can be achieved.

[0084] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0085] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0087] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0089] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0090] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining dynamically switching satellite clusters, characterized in that, include: Determine the clustering situation within the communication coverage area of ​​the gateway station at the current moment, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; Calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determine the corresponding first mean and first variance based on the data transmission volume, wherein the first variance represents the load balancing degree among the cluster heads of each first cluster at the current time; Obtain the second mean and second variance at each historical time point, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at the historical time point; The first mean, the first variance, multiple second means, and multiple second variances are input into a pre-trained probability prediction model, and the probability prediction model is used to output the first probability of re-clustering at the current time. The probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function. Based on the first probability and a pre-set first probability threshold, determine whether re-clustering is needed at the next moment.

2. The method according to claim 1, characterized in that, Based on the first probability and a pre-set first probability threshold, determine whether re-clustering is needed in the next moment, including: If the first probability is greater than the first probability threshold, it indicates that re-clustering is required at the next time step; and If the first probability is less than or equal to the first probability threshold, it means that no re-clustering is required at the next time step.

3. The method according to claim 2, characterized in that, When the first probability is greater than the first probability threshold, the re-clustering operation is required, including: Identify multiple first satellites located within the communication coverage area of ​​the gateway station; Using ephemeris information, obtain multiple first satellite information corresponding to each first satellite, and based on the multiple first satellite information, determine the first vector corresponding to each first satellite; Multiple first vectors are input into a pre-trained conversion model, and a second vector corresponding to each of the first satellites is output based on the conversion model; and Based on the multiple second vectors, the clustering situation at the next time step is determined.

4. The method according to claim 3, characterized in that, The operation of determining the clustering situation at the next time step based on the multiple second vectors includes: A matrix is ​​constructed based on the plurality of second vectors, and the matrix is ​​input into a pre-trained cluster head determination model. The second probability and the third probability of each first satellite as a cluster head are determined based on the cluster head determination model, wherein the second probability represents the probability of the first satellite as the cluster head, and the third probability represents the probability of the first satellite not being the cluster head. If the second probability is greater than a preset second probability threshold, then a plurality of second satellites are identified as cluster heads among the plurality of first satellites; In the case of determining the plurality of second satellites, determine a plurality of third satellites among the plurality of first satellites, in addition to the plurality of second satellites; and The multiple second satellites and the third satellite are clustered.

5. The method according to claim 4, characterized in that, The clustering operation for the plurality of second satellites and the third satellite includes: Determine the channel bandwidth and signal-to-noise ratio between each third satellite and each second satellite, and calculate the corresponding channel capacity based on the channel bandwidth and the signal-to-noise ratio; Determine the distances between each of the third satellites and each of the second satellites; Based on the distance and the channel capacity, calculate the matching degree between each third satellite and each second satellite; and Based on the matching degree, the plurality of second satellites and the third satellite are clustered.

6. The method according to claim 5, characterized in that, Also includes: If the satellite distribution within the communication coverage area of ​​the gateway station changes, calculate the matching degree between the fourth satellite that caused the change in satellite distribution and the cluster heads of each second cluster; as well as The fourth satellite is added to the second cluster with the highest matching degree.

7. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, a processor performs the method according to any one of claims 1 to 6.

8. A device for determining dynamically switching satellite clusters, characterized in that, include: The clustering situation determination module is used to determine the clustering situation within the communication coverage area of ​​the gateway station at the current time, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; The first calculation module is used to calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and to determine the corresponding first mean and first variance based on the data transmission volume. The first data acquisition module is used to acquire the second mean and the second variance at each historical time, wherein the second variance represents the load balancing degree between the cluster heads of each second cluster at the historical time. A probability output module is used to input the first mean, the first variance, multiple second means, and multiple second variances into a pre-trained probability prediction model, and to output the first probability of re-clustering at the current time using the probability prediction model, wherein the probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function; and The first judgment module is used to determine whether re-clustering is needed at the next moment based on the first probability and a preset first probability threshold.

9. The apparatus according to claim 8, characterized in that, The first judgment module includes: The first clustering module is used to indicate that re-clustering is needed at the next time step when the first probability is greater than the first probability threshold; and The second clustering module is used to indicate that no re-clustering is needed in the next time step if the first probability is less than or equal to the first probability threshold.

10. A device for determining dynamically switching satellite clusters, characterized in that, include: processor; as well as A memory, connected to the processor, for providing the processor with instructions to perform the following processing steps: Determine the clustering situation within the communication coverage area of ​​the gateway station at the current moment, wherein the clustering situation represents the satellite distribution of each first cluster within the communication coverage area; Calculate the data transmission volume between the cluster head of each first cluster and the gateway station at the current time, and determine the corresponding first mean and first variance based on the data transmission volume; Obtain the second mean and second variance at each historical time point, wherein the second variance represents the load balancing degree among the cluster heads of each second cluster at the historical time point; The first mean, the first variance, multiple second means, and multiple second variances are input into a pre-trained probability prediction model, and the probability prediction model is used to output the first probability of re-clustering at the current time. The probability prediction model includes an RNN model, a fully connected layer, and a sigmoid function. Based on the first probability and a pre-set first probability threshold, determine whether re-clustering is needed at the next moment.

Images