A satellite-ground joint edge computing method

Abstract

The present invention discloses a satellite-ground joint edge computing method, which relates to the field of wireless communications. The system consists of a plurality of ground base stations and low-orbit satellites equipped with edge computing servers, providing low-latency computing unloading services for ground terminals and space terminals with limited computing. The ground terminal and the space terminal configure the transmission power and select a suitable edge computing server according to the environmental information and computing task information, and unload the computing task through the uplink. After the ground base station and the low-orbit satellite receive the data signal of the corresponding terminal, they use the receiving beam to decode it and send it to its edge computing server to perform task calculation, and allocate corresponding computing resources to it. After the computing task is completed, the calculation result is transmitted back to the corresponding terminal through the downlink. The present invention provides an effective satellite-ground joint edge computing method for the 6G space-ground integrated network.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a satellite-ground combined edge computing method. Background Art

[0002] In recent years, with the rapid development of the Internet of Things industry, the number of IoT terminals has exploded, and it is expected that by 2030, the number of IoT terminals will reach hundreds of billions. In this case, the current 5G communication technology can no longer fully match the rapidly growing terminal access requirements. In order to meet the upcoming challenges, academia and industry have begun research on 6G mobile communication networks.

[0003] At present, with the continuous increase in global terminal connections and data traffic, ground networks and computing alone can no longer effectively meet the needs of efficient and low-latency processing of massive data. In this case, 6G will be an integrated network with deep integration of ground networks and satellite networks. Specifically, the 6G network will further integrate satellite communications, big data, and artificial intelligence technologies on the basis of 5G technology to provide a wider coverage, greater communication capacity, and lower transmission latency. In addition, more and more emerging business scenarios, such as driverless cars, augmented reality, and industrial Internet of Things, have also put forward higher requirements for the communication and computing of massive data. Therefore, it is necessary to provide an effective communication and computing fusion solution for the 6G satellite-ground joint network.

[0004] By comprehensively considering the characteristics of ground communication, satellite-to-ground communication and inter-satellite links, integrating the edge computing resources of ground networks and low-orbit satellite networks, a satellite-to-ground joint edge computing method is designed for 6G networks. This can effectively provide network connection and computing power support for resource-constrained areas, meet the application scenarios of advanced IoT terminals on the ground and in space, and significantly reduce the energy consumption of the satellite-to-ground joint system. Summary of the invention

[0005] The purpose of this invention is to solve the problems of limited computing resources, communication congestion, excessive latency, and high energy consumption of ground terminals and space terminals in 6G networks, and proposes a satellite-ground joint edge computing method.

[0006] The specific technical solutions adopted by the present invention are as follows:

[0007] The present invention provides a satellite-ground joint edge computing method, comprising the following steps:

[0008] 1) Deploy M edge computing servers and antennas in the satellite-to-ground computing system. of ground base stations and N edge computing servers with antennas The low-orbit satellite also includes K computationally constrained single-antenna ground terminals and L computationally constrained single-antenna space terminals; obtain the ground channel state information h from the kth ground terminal to the mth ground base station k,m , the satellite-to-ground channel state information g from the kth ground terminal to the nth low-orbit satellite k,n , the intersatellite link information between the space terminal and the low-orbit satellite includes the transmitter optical efficiency of the lth space terminal Transmitter Aperture Gain Receiver optical efficiency of the nth low-orbit satellite Receiver Aperture Gain Transmitter pointing loss coefficient from the lth space terminal to the nth low-orbit satellite and the receiver pointing loss coefficient

[0009] 2) The ground terminal and the space terminal select a ground base station or a low-orbit satellite for computing task offloading using an offloading object selection strategy according to the environmental information and computing task information, wherein the computing task information includes the computing task amount and task complexity; the computing task amount d of the kth ground terminal is obtained. k , task complexity c k , the computing task of the lth space terminal Task complexity

[0010] 3) According to the offloading object in the offloading object selection strategy, the ground terminal and the space terminal use a transmission power allocation method to construct a transmission signal and offload the computing task through the uplink;

[0011] 4) A joint receiving beam and resource allocation method is used to enable the ground base station and the low-orbit satellite to receive the signal from the ground terminal; then serial interference elimination is performed and the corresponding receiving beam is designed for decoding, and the low-orbit satellite decodes the optical signal from the space terminal, and obtains the mission data after decoding; the ground base station and the low-orbit satellite send the mission data to their respective edge computing servers, and allocate computing resources to the edge computing servers;

[0012] 5) Based on the allocated computing resources, the edge computing server performs computing tasks and transmits the computing results back to the corresponding terminal after completing the calculation.

[0013] Based on the above technical solution, some of the steps can be implemented in the following preferred manner:

[0014] Preferably, in step 2), the uninstall object selection strategy is:

[0015] a) Initialize the unloading object selection flag α of the ground terminal according to the distance between the ground terminal and the ground base station and the low-orbit satellite k,m ∈{0,1} and β k,n ∈{0,1}, initialize the unloading object selection flag γ of the space terminal according to the distance between the space terminal and the low-orbit satellite l,n ∈{0,1}; where α k,m =1 means that the kth ground terminal offloads the computing task to the mth ground base station, β k,n =1 means that the kth ground terminal offloads the computing task to the nth low-orbit satellite, γ l,n =1 means that the lth space terminal offloads the computing task to the nth low-orbit satellite; initialize the ground terminal transmission power and space terminal transmission power in is the maximum transmission power of the kth ground terminal, is the maximum transmission power of the lth space terminal; the receiving beam of the mth ground base station to the kth ground terminal is initialized as The receiving beam of the nth low-orbit satellite to the kth ground terminal is where ||·|| represents the second norm of the vector; the computing resources allocated to the kth ground terminal by the edge computing server of the mth ground base station are initialized as Initialize the computing resources allocated by the edge computing server of the nth low-orbit satellite to the kth ground terminal as Initialize the computing resources allocated by the edge computing server of the nth low-orbit satellite to the lth space terminal as in represents the total computing resources of the edge computing server of the mth ground base station, represents the total computing resources of the edge computing server of the nth low-orbit satellite;

[0016] b) Calculate the weighted total energy consumption required by all terminals to complete the computing task in is the weight of the kth ground terminal, is the weight of the lth space terminal; is the energy consumption of the kth ground terminal offloading the task to the ground base station, where The energy coefficient of the edge computing server for the mth ground base station, is the data transmission rate from the kth ground terminal to the mth ground base station, where B 1 is the channel bandwidth of terrestrial transmission, is the variance of the Gaussian white noise transmitted from the ground; is the energy consumption of the kth ground terminal offloading the task to the low-orbit satellite, where is the energy coefficient of the nth low-orbit satellite edge computing server, is the data transmission rate from the kth ground terminal to the nth low-orbit satellite, where B 2 is the channel bandwidth of satellite-to-ground transmission, is the variance of Gaussian white noise transmitted from satellite to ground; is the energy consumption of the lth space terminal to offload the task to the low-orbit satellite, where is the data transmission rate from the lth space terminal to the nth low-orbit satellite, where B 3 is the channel bandwidth of intersatellite transmission, is the variance of Gaussian white noise for intersatellite transmission, λ is the wavelength of optical communication, φ l,n is the distance between the lth space terminal and the nth low-orbit satellite;

[0017] c) According to the task offloading strategy restrictions, and

[0018] d) According to the computing resource constraints of the edge computing server, let and

[0019] e) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and in is the distance between the kth ground terminal and the nth low-orbit satellite, μ is the speed of light, is the maximum tolerable delay for the kth ground terminal to complete the computing task, The maximum tolerable delay for the lth space terminal to complete the computing task;

[0020] f) Use the branch and bound method to find the solution that minimizes the weighted total energy consumption E of all terminals to complete the task calculation, that is, to obtain the offloading object selection flag α k,m ,β k,n ,γ l,n .

[0021] Preferably, in step 3), the transmission power allocation method is:

[0022] a) According to the result of step 2), obtain and update the unloading object selection flag α of the ground terminal and the space terminal k,m ,β k,n ,γ l,n ;

[0023] b) Introducing auxiliary variables in renew

[0024] c) Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the kth ground terminal to offload the task to the ground base station: Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the k′th ground terminal to offload the task to the low-orbit satellite:

[0025] d) Introducing auxiliary variables and make and Introducing intermediate variables and make and Calculate the weighted total energy consumption of all terminals to complete the computing tasks:

[0026]

[0027] e) According to the transmission power limit of ground terminals and space terminals, and

[0028] f) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and

[0029] g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing task, that is, the ground terminal transmission power p k and the space terminal transmission power q l .

[0030] Preferably, in step 4), the joint receive beam and resource allocation method is:

[0031] a) According to the result of step 3), obtain and update the transmission power p of the ground terminal k and uninstall object selection flag α k,m ,β k,n , the transmitting power q of the space terminal l and Unload Object Selection Flags¶ l,n ;

[0032] b) Introducing auxiliary variables and make and Calculate the weighted total energy consumption of all terminals to complete the computing tasks:

[0033]

[0034] c) Introducing auxiliary variables and make

[0035] d) Define intermediate variables and make in X k,m , Y k,n , The value in the previous iteration, tr(·) represents the trace of the matrix;

[0036] e) According to the delay requirement of the ground terminal to complete the computing task,

[0037] f) According to the computing resource limitation of the edge computing server, and

[0038] g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing tasks, that is, obtain the computing resources allocated to the ground terminals by the edge computing server of the ground base station Get the computing resources allocated to the ground terminal by the edge computing server of the low-orbit satellite Get the computing resources allocated to the space terminal by the edge computing server of the low-orbit satellite To W k,m and V k,n Perform eigenvalue decomposition respectively to obtain the receiving beam w of the ground base station and the low-orbit satellite k,m and v k,n .

[0039] Preferably, the environmental information in step 2) includes ground channel state information, satellite-to-ground channel state information and inter-satellite link information.

[0040] The beneficial effects of the present invention are as follows: the proposed satellite-ground joint edge computing method solves a series of problems caused by limited communication between ground terminals and space terminals, lack of computing resources, high task computing delay, and high energy consumption in the 6G space-ground integration scenario. The proposed joint design of the offloading strategy, transmission power, receiving beam, and optimization algorithm for computing resource allocation has the advantages of low complexity, low energy consumption, reasonable computing resource allocation, and computing delay that meets terminal requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a system block diagram of the satellite-ground joint edge computing method;

[0042] Figure 2 It is a comparison of the weighted total energy consumption of the proposed method when the number of ground base stations and low-orbit satellites is different and the terminal delay requirements are different (the number of ground base stations and low-orbit satellites is the same and is 3, 4 and 5 respectively);

[0043] Figure 3 It is a comparison of the weighted total energy consumption performance of the terminal to complete task calculations under different satellite-ground combined edge computing methods. DETAILED DESCRIPTION

[0044] In this embodiment, the system block diagram of the satellite-ground combined edge computing method is as follows: Figure 1 As shown in the figure, the system consists of M multi-antenna ground base stations equipped with edge computing servers and N low-orbit satellites equipped with edge computing servers, providing low-latency computing offloading services for K computing-constrained ground terminals and L computing-constrained space terminals. The ground terminal and the space terminal configure the transmission power and select the appropriate edge computing server according to the environmental information and computing task information, and offload the computing task through the uplink. After the ground base station and the low-orbit satellite receive the data signal from the corresponding terminal, they design the receiving beam to decode and send it to its edge computing server to perform the task calculation, and allocate the corresponding computing resources to it. After the computing task is completed, the calculation result is transmitted back to the corresponding terminal through the downlink.

[0045] The specific technical solutions adopted in this embodiment are as follows:

[0046] A satellite-ground joint edge computing method comprises the following steps:

[0047] 1) Deploy M edge computing servers and antennas in the satellite-to-ground computing system. of ground base stations and N edge computing servers with antennas A low-orbit satellite has K computationally constrained single-antenna ground terminals and L computationally constrained single-antenna space terminals. It is necessary to select the corresponding edge computing server to complete its own computing tasks. Among them, the ground channel state information from the kth ground terminal to the mth ground base station is h k,m , the satellite-to-ground channel state information from the kth ground terminal to the nth low-orbit satellite is g k,n The intersatellite link information between the space terminal and the low-orbit satellite includes the transmitter optical efficiency of the lth space terminal: The transmitter aperture gain is The receiver optical efficiency of the nth low-orbit satellite is The receiver aperture gain is The transmitter pointing loss coefficient and receiver pointing loss coefficient from the lth space terminal to the nth low-orbit satellite are and

[0048] 2) The ground terminal and the space terminal select a ground base station or a low-orbit satellite for computing task offloading using an offloading object selection strategy based on environmental information (ground channel status information, satellite-to-ground channel status information, and inter-satellite link information) and computing task information. The computing task information includes computing task volume and task complexity. The computing task volume of the kth ground terminal is d k , the task complexity is c k , the computing task of the lth space terminal is The task complexity is

[0049] In this embodiment, in the above step 2), one uninstallation object selection strategy is:

[0050] a) Initialize the unloading object selection flag α of the ground terminal according to the distance between the ground terminal and the ground base station and the low-orbit satellite k,m ∈{0,1},β k,n ∈{0,1}, initialize the unloading object selection flag γ of the space terminal according to the distance between the space terminal and the low-orbit satellite l,n ∈{0,1}, where α k,m =1 means that the kth ground terminal offloads the computing task to the mth ground base station, β k,n =1 means that the kth ground terminal offloads the computing task to the nth low-orbit satellite, γ l,n =1 means that the lth space terminal offloads the computing task to the nth low-orbit satellite; initialize the ground terminal transmission power Space terminal transmission power in is the maximum transmission power of the kth ground terminal, is the maximum transmission power of the lth space terminal; the receiving beam of the mth ground base station to the kth ground terminal is initialized as The receiving beam of the nth low-orbit satellite to the kth ground terminal is where ||·|| represents the second norm of the vector; the computing resources allocated to the kth ground terminal by the edge computing server of the mth ground base station are initialized as The computing resources allocated by the edge computing server of the nth low-orbit satellite to the kth ground terminal are The computing resources allocated by the edge computing server of the nth low-orbit satellite to the lth space terminal are in represents the total computing resources of the edge computing server of the mth ground base station, Represents the total computing resources of the edge computing server of the nth low-orbit satellite.

[0051] b) Calculate the weighted total energy consumption required by all terminals to complete the computing task in is the weight of the kth ground terminal, is the weight of the lth space terminal. is the energy consumption of the kth ground terminal offloading the task to the ground base station, where The energy coefficient of the edge computing server for the mth ground base station, is the data transmission rate from the kth ground terminal to the mth ground base station, where B 1 is the channel bandwidth of terrestrial transmission, is the variance of the Gaussian white noise transmitted from the ground. is the energy consumption of the kth ground terminal offloading the task to the low-orbit satellite, where is the energy coefficient of the nth low-orbit satellite edge computing server, is the data transmission rate from the kth ground terminal to the nth low-orbit satellite, where B 2 is the channel bandwidth of satellite-to-ground transmission, is the variance of Gaussian white noise transmitted from satellite to ground. is the energy consumption of the lth space terminal to offload the task to the low-orbit satellite, where is the data transmission rate from the lth space terminal to the nth low-orbit satellite, where B 3 is the channel bandwidth of intersatellite transmission, is the variance of Gaussian white noise for intersatellite transmission, λ is the wavelength of optical communication, φ l,n is the distance between the lth space terminal and the nth low-orbit satellite;

[0052] c) According to the task offloading strategy restrictions, and

[0053] d) According to the computing resource constraints of the edge computing server, let and

[0054] e) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and in is the distance between the kth ground terminal and the nth low-orbit satellite, μ is the speed of light, is the maximum tolerable delay for the kth ground terminal to complete the computing task, The maximum tolerable delay for the lth space terminal to complete the computing task;

[0055] f) Use the branch and bound method to find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing task, that is, to obtain the offloading object selection flag α k,m ,β k,n ,γ l,n .

[0056] 3) According to the offloading object, the ground terminal and the space terminal use a transmission power allocation method to construct a transmission signal and offload the computing task through the uplink;

[0057] In this embodiment, in the above step 3), a transmission power allocation method is:

[0058] a) Get and update the unloading object selection flag of the ground terminal and the space terminal α k,m ,β k,n ,γ l,n ;

[0059] b) Introducing auxiliary variables in renew

[0060] c) Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the kth ground terminal to offload the task to the ground base station: Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the k′th ground terminal to offload the task to the low-orbit satellite:

[0061] d) Introducing auxiliary variables and make and Introducing intermediate variables and make and Calculate the weighted total energy consumption of all terminals to complete the task calculation:

[0062] e) According to the transmission power limit of ground terminals and space terminals, and

[0063] f) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and

[0064] g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing task, that is, the ground terminal transmission power p k and the space terminal transmission power q l .

[0065] 4) Using a joint receiving beam and resource allocation method, after receiving the signal from the ground terminal, the ground base station and the low-orbit satellite perform serial interference elimination and design the corresponding receiving beam for decoding. At the same time, the low-orbit satellite decodes the optical signal from the space terminal. After decoding, the ground base station and the low-orbit satellite send the mission data to the edge computing server and allocate computing resources to it.

[0066] In this embodiment, in the above step 4), a joint receiving beam and resource allocation method is:

[0067] a) Obtain and update the transmission power p of the ground terminal k and uninstall object selection flag α k,m ,β k,n , the transmission power q of the space terminal l and Unload Object Selection Flags¶ l,n ;

[0068] b) Introducing auxiliary variables and make and Calculate the weighted total energy consumption of all terminals to complete the task calculation:

[0069] c) Introducing auxiliary variables and make

[0070] d) Define intermediate variables and make in X k,m , Y k,n , The value in the previous iteration, tr(·) represents the trace of the matrix;

[0071] e) According to the delay requirement of the ground terminal to complete the computing task,

[0072] f) According to the computing resource limitation of the edge computing server, and

[0073] g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing tasks, that is, obtain the computing resources allocated to the ground terminals by the edge computing server of the ground base station Get the computing resources allocated to the ground terminal by the edge computing server of the low-orbit satellite Get the computing resources allocated to the space terminal by the edge computing server of the low-orbit satellite To W k,m and V k,n Perform eigenvalue decomposition respectively to obtain the receiving beam w of the ground base station and the low-orbit satellite k,m , v k,n ;

[0074] 5) Based on the allocated computing resources, the edge computing server performs computing tasks and transmits the computing results back to the corresponding terminal after completing the calculation.

[0075] Computer simulations show that Figure 2 As shown in the figure, the present invention proposes a satellite-ground joint edge computing method. The higher the terminal's latency requirement, the greater the weighted energy consumption to complete the computing task. Moreover, with the increase in ground base stations and low-orbit satellites, the weighted energy consumption is significantly reduced. In addition, Figure 3 It shows that the method proposed in the present invention has obvious performance advantages over the fixed receiving beam, fixed maximum transmit power, random unloading strategy and average computing resource method, and has a large performance gain under different terminal delay requirements. Therefore, the present invention provides an effective satellite-ground joint edge computing method for the space-ground integrated network with limited computing resources in the 6G scenario.

[0076] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A satellite-ground joint edge computing method, characterized in that The steps include: 1) Deploy M edge computing servers and antennas in the satellite-to-ground computing system. of ground base stations and N edge computing servers with antennas The low-orbit satellite also includes K computationally constrained single-antenna ground terminals and L computationally constrained single-antenna space terminals; obtain the ground channel state information h from the kth ground terminal to the mth ground base station k,m , the satellite-to-ground channel state information g from the kth ground terminal to the nth low-orbit satellite k,n , the intersatellite link information between the space terminal and the low-orbit satellite includes the transmitter optical efficiency of the lth space terminal Transmitter Aperture Gain Receiver optical efficiency of the nth low-orbit satellite Receiver Aperture Gain Transmitter pointing loss coefficient from the lth space terminal to the nth low-orbit satellite and the receiver pointing loss coefficient 2) The ground terminal and the space terminal select a ground base station or a low-orbit satellite for computing task offloading using an offloading object selection strategy based on environmental information and computing task information, wherein the computing task information includes computing task volume and task complexity; Get the computing task volume d of the kth ground terminal k , task complexity c k , the computing task of the lth space terminal Task complexity 3) According to the offloading object in the offloading object selection strategy, the ground terminal and the space terminal use a transmission power allocation method to construct a transmission signal and offload the computing task through the uplink; 4) Using a joint receiving beam and resource allocation method to enable ground base stations and low-orbit satellites to receive signals from ground terminals; Subsequently, serial interference elimination is performed and the corresponding receiving beam is designed for decoding. At the same time, the low-orbit satellite decodes the optical signal from the space terminal, and obtains the mission data after decoding. The ground base station and the low-orbit satellite send the mission data to their respective edge computing servers and allocate computing resources to the edge computing servers. 5) According to the allocated computing resources, the edge computing server performs the computing task and transmits the computing result back to the corresponding terminal after completing the calculation; In step 2), the uninstall object selection strategy is: a) Initialize the unloading object selection flag α of the ground terminal according to the distance between the ground terminal and the ground base station and the low-orbit satellite k,m ∈{0,1} and β k,n ∈{0,1}, initialize the unloading object selection flag γ of the space terminal according to the distance between the space terminal and the low-orbit satellite l,n ∈{0,1}; where α k,m =1 means that the kth ground terminal offloads the computing task to the mth ground base station, β k,n =1 means that the kth ground terminal offloads the computing task to the nth low-orbit satellite, γ l,n =1 means that the lth space terminal offloads the computing task to the nth low-orbit satellite; initialize the ground terminal transmission power and space terminal transmission power in is the maximum transmission power of the kth ground terminal, is the maximum transmission power of the lth space terminal; the receiving beam of the mth ground base station to the kth ground terminal is initialized as The receiving beam of the nth low-orbit satellite to the kth ground terminal is where ||·|| represents the second norm of the vector; the computing resources allocated to the kth ground terminal by the edge computing server of the mth ground base station are initialized as Initialize the computing resources allocated by the edge computing server of the nth low-orbit satellite to the kth ground terminal as Initialize the computing resources allocated by the edge computing server of the nth low-orbit satellite to the lth space terminal as in represents the total computing resources of the edge computing server of the mth ground base station, represents the total computing resources of the edge computing server of the nth low-orbit satellite; b) Calculate the weighted total energy consumption required by all terminals to complete the computing task in is the weight of the kth ground terminal, is the weight of the lth space terminal; is the energy consumption of the kth ground terminal offloading the task to the ground base station, where The energy coefficient of the edge computing server for the mth ground base station, is the data transmission rate from the kth ground terminal to the mth ground base station, where B1 is the channel bandwidth of the ground transmission, is the variance of the Gaussian white noise transmitted from the ground, p i is the transmission power of the i-th ground terminal; is the energy consumption of the kth ground terminal offloading the task to the low-orbit satellite, where is the energy coefficient of the nth low-orbit satellite edge computing server, is the data transmission rate from the kth ground terminal to the nth low-orbit satellite, where B2 is the channel bandwidth of satellite-to-ground transmission, is the variance of Gaussian white noise transmitted from satellite to ground; is the energy consumption of the lth space terminal to offload the task to the low-orbit satellite, where is the data transmission rate from the lth space terminal to the nth low-orbit satellite, where B3 is the channel bandwidth of inter-satellite transmission, is the variance of Gaussian white noise for intersatellite transmission, λ is the wavelength of optical communication, φ l,n is the distance between the lth space terminal and the nth low-orbit satellite; c) According to the task offloading strategy restrictions, and d) According to the computing resource constraints of the edge computing server, let and e) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and in is the distance between the kth ground terminal and the nth low-orbit satellite, μ is the speed of light, is the maximum tolerable delay for the kth ground terminal to complete the computing task, The maximum tolerable delay for the lth space terminal to complete the computing task; f) Use the branch and bound method to find the solution that minimizes the weighted total energy consumption E of all terminals to complete the task calculation, that is, to obtain the offloading object selection flag α k,m ,β k,n ,γ l,n ; In step 3), the transmission power allocation method is: a) According to the result of step 2), obtain and update the unloading object selection flag α of the ground terminal and the space terminal k,m ,β k,n ,γ l,n ; b) Introducing auxiliary variables in renew c) Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the kth ground terminal to offload the task to the ground base station: Introducing auxiliary variables The Gaussian elimination method is used to solve the transmission power of the k′th ground terminal to offload the task to the low-orbit satellite: d) Introducing auxiliary variables and make and Introducing intermediate variables and make and Calculate the weighted total energy consumption of all terminals to complete the computing tasks: e) According to the transmission power limit of ground terminals and space terminals, and f) According to the delay requirements of the ground terminal and the space terminal to complete the computing task, let and g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing task, that is, the ground terminal transmission power p k and the space terminal transmission power q l ; In step 4), the joint receive beam and resource allocation method is: a) According to the result of step 3), obtain and update the transmission power p of the ground terminal k and uninstall object selection flag α k,m ,β k,n , the transmitting power q of the space terminal l and Unload Object Selection Flags¶ l,n ; b) Introducing auxiliary variables and make and Calculate the weighted total energy consumption of all terminals to complete the computing tasks: c) Introducing auxiliary variables and make d) Define intermediate variables and make in X k,m , Y k,n , The value in the previous iteration, tr(·) represents the trace of the matrix; e) According to the delay requirement of the ground terminal to complete the computing task, f) According to the computing resource limitation of the edge computing server, and g) Use the interior point method or directly call the CVX toolkit to solve and find the solution that minimizes the weighted total energy consumption E of all terminals to complete the computing tasks, that is, obtain the computing resources allocated to the ground terminals by the edge computing server of the ground base station Get the computing resources allocated to the ground terminal by the edge computing server of the low-orbit satellite Get the computing resources allocated to the space terminal by the edge computing server of the low-orbit satellite To W k,m and V k,n Perform eigenvalue decomposition respectively to obtain the receiving beam w of the ground base station and the low-orbit satellite k,m and v k,n .

2. The satellite-ground combined edge computing method according to claim 1, characterized in that: The environmental information in step 2) includes ground channel status information, satellite-to-ground channel status information and inter-satellite link information.

Images