Satellite-ground collaborative network access method and related equipment

Abstract

The present application provides a satellite-ground collaborative network access method and related equipment. The satellite-ground collaborative network access method includes: determining a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model; in response to determining an access request of a user to be accessed, determining historical transmission data of the user to be accessed, and determining the bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network; determining a communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any communication link, and accessing the satellite-ground collaborative network based on the communication link.

Description

Technical Field

[0001] The present application relates to the field of satellite-ground collaborative network technology, and in particular to a satellite-ground collaborative network access method and related equipment. Background Art

[0002] Transmission lines stretch for thousands of miles, creating complex corridor environments and potentially presenting a variety of potential hazards at any given time. Transmission lines in some areas are limited by cost and transmission speed issues. Therefore, we employ methods such as time-lapse photography and short-duration video to monitor transmission lines. However, existing transmission line monitoring suffers from low intelligence, low efficiency, high cost, low flexibility, and weak security. Therefore, the number of terminals used in online transmission line monitoring technology will be enormous and dynamically deployed in the future, and the monitored and collected content will tend to be presented in high-definition video format. This places high demands on the coverage, flexibility, and communication operation costs of the transmission methods. Satellite-ground collaborative networks can meet the requirements for large-scale data transmission, such as high-definition video and images, required for critical area monitoring. However, due to inadequate network access, satellite-ground collaborative networks face limited available network resources. Summary of the Invention

[0003] In view of this, the purpose of this application is to propose a satellite-ground collaborative network access method and related equipment.

[0004] Based on the above objectives, the present application provides a satellite-ground collaborative network access method, which is characterized by including:

[0005] Determining a pre-built satellite-ground collaborative network model and a bandwidth capacity of any communication link in the satellite-ground collaborative network model;

[0006] In response to determining an access request from a user to be accessed, determining historical transmission data of the user to be accessed, and determining a bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network;

[0007] According to the bandwidth occupancy and the bandwidth capacity of any of the communication links, a communication link for accessing the satellite-ground collaborative network is determined, and the satellite-ground collaborative network is accessed according to the communication link.

[0008] Optionally, the construction of the satellite-ground collaborative network includes:

[0009] Obtaining the functions and service characteristics of the transmission line of the system to be connected; wherein the target system is a monitoring system;

[0010] The satellite-ground collaborative network is constructed according to the functions and the service characteristics; wherein the satellite-ground collaborative network includes a macro base station, a ground small base station and a satellite-ground small base station, and communication services are provided for the target system through the macro base station, the ground small base station and the satellite-ground small base station.

[0011] Optionally, the satellite-ground coordinated network includes a first cellular cell, a second cellular cell, and a third cellular cell, the first cellular cell is determined according to the macro base station, the second cellular cell is determined according to the ground small base station, and the third cellular cell is determined according to the satellite-ground small base station;

[0012] Determining the bandwidth capacity of any type of communication link in the satellite-terrestrial collaborative network model includes:

[0013] Determining a channel coefficient of any cell and a distance between any cell of any user in the system to be accessed;

[0014] The bandwidth capacity of the communication link of any type is determined based on the channel coefficient and the separation distance.

[0015] Optionally, the training of the deep learning network includes:

[0016] Obtain historical transmission data of the network to be accessed, and train the deep learning network based on the historical transmission data; wherein the deep learning network updates data based on a reward value, and the reward value is determined based on a sigmoid function.

[0017] Optionally, determining the communication link for accessing the satellite-ground collaborative network according to the bandwidth occupancy and the bandwidth capacity of any of the communication links includes:

[0018] The bandwidth occupancy is compared with the bandwidth capacity of any of the communication links, and the communication link with the same bandwidth capacity as the bandwidth occupancy is determined as the communication link accessing the satellite-ground coordinated network.

[0019] Optionally, the method further includes:

[0020] Determining a channel observation value according to a communication link accessed to the satellite-ground collaborative network;

[0021] A reward value is determined according to the channel observation value, and the deep learning network is updated.

[0022] Based on the same inventive concept, an embodiment of the present application further provides a satellite-ground collaborative network access device, including:

[0023] a bandwidth capacity determination module configured to determine a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model;

[0024] a bandwidth occupancy calculation module configured to, in response to determining an access request from a user to be accessed, determine historical transmission data of the user to be accessed, and determine the bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network;

[0025] The access module is configured to determine a communication link for accessing the satellite-ground collaborative network according to the bandwidth occupancy and the bandwidth capacity of any of the communication links, and access the satellite-ground collaborative network according to the communication link.

[0026] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor, characterized in that when the processor executes the program, it implements the satellite-ground collaborative network access method as described in any one of the above.

[0027] Based on the same inventive concept, an embodiment of the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, characterized in that the computer instructions are used to enable a computer to execute any of the above-mentioned satellite-ground collaborative network access methods.

[0028] Based on the same inventive concept, an embodiment of the present application further provides a computer program product, including computer program instructions. When the computer program instructions are executed on a computer, the computer executes any of the above-mentioned satellite-ground collaborative network access methods.

[0029] From the above, it can be seen that the satellite-ground collaborative network access method provided by the present application includes: determining a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model; in response to determining the access request of the user to be accessed, determining the historical transmission data of the user to be accessed, and determining the bandwidth occupancy of the access request based on the historical transmission data and the pre-trained deep learning network; determining the communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any communication link, and accessing the satellite-ground collaborative network based on the communication link. In view of the problems of low intelligence, low efficiency, high cost, low flexibility and weak security in the existing transmission line monitoring, the deep learning satellite-ground collaborative network multiple access algorithm can play a better role in high-load scenarios, and is significantly better than the mainstream multiple access algorithm in terms of link resource utilization and system throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in this application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are merely embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0031] Figure 1 A schematic diagram of a flow chart of a satellite-ground collaborative network access method according to an embodiment of the present application;

[0032] Figure 2 A schematic diagram of a device for a satellite-ground collaborative network access method according to an embodiment of the present application;

[0033] Figure 3 This is a schematic structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of this application more clear, this application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0035] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should have the usual meanings understood by people with ordinary skills in the field to which this application belongs. The "first", "second" and similar words used in the embodiments of the present application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0036] As mentioned in the background technology section, transmission lines stretch for thousands of miles, creating complex corridor environments and potentially creating various potential hazards at any given time. Transmission lines in some areas are limited by cost and transmission speed issues. Therefore, we employ methods such as time-lapse photography and short-duration video to monitor transmission lines. However, existing transmission line monitoring suffers from low intelligence, low efficiency, high cost, low flexibility, and weak security. Therefore, the number of terminals used in online transmission line monitoring technology will be enormous and dynamically deployed in the future, and the monitoring and acquisition content will tend to be presented in the form of high-definition video. This places high demands on the coverage, flexibility, and communication operation costs of the transmission methods. Satellite-ground collaborative networks can meet the requirements for large-scale data transmission, such as high-definition video and images, required for monitoring in critical areas. However, due to inadequate network access, satellite-ground collaborative networks face the problem of limited available network resources.

[0037] In view of this, embodiments of the present application provide a satellite-ground collaborative network access method, device, electronic device, and storage medium.

[0038] Specifically, it includes: (1) a satellite-ground collaborative network architecture based on transmission line monitoring, which consists of various monitoring terminals, base stations, power grid data centers, and communication channels connecting them. The monitoring terminal contains various monitoring devices, and the communication channels connecting the monitoring terminal and the power grid monitoring center constitute the access layer network. The base station is composed of various base stations, which transmit data through the ground mobile network and the low-orbit satellite network. The monitoring terminal is connected to the network through the base station, and the monitoring data is transmitted back to the data center in real time through the uplink of "terminal-base station-core network-power grid monitoring center". It can also perform dynamic monitoring of the transmission line. Among them, the base station provides a satellite-ground collaborative network (ground network / satellite) to transmit power grid data.

[0039] (2) The satellite-ground collaborative network model is divided into N cells. Since all cells share the same frequency resource pool, we use frequency multiplexing between cells and divide the available frequency bandwidth of each cell (e.g., base station) into L sub-channels.

[0040] (3) Grid users are considered independent units accessing satellite links, and an independent multi-agent is constructed for each user. This ensures that users interact with the network environment and independently update their policies. The access algorithm can leverage historical experience with random service arrivals to make each user's delivery process regular and predictable. First, an interactive environment based on a deep learning model is constructed. Then, we train and provide feedback on the model, temporalize the time dimension, and analyze the action, state, and reward design in a multi-access scenario.

[0041] (4) The deep learning framework used consists of two networks with the same structure but different parameters. The valuation network and the target value network are used to approximate the action-state value function. The target value network is used to eliminate data correlation and ensure the stability of algorithm learning. The network we use is a deep residual neural network with 32 hidden layers, each with 128 neurons. The experience replay mechanism is used to break the data correlation, that is, the experience gained by the user in interacting with the environment at each time is stored in the experience replay pool. Taking into account the interference of external random factors, random perturbations are added to the dynamic changes of the two satellite service loads.

[0042] like Figure 1 As shown, the satellite-ground collaborative network access method includes: step S102, determining a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model;

[0043] Step S104: In response to determining an access request from a user to be accessed, determining historical transmission data of the user to be accessed, and determining a bandwidth usage of the access request based on the historical transmission data and a pre-trained deep learning network;

[0044] Step S106: Determine a communication link for accessing the satellite-ground collaborative network according to the bandwidth occupancy and the bandwidth capacity of any of the communication links, and access the satellite-ground collaborative network according to the communication link.

[0045] In step S102, the construction of the satellite-ground collaborative network includes:

[0046] Obtain the functions and service characteristics of the transmission line of the system to be connected; (the target system is the monitoring system)

[0047] The satellite-ground collaborative network is constructed according to the functions and the service characteristics; wherein the satellite-ground collaborative network includes a macro base station, a ground small base station and a satellite-ground small base station, and communication services are provided for the target system through the macro base station, the ground small base station and the satellite-ground small base station.

[0048] In some optional embodiments, the satellite-ground collaborative network includes a first cellular cell, a second cellular cell, and a third cellular cell. The first cellular cell is determined based on the macro base station, the second cellular cell is determined based on the ground small base station, and the third cellular cell is determined based on the satellite-ground small base station.

[0049] Therefore, in step S102, determining the bandwidth capacity of any type of communication link in the satellite-terrestrial collaborative network model includes:

[0050] Determining a channel coefficient of any cell and a distance between any cell of any user in the system to be accessed;

[0051] The bandwidth capacity of the communication link of any type is determined based on the channel coefficient and the separation distance.

[0052] In step S104, the training of the deep learning network includes: obtaining historical transmission data of the network to be accessed, and training the deep learning network according to the historical transmission data; wherein the deep learning network updates data according to a reward value, and the reward value is determined according to a sigmoid function.

[0053] In step S106, determining the communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any of the communication links includes: comparing the bandwidth occupancy with the bandwidth capacity of any of the communication links, and determining the communication link having the same bandwidth capacity as the bandwidth occupancy as the communication link for accessing the satellite-ground collaborative network.

[0054] In some optional implementations, the satellite-ground collaborative network access method further includes: determining a channel observation value based on a communication link accessing the satellite-ground collaborative network; determining a reward value based on the channel observation value, and updating the deep learning network.

[0055] In some optional implementations, a satellite-ground collaborative network model primarily consists of macro base stations, traditional terrestrial small base stations, and satellite-ground small base stations. Each small base station provides traffic offloading and coverage enhancement. Small base stations are connected to the core network via wired or wireless backhaul. Small base stations are categorized as traditional terrestrial small base stations and low-orbit satellite small base stations, respectively, using backhaul based on traditional terrestrial networks and low-orbit satellite networks.

[0056] In some optional implementations, the model has a total of N cells, where the number of cells generally corresponds to the number of base stations, N = {0, 1, …, N}. Therefore, in the current satellite-ground collaborative network model, there are three types of cell sets, where n = 0 represents a macro base station cell. 1 ≤ n ≤ N′ represents a traditional terrestrial small cell. N′ ≤ n ≤ N represents a low-orbit satellite small base station. We use (N + 1) × K and a binary matrix B to represent the locations of K users and the links between each cell. bn,k = 1 indicates that user k is within the coverage of cell n, and bn,k = 0 indicates that user k is not within the coverage of cell n.

[0057] Since all cells share the same frequency resource pool, we consider frequency multiplexing between cells. This divides the available frequency bandwidth of each cell into L subchannels. To better represent user association and subchannel allocation, we use a (N+1)×K×L binary matrix Y to represent the channel allocation. xn,k,l=1 means that user k accesses cell n through subchannel l, otherwise xn,k,l=0. Cell n receives the signal from user k through subchannel l, which is calculated as follows:

[0058]

[0059] Where qv represents the user transmission power; tk represents the signal sent by user k per unit energy; Represents the channel coefficient. According to the characteristics of channel fading, the base station and monitoring terminal in the scene remain relatively fixed. The power fading in the channel includes large-scale fading and small-scale fading. Large-scale fading is shadow fading and path loss, and small-scale fading is Rayleigh fading. n,k,l Communicate with user k in cell n via subchannel 1 / cell n provides services to user k via subchannel 1, n is the random noise of cell n.

[0060] In some optional implementations, according to the channel coefficient Determining the bandwidth capacity of cell n serving user k through subchannel l, and further determining the total bandwidth capacity of cell n based on the bandwidth capacity of cell n serving user k through subchannel l, specifically includes:

[0061] Channel coefficient Determine according to the following formula:

[0062]

[0063] Among them, α n,k,l is a complex Gaussian variable that obeys a normal distribution; β n,k,l is the shadow attenuation that obeys the log-normal distribution; (e n,k ) -γ represents the path loss; e n,k is the distance between user j and cell m. α is the path loss coefficient. According to Shannon’s formula, we can obtain the bandwidth capacity of cell n serving user k through subchannel l as follows:

[0064]

[0065] Furthermore, the total bandwidth capacity of cell n is calculated according to the following formula:

[0066]

[0067] Since the total data rate of all users connected to each cell cannot exceed the backhaul capacity limit of the base station, we use D n Denotes the backhaul capacity of cell n. For macro base station cells and traditional small base station cells in the cellular cluster with 0≤n≤N', D n Limited by the capacity of the ground backhaul network. Generally speaking, the bandwidth of the satellite downlink is greater than that of the satellite uplink. For a cellular cluster with N'+1≤n≤N satellite-to-ground small base station cells, Dn is determined by the satellite-to-ground uplink, y n,k,l User k in cell n communicates via subchannel 1 / cell n provides services to user k via subchannel 1.

[0068] In some optional implementations, constructing a deep learning-based satellite-ground collaborative network multiple access algorithm specifically includes:

[0069] An independent multi-agent is constructed for each user. This ensures that users interact with the network environment and independently update their policies. The access algorithm leverages historical experience with random service arrivals to make each user's transmission process regular and predictable. This effectively improves the utilization of the satellite-to-ground link.

[0070] We denote the action space as b o (t)∈{0,1}. It represents the action taken by user o at time slot t, which is described as follows:

[0071] (1) When b o When (t) = 0, user o takes the action of not sending a data packet at time slot t;

[0072] (2) When b o When (t) = 1, user o takes the action of sending a data packet at time slot t;

[0073] To simulate different network load intensities, we set up a packet queue for each user to store the generated packets. The size of each packet occupies exactly one time slot. The time interval between the generation of two packets follows the Poisson distribution, which simulates the randomness of service generation. The queue state s p Indicates the service stacking status of each user. p >0 means the user wants to send another service.

[0074] The state space we design must account for and accurately describe all possibilities. We use the state space to describe the occupancy of channel resources. After a user takes action at time slot t and observes the environment, the channel has three possible states: s(t)∈{s,d,e}. These three states are explained in detail below:

[0075] (1) s indicates success. Only one user sends a service and successfully occupies the channel;

[0076] (2) d represents a collision. Multiple users send services at the same time, resulting in a collision.

[0077] (3) e indicates idle. At this time, no user sends services and the channel resources are wasted.

[0078] After each time slot transmission, all users can immediately obtain the channel observation value f. Therefore, when there is a data packet to be sent, that is, s p >0, the channel observation value at time t is defined by the set f(t), that is, f(t)∈{0,1,…}. The specific interpretation of this set is as follows:

[0079] (1) When f(t) = 0, it means the channel is idle;

[0080] (2) When f(t) = 1, it means that the channel has been successfully occupied;

[0081] (3) When f(t)>1 (i.e., f(t)=2,3,…,n), it indicates that the channel is in a collision state.

[0082] We assume that all users in the network are independent of each other and operate in a distributed manner. Each accessing user independently observes the network environment, independently trains the network, and independently makes decisions. The parameters in the algorithm utilize the approximate action value function in deep learning. The input of the deep learning model is the state s(t) at different times, and the output is the approximate p value corresponding to different states.

[0083] Since the entire network has time periods, and the service queue state s is introduced p The purpose is to study network scenarios under different load intensities. Therefore, we define action observation as follows: Divide d(t+1) into five scenarios. When the access user takes action b(t), the state s(t) will transition to the state s(t+1). At the same time, it will generate a reward r(t+1). The state at the next moment is recorded as N is the length of the state history that the user must track. In a time-invariant network (a network whose characteristics do not change over time, meaning the network's output is determined solely by current and past inputs, independent of the timing of the inputs), a larger state history length N leads to more accurate user decisions. However, this also increases the complexity of network training and slows convergence.

[0084] We use deep learning algorithms to optimize the reward value of the multi-access algorithm. With the goal of maximizing network throughput, we design corresponding reward values ​​for the five cases of d(t+1). The reward values ​​for the five cases of d(t+1) are shown in the following table. If the action we choose helps improve the throughput performance of the network system, it will receive a positive reward. Reward / penalty value

[0085] (1) When there is a service waiting to be sent in the queue, if the access user Oi performs a send operation and no other access user sends a service, the channel is successfully occupied and the reward value is 1;

[0086] (2) When there are services waiting to be sent in the queue, if the access user Oi takes the sending operation and another access user has also sent a service, a collision occurs and a penalty of -1 is given;

[0087] (3) Even if there are no services waiting to be sent in the queue, some users may still perform the sending operation. Although this does not affect the actual situation, the user will still be penalized with -1.

[0088] (4) When the access user Oi does not take any sending action without considering the service queue status, if other access users send services, resulting in the channel being in a non-idle state, this indicates that the access user Oi has successfully retreated, and the reward value is 1;

[0089] (5) When the access user Oi does not take the sending action without considering the service queue status, if other access users do not send services either, the channel is in the idle state. We need to give a penalty to make the access user take more active sending actions in the future.

[0090] We have proposed various penalty value design methods. If we directly assign a specific penalty value, such as -1 or -2, the accessing user will not be able to find the correct learning target. This will result in the accessing user not being able to find the correct learning target. If the penalty value is too large, the accessing user will adopt a passive strategy of not sending services, causing the channel to be idle most of the time. Therefore, based on the queuing situation of the services in the queue, we use a smoother S-shaped growth curve (sigmoid function) to design the penalty value. The specific calculation formula is as follows:

[0091] r(x)=-k*S(x)

[0092] Where x = s p -c. k is the degree coefficient; x is the queue length point; c is the offset; they can adjust the penalty value in time according to the business intensity and queue length.

[0093] In some optional implementations, the deep learning network framework consists of two networks with identical structures but different parameters. A valuation network and a target value network are used to approximate the action-state value function. The target value network is used to eliminate data correlation and ensure the stability of the algorithm learning. The network used is a deep residual neural network with 32 hidden layers, each with 128 neurons.

[0094] In addition, we also use the experience replay mechanism to break the data correlation, that is, the experience gained by the user in interacting with the environment at each time t is stored in the experience replay pool F N During the training network step, we randomly extract a certain number of experiences from the experience replay pool.

[0095] In some optional implementations, the present application also provides a verification method. Specifically, the experiment uses link utilization and throughput rate as the performance measurement criteria of the algorithm. s , Idle D d and collision D c Three states. The total number of time slots in the experiment is O, that is, O = D s +D d +D c .

[0096] The total number of data packets generated by the entire network system, Pall, includes the successfully sent data packets, P s , the data packets P that are discarded due to collision c and the unsent packets P that are accumulated in the queue q The calculation formula is as follows.

[0097] P all =P s +P c +P q (6)

[0098] The calculation formula of link utilization lu is as follows:

[0099] lu=P s / O (7)

[0100] Among them, P s is the data packet sent successfully, and O is the total number of time slots.

[0101] The throughput is calculated as follows:

[0102] Throughput=P s / P all (8)

[0103] Pall is the total number of data packets generated by the network system.

[0104] When the network load is low, we use throughput as the performance metric. When the network load is high, a large number of services are backlogged in the service queue, rendering throughput meaningless. Therefore, under heavy load, we use link utilization as the performance metric. To better investigate the algorithm's performance, we also break down the aforementioned statistics during the experiment. Every 0 / 10 time slot, we count the number of successfully transmitted packets, the number of packets that collided, and the cumulative number of packets generated by the system. These metrics reflect the training speed of the proposed algorithm.

[0105] The deep learning model consists of recurrent layers, with each network consisting of a GRU with 128 neurons. It includes five hidden layers, a hybrid network, and a hypernetwork. The hidden layer of the hypernetwork also contains 128 neurons. To prevent falling into local optima, each network uses a greedy algorithm to select actions, with the greedy algorithm parameter ∈ set to 1 to 0.001 and a decay multiplier of 0.885. The experience replay size is set to 1000. The sampling batch size for deep learning training is set to 16. The learning rate of the Adam optimizer is 1×10⁻⁶. The target network is updated every 100 training steps. All experiments were run for 40,000 time steps, i.e., 40,000 time slots were computed.

[0106] To demonstrate the advantages of our proposed multi-access algorithm, we selected currently mainstream multi-access algorithms, such as the non-orthogonal multi-access-assisted federated learning algorithm and the MTCD algorithm. We first simulated a full-buffer service scenario, where every user has a data packet to send in every time slot. Furthermore, in this scenario, the service is non-random. We then calculated the link utilization of our proposed algorithm and the comparison algorithms for 5, 10, 15, 20, 25, 30, 35, 40, and 45 users, respectively.

[0107] From the above, it can be seen that the satellite-ground collaborative network access method provided by the present application includes: determining a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model; in response to determining the access request of the user to be accessed, determining the historical transmission data of the user to be accessed, and determining the bandwidth occupancy of the access request based on the historical transmission data and the pre-trained deep learning network; determining the communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any communication link, and accessing the satellite-ground collaborative network based on the communication link. In view of the problems of low intelligence, low efficiency, high cost, low flexibility and weak security in the existing transmission line monitoring, the deep learning satellite-ground collaborative network multiple access algorithm can play a better role in high-load scenarios, and is significantly better than the mainstream multiple access algorithm in terms of link resource utilization and system throughput.

[0108] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied in a distributed scenario and performed by multiple devices working together. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the method.

[0109] It should be noted that the above description is limited to some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a satellite-ground collaborative network access device.

[0111] refer to Figure 2 The satellite-ground collaborative network access device includes:

[0112] The bandwidth capacity determination module 202 is configured to determine a pre-built satellite-ground collaborative network model and the bandwidth capacity of any communication link in the satellite-ground collaborative network model;

[0113] The bandwidth occupancy calculation module 204 is configured to determine, in response to determining an access request from a user to be connected, historical transmission data of the user to be connected, and determine the bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network;

[0114] The access module 206 is configured to determine a communication link for accessing the satellite-ground coordination network according to the bandwidth occupancy and the bandwidth capacity of any of the communication links, and access the satellite-ground coordination network according to the communication link.

[0115] For the convenience of description, the above devices are described as being divided into various modules according to their functions. Of course, when implementing this application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0116] The apparatus of the above embodiment is used to implement the corresponding satellite-ground collaborative network access method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0117] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements the satellite-ground collaborative network access method described in any of the above embodiments.

[0118] Figure 3 10 is a schematic diagram showing a more specific hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other within the device via the bus 1050.

[0119] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0120] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0121] The input / output interface 1030 is used to connect input / output modules to implement information input and output. The input / output modules can be configured as components within the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and output devices may include a display, speaker, vibrator, indicator light, etc.

[0122] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WiFi, Bluetooth, etc.).

[0123] The bus 1050 comprises a path for transmitting information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).

[0124] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0125] The electronic device of the above embodiment is used to implement the corresponding satellite-ground collaborative network access method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0126] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the satellite-ground collaborative network access method described in any of the above embodiments.

[0127] The computer-readable media of this embodiment include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0128] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the satellite-ground collaborative network access method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0129] Based on the same inventive concept, corresponding to the satellite-ground collaborative network access method described in any of the above embodiments, the present disclosure further provides a computer program product comprising computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer to cause the computer and / or the processor to perform the satellite-ground collaborative network access method. For the execution entities corresponding to the steps in each embodiment of the satellite-ground collaborative network access method, the processors executing the corresponding steps can belong to the corresponding execution entities.

[0130] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the satellite-ground collaborative network access method as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0131] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. Within the scope of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0132] In addition, for simplicity of description and discussion, and in order not to make the embodiment of the application difficult to understand, the known power supply / ground connection with integrated circuit (IC) chip and other components may or may not be shown in the accompanying drawings provided. In addition, the device can be shown in the form of a block diagram to avoid making the embodiment of the application difficult to understand, and this also takes into account the following fact, that is, the details of the embodiment of these block diagram devices are highly dependent on the platform to be implemented in the embodiment of the application (that is, these details should be fully within the scope of understanding of those skilled in the art). When specific details (for example, circuit) are set forth to describe exemplary embodiments of the application, it will be apparent to those skilled in the art that the embodiment of the application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0133] Although the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may utilize the embodiments discussed.

[0134] The embodiments of the present application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of this application.

Claims

1. A satellite-ground collaborative network access method, characterized in that: include: Determining a pre-constructed satellite-ground coordinated network and the bandwidth capacity of any communication link in the satellite-ground coordinated network; wherein the satellite-ground coordinated network includes a first cell, a second cell, a third cell, a macro base station, a terrestrial small base station, and a satellite-ground small base station, the first cell is determined based on the macro base station, the second cell is determined based on the terrestrial small base station, and the third cell is determined based on the satellite-ground small base station. Constructing the satellite-ground coordinated network includes: Acquire functions and service characteristics of the transmission line of the system to be accessed; and construct the satellite-ground collaborative network according to the functions and service characteristics; Determining the bandwidth capacity of any of the communication links in the satellite-ground coordinated network includes: Determining a channel coefficient of any cellular cell and a distance between any user in the system to be accessed and any cellular cell; determining a bandwidth capacity of any of the communication links based on the channel coefficient and the separation distance; In response to determining an access request from a user to be accessed, determining historical transmission data of the user to be accessed, and determining a bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network; Determining a communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any of the communication links, and accessing the satellite-ground collaborative network based on the communication link; this includes: comparing the bandwidth occupancy with the bandwidth capacity of any of the communication links, and determining the communication link having the same bandwidth capacity as the bandwidth occupancy as the communication link for accessing the satellite-ground collaborative network.

2. The method according to claim 1, characterized in that The system to be connected is a monitoring system; The satellite-ground collaborative network is constructed according to the functions and the service characteristics; wherein, communication services are provided for the system to be accessed through the macro base station, the ground small base station and the satellite-ground small base station.

3. The method according to claim 1, characterized in that The training of the deep learning network includes: Obtain historical transmission data of the network to be accessed, and train the deep learning network based on the historical transmission data; wherein the deep learning network updates data based on a reward value, and the reward value is determined based on a sigmoid function.

4. The method according to claim 1, wherein The method further comprises: Determining a channel observation value according to a communication link accessed to the satellite-ground collaborative network; A reward value is determined according to the channel observation value, and the deep learning network is updated.

5. A satellite-ground collaborative network access device, characterized in that: include: A bandwidth capacity determination module is configured to determine a pre-constructed satellite-ground coordinated network and the bandwidth capacity of any communication link in the satellite-ground coordinated network; wherein the satellite-ground coordinated network includes a first cell, a second cell, a third cell, a macro base station, a terrestrial small base station, and a satellite-ground small base station, the first cell is determined based on the macro base station, the second cell is determined based on the terrestrial small base station, and the third cell is determined based on the satellite-ground small base station, and construction of the satellite-ground coordinated network includes: Acquire functions and service characteristics of the transmission line of the system to be accessed; and construct the satellite-ground collaborative network according to the functions and service characteristics; Determining the bandwidth capacity of any of the communication links in the satellite-ground coordinated network includes: Determining a channel coefficient of any cellular cell and a distance between any user in the system to be accessed and any cellular cell; determining a bandwidth capacity of any of the communication links based on the channel coefficient and the separation distance; a bandwidth occupancy calculation module configured to, in response to determining an access request from a user to be accessed, determine historical transmission data of the user to be accessed, and determine the bandwidth occupancy of the access request based on the historical transmission data and a pre-trained deep learning network; An access module is configured to determine a communication link for accessing the satellite-ground collaborative network based on the bandwidth occupancy and the bandwidth capacity of any of the communication links, and access the satellite-ground collaborative network based on the communication link; the access module includes: comparing the bandwidth occupancy with the bandwidth capacity of any of the communication links, and determining the communication link having the same bandwidth capacity as the bandwidth occupancy as the communication link for accessing the satellite-ground collaborative network.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method according to any one of claims 1 to 4 when executing the computer program. 7 . A non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method according to claim 1 .

8. A computer program product comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 4.

Images