An emergency information service system and method based on integration of satellite communication and ground mobile communication

Abstract

The application provides an emergency information service system and method based on satellite communication and ground mobile communication fusion, the system comprises: a high flux satellite fixed station terminal, a Ku band 0.6 meter VSAT terminal is adopted, an ACU and a MODEM are integrated, a three-axis structure and a carbon fiber antenna are provided; a ground mobile communication subsystem, comprising a 4G skin base station, a security gateway and a signaling gateway; a multi-WAN port routing device realizes dynamic load balancing of satellite and ground broadband; an emergency power supply system is composed of a generator and a power bank to form a three-level power supply; a network management and monitoring system supports remote configuration and AI optimization. The method comprises the steps of disaster monitoring, system activation, network establishment, signal coverage, communication service and management and maintenance. The application combines AI driven network optimization to quickly establish a reliable communication network in an emergency scene through satellite and ground communication fusion, has the characteristics of wind load resistance, low time delay, intelligent scheduling and the like, and significantly improves the emergency communication guarantee capability.

Description

Technical Field

[0001] This invention belongs to the field of emergency communication technology, specifically relating to an emergency information service system and method based on the integration of satellite communication and terrestrial mobile communication, which is particularly suitable for emergency communication support for village-level units in areas prone to natural disasters. Background Technology

[0002] During natural disasters such as earthquakes, floods, and mudslides, the "three disruptions"—roads, internet, and power—often occur, paralyzing traditional terrestrial communication networks. This is especially true in remote mountainous and rural areas where communication infrastructure is weak, making it difficult to obtain disaster information and contact the outside world in a timely manner. While satellite communication can provide emergency communication services, it suffers from problems such as complex terminal equipment, high costs, and insufficient integration with terrestrial mobile communication networks, failing to meet the actual needs of village-level emergency communication.

[0003] The existing technology has the following problems: insufficient coverage of conventional base stations in remote areas and poor disaster resistance; power supply interruption during disasters causes communication equipment to malfunction; high construction cost and difficult maintenance of emergency communication systems; lack of effective satellite and ground communication integration solutions; and decentralized management and slow response speed of emergency communication systems.

[0004] Therefore, designing an emergency information service system that can provide reliable communication services in emergency situations such as disasters, achieving efficient communication support in disaster scenarios, and meeting the demand for broadband and real-time communication in emergency situations is an urgent task to ensure emergency communication under village-level disaster conditions. Summary of the Invention

[0005] In view of this, the present invention provides an emergency information service method and system based on the integration of satellite communication and terrestrial mobile communication. By organically combining high-throughput satellite fixed terminals and 4G pico base stations, a stable, reliable, and low-cost village-level emergency communication network is constructed. Furthermore, the network performance is comprehensively evaluated through an AI large model, and the network is adaptively adjusted and optimized to solve the communication guarantee problem under disaster conditions.

[0006] This application provides an emergency information service method and system based on the fusion of satellite communication and terrestrial mobile communication. It constructs a distributed cloud-edge collaborative transmission line defect identification platform, deploying inference and semi-automatic annotation systems at the edge, and building a multi-model incremental training system in the cloud using digital twin technology. The central cloud and edge nodes cooperate with each other, and incremental learning improves annotation efficiency and model accuracy. Simultaneously, samples are annotated manually or by other systems, and the model is re-incrementally trained. If the incrementally trained model shows a significant improvement in accuracy compared to the original model, it can be deployed to the edge nodes to update the original model. A crowdsourcing mechanism is used to implement the pipeline task, dividing the semi-automatic annotation task into multiple pipeline sub-tasks. This pipeline operation not only makes the annotation work more efficient but also integrates the characteristics of different models, resulting in more accurate annotation results.

[0007] In a first aspect, embodiments of this application provide an emergency information service system based on the convergence of satellite communication and terrestrial mobile communication, the system comprising:

[0008] High-throughput satellite fixed station terminal: adopts Ku-band 0.6-meter land-based automatic VSAT terminal, integrates antenna control unit (ACU) and modem, features a triaxial structure design and high-strength carbon fiber composite antenna surface, and realizes satellite broadband access;

[0009] The terrestrial mobile communication subsystem includes a 4G pico base station, a pico base station gateway, and a network management system. The pico base station gateway includes a security gateway and a signaling gateway, and provides LTE terminal access services.

[0010] Multi-WAN port routing device: configured to simultaneously access satellite broadband and terrestrial broadband, and to achieve dynamic load balancing;

[0011] Emergency power system: including small generators and high-power power banks, forming a three-level power supply protection architecture to ensure equipment power supply in the event of a power outage;

[0012] Network management and monitoring system: including satellite fixed station network management and pico base station network management, supports remote parameter configuration and status monitoring, and uses AI big data models to comprehensively evaluate network performance and adaptively adjust and optimize the network.

[0013] Optionally, in one implementation of the first aspect of the present invention, the high-throughput satellite fixed station terminal includes:

[0014] The three-axis stabilization structure consists of an azimuth axis cross roller bearing, a pitch axis harmonic reducer, and a polarization axis stepper motor, achieving a pointing error of ≤0.3° under wind load conditions of level 6. The antenna surface is made of carbon fiber composite material with a thickness of 2.5mm and a tensile strength of ≥800MPa, with a total weight of <15kg. The ACU and MODEM are integrated into the antenna feed through a common aperture design, resulting in a signal insertion loss of <0.5dB. It supports adaptive coding and modulation from QPSK to 64APSK and has a built-in TCP acceleration engine optimized with the SCPS-TP protocol, enabling end-to-end transmission latency of <800ms in GEO satellite scenarios.

[0015] The 4G pico base station in the terrestrial mobile communication subsystem adopts a software-defined radio (SDR) architecture, supports dynamic spectrum sharing (DSS) technology, and can simultaneously access LTE Band 3 / 5 / 8 frequency bands within a 10MHz bandwidth. The transmit power is adjustable from 0.1W to 5W, and the coverage radius is 70-100 meters. The security gateway adopts a hardware encryption module, supports the IPSec protocol and the national cryptographic SM4 algorithm, and has a throughput of ≥200Mbps. The signaling gateway dynamically selects MME nodes through the S1-flex interface, and the signaling processing capacity is ≥200 EPS / s.

[0016] Optionally, in one implementation of the first aspect of the present invention, the emergency power supply system includes:

[0017] The small generator has a rated power of 2kVA, adopts frequency conversion technology, has a fuel consumption of <0.3L / kWh, and supports 48 hours of continuous operation; the high-power power bank uses a lithium iron phosphate battery pack with a capacity of 20kWh and an operating temperature range of -20℃ to 60℃; in the event of a mains power outage, the UPS switches power supply within <10ms, and the generator starts within 30 seconds; the multi-WAN port routing device:

[0018] It has a built-in multipath transmission protocol MPTCP, which dynamically allocates traffic weights based on satellite link latency and terrestrial broadband packet loss rate; it assigns DSCP 46 priority to voice services through the DiffServ QoS mechanism, ensuring a MOS value ≥ 3.5; and it integrates an 8GB SSD local cache for storing emergency command information.

[0019] The network management and monitoring system adopts dual-channel monitoring: satellite network management and base station network management;

[0020] The satellite network management system monitors EIRP and G / T values ​​in real time and supports remote adjustment of the symbol rate. The base station network management system visualizes user distribution and traffic heatmaps and supports one-click triggering of the SON self-organizing network function. Alarm linkage: When the satellite link packet loss rate is greater than 5% for 5 consecutive minutes, the base station power is automatically reduced and a work order is pushed to the operation and maintenance APP.

[0021] Optionally, in one implementation of the first aspect of the present invention, the step of comprehensively evaluating network performance through a large AI model and adaptively adjusting and optimizing the network includes:

[0022] S1, set different priorities for satellite network management and base station network management;

[0023] S2, use the network management system with higher priority as the first communication link and the network management system with lower priority as the second communication link;

[0024] S3, monitor the network management performance parameters of the first communication link, including: signal strength, transmission delay, and packet loss rate;

[0025] S4, Perform time series performance prediction on the parameter data using an AI large model;

[0026] S5, compare the time series performance prediction with each preset corresponding reference value, and comprehensively evaluate each comparison result to obtain the first evaluation value;

[0027] S6, When the evaluated value is greater than or equal to the preset threshold, return to S3;

[0028] S7, otherwise, switch the network management priority, return to steps S2-S5, and use the result as the second evaluation value;

[0029] S8, compare the first evaluation value with the second evaluation value, and when the first evaluation value is less than the second evaluation value, optimize the communication link of the network management system;

[0030] S9, when the first evaluation value is greater than or equal to the second evaluation value, switch the network management priority and optimize the network management communication link.

[0031] Optionally, in one implementation of the first aspect of the present invention, the AI ​​large model adopts a multi-module fusion hybrid neural network architecture, combining time series prediction and reinforcement learning optimization capabilities. The network architecture includes: an input layer, a feature extraction module, a performance prediction module, an evaluation and decision module, and an output layer.

[0032] The input layer is used to receive real-time monitored network management performance parameters, and the data format is a sliding window of data from the past 60 seconds.

[0033] The feature extraction module is a CNN-LSTM hybrid layer that combines an attention mechanism. The CNN branch is used to extract spatial local features of the parameters, including abrupt change patterns in signal intensity; the LSTM branch is used to capture temporal dependencies, including periodic fluctuations in packet loss rate; and the attention mechanism is used to dynamically weight the importance of different parameters.

[0034] The performance prediction module includes a time series prediction sub-network, which uses Transformer to predict the performance trend in the next 5-10 seconds and outputs predicted values ​​for latency and packet loss rate.

[0035] The evaluation and decision-making module is used for reinforcement learning agents and includes state, action, and reward function attributes. The state attributes include: current predicted value, historical evaluation value, and link priority. The action attributes include: switching priority, adjusting link weight, and triggering optimization strategy. The reward function attributes are dynamically generated based on the objective function.

[0036] The output layer is used to generate evaluation values ​​and optimization instructions;

[0037] The objective function needs to balance performance stability and resource utilization, and is divided into two parts: prediction and optimization.

[0038] ,

[0039]

[0040] ,

[0041] ,

[0042] Among them, the weighted mean square error Indicates delay Indicator, Mean Absolute Error Indicates packet loss rate index, Indicates the strength of the loss-smoothed signal. outliers These represent the adjustment factors for the corresponding indicators. These are the ordinal number and the total number of the sampled data, respectively;

[0043] ,

[0044] in, This represents the ratio of the current link quality to a threshold. Indicates switching costs, This indicates a fairness measure to prevent low-priority links from being idle for extended periods. These represent the corresponding weights.

[0045] Optionally, in one implementation of the first aspect of the present invention, the AI ​​large model includes a data processing module, a model training module, an inference module, and an optimization control module. The modules work together to achieve real-time monitoring, prediction, and optimization of network performance.

[0046] The data processing module is used to collect and preprocess network performance parameters and input them into the AI ​​model.

[0047] The model training module uses historical data to train the model, learn the relationship between network performance and parameters, and generate optimization suggestions.

[0048] The inference module then predicts the current network state based on the trained model and outputs an optimization strategy.

[0049] The optimization control module dynamically adjusts network parameters based on the inference results, including priority allocation and link switching, to improve network performance.

[0050] Secondly, embodiments of this application provide an emergency information service method based on the convergence of satellite communication and terrestrial mobile communication, applied to an emergency information service system based on the convergence of satellite communication and terrestrial mobile communication as described in the first aspect, comprising:

[0051] Disaster monitoring steps: The operator's monitoring system detected an alarm indicating that a regular village-level base station had gone out of service;

[0052] System activation steps: Activate the village-level broadband satellite emergency communication station automatically or manually;

[0053] Network establishment steps: High-throughput satellite fixed terminals establish satellite links, and pico base stations access the core network through the satellite network;

[0054] Signal coverage steps: Pico base stations provide 4G network signal coverage to the disaster-stricken areas;

[0055] Communication service steps: The mobile phones of the affected people are automatically connected to the emergency network to enable communication with the outside world;

[0056] Management and maintenance steps: Remotely monitor and manage the equipment through the network management system, and comprehensively evaluate the network performance through AI big data models to adaptively adjust and optimize the network.

[0057] Optionally, in one implementation of the second aspect of the present invention, establishing a satellite communication link includes automatically aligning with the AsiaSat 6D high-throughput satellite, establishing a 50Gbps communication bandwidth, and providing regional 4G coverage through pico base stations includes using an external antenna to achieve precise coverage of 70-100 meters.

[0058] Thirdly, embodiments of this application provide an electronic device, including:

[0059] processor;

[0060] Memory used to store processor-executable instructions;

[0061] The processor is configured to implement, when executing the instructions, an emergency information service method based on the fusion of satellite communication and terrestrial mobile communication as described in the second aspect.

[0062] Fourthly, embodiments of this application provide a computer-readable storage medium storing a program that instructs a device to execute an emergency information service method based on the fusion of satellite communication and terrestrial mobile communication as described in the second aspect.

[0063] This application provides an emergency information service system and method based on the integration of satellite communication and terrestrial mobile communication. The system includes: a high-throughput satellite fixed station terminal, using a Ku-band 0.6-meter VSAT terminal, integrating an ACU and a modem, with a triaxial structure and a carbon fiber antenna; a terrestrial mobile communication subsystem, including a 4G picocell, a security gateway, and a signaling gateway; a multi-WAN port routing device to achieve dynamic load balancing between satellite and terrestrial broadband; an emergency power supply system using a generator and power banks to form a three-level power supply; and a network management and monitoring system supporting remote configuration and AI optimization. The method includes disaster monitoring, system activation, network establishment, signal coverage, communication services, and management and maintenance steps. This invention, through the integration of satellite and terrestrial communication and combined with AI-driven network optimization, rapidly establishes a reliable communication network in emergency scenarios, possessing characteristics such as wind resistance, low latency, and intelligent scheduling, significantly improving emergency communication support capabilities.

[0064] Beneficial effects:

[0065] (1) High reliability communication guarantee: Satellite and ground dual links are integrated, and dynamic load balancing is achieved through multi-WAN port routing equipment to ensure uninterrupted communication in extreme disaster environments.

[0066] (2) The high-throughput satellite terminal adopts a three-axis stabilization structure and carbon fiber antenna, which can withstand wind load of level 6, and the pointing error is ≤0.3°, making it suitable for harsh environments.

[0067] (3) Rapid deployment and flexible coverage: 4G pico base stations support dynamic spectrum sharing (DSS) with a coverage radius of 70-100 meters, which can quickly establish emergency networks to meet the communication needs of disaster areas.

[0068] (4) Automatically aligns with satellites, with a high-speed bandwidth of 50Gbps, and combined with the external antenna of the pico base station to achieve precise coverage and improve rescue efficiency.

[0069] (5) Intelligent management and optimization: AI big data model monitors network performance (signal strength, latency, packet loss rate) in real time, predicts trends in the next 5-10 seconds, dynamically adjusts priorities and link weights, and optimizes communication quality.

[0070] (6) The network management system supports remote monitoring and automatically triggers alarm linkage (such as power reduction when packet loss rate > 5%), improving operation and maintenance efficiency.

[0071] (7) High-efficiency energy guarantee: Three-level power supply architecture (generator + large-capacity lithium iron battery + UPS) supports 48 hours of continuous operation and ensures stable power supply to equipment in the event of power failure.

[0072] (8) Security and adaptability: The security gateway supports the national standard SM4 encryption, with a throughput of ≥200Mbps, ensuring data transmission security. Attached Figure Description

[0073] Figure 1 This is a schematic diagram of an emergency information service system based on the integration of satellite communication and terrestrial mobile communication, provided as an embodiment of this application.

[0074] Figure 2 This is a schematic diagram illustrating the process of evaluating network performance using a large AI model, as provided in one embodiment of this application.

[0075] Figure 3 This is a diagram of the AI ​​large model network architecture provided in one embodiment of this application.

[0076] Figure 4 This is a schematic diagram of an emergency information service system method based on the fusion of satellite communication and terrestrial mobile communication, provided as an embodiment of this application.

[0077] Figure 5 A schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0078] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0079] It should be noted that, in the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0080] It should be noted that in the embodiments of this application, the terms "first," "second," etc., are used only for descriptive purposes and should not be construed as indicating or implying relative importance, nor as indicating or implying order. Features specified as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0081] Based on the embodiments described in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0082] Example 1

[0083] Figure 1 This is a schematic diagram of an emergency information service system based on the integration of satellite communication and terrestrial mobile communication, provided as an embodiment of this application.

[0084] like Figure 1 As shown, an emergency information service system based on the integration of satellite communication and terrestrial mobile communication is disclosed. The system includes:

[0085] High-throughput satellite fixed station terminal: adopts Ku-band 0.6-meter land-based automatic VSAT terminal, integrates antenna control unit (ACU) and modem, features a triaxial structure design and high-strength carbon fiber composite antenna surface, and realizes satellite broadband access;

[0086] The terrestrial mobile communication subsystem includes a 4G pico base station, a pico base station gateway, and a network management system. The pico base station gateway includes a security gateway and a signaling gateway, and provides LTE terminal access services.

[0087] Multi-WAN port routing device: configured to simultaneously access satellite broadband and terrestrial broadband, and to achieve dynamic load balancing;

[0088] Emergency power system: including small generators and high-power power banks, forming a three-level power supply protection architecture to ensure equipment power supply in the event of a power outage;

[0089] Network management and monitoring system: including satellite fixed station network management and pico base station network management, supports remote parameter configuration and status monitoring, and uses AI big data models to comprehensively evaluate network performance and adaptively adjust and optimize the network.

[0090] It is understood that, in this embodiment, the high-throughput satellite fixed station terminal includes:

[0091] The three-axis stabilization structure consists of an azimuth axis cross roller bearing, a pitch axis harmonic reducer, and a polarization axis stepper motor, achieving a pointing error of ≤0.3° under wind load conditions of level 6. The antenna surface is made of carbon fiber composite material with a thickness of 2.5mm and a tensile strength of ≥800MPa, with a total weight of <15kg. The ACU and MODEM are integrated into the antenna feed through a common aperture design, resulting in a signal insertion loss of <0.5dB. It supports adaptive coding and modulation from QPSK to 64APSK and has a built-in TCP acceleration engine optimized with the SCPS-TP protocol, enabling end-to-end transmission latency of <800ms in GEO satellite scenarios.

[0092] The 4G pico base station in the terrestrial mobile communication subsystem adopts a software-defined radio (SDR) architecture, supports dynamic spectrum sharing (DSS) technology, and can simultaneously access LTE Band 3 / 5 / 8 frequency bands within a 10MHz bandwidth. The transmit power is adjustable from 0.1W to 5W, and the coverage radius is 70-100 meters. The security gateway adopts a hardware encryption module, supports the IPSec protocol and the national cryptographic SM4 algorithm, and has a throughput of ≥200Mbps. The signaling gateway dynamically selects MME nodes through the S1-flex interface, and the signaling processing capacity is ≥200 EPS / s.

[0093] The emergency power supply system includes:

[0094] The small generator has a rated power of 2kVA, adopts frequency conversion technology, has a fuel consumption of <0.3L / kWh, and supports 48 hours of continuous operation; the high-power power bank uses a lithium iron phosphate battery pack with a capacity of 20kWh and an operating temperature range of -20℃ to 60℃; in the event of a mains power outage, the UPS switches power supply within <10ms, and the generator starts within 30 seconds; the multi-WAN port routing device:

[0095] It has a built-in multipath transmission protocol MPTCP, which dynamically allocates traffic weights based on satellite link latency and terrestrial broadband packet loss rate; it assigns DSCP 46 priority to voice services through the DiffServ QoS mechanism, ensuring a MOS value ≥ 3.5; and it integrates an 8GB SSD local cache for storing emergency command information.

[0096] The network management and monitoring system adopts dual-channel monitoring: satellite network management and base station network management;

[0097] The satellite network management system monitors EIRP and G / T values ​​in real time and supports remote adjustment of the symbol rate. The base station network management system visualizes user distribution and traffic heatmaps and supports one-click triggering of the SON self-organizing network function. Alarm linkage: When the satellite link packet loss rate is greater than 5% for 5 consecutive minutes, the base station power is automatically reduced and a work order is pushed to the operation and maintenance APP.

[0098] Figure 2 This is a schematic diagram illustrating the process of evaluating network performance using a large AI model, as provided in one embodiment of this application. Figure 2 As shown, the comprehensive evaluation of network performance using a large AI model, and the adaptive adjustment and optimization of the network, include:

[0099] S1, set different priorities for satellite network management and base station network management;

[0100] S2, use the network management system with higher priority as the first communication link and the network management system with lower priority as the second communication link;

[0101] S3, monitor the network management performance parameters of the first communication link, including: signal strength, transmission delay, and packet loss rate;

[0102] S4, Perform time series performance prediction on the parameter data using an AI large model;

[0103] S5, compare the time series performance prediction with each preset corresponding reference value, and comprehensively evaluate each comparison result to obtain the first evaluation value;

[0104] S6, When the evaluated value is greater than or equal to the preset threshold, return to S3;

[0105] S7, otherwise, switch the network management priority, return to steps S2-S5, and use the result as the second evaluation value;

[0106] S8, compare the first evaluation value with the second evaluation value, and when the first evaluation value is less than the second evaluation value, optimize the communication link of the network management system;

[0107] S9, when the first evaluation value is greater than or equal to the second evaluation value, switch the network management priority and optimize the network management communication link.

[0108] Figure 3 This is a diagram of the AI ​​large model network architecture provided in one embodiment of this application.

[0109] like Figure 3As shown, the AI ​​large model adopts a multi-module fusion hybrid neural network architecture, combining time series prediction and reinforcement learning optimization capabilities. The network architecture includes: an input layer, a feature extraction module, a performance prediction module, an evaluation and decision module, and an output layer.

[0110] The input layer is used to receive real-time monitored network management performance parameters, with the data format being a sliding window of data from the past 60 seconds.

[0111] Specifically, the input layer receives real-time monitored network management performance parameters in a sliding window format representing the data from the past 60 seconds. This design captures dynamic trends in time-series data, providing high-quality input data for subsequent modules.

[0112] The feature extraction module is a CNN-LSTM hybrid layer that combines an attention mechanism. The CNN branch is used to extract spatial local features of the parameters, including abrupt change patterns in signal intensity; the LSTM branch is used to capture temporal dependencies, including periodic fluctuations in packet loss rate; and the attention mechanism is used to dynamically weight the importance of different parameters.

[0113] Specifically, the feature extraction module employs a CNN-LSTM hybrid layer incorporating an attention mechanism. The CNN branch extracts spatial local features of the parameters, such as abrupt changes in signal intensity; while the LSTM branch captures temporal dependencies, such as periodic fluctuations in packet loss rate. The attention mechanism further dynamically weights the importance of different parameters, thereby enhancing the model's ability to focus on key features.

[0114] The performance prediction module includes a time series prediction sub-network, which uses Transformer to predict the performance trend for the next 5-10 seconds and outputs predicted values ​​for latency and packet loss rate.

[0115] Specifically, the performance prediction module includes a time series prediction sub-network that uses a Transformer model to predict performance trends over the next 5-10 seconds and outputs predicted values ​​for latency and packet loss rate. The Transformer model effectively captures long-term and short-term dependencies through a self-attention mechanism, while avoiding the gradient vanishing problem of traditional RNN models.

[0116] The evaluation and decision-making module is used for reinforcement learning agents and includes state, action, and reward function attributes. The state attributes include: current predicted value, historical evaluation value, and link priority. The action attributes include: switching priority, adjusting link weights, and triggering optimization strategies. The reward function attributes are dynamically generated based on the objective function.

[0117] Specifically, this module is the core of the reinforcement learning agent, including state, action, and reward function attributes. State attributes include the current predicted value, historical evaluation values, and link priority; action attributes include switching priorities, adjusting link weights, and triggering optimization strategies; reward function attributes are dynamically generated based on the objective function.

[0118] The output layer is used to generate evaluation values ​​and optimization instructions. Specifically, the output layer generates evaluation values ​​and optimization instructions to guide the allocation and optimization of actual network resources.

[0119] The objective function needs to balance performance stability and resource utilization, and is divided into two parts: prediction and optimization.

[0120] The target loss function for time series prediction is:

[0121] ,

[0122]

[0123] ,

[0124] ,

[0125] Among them, the weighted mean square error Indicates delay Indicator, Mean Absolute Error Indicates packet loss rate index, Indicates the strength of the loss-smoothed signal. outliers These represent the adjustment factors for the corresponding indicators. These are the ordinal number and the total number of the sampled data, respectively;

[0126] The optimal decision objective function is:

[0127] ,

[0128] in, This represents the ratio of the current link quality to a threshold. Indicates switching costs, This indicates a fairness measure to prevent low-priority links from being idle for extended periods. These represent the corresponding weights.

[0129] Specifically, the objective function needs to balance performance stability and resource utilization, and is divided into two parts: prediction and optimization. The time series prediction objective loss function includes weighted mean square error (for latency metrics) and mean absolute error (for packet loss rate), while also considering loss smoothing of outliers in signal strength. This design ensures a good balance between prediction accuracy and stability in the model.

[0130] This model design fully combines the advantages of CNN and LSTM, leveraging the powerful spatial feature extraction capabilities of CNN and the time series modeling capabilities of LSTM. It also incorporates a Transformer model to further enhance the ability to capture long- and short-term dependencies. Furthermore, the introduction of a reinforcement learning module allows the model to dynamically adjust its strategy to optimize network performance and improve resource utilization.

[0131] The AI ​​big model includes a data processing module, a model training module, an inference module, and an optimization and control module. These modules work together to achieve real-time monitoring, prediction, and optimization of network performance.

[0132] The data processing module is used to collect and preprocess network performance parameters and input them into the AI ​​model.

[0133] The model training module uses historical data to train the model, learn the relationship between network performance and parameters, and generate optimization suggestions.

[0134] The inference module then predicts the current network state based on the trained model and outputs an optimization strategy.

[0135] The optimization control module dynamically adjusts network parameters based on the inference results, including priority allocation and link switching, to improve network performance.

[0136] The AI ​​big model achieves efficient prediction and optimization of network management performance parameters through multi-module fusion, and its design has high innovation and practicality in both theory and practice.

[0137] Example 2

[0138] like Figure 4 As shown, this application provides an emergency information service method based on the convergence of satellite communication and terrestrial mobile communication, applied to an emergency information service system based on the convergence of satellite communication and terrestrial mobile communication as described in Embodiment 1, comprising:

[0139] Disaster monitoring steps: The operator's monitoring system detected an alarm indicating that a regular village-level base station had gone out of service;

[0140] System activation steps: Activate the village-level broadband satellite emergency communication station automatically or manually;

[0141] Network establishment steps: High-throughput satellite fixed terminals establish satellite links, and pico base stations access the core network through the satellite network;

[0142] Signal coverage steps: Pico base stations provide 4G network signal coverage to the disaster-stricken areas;

[0143] Communication service steps: The mobile phones of the affected people are automatically connected to the emergency network to enable communication with the outside world;

[0144] Management and maintenance steps: Remotely monitor and manage the equipment through the network management system, and comprehensively evaluate the network performance through AI big data models to adaptively adjust and optimize the network.

[0145] It is understood that in this embodiment, establishing a satellite communication link includes automatically aligning with the AsiaSat 6D high-throughput satellite, establishing a 50Gbps communication bandwidth, and providing regional 4G coverage through pico base stations, including using external antennas to achieve precise coverage of 70-100 meters.

[0146] This application provides an emergency information service system and method based on the integration of satellite communication and terrestrial mobile communication. The system includes: a high-throughput satellite fixed station terminal, using a Ku-band 0.6-meter VSAT terminal, integrating an ACU and a modem, with a triaxial structure and a carbon fiber antenna; a terrestrial mobile communication subsystem, including a 4G picocell, a security gateway, and a signaling gateway; a multi-WAN port routing device to achieve dynamic load balancing between satellite and terrestrial broadband; an emergency power supply system using a generator and power banks to form a three-level power supply; and a network management and monitoring system supporting remote configuration and AI optimization. The method includes disaster monitoring, system activation, network establishment, signal coverage, communication services, and management and maintenance steps. This invention, through the integration of satellite and terrestrial communication and combined with AI-driven network optimization, rapidly establishes a reliable communication network in emergency scenarios, possessing characteristics such as wind resistance, low latency, and intelligent scheduling, significantly improving emergency communication support capabilities.

[0147] Figure 5 This is an electronic device provided in one embodiment of this application. For example... Figure 5 As shown, the electronic device includes at least the following components: processor 101 and memory 100, communication interface 103, and bus 102.

[0148] In this embodiment, the memory 100 is used to store executable instructions of the processor 101, which, when configured to execute instructions, implements... Figure 5 The image shows a device module for providing emergency information services based on the integration of satellite communication and terrestrial mobile communication.

[0149] In embodiments of this application, a computer-readable storage medium includes instructions that instruct a device to perform the method as described in the first aspect. For example, the instructions instruct the device to perform... Figure 4 The method is shown in the process steps.

[0150] In one embodiment of this application, the program operating in the electronic device may be a program that controls a central processing unit (CPU) or similar device to achieve the functions of the above-described embodiments of the present invention (a program that enables the computer to function). The information processed by these devices is then temporarily stored in random access memory (RAM) during processing, and subsequently stored in various ROMs such as read-only memory (FlashROM) or hard disk drives (HDDs), and read, corrected, and written by the CPU as needed.

[0151] It should be noted that a portion of the electronic device described above can also be implemented using a computer. In this case, the program for implementing the control function can be recorded on a computer-readable recording medium, and the program recorded on the recording medium can be read into the computer and executed.

[0152] It should be noted that the term "computer" as used here refers to a computer built into an electronic device, employing hardware including an operating system and peripheral devices. Furthermore, "computer-readable recording media" refers to removable media such as floppy disks, magneto-optical disks, ROMs, and CD-ROMs, as well as storage devices such as hard drives built into a computer.

[0153] Furthermore, a "computer-readable recording medium" can include: a medium that dynamically stores a program for a short period of time, such as a communication line used when transmitting a program via a network such as the Internet or a communication line such as a telephone line; or a medium that stores a program for a fixed period of time, such as volatile memory inside a computer that serves as a server or client in this case. In addition, the aforementioned program can be a program used to implement the above-mentioned functions, or it can be a program that can implement the above-mentioned functions by combining with programs already recorded in the computer.

[0154] Furthermore, the electronic device in the above embodiments can also be implemented as an assembly (device group) composed of multiple devices. Each device constituting the device group can possess some or all of the functions or functional blocks of the electronic device in the above embodiments. As a device group, it is sufficient to have all the functions or functional blocks of the electronic device.

[0155] Those skilled in the art should recognize that the above embodiments are only used to illustrate this application and are not intended to limit this application. Any appropriate changes and variations made to the above embodiments within the essential spirit and scope of this application fall within the scope of protection claimed in this application.

Claims

1. An emergency information service system based on the integration of satellite communication and terrestrial mobile communication, characterized in that, The system comprises: High-throughput satellite fixed station terminal: Ku-band 0.6-meter land automatic VSAT terminal, integrated antenna control unit ACU and modem MODEM, with three-axis structure design and high-strength carbon fiber composite antenna surface, realizing satellite broadband access; Ground mobile communication subsystem: including 4G skin base station, skin base station gateway and network management system, the skin base station gateway includes security gateway and signaling gateway, providing LTE terminal access service; Multi-WAN port routing device: configured to access satellite broadband and ground broadband at the same time, and realize dynamic load balancing; Emergency power supply system: including small generator and high-power power bank, forming a three-level power supply guarantee architecture to ensure equipment power supply in power failure; Network management and monitoring system: including satellite fixed station network management and skin base station network management, supporting remote parameter configuration and state monitoring, and comprehensively evaluating network performance through AI large model, and adaptively adjusting and optimizing the network; The AI large model adopts a hybrid neural network architecture with multi-module fusion, combining time series prediction and reinforcement learning optimization capabilities, and the network architecture comprises an input layer, a feature extraction module, a performance prediction module, an evaluation and decision module, and an output layer. The input layer is used to receive real-time monitored network management performance parameters, and the data format is a sliding window data of the past 60 seconds. The feature extraction module is a CNN-LSTM hybrid layer combined with an attention mechanism, wherein the CNN branch is used to extract spatial local features of the parameters, the spatial local features include mutation patterns of signal strength, the LSTM branch is used to capture time dependence, including periodic fluctuations of packet loss rate, and the attention mechanism is used to dynamically weight the importance of different parameters. The performance prediction module includes a time series prediction subnetwork, which is used to predict the performance trend of the next 5-10 seconds using a Transformer, and outputs the predicted values of delay and packet loss rate. The evaluation and decision module is used for a reinforcement learning agent, including state, action and reward function attributes, wherein the state attributes include current predicted values, historical evaluation values and link priority, the action attributes include switching priority, adjusting link weight and triggering optimization strategy, and the reward function attribute is dynamically generated based on the objective function; The output layer is used to generate evaluation values and optimization instructions; The objective function needs to balance performance stability and resource utilization, and is divided into prediction and optimization parts: The time series prediction target loss function is: ， ， ， ， wherein the weighted mean square error denotes the delay denotes the mean absolute error denotes the packet loss rate denotes the index, denotes the loss smoothed signal strength denotes the outlier, denote the adjustment factors for the respective indices, are the ordinal and total number of the sampled data, respectively; The optimization decision target function is: ， wherein, represents a ratio of the current link quality to a threshold value, represents a handover cost, represents a fairness term to avoid long-term idling of low-priority links, respectively represent corresponding weights.

2. The emergency information service system based on the fusion of satellite communication and terrestrial mobile communication according to claim 1, characterized in that, The high-throughput satellite fixed station terminal comprises: The three-axis stable structure is composed of an azimuth axis crossed roller bearing, a pitch axis harmonic reducer, and a polarization axis stepping motor, and the pointing error is less than or equal to 0.3° under the condition of 6-grade wind load; the antenna surface is made of carbon fiber composite material with a thickness of 2.5 mm and a tensile strength greater than or equal to 800 MPa, and the total weight is less than 15 kg; the ACU and the MODEM are integrated in the antenna feed source through a common aperture design, and the signal insertion loss is less than 0.5 dB; adaptive coding and modulation from QPSK to 64APSK are supported, and a TCP acceleration engine optimized by SCPS-TP protocol is built in, so that the end-to-end transmission delay in the GEO satellite scenario is less than 800 ms; The 4G skin base station in the ground mobile communication subsystem adopts a software-defined radio (SDR) architecture and supports dynamic spectrum sharing (DSS) technology, which can simultaneously access the LTE Band 3 / 5 / 8 frequency band within a 10 MHz bandwidth, with a transmission power of 0.1 W-5 W adjustable and a coverage radius of 70-100 meters; the security gateway uses a hardware encryption module, supports the IPSec protocol and the SM4 algorithm, and has a throughput of greater than or equal to 200 Mbps; the signaling gateway dynamically selects the MME node through the S1-flex interface, and the signaling processing capacity is greater than or equal to 200 EPS / s.

3. The emergency information service system based on the fusion of satellite communication and terrestrial mobile communication according to claim 2, characterized in that, The emergency power supply system comprises: A small generator with a rated power of 2 kVA uses frequency conversion technology, consumes less than 0.3 L / kWh, and supports 48 hours of continuous operation; a high-power power bank uses a lithium iron phosphate battery pack with a capacity of 20 kWh and an operating temperature range of -20°C to 60°C; when the power supply is interrupted, the UPS switches the power supply within less than 10 ms, and the generator starts within 30 seconds; the multi-WAN port routing device: Built-in multi-path transmission protocol MPTCP dynamically allocates flow weight according to satellite link latency and ground broadband packet loss rate; through the DiffServ QoS mechanism, voice services are assigned a DSCP 46 priority to ensure a MOS value of greater than or equal to 3.5; an 8GB SSD local cache is integrated for storing emergency command information; The network management and monitoring system uses dual-channel monitoring: satellite network management and base station network management; The satellite network management: real-time monitoring of EIRP, G / T value radio frequency parameters, and support for remote adjustment of symbol rate; the base station network management: visual presentation of user distribution and traffic heat map, and support for one-key triggering of SON self-organizing network function; alarm linkage: when the satellite link packet loss rate is greater than 5% for more than 5 minutes, the base station power is automatically reduced and a work order is pushed to the operation and maintenance APP.

4. The emergency information service system based on the fusion of satellite communication and terrestrial mobile communication according to claim 3, characterized in that, The network performance is comprehensively evaluated by an AI large model, and the network is adaptively adjusted and optimized, including: S1, setting different priorities for the satellite network management and the base station network management; S2, taking the network management with a higher priority as the first communication link and the network management with a lower priority as the second communication link; S3, monitoring the parameter data of the network performance of the first communication link, including signal strength, transmission delay, and packet loss rate; S4, performing time series performance prediction on the parameter data by an AI large model; S5, compare the time series performance prediction with each preset corresponding reference value, and comprehensively evaluate each comparison result to obtain a first evaluation value; S6, when the evaluation value is greater than or equal to a preset threshold value, return to S3; S7, otherwise, switch the network management priority, return to steps S2-S5, and obtain a result as a second evaluation value; S8, compare the sizes of the first evaluation value and the second evaluation value, and optimize the communication link of the network management when the first evaluation value is less than the second evaluation value; S9, when the first evaluation value is greater than or equal to the second evaluation value, switch the network management priority and optimize the network communication link.

5. The emergency information service system based on the fusion of satellite communication and terrestrial mobile communication according to claim 4, characterized in that, The AI large model includes a data processing module, a model training module, an inference module, and an optimization control module, which cooperate with each other to realize real-time monitoring, prediction, and optimization of network performance. The data processing module is used to collect and preprocess network performance parameters and input them into the AI model. The model training module uses historical data to train the model to learn the relationship between network performance and parameters and generate optimization suggestions. The inference module predicts the current network state based on the trained model and outputs optimization strategies. The optimization control module dynamically adjusts network parameters, including priority allocation and link switching, to improve network performance.

6. An emergency information service method based on the integration of satellite communication and terrestrial mobile communication, applied to the emergency information service system based on the integration of satellite communication and terrestrial mobile communication according to any one of claims 1 to 5, characterized in that, The disaster monitoring step: the operator monitoring system detects a village-level conventional base station service withdrawal alarm; The system activation step: automatically or manually activating a village-level broadband satellite emergency communication station; The network establishment step: a high-throughput satellite fixed terminal establishes a satellite link, and a skin base station accesses a core network through a satellite network; The signal coverage step: the skin base station provides 4G network signal coverage for the disaster area; The communication service step: the affected people's mobile phones automatically access the emergency network to realize communication with the outside world; The management and maintenance step: remotely monitoring and managing the equipment through the network management system, and comprehensively evaluating the network performance through the AI large model to adaptively adjust and optimize the network. Establishing a satellite communication link includes automatically aligning the Asia-Pacific 6D high-throughput satellite, establishing a 50Gbps communication bandwidth, and providing regional 4G coverage through the skin base station, including using an external antenna to achieve 70-100 meter accurate coverage.

7. The emergency information service method based on the fusion of satellite communication and terrestrial mobile communication according to claim 6, characterized in that, The processor; 8. An electronic device, comprising: Memory for storing processor-executable instructions; When the processor is configured to execute the instructions, it realizes the emergency information service method based on the fusion of satellite communication and ground mobile communication according to any one of claims 6-7. The computer readable storage medium stores a program, which instructs the device to execute the emergency information service method based on the fusion of satellite communication and ground mobile communication according to any one of claims 6-7. ​ 9. A computer-readable storage medium, characterized in that, ​

Images