An emergency communication method and system fusing a satellite network and a ground base station network

Abstract

The application belongs to the technical field of emergency communication, and discloses an emergency communication method and system fusing a satellite network and a ground base station network; the method comprises the following steps: acquiring satellite data and ground data; performing spectrum allocation through federated learning; positioning blind area coordinates and survivor base station coordinates according to the satellite data, and adjusting the beam direction and width in combination with MIMO technology; selecting the survivor base station as a relay base station; acquiring a channel state, dynamically adapting modulation and coding through a dynamic switching strategy according to the spectrum allocation result and the beam gain value; constructing a comprehensive link weight model according to the link delay of the ground data, the spectrum allocation result and the dynamic switching strategy, and generating an optimal path by using an improved Dijkstra algorithm; and the relay base station is also powered through a hybrid power supply strategy; the application can significantly improve the fast self-healing capability of the communication network to node failure and link interruption and the service transmission reliability.

Description

Technical Field

[0001] The present invention relates to the field of emergency communication technology, and more specifically, to an emergency communication method and system integrating a satellite network and a ground base station network. Background Art

[0002] In emergency communication scenarios, traditional ground base station networks are vulnerable to natural disasters such as earthquakes, typhoons, or emergencies, which can lead to infrastructure damage, power outages, or link congestion, making it difficult to ensure the continuous transmission of critical services.

[0003] A Chinese patent with authorization announcement number CN109495159B discloses an emergency communication system and method based on satellite communication. The system includes a user terminal, a receiving terminal, a ground relay station and a communication satellite. The ground relay station includes a sending ground station and a receiving ground station. The number of sending ground stations is at least two, and the user terminal is communicatively connected to the at least two sending ground stations; the user terminal is configured with a communication link selection module and a detection module for detecting the communication quality between the at least two sending ground stations and the communication satellite; the detection module sends specific data signals representing each communication link to the target receiving ground station via the at least two sending ground stations and the communication satellite, receives feedback signals from the target receiving ground station, and determines the communication quality of each communication link based on the feedback signals; the link selection module is data-connected to the detection module, and selects one of the communication links for communication based on the communication quality of each communication link to ensure the normal progress and communication quality of the communication.

[0004] Although the above method can meet most scenarios, research and practical application of the above method and existing technology have found that the above method and existing technology have at least the following defects:

[0005] Relying on fixed routing strategies, it is unable to respond quickly to node failures or link interruptions.

[0006] In view of this, the present invention proposes an emergency communication method and system that integrates a satellite network and a ground base station network to solve the above problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: an emergency communication method integrating a satellite network and a ground base station network, comprising the following steps:

[0008] Acquire satellite data and ground data; allocate spectrum through federated learning;

[0009] Based on satellite data, the coordinates of the blind spot and the surviving base station are located, and the beam direction and width are adjusted in combination with MIMO technology; the surviving base station is selected as the relay base station;

[0010] Obtain channel status and dynamically adapt modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values;

[0011] Based on ground data, spectrum allocation results, and link delays of dynamic switching strategies, a comprehensive link weight model is constructed, and the improved Dijkstra algorithm is used to generate the optimal path.

[0012] Based on satellite data and ground data, combined with the actual load of the relay base station, a hybrid power supply strategy is used to power the relay base station.

[0013] Furthermore, methods for spectrum allocation through federated learning include:

[0014] Acquire satellite data and ground data and perform standardized processing;

[0015] Analyze satellite data and ground data to obtain spectrum status characteristics;

[0016] The central server uses the spectrum status characteristics as input to the spectrum allocation model to obtain the initial spectrum allocation result;

[0017] The edge nodes, composed of satellites and ground base stations, use locally collected data as input, calculate the error between the initial spectrum allocation result and the actual allocation result, and update the model parameters of the spectrum allocation model using the gradient descent algorithm to generate gradient increments.

[0018] The central server obtains the gradient increments, aggregates the gradient increments using the FedAvg algorithm, generates globally updated model parameters, broadcasts the updated model parameters to all edge nodes, and starts the next round of iteration;

[0019] The model parameters corresponding to the minimum error are obtained, and the edge node obtains the updated initial spectrum allocation result through the corresponding spectrum allocation model, and generates the spectrum allocation result in combination with local constraints.

[0020] Furthermore, the method for obtaining the spectrum state characteristics includes:

[0021] Counting the ratio of the duration of a preset frequency band occupied by a signal within a preset time to the total duration in the satellite data to obtain the frequency band occupancy rate; performing Fourier transform on the received signal to obtain a frequency domain signal, and obtaining the frequency component distribution of the frequency domain signal within the preset frequency band;

[0022] Obtaining the power spectral density of the interference signal in the satellite data within a preset frequency band, and integrating the power spectral density within the preset frequency band to obtain the interference signal power;

[0023] Count the actual number of users accessing the ground data within a preset time, and calculate the load rate based on the maximum number of users accessing the base station.

[0024] The frequency band occupancy, frequency component distribution, interference signal power and load rate are spliced ​​together to obtain the spectrum status characteristics.

[0025] Furthermore, the method for locating the blind spot coordinates includes:

[0026] Calculate the distance between the base station location coordinates and the coordinates to be judged. Combined with the signal strength, if the distance between the coordinates to be judged and the base station is less than the maximum effective distance, but the corresponding signal strength is lower than the signal strength threshold, then preliminarily determine that the coordinates to be judged are a communication blind spot. Traverse all coordinates to be judged, obtain the communication blind spot set, and repeatedly judge the communication blind spot set to obtain the blind spot coordinates.

[0027] Furthermore, the method for adjusting the beam direction includes:

[0028] Obtain blind spot coordinates, elevation data and satellite attitude data;

[0029] The blind spot coordinates are converted into geocentric rectangular coordinates based on the curvature radius of the circle, and the angle between the position vector of the geocentric rectangular coordinates and the horizontal position relative to the satellite is calculated.

[0030] Then convert the blind spot coordinates into azimuth and elevation angles in the satellite coordinate system;

[0031] Based on the geometric model of the phased array antenna, calculate the phase offset of each element of the phased array antenna array;

[0032] Receive the pilot signal from the user terminal and calculate the deviation between the actual beam center and the target area;

[0033] The gradient descent method is used to adjust the phase offset to minimize the standard deviation of the signal intensity at the edge of the target area;

[0034] Minimum mean square error (MMSE) precoding is used to obtain a precoding matrix. Combined with beam pointing constraints, azimuth and elevation control parameters are embedded in the precoding matrix to adjust the beam direction.

[0035] Furthermore, the method for adjusting the beam width includes:

[0036] The target area is calculated through satellite remote sensing and the user density is counted. When the target area is smaller than the preset area threshold and the user density is greater than the preset density threshold, the N×N center area array elements are activated to increase the equivalent aperture and narrow the half-power beamwidth to a preset narrow angle range. Otherwise, all array elements are activated, introducing random phase perturbations within the preset phase range and expanding the half-power beamwidth to a preset wide angle range.

[0037] Furthermore, the method for dynamically adapting modulation and coding through a dynamic switching strategy includes:

[0038] Calculate the net signal power by the receiving end signal strength and the receiving antenna gain; calculate the noise power by combining the noise power spectrum density and bandwidth; and obtain the channel signal-to-noise ratio based on the net signal power and noise power.

[0039] The rain attenuation value is calculated based on the rainfall rate; the coding rate is selected based on the channel signal-to-noise ratio through a three-level coding redundancy strategy:

[0040] The beam gain value is calculated based on the antenna efficiency, antenna aperture area and operating wavelength;

[0041] According to the channel signal-to-noise ratio, coding rate and beam gain value, the corresponding modulation mode is matched based on the dynamic matching model;

[0042] Collect Ka-band and L-band dual-band signals, use dual-band signal merging technology to merge the signals, and calculate the merged signal based on the channel signal-to-noise ratio and the received signal of the corresponding frequency band;

[0043] Dynamically adjust the signal transmission power according to the rain attenuation value.

[0044] Furthermore, the method for selecting the coding rate through the three-level coding redundancy strategy includes:

[0045] The coding rate is allocated based on a preset redundancy strategy according to the rain attenuation value and the channel signal-to-noise ratio (SNR). The preset redundancy strategy includes: when the rain attenuation value is less than a preset low rain attenuation threshold and the channel signal-to-noise ratio (SNR) is greater than a preset high channel signal-to-noise ratio (SNR) threshold, the coding rate is a first-level coding redundancy value; when the rain attenuation value is not less than a preset low rain attenuation threshold and less than a preset high rain attenuation threshold, or when the channel signal-to-noise ratio (SNR) is not less than a preset low channel signal-to-noise ratio (SNR) threshold and less than a preset high channel signal-to-noise ratio (SNR) threshold, the coding rate is a second-level coding redundancy value; when the rain attenuation value is not less than a preset high rain attenuation threshold and the channel signal-to-noise ratio (SNR) is less than a preset low channel signal-to-noise ratio (SNR) threshold, the coding rate is a third-level coding redundancy value; and when the channel signal-to-noise ratio (SNR) exceeds the high channel signal-to-noise ratio (SNR) threshold for W consecutive times, the coding rate is switched to a preset backup coding rate value.

[0046] Furthermore, the method for matching the corresponding modulation mode according to the dynamic matching model includes:

[0047] When the channel signal-to-noise ratio is greater than the first receiving signal-to-noise ratio, 64QAM modulation is used, and the first receiving signal-to-noise ratio is calculated by combining the preset theoretical first receiving signal-to-noise ratio and the beam gain;

[0048] When the channel signal-to-noise ratio is not less than the second receiving signal-to-noise ratio and not greater than the first receiving signal-to-noise ratio, 16QAM modulation is adopted, and the second receiving signal-to-noise ratio is calculated by combining the preset theoretical second receiving signal-to-noise ratio and the beam gain;

[0049] When the channel signal-to-noise ratio is not less than the third receiving signal-to-noise ratio and less than the second receiving signal-to-noise ratio, QPSK modulation is adopted, and the third receiving signal-to-noise ratio is obtained by calculating the preset theoretical third receiving signal-to-noise ratio and the beam gain;

[0050] When the channel signal-to-noise ratio is less than the third receiving signal-to-noise ratio, the BPSK modulation mode is adopted;

[0051] When the modulation mode is m, the preset hysteresis margin is superimposed on the preset minimum signal-to-noise ratio threshold corresponding to the modulation mode m to obtain an updated signal-to-noise ratio threshold. When the channel signal-to-noise ratio exceeds the updated signal-to-noise ratio threshold, the modulation mode is switched, where m is 64QAM, 16QAM, QPSK, and BPSK.

[0052] Furthermore, the method for generating the optimal path includes:

[0053] A network topology diagram consisting of satellites, ground base stations, and user terminals is constructed. A composite weight function is constructed based on link delay, bandwidth, and power outage risk to calculate the initial weight of each link. Link delay is calculated based on transmission delay, propagation delay, processing delay, and retransmission delay. Power outage risk is calculated based on remaining power, preset critical power, total average power consumption, and total power consumption.

[0054] Each node deploys a heartbeat packet sending module to send heartbeat packets to adjacent nodes at a preset time. At the same time, it regularly evaluates the optimal path and triggers the optimal path recalculation when a preset emergency occurs.

[0055] Take the satellite as the source node and the user terminal as the target node. Select the node u closest to the source node from the unvisited nodes and traverse its neighbor nodes v. If the new distance from u to v is shorter, update the distance and record the predecessor node.

[0056] During the iteration process, a heap list containing K shortest paths is maintained. When the weight of the new path is less than the maximum weight of the heap list, it is replaced and updated until all nodes are visited. The path with the smallest weight is taken as the optimal path.

[0057] Furthermore, the method for powering the relay base station using a hybrid power supply strategy includes:

[0058] If the solar output power P is not lower than the sum of the device's real-time power consumption Q and the battery charging power, the solar power is triggered to directly power the load, and the remaining power is used to charge the battery;

[0059] If P is lower than Q or the remaining battery power E is greater than the preset critical power, the solar energy and battery combination is triggered to power the load; the preset critical power is calculated based on Q and the preset minimum guarantee time;

[0060] If P is lower than the preset power threshold or E is not greater than the preset critical power, the battery will be used for power supply only. When E reaches the minimum protection power, a power-off warning will be triggered.

[0061] The waste heat generated by the device's heat dissipation is recovered through the thermoelectric power generation sheet, and the recovered power is added to the battery charging circuit. The recovered power is calculated through the temperature difference efficiency, device temperature, ambient temperature and heat dissipation area.

[0062] An emergency communication system integrating a satellite network and a ground base station network, implementing an emergency communication method integrating a satellite network and a ground base station network, comprising:

[0063] Spectrum allocation module: obtains satellite data and ground data; allocates spectrum through federated learning;

[0064] Beam adjustment module: locates the blind spot coordinates and the coordinates of the surviving base stations based on satellite data, and adjusts the beam direction and width in combination with MIMO technology; selects the surviving base station as the relay base station;

[0065] Coding and modulation module: obtains channel status and dynamically adapts modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values;

[0066] Path planning module: Based on ground data, spectrum allocation results, and link delay of dynamic switching strategies, a comprehensive link weight model is constructed and the improved Dijkstra algorithm is used to generate the optimal path;

[0067] Intelligent power supply module: Based on satellite data and ground data, combined with the actual load of the relay base station, it supplies power to the relay base station through a hybrid power supply strategy.

[0068] The technical effects and advantages of the present invention's emergency communication method and system integrating satellite network and ground base station network are as follows:

[0069] The present invention constructs a comprehensive link weight model by integrating real-time status data such as ground base station load, spectrum allocation results, and dynamic modulation and coding link delay, and uses an improved Dijkstra algorithm to dynamically generate the optimal path. This solves the technical problem of existing technologies relying on preset routing rules and being unable to perceive network status in real time. When a ground base station is damaged, a satellite link is interrupted by rain attenuation interference, or a relay node fails due to excessive load in an emergency scenario, the present invention can obtain node status and link parameters in real time through data acquisition. Based on the comprehensive link weight model, the failed nodes or links are immediately eliminated and a hybrid path containing satellite direct connection and ground relay jump is recalculated, avoiding communication interruption or congestion caused by the fixed routing strategy's inability to perceive node failure. At the same time, the dynamic screening and hybrid power supply strategy of the relay base station ensures the continuous operation of key ground nodes. Combined with the wide-area coverage capability of the satellite network, a satellite blind spot filling-ground collaboration-dynamic routing anti-destruction disaster recovery mechanism is formed, enabling the emergency communication system to maintain end-to-end connectivity in extreme environments, significantly improving the communication network's rapid self-healing capabilities for node failure and link interruption and the reliability of service transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a flow chart of an emergency communication method integrating a satellite network and a ground base station network according to embodiment 1 of the present invention;

[0071] Figure 2 This is a data flow diagram of Example 1 of the present invention;

[0072] Figure 3 This is a data flow diagram of Example 2 of the present invention;

[0073] Figure 4 This is a schematic diagram of an emergency communication system integrating a satellite network and a ground base station network according to Example 3 of the present invention. DETAILED DESCRIPTION

[0074] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0075] Example 1:

[0076] See also Figure 1 and Figure 2 As shown, this embodiment provides an emergency communication method integrating a satellite network and a ground base station network, comprising the following steps:

[0077] Acquire satellite data and ground data; allocate spectrum through federated learning;

[0078] Obtaining satellite data and ground data and allocating spectrum through federated learning can dynamically aggregate multi-dimensional status information of satellite and ground networks while protecting data privacy. This can adaptively optimize the allocation strategy of spectrum resources between satellite links and ground links in response to the complex and changeable channel environment and heterogeneous network characteristics in emergency communications, achieve the coordinated complementarity of high-speed transmission in high-frequency bands (such as Ka-band) and high-reliability transmission in low-frequency bands (such as L-band), improve spectrum utilization efficiency while ensuring link stability for key services, and provide core resource scheduling support for the full-area coverage and disaster resistance of the integrated network in emergency scenarios.

[0079] Methods for spectrum allocation through federated learning include:

[0080] Acquire satellite data and ground data and perform standardized processing;

[0081] Counting the ratio of the duration of a preset frequency band occupied by a signal within a preset time to the total duration in the satellite data to obtain the frequency band occupancy rate; performing Fourier transform on the received signal to obtain a frequency domain signal, and obtaining the frequency component distribution of the frequency domain signal within the preset frequency band;

[0082] Obtaining the power spectral density of the interference signal in the satellite data within a preset frequency band, and integrating the power spectral density within the preset frequency band to obtain the interference signal power;

[0083] Count the actual number of users accessing the ground data within a preset time, and calculate the load rate based on the maximum number of users accessing the base station.

[0084] The frequency band occupancy, frequency component distribution, interference signal power and load rate are spliced ​​together to obtain the spectrum status characteristics;

[0085] The central server uses the spectrum status characteristics as input to the spectrum allocation model to obtain the initial spectrum allocation result;

[0086] The edge nodes, composed of satellites and ground base stations, use locally collected data as input, calculate the error between the initial spectrum allocation results predicted by the spectrum allocation model and the actual allocation results, and update the model parameters using the gradient descent algorithm to generate gradient increments.

[0087] The central server obtains the gradient increments, aggregates the gradient increments using the FedAvg algorithm, generates globally updated model parameters, broadcasts the updated model parameters to all edge nodes, and starts the next round of iteration;

[0088] The model parameters corresponding to the minimum error are obtained, and the edge node obtains the updated initial spectrum allocation result through the corresponding spectrum allocation model, and generates the spectrum allocation result in combination with local constraints.

[0089] The satellite locates the blind spot coordinates based on satellite data, and adopts multi-user MIMO technology to adjust the beam direction and width according to the blind spot coordinates. The satellite locates the blind spot coordinates based on satellite data and adopts multi-user MIMO technology to adjust the satellite beam direction and width. It can accurately compensate for the blind spots of ground base station coverage caused by terrain obstruction and equipment damage in emergency communications by dynamically focusing the satellite beam energy to the target area, and realize the coordination of satellite wide-area coverage blind spot filling and ground base station regional access. Multi-user MIMO technology improves spectrum efficiency through spatial division multiplexing, enhances the capacity of high-density user areas in narrow beam mode, and expands the coverage of sparse areas in wide beam mode. At the same time, it suppresses co-channel interference and resists channel fading such as rain fade, providing on-demand coverage and intelligent energy aggregation physical layer support for the converged network, ensuring blind spot-free communication and stable transmission of multi-user concurrent services in emergency scenarios.

[0090] Methods for locating blind spot coordinates include:

[0091] Calculate the distance between the base station location coordinates and the coordinates to be judged. Combined with the signal strength, if the distance between the coordinates to be judged and the base station is less than the maximum effective distance, but the corresponding signal strength is lower than the signal strength threshold, then preliminarily determine that the coordinates to be judged are a communication blind spot. Traverse all coordinates to be judged, obtain the communication blind spot set, and repeatedly judge the communication blind spot set to obtain the blind spot coordinates.

[0092] Methods for adjusting the beam direction include:

[0093] Obtain blind spot coordinates, elevation data and satellite attitude data;

[0094] The communication blind spot coordinates are converted into geocentric rectangular coordinates in combination with the curvature radius of the circle; the position vector of the blind spot coordinates relative to the satellite and the horizontal angle of the blind spot coordinates relative to the satellite in the satellite orbit plane are calculated based on the geocentric rectangular coordinates;

[0095] Convert the blind area coordinates into azimuth and elevation angles in the satellite coordinate system;

[0096] Based on the geometric model of the phased array antenna, calculate the phase offset of each element of the phased array antenna array;

[0097] Receive the pilot signal from the user terminal and calculate the deviation between the actual beam center and the target area;

[0098] The gradient descent method is used to adjust the phase offset to minimize the standard deviation of the signal intensity at the edge of the target area;

[0099] Minimum mean square error (MMSE) precoding is used to obtain a precoding matrix. Combined with beam pointing constraints, azimuth and elevation control parameters are embedded in the precoding matrix to adjust the beam direction.

[0100] Methods for adjusting the beamwidth include:

[0101] The target area is calculated through satellite remote sensing and the user density is counted. When the target area is smaller than a preset area threshold and the user density is greater than a preset density threshold, the N×N array elements in the central area are activated, increasing the equivalent aperture and narrowing the half-power beamwidth to a preset narrow angle range. Otherwise, all array elements are activated, introducing random phase perturbations within a preset phase range (e.g., ±π / 4) to expand the half-power beamwidth to a preset wide angle range.

[0102] Obtain channel status and dynamically adapt modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values; obtain channel status and combine spectrum allocation results and beam gain values, and dynamically adapt modulation and coding through dynamic switching strategies, which can respond in real time to channel quality fluctuations between satellite links and ground links in emergency communications, such as signal attenuation caused by rain fade and multipath interference caused by obstacles, while ensuring transmission reliability and maximizing spectrum utilization efficiency; this mechanism supports satellites and ground base stations to flexibly adjust transmission parameters according to real-time channel status. For example, when the satellite link uses 64QAM high-speed transmission at a high signal-to-noise ratio, the ground base station switches to BPSK with strong noise resistance in an interference environment, achieving a dynamic balance between anti-fading capability and spectrum efficiency, providing differentiated transmission guarantees for multiple types of services such as voice, video, and data in emergency scenarios, and ensuring stable end-to-end communication quality of the converged network in complex environments.

[0103] The method of dynamically adapting modulation and coding through a dynamic switching strategy includes:

[0104] Calculate the net signal power by the receiving end signal strength and the receiving antenna gain; calculate the noise power by combining the noise power spectrum density and bandwidth; and obtain the channel signal-to-noise ratio based on the net signal power and noise power.

[0105] The rain attenuation value is calculated based on the rainfall rate; the coding rate is selected based on the channel signal-to-noise ratio through a three-level coding redundancy strategy:

[0106] The method of selecting the coding rate through the three-level coding redundancy strategy includes:

[0107] The coding rate is allocated based on the preset redundancy strategy according to the rain attenuation value and the channel signal-to-noise ratio. In the preset redundancy strategy, when the rain attenuation value is less than the preset low rain attenuation threshold and the channel signal-to-noise ratio is greater than the preset high channel signal-to-noise ratio threshold, the coding rate is a first-level coding redundancy value, such as 0.8. When the rain attenuation value is not less than the preset low rain attenuation threshold and less than the preset high rain attenuation threshold, or when the channel signal-to-noise ratio is not less than the preset low channel signal-to-noise ratio threshold and less than the preset high channel signal-to-noise ratio threshold, the coding rate is a second-level coding redundancy value, such as 0.6. When the rain attenuation value is not less than the preset high rain attenuation threshold and the channel signal-to-noise ratio is less than the preset low channel signal-to-noise ratio threshold, the coding rate is a third-level coding redundancy value, such as 0.4. When the channel signal-to-noise ratio exceeds the high channel signal-to-noise ratio threshold for W consecutive times, the coding rate is switched to the preset backup coding rate value.

[0108] The rain attenuation value is calculated based on the rainfall rate and combined with the channel signal-to-noise ratio (SNR). A three-level coding redundancy strategy dynamically selects the coding rate. A low redundancy rate is used to ensure efficiency in low rain attenuation and high SNR conditions, a medium redundancy rate is used to balance performance in medium rain attenuation or SNR conditions, and a high redundancy rate is used to ensure reliability in high rain attenuation and low SNR conditions. When the channel remains good, the system switches to a preset backup code rate to optimize transmission efficiency. This system dynamically adapts to the signal attenuation and noise interference levels in different scenarios, addressing the heterogeneous channel characteristics of emergency communications, where satellite links are susceptible to rain attenuation and terrestrial links are susceptible to multipath interference. When extreme weather conditions such as heavy rain cause severe rain attenuation in satellite links, high-redundancy coding is used to enhance signal anti-interference capabilities and reduce the bit error rate to ensure the transmission of critical signaling, such as rescue instructions. When channel conditions improve, coding redundancy is automatically reduced, improving spectrum efficiency to support high-bandwidth services such as HD video backhaul. This avoids the over-protection or unreliability issues of fixed coding strategies in complex environments. This system provides converged networks with intelligent coding adaptation that balances transmission reliability and efficiency, ensuring stable transmission of multiple services and optimized resource utilization in emergency scenarios.

[0109] The beam gain value is calculated based on the antenna efficiency, antenna aperture area and operating wavelength;

[0110] According to the channel signal-to-noise ratio, coding rate and beam gain value, the corresponding modulation mode is matched based on the dynamic matching model;

[0111] Methods for matching corresponding modulation modes based on a dynamic matching model include:

[0112] When the channel signal-to-noise ratio is greater than the first receiving signal-to-noise ratio, 64QAM modulation is used, and the first receiving signal-to-noise ratio is calculated by combining the preset theoretical first receiving signal-to-noise ratio and the beam gain;

[0113] When the channel signal-to-noise ratio is not less than the second receiving signal-to-noise ratio and not greater than the first receiving signal-to-noise ratio, 16QAM modulation is adopted, and the second receiving signal-to-noise ratio is calculated by combining the preset theoretical second receiving signal-to-noise ratio and the beam gain;

[0114] When the channel signal-to-noise ratio is not less than the third receiving signal-to-noise ratio and less than the second receiving signal-to-noise ratio, QPSK modulation is adopted, and the third receiving signal-to-noise ratio is obtained by calculating the preset theoretical third receiving signal-to-noise ratio and the beam gain;

[0115] When the channel signal-to-noise ratio is less than the third receiving signal-to-noise ratio, the BPSK modulation mode is adopted;

[0116] When the modulation mode is m, a preset hysteresis margin is superimposed on the preset minimum signal-to-noise ratio threshold corresponding to the modulation mode m. The positive or negative value of the preset hysteresis margin can be adjusted according to the actual situation to obtain an updated signal-to-noise ratio threshold. When the channel signal-to-noise ratio exceeds the updated signal-to-noise ratio threshold, the modulation mode is switched, where m is 64QAM, 16QAM, QPSK, and BPSK.

[0117] Based on a dynamic matching model and combined with beam gain, the modulation scheme is dynamically adjusted. For example, 64QAM is used to improve transmission rates when the signal-to-noise ratio is high, and BPSK is switched to ensure signal noise immunity when the signal-to-noise ratio is low. A preset hysteresis margin is added to prevent frequent modulation oscillations. This allows for fine-grained adaptation to the heterogeneous channel characteristics of satellite and terrestrial networks (such as the time-varying rain attenuation characteristics of satellite links and multipath fading of terrestrial links). When the satellite link signal-to-noise ratio decreases due to rain attenuation, the scheme automatically switches from higher-order modulation to lower-order modulation, such as from 16QAM to QPSK. This sacrifices some spectral efficiency in exchange for signal reliability, ensuring uninterrupted service for high-traffic services such as real-time video in disaster areas. When the signal-to-noise ratio in the area covered by the ground base station steadily improves, a preset minimum signal-to-noise ratio threshold, combined with a preset hysteresis margin, is used to determine a smooth switch to higher-order modulation, such as from QPSK to 16QAM. This avoids repeated switching caused by noise fluctuations and achieves a balance between transmission efficiency and switching stability. This mechanism enables the converged network to intelligently select the optimal modulation method based on the real-time channel status, taking into account both high-speed transmission and anti-interference capabilities in extreme emergency scenarios, providing physical layer guarantees for seamless switching of multi-mode terminals between satellite and terrestrial networks, and enhancing the robustness and spectrum utilization efficiency of the overall communication system.

[0118] Simultaneously receives Ka-band (high frequency band, 26.5-40GHz) and L-band (low frequency band, 1-2GHz) signals, using dual-band signal merging technology to combine the signals. The combined signal is calculated based on the channel signal-to-noise ratio and the received signal of the corresponding frequency band;

[0119] The signal transmission power is dynamically adjusted according to the rain attenuation value, where the signal transmission power does not exceed the maximum power limit of the satellite transponder (such as 50dBm) and meets the power spectrum density of the ground base station receiver.

[0120] Based on ground data, spectrum allocation results, and link delays of dynamic switching strategies, a comprehensive link weight model is constructed, and the improved Dijkstra algorithm is used to generate the optimal path.

[0121] A comprehensive link weight model is constructed based on ground data, spectrum allocation results, and link latency from dynamic switching strategies. An improved Dijkstra algorithm is then used to generate the optimal path. This model dynamically adapts to the real-time state of heterogeneous networks, addressing the high propagation latency of satellite links and potential localized failures and load imbalances in ground base station links during emergency communications. When satellite link latency is high but ground base stations are damaged, direct satellite links are prioritized to ensure full coverage. When ground relay base stations are available, the comprehensive link weight model comprehensively assesses path reliability and efficiency, avoiding high-latency or high-failure-risk links and achieving collaborative routing optimization in satellite-ground hybrid networks. This mechanism supports dynamic changes in network topology during emergency scenarios, ensuring that critical services such as rescue command signaling and data backhaul from disaster areas are transmitted with minimal latency and maximum reliability. It also balances network load, improving the overall resource utilization efficiency and disaster resilience of the converged network.

[0122] Methods for generating optimal paths include:

[0123] Build a network topology diagram including satellites, ground base stations and user terminals ,in, is a set of nodes; is the link set;

[0124] Calculate the transmission delay based on link transmission with a preset data block size and dynamic switching strategy; calculate the propagation delay based on the propagation distance and propagation speed; obtain the processing delay through measurement; calculate the number of retransmissions based on the bit error rate, calculate the retransmission delay based on the number of retransmissions and the transmission delay, and calculate the link delay based on the transmission delay, propagation delay, processing delay, and retransmission delay;

[0125] The power outage risk is calculated based on the remaining power, preset critical power, total average power consumption, and total power consumption. The total average power consumption is calculated based on the average power consumption and demand time, and the total power consumption is calculated based on the average power consumption and demand time. The demand time is calculated based on the historical power outage demand time.

[0126] The initial weight of each link is calculated based on a composite weight function constructed from link delay, bandwidth, and power outage risk; the bandwidth can be obtained based on the spectrum allocation results.

[0127] A heartbeat packet sending module is deployed at each node, sending heartbeat packets to adjacent nodes at preset intervals, such as every 5 seconds, to monitor node status. At the same time, satellites and ground base stations periodically report link status data to a central management node for subsequent weight updates.

[0128] Regularly re-evaluate the optimal path in the network according to a preset timer; for example, automatically start the path calculation process every 30 seconds;

[0129] When a preset emergency event is detected, such as node failure, including base station power outage, node failure to reply to heartbeat packets on time; link interruption, such as optical cable break causing link unreachability or service priority change, the optimal path recalculation is immediately triggered.

[0130] Define the satellite as the source node of the transmission and the user terminal as the target node;

[0131] Select the node u with the shortest distance to the source node from the unvisited nodes as the current node;

[0132] Traverse all neighbor nodes v of the current node u and calculate the new distance from the source node to v via u;

[0133] If the new distance is less than the distance from the source node to the neighbor node v, then update the distance from the source node to the neighbor node v to the new distance, and set the predecessor node of the neighbor node v to u; record the current shortest distance from the source node to each node, as well as the predecessor node of each node;

[0134] When updating the distance and path in each iteration, a list of K shortest paths is maintained. The paths and their weights are stored in a heap data structure, sorted by weight from small to large. When the weight of a new path is less than the maximum weight path in the list, the path is replaced.

[0135] Repeat the above iterative process until all nodes are visited, and take the path with the smallest weight as the optimal path.

[0136] Based on satellite data and ground data, combined with the actual load of the relay base station, a hybrid power supply strategy is used to power the relay base station.

[0137] By combining satellite data with the actual load of the relay base station, a hybrid power supply strategy is formulated. This can provide continuous and stable power guarantee for the relay base station in response to the possible paralysis of the ground power grid and the interruption of traditional mains power supply in emergency scenarios: solar power supply is prioritized in areas with sufficient sunshine to reduce energy consumption dependence, and at night or in rainy environments, it is switched to battery or satellite-assisted power supply mode to avoid relay node failure due to power outages; at the same time, the power supply power is dynamically adjusted according to the base station load to ensure the continuous operation of the relay base station as a key hub for the integration of satellite and ground networks, maintain the local connectivity of the ground network and enhance the collaborative efficiency with the satellite link, providing underlying physical support for the emergency communication system's resistance to damage and long-term endurance in extreme environments, and ensuring the uninterrupted transmission of core services such as rescue command and data backhaul from disaster areas.

[0138] Methods for powering relay base stations using a hybrid power supply strategy include:

[0139] If the solar output power is not lower than the sum of the device's real-time power consumption and the battery charging power, the solar power is triggered to directly power the load, and the remaining power is used to charge the battery;

[0140] If the solar power output is lower than the device's real-time power consumption or the remaining battery power is greater than the preset critical power level, a combination of solar power and battery power is triggered to power the load, with solar power being the primary source of power and the battery providing the remaining power. The preset critical power level is calculated based on the device's real-time power consumption and the preset minimum guaranteed time.

[0141] If the solar output power is lower than the preset power threshold or the remaining battery power is not greater than the preset critical power, the system will be powered only by the battery. When the remaining battery power reaches the minimum protection power, a power-off warning will be triggered.

[0142] The waste heat generated by the equipment's heat dissipation is recovered through the thermoelectric power generation sheet, and the recovered power is added to the battery charging circuit. The recovered power is calculated through the temperature difference efficiency, equipment temperature, ambient temperature and heat dissipation area. Among them, the temperature difference efficiency is calculated by the ratio of actual output energy to available thermal energy.

[0143] Example 2:

[0144] See also Figure 3 As shown, this embodiment provides a beam dynamic adjustment method applied to Example 1, including the following steps:

[0145] Define the state space: The state vector includes user status, such as location coordinates, mobile speed, and service type (voice / video); satellite status, including current beam direction, beam width, and remaining satellite fuel; and environmental status, including rain attenuation distribution and co-channel interference power spectrum;

[0146] Define the action space: The actions include beam adjustment parameters such as direction increment and width scaling factor, and power allocation strategies such as transmit power gain in the target area;

[0147] For each user i, we extract the location sequence of the past T = 10 time steps and use the LSTM network to extract movement trend features, such as trajectory prediction direction and speed change rate.

[0148] Aggregate all user locations to generate a heatmap feature (Heatmap) to reflect the current user-dense areas, such as the locations of temporary resettlement sites in disaster areas.

[0149] The beam direction is converted into a unit vector in the satellite coordinate system, and the angle between it and the user position vector is calculated to measure the coverage deviation.

[0150] Mapping beam width to coverage area as a coverage capability indicator;

[0151] The reward function is designed based on coverage, stability, and transmission efficiency. Coverage is calculated based on the service priority of the current number of covered users, the total number of users, and the predefined number of users in the blind spot. Stability is calculated based on the adjustment amount of the beam direction. Transmission efficiency is calculated based on the actual transmission rate, the theoretical maximum rate, the adjusted energy consumption, and the energy consumption threshold.

[0152] The policy network is designed using an actor-critic architecture. The state vector in the state space is used as the input of the policy network. The actor network outputs the probability distribution of beam adjustment actions, and the critic network evaluates the state-action value.

[0153] Through training with historical simulation data and real-time emergency scenario data, an optimized policy network is obtained, and the action is adjusted based on the optimized policy network output.

[0154] Example 3:

[0155] See also Figure 4 As shown, this embodiment provides an emergency communication system integrating a satellite network and a ground base station network, including:

[0156] Spectrum allocation module: obtains satellite data and ground data; allocates spectrum through federated learning;

[0157] Beam adjustment module: locates the blind spot coordinates and the coordinates of the surviving base stations based on satellite data, and adjusts the beam direction and width in combination with MIMO technology; selects the surviving base station as the relay base station;

[0158] Coding and modulation module: obtains channel status and dynamically adapts modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values;

[0159] Path planning module: Based on ground data, spectrum allocation results, and link delay of dynamic switching strategies, a comprehensive link weight model is constructed and the improved Dijkstra algorithm is used to generate the optimal path;

[0160] Intelligent power supply module: Based on satellite data and ground data, combined with the actual load of the relay base station, it supplies power to the relay base station through a hybrid power supply strategy.

[0161] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0162] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An emergency communication method integrating a satellite network and a ground base station network, characterized in that: The steps include: Obtain satellite and ground data; allocate spectrum through federated learning: Obtain satellite and ground data and perform standardization; Analyze the satellite and ground data to obtain spectrum state characteristics; The central server uses the spectrum state characteristics as input to the spectrum allocation model to obtain the initial spectrum allocation results; The edge nodes composed of satellites and ground base stations use locally collected data as input, calculate the error between the initial spectrum allocation results and the actual allocation results, and update the model parameters of the spectrum allocation model using the gradient descent algorithm to generate gradient increments; The central server uses the FedAvg algorithm to aggregate gradient increments, generate globally updated model parameters, and broadcast them to all edge nodes for iteration. Finally, the edge nodes generate spectrum allocation results based on local constraints. Based on satellite data, the coordinates of the blind spot and the surviving base station are located, and the beam direction and width are adjusted in combination with MIMO technology: Obtain blind spot coordinates, elevation data and satellite attitude data; The blind spot coordinates are converted into geocentric rectangular coordinates based on the curvature radius of the circle, and the angle between the position vector of the geocentric rectangular coordinates and the horizontal position relative to the satellite is calculated. Then convert the blind spot coordinates into azimuth and elevation angles in the satellite coordinate system; Based on the geometric model of the phased array antenna, calculate the phase offset of each element of the phased array antenna array; Receive the pilot signal from the user terminal and calculate the deviation between the actual beam center and the target area; The gradient descent method is used to adjust the phase offset to minimize the standard deviation of the signal intensity at the edge of the target area; Minimum mean square error (MSS) precoding is used to obtain a precoding matrix. Combined with beam pointing constraints, azimuth and elevation control parameters are embedded in the precoding matrix to adjust the beam direction. The target area is calculated through satellite remote sensing and the user density is counted. When the target area is smaller than a preset area threshold and the user density is greater than a preset density threshold, the N×N array elements in the central area are activated to increase the equivalent aperture and narrow the half-power beamwidth to a preset narrow angle range. Otherwise, all array elements are activated, introducing random phase perturbations within a preset phase range and expanding the half-power beamwidth to a preset wide angle range. Select the surviving base station as the relay base station; Obtain channel status and dynamically adapt modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values; Based on ground data, spectrum allocation results, and link delays from dynamic switching strategies, a comprehensive link weight model is constructed. An improved Dijkstra algorithm that maintains a heap list of K shortest paths is used to generate the optimal path. Based on satellite data and ground data, combined with the actual load of the relay base station, a hybrid power supply strategy is used to power the relay base station.

2. The emergency communication method integrating satellite network and ground base station network according to claim 1, characterized in that: Methods for obtaining spectrum state characteristics include: Counting the ratio of the duration of a preset frequency band occupied by a signal within a preset time to the total duration in the satellite data to obtain the frequency band occupancy rate; performing Fourier transform on the received signal to obtain a frequency domain signal, and obtaining the frequency component distribution of the frequency domain signal within the preset frequency band; Obtaining the power spectral density of the interference signal in the satellite data within a preset frequency band, and integrating the power spectral density within the preset frequency band to obtain the interference signal power; Count the actual number of users accessing the ground data within a preset time, and calculate the load rate based on the maximum number of users accessing the base station. The frequency band occupancy, frequency component distribution, interference signal power and load rate are spliced ​​together to obtain the spectrum status characteristics.

3. The emergency communication method integrating satellite network and ground base station network according to claim 1, characterized in that: Methods for locating blind spot coordinates include: Calculate the distance between the base station location coordinates and the coordinates to be judged. Combined with the signal strength, if the distance between the coordinates to be judged and the base station is less than the maximum effective distance, but the corresponding signal strength is lower than the signal strength threshold, then preliminarily determine that the coordinates to be judged are a communication blind spot. Traverse all coordinates to be judged, obtain the communication blind spot set, and repeatedly judge the communication blind spot set to obtain the blind spot coordinates.

4. The emergency communication method integrating satellite network and ground base station network according to claim 1, characterized in that: The method of dynamically adapting modulation and coding through a dynamic switching strategy includes: Calculate the net signal power by the receiving end signal strength and the receiving antenna gain; calculate the noise power by combining the noise power spectrum density and bandwidth; and obtain the channel signal-to-noise ratio based on the net signal power and noise power. The rain attenuation value is calculated based on the rainfall rate; the coding rate is selected based on the channel signal-to-noise ratio through a three-level coding redundancy strategy: The beam gain value is calculated based on the antenna efficiency, antenna aperture area and operating wavelength; According to the channel signal-to-noise ratio, coding rate and beam gain value, the corresponding modulation mode is matched based on the dynamic matching model; Collect Ka-band and L-band dual-band signals, use dual-band signal merging technology to merge the signals, and calculate the merged signal based on the channel signal-to-noise ratio and the received signal of the corresponding frequency band; Dynamically adjust the signal transmission power according to the rain attenuation value.

5. The emergency communication method integrating satellite network and ground base station network according to claim 4, characterized in that: The method of selecting the coding rate through the three-level coding redundancy strategy includes: The coding rate is allocated based on a preset redundancy strategy according to the rain attenuation value and the channel signal-to-noise ratio (SNR). The preset redundancy strategy includes: when the rain attenuation value is less than a preset low rain attenuation threshold and the channel signal-to-noise ratio (SNR) is greater than a preset high channel signal-to-noise ratio (SNR) threshold, the coding rate is a first-level coding redundancy value; when the rain attenuation value is not less than a preset low rain attenuation threshold and less than a preset high rain attenuation threshold, or when the channel signal-to-noise ratio (SNR) is not less than a preset low channel signal-to-noise ratio (SNR) threshold and less than a preset high channel signal-to-noise ratio (SNR) threshold, the coding rate is a second-level coding redundancy value; when the rain attenuation value is not less than a preset high rain attenuation threshold and the channel signal-to-noise ratio (SNR) is less than a preset low channel signal-to-noise ratio (SNR) threshold, the coding rate is a third-level coding redundancy value; and when the channel signal-to-noise ratio (SNR) exceeds the high channel signal-to-noise ratio (SNR) threshold for W consecutive times, the coding rate is switched to a preset backup coding rate value.

6. The emergency communication method integrating satellite network and ground base station network according to claim 4, characterized in that: The method for matching the corresponding modulation mode according to the dynamic matching model includes: When the channel signal-to-noise ratio is greater than the first receiving signal-to-noise ratio, 64QAM modulation is used, and the first receiving signal-to-noise ratio is calculated by combining the preset theoretical first receiving signal-to-noise ratio and the beam gain; When the channel signal-to-noise ratio is not less than the second receiving signal-to-noise ratio and not greater than the first receiving signal-to-noise ratio, 16QAM modulation is adopted, and the second receiving signal-to-noise ratio is calculated by combining the preset theoretical second receiving signal-to-noise ratio and the beam gain; When the channel signal-to-noise ratio is not less than the third receiving signal-to-noise ratio and less than the second receiving signal-to-noise ratio, QPSK modulation is adopted, and the third receiving signal-to-noise ratio is obtained by calculating the preset theoretical third receiving signal-to-noise ratio and the beam gain; When the channel signal-to-noise ratio is less than the third receiving signal-to-noise ratio, the BPSK modulation mode is adopted; When the modulation mode is m, the preset hysteresis margin is superimposed on the preset minimum signal-to-noise ratio threshold corresponding to the modulation mode m to obtain an updated signal-to-noise ratio threshold. When the channel signal-to-noise ratio exceeds the updated signal-to-noise ratio threshold, the modulation mode is switched, where m is 64QAM, 16QAM, QPSK, and BPSK.

7. The emergency communication method integrating satellite network and ground base station network according to claim 1, characterized in that: Methods for generating optimal paths include: A network topology diagram consisting of satellites, ground base stations, and user terminals is constructed. A composite weight function is constructed based on link latency, bandwidth, and power outage risk to calculate the initial weight of each link. Link latency is calculated based on transmission delay, propagation delay, processing delay, and retransmission delay. Power outage risk is calculated based on remaining power, preset critical power, total average power consumption, and total power consumption. Each node deploys a heartbeat packet sending module to send heartbeat packets to adjacent nodes at a preset time. At the same time, it regularly evaluates the optimal path and triggers the optimal path recalculation when a preset emergency occurs. Take the satellite as the source node and the user terminal as the target node. Select the node u closest to the source node from the unvisited nodes and traverse its neighbor nodes v. If the new distance from u to v is shorter, update the distance and record the predecessor node. During the iteration process, a heap list containing K shortest paths is maintained. When the weight of the new path is less than the maximum weight of the heap list, it is replaced and updated until all nodes are visited. The path with the smallest weight is taken as the optimal path.

8. The emergency communication method integrating satellite network and ground base station network according to claim 1, characterized in that: Methods for powering relay base stations using a hybrid power supply strategy include: If the solar output power P is not lower than the sum of the device's real-time power consumption Q and the battery charging power, the solar power is triggered to directly power the load, and the remaining power is used to charge the battery; If P is lower than Q or the remaining battery power E is greater than the preset critical power, the solar energy and battery combination is triggered to power the load; the preset critical power is calculated based on Q and the preset minimum guarantee time; If P is lower than the preset power threshold or E is not greater than the preset critical power, the battery will be used for power supply only. When E reaches the minimum protection power, a power-off warning will be triggered. The waste heat generated by the device's heat dissipation is recovered through the thermoelectric power generation sheet, and the recovered power is added to the battery charging circuit. The recovered power is calculated through the temperature difference efficiency, device temperature, ambient temperature and heat dissipation area.

9. An emergency communication system integrating a satellite network and a ground base station network, implementing the emergency communication method integrating a satellite network and a ground base station network as claimed in any one of claims 1 to 8, characterized in that: include: Spectrum allocation module: obtains satellite data and ground data; allocates spectrum through federated learning; Beam adjustment module: locates the blind spot coordinates and surviving base station coordinates based on satellite data, and adjusts the beam direction and width in combination with MIMO technology; Select the surviving base station as the relay base station; Coding and modulation module: obtains channel status and dynamically adapts modulation and coding through dynamic switching strategies based on spectrum allocation results and beam gain values; Path planning module: Based on ground data, spectrum allocation results, and link delay of dynamic switching strategies, a comprehensive link weight model is constructed and the improved Dijkstra algorithm is used to generate the optimal path; Intelligent power supply module: Based on satellite data and ground data, combined with the actual load of the relay base station, it supplies power to the relay base station through a hybrid power supply strategy.

Images