A multi-data fusion-based integrated sensing and communication coordination method for low-altitude airspace

Abstract

This invention discloses a multi-data fusion-based integrated sensing and communication coordination method for low-altitude airspace perception, belonging to the field of multi-data fusion technology. The method includes the following steps: design and generation of sensing fusion signals; acquisition and processing of target sensing signals; processing and information extraction of communication signals; collaborative decision-making and optimization of communication and sensing; differentiated service delivery and closed-loop operation and maintenance. This invention adds another functional module to the existing sensing and communication systems, differing from current integrated sensing technologies that mostly adopt a sensing-centric or communication-centric design approach. Through an incremental overlay scheme, sensing and communication functions share the same radio frequency front-end and signal processing platform, significantly reducing the number and cost of hardware devices, simplifying system deployment, and providing reliable communication support to facilitate the development of various low-altitude services, thus promoting the development of the low-altitude economy.

Description

Technical Field

[0001] This invention relates to the field of multi-data fusion, specifically to a sensor-integrated low-altitude airspace perception and communication collaboration method based on multi-data fusion. Background Technology

[0002] Sensing and communication integration technology is an important development direction in the low-altitude economy. It deeply integrates communication and sensing functions, providing strong support for the development of the low-altitude economy. This technology forms a collaborative network system by integrating key infrastructure such as communication base stations, satellite communication and positioning, and drones, providing seamless communication and high-precision sensing services. In the low-altitude economy, sensing and communication integration technology is mainly used for low-altitude security, drone supervision, etc., providing comprehensive support for applications in the low-altitude economic field. At present, the low-altitude airspace (usually referring to 0-1000 meters) is becoming the core space for new economic forms such as logistics distribution, urban air traffic (UAM), and emergency rescue. With the rapid development of the low-altitude economy, especially the explosive growth of drone application scenarios, higher requirements are placed on low-altitude airspace sensing and communication systems. The surge in the number of drones has also made low-altitude air routes increasingly congested, placing more stringent standards on target detection accuracy, positioning real-time performance, and communication reliability. However, traditional single sensing and communication technologies can no longer meet the current needs of low-altitude operations. Furthermore, the traditional model adopts independent design architectures for sensing and communication, with sensing and communication devices implementing independent functions. This leads to competition between the sensing and communication systems in terms of resource application and processing. At the same time, the independently deployed sensing and communication systems will also interfere with each other, affecting the effective performance of each. Summary of the Invention

[0003] This invention provides a multi-data fusion-based integrated sensing and communication collaborative method for low-altitude airspace perception. It can effectively solve the problems mentioned in the background that traditional single sensing and communication technologies can no longer meet the current low-altitude service requirements. Furthermore, the traditional model adopts independent design architectures for sensing and communication, with sensing and communication devices implementing independent functions. This leads to competition between the sensing and communication systems in terms of resource application and processing. At the same time, the independently deployed sensing and communication systems also interfere with each other, affecting the effective performance of each.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a multi-data fusion-based integrated low-altitude airspace perception and communication coordination method, which achieves integrated collaborative application by deeply integrating communication and perception functions, including the following steps: Step S1, Design and generation of synesthetic fusion signals; Step S2: Acquisition and processing of target perception signals; Step S3: Processing of communication signals and extraction of information; Step S4: Communication and sensing collaborative decision-making and optimization; Step S5: Differentiated service delivery and closed-loop operation and maintenance.

[0005] According to the above technical solution, in step S1, the electromagnetic signals in the target airspace are comprehensively monitored by a wide-spectrum scanning receiver to obtain key parameters of the area and real-time meteorological information. In addition, airspace management information needs to be collected, and the collected wide-spectrum data is processed to identify different types of signal sources. The situation assessment results are output in a standardized data format. Based on the situation assessment results, the parameters of the sensing fusion waveform are designed and optimized. The waveform parameter design requires first determining the carrier center frequency and signal bandwidth. The waveform parameter optimization adopts a multi-objective optimization algorithm, with communication capacity and sensing accuracy as optimization objectives, and hardware limitations, power constraints, and spectrum regulations as optimization constraints. The Pareto optimal combination of waveform parameters is found through iterative search. To further improve the synergistic optimization effect of communication performance and sensing performance, the joint optimization objective function of communication and sensing is constructed as follows: in, Represents the set of waveform parameters for synesthetic fusion. Represents the communication capacity function. Represents the perception error function. This represents the cross-interference term between communication and sensing. These are the weighting coefficients; To achieve adaptive adjustment under different application scenarios, the weight coefficients adopt a dynamic update mechanism, and their expression is as follows: in, Indicates the first Environmental state quantities corresponding to class performance indicators These are adjustment parameters used to control the sensitivity to weight changes.

[0006] According to the above technical solution, in step S1, after completing the waveform parameter design, the generation of the fused waveform and the configuration of the radio frequency front end are carried out. First, the communication baseband processor generates the communication data stream to be transmitted. After channel coding and modulation mapping, a baseband communication signal is formed. At the same time, the sensing and detection unit generates a broadband frequency modulation signal as a carrier. Then, the communication signal is modulated onto the sensing carrier through a waveform superposition algorithm to form the final fused waveform. The RF front-end configuration includes transmit power setting, antenna array pointing adjustment, and beamforming weight calculation. Once configured, the RF front-end can start transmitting inductive fusion signals and simultaneously receive target echoes and communication signals. Synesthesia fusion signal can be represented as: in, Represents the communication sub-signal, Indicates the sensing sub-signal. and For dynamic weighting coefficients, This is the phase adjustment term.

[0007] According to the above technical solution, in step S2, the reflected echo signal from the low-altitude target and the direct signal from the communication terminal are collected simultaneously by the receiving antenna array. The received signal contains the target reflected echo, the communication signal transmitted by the communication terminal, various interference signals and noise. The collected signal needs to be processed. A two-stage separation scheme based on code domain separation and time-frequency analysis is adopted to separate the pure target sensing echo signal from the mixed received signal. Then, the signal-to-noise ratio is improved by pulse coherent accumulation, and the time delay and Doppler frequency shift of the echo are accurately obtained.

[0008] According to the above technical solution, S2 performs target parameter estimation and detection and identification based on the extracted target echo signal. The target parameter estimation includes three dimensions: distance estimation, velocity estimation, and angle estimation. The target detection uses a constant false alarm rate (CFAR) detection algorithm to detect the presence of targets while maintaining a preset false alarm probability. Detected targets need to be individually identified. For UAV targets, the type and size of the UAV are determined by analyzing the micro-Doppler modulation features generated by its rotor blades. The detection and identification results are output in the form of a target state vector. To improve the separation accuracy of communication signals and sensing echo signals, a joint sparse optimization model is constructed as follows: in, In order to receive signals, and These are feature dictionaries for communication signals and sensing signals, respectively. and represents the corresponding sparse representation coefficients.

[0009] According to the above technical solution, in step S3, while completing the separation of sensing signals, communication signals are separated from the mixed signals and demodulated. Pilot-assisted channel estimation and interference cancellation techniques are used to separate the communication signals. The separated communication signals are demodulated after channel estimation. It is also necessary to evaluate the performance quality of the communication link and optimize the link adaptation based on the evaluation results. The communication performance status under the current channel conditions can be obtained by analyzing the signal indicators at the receiving end.

[0010] According to the above technical solution, in step S4, the sensing information from multiple integrated sensing base stations is fused to form a continuous tracking trajectory for low-altitude targets. Each base station first completes target detection and parameter estimation locally to obtain a local target state estimate. Then, the local estimation results are uploaded to the fusion center and globally fused through a distributed fusion algorithm. The fusion center integrates the observation information and eliminates measurement redundancy.

[0011] According to the above technical solution, S4 improves overall performance through distributed cooperation, including two aspects: cooperative perception and cooperative communication. Cooperative perception improves perception performance by having multiple base stations jointly observe the same target and utilizing spatial diversity and geometric diversity. Cooperative communication improves perception performance by having multiple base stations cooperate in transmission and reception. Simultaneously, based on performance monitoring results and service change trends, network parameters and working modes are dynamically adjusted to achieve network optimization. Optimization includes cell parameter adjustment, load balancing, and interference coordination. The result of multi-base station fusion can be expressed as: in, Indicates the first Target state estimation results for each base station This indicates the uncertainty of the corresponding estimate.

[0012] According to the above technical solution, in step S5, the processed perception information is packaged into standardized service products and distributed to various terminals. The perception information service includes multiple product forms, including real-time target situation products, target detailed information products, historical data analysis products, and early warning products. The service distribution adopts a service-oriented architecture, and various types of sensing information are provided to the outside world in the form of services. Application terminals subscribe to and obtain sensing services through standard interfaces, and have functions such as user authentication, permission management, and service billing. Quality assurance is achieved by configuring different QoS parameters for different types of services, including priority, maximum latency, minimum bandwidth, and bit error rate threshold. This enables the scheduler to make scheduling decisions based on the QoS parameters of the services, allowing high-priority services to obtain resources first.

[0013] According to the above technical solution, S5 also requires continuous operation and maintenance management and performance optimization of the integrated sensing system, including equipment status monitoring, fault alarm handling, software upgrade and update, security policy execution, and performance optimization based on the analysis results of operating data, continuously improving system parameters and working algorithms; To achieve adaptive optimization of system operation, a reinforcement learning update mechanism is introduced, the expression of which is: in, Indicates the system status. Indicates the scheduling strategy. Represents the reward function, For learning rate, Discount factor; The relevant parameters in the above model can be dynamically updated using historical data and online operational data to adapt to changes in the low-altitude environment and business needs.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention adds another functional module to the existing sensing and communication systems, which differs from the current integrated sensing technology that mostly adopts a sensing-centric or communication-centric design approach. Through an incremental overlay scheme, the sensing and communication functions share the same radio frequency front-end and signal processing platform, significantly reducing the number and cost of hardware devices and simplifying system deployment. In terms of spectrum resource utilization, the sensing and communication signals share the same frequency band, and the effective coexistence of the two signals is achieved through precise waveform design, which significantly improves spectrum efficiency. In terms of response speed, the deep integration of sensing and communication functions can be used for communication link optimization in real time when target information is detected, without the need for additional information interaction delay. In addition, by optimizing resource allocation and task scheduling globally under a unified framework, the best balance between sensing performance and communication performance can be achieved, providing strong technical support for the safe management and efficient utilization of low-altitude airspace. Furthermore, through high-precision target perception capabilities, it is also possible to effectively monitor low-altitude aircraft, promptly detect and handle anomalies, and at the same time, through reliable communication support capabilities, facilitate the development of various low-altitude services and promote the development of the low-altitude economy.

[0015] Furthermore, in the process of designing the fusion waveform, this invention introduces a joint optimization mechanism based on adaptive weight adjustment, which dynamically adjusts the weight coefficients of communication performance and sensing performance according to the real-time environmental perception results and communication service requirements, thereby achieving adaptive performance balance under different application scenarios. Meanwhile, a sensing signal mapping method based on time-frequency resource block is proposed, which embeds communication data symbols and sensing detection signals in a structured manner under a unified resource framework. By dynamically allocating resource units, low-interference coordination between communication and sensing in the same waveform is achieved. In addition, an online optimization mechanism based on feedback closed loop is introduced. By continuously collecting sensing errors and communication link quality indicators, waveform parameters and resource scheduling strategies are corrected in real time, thereby improving the robustness and adaptability of the system in complex low-altitude environments. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0017] In the attached diagram: Figure 1 This is a flowchart of the steps of the collaborative method of the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0019] Example: Figure 1 As shown, this invention provides a technical solution: a multi-data fusion-based integrated low-altitude airspace perception and communication coordination method. This method achieves integrated collaborative applications by deeply integrating communication and perception functions, and includes the following steps: Step S1, Design and generation of synesthetic fusion signals; Step S2: Acquisition and processing of target perception signals; Step S3: Processing of communication signals and extraction of information; Step S4: Communication and sensing collaborative decision-making and optimization; Step S5: Differentiated service delivery and closed-loop operation and maintenance.

[0020] Based on the above technical solution, S1, a wide-spectrum scanning receiver is used to comprehensively monitor electromagnetic signals in the target airspace and obtain key parameters of the area, including spectrum occupancy, interference signal distribution, and noise floor level. At the same time, real-time meteorological information, including wind speed and direction, temperature and air pressure, and precipitation, is obtained by connecting to a meteorological data interface. In addition, airspace management information needs to be collected. The collected wide-spectrum data is processed using a deep learning-based spectrum analysis algorithm to identify different types of signal sources. The situation assessment results are output in a standardized data format. Furthermore, in the multi-objective optimization process, a dynamic weight adjustment strategy is introduced, where the weight of the communication performance index is α and the weight of the perception performance index is β, where α+β=1, and the weight is adaptively adjusted according to real-time business requirements and environmental complexity. When the communication service load is high, the α weight is increased to prioritize communication capacity; when the target density is high or the environmental complexity increases, the β weight is increased to enhance perception accuracy. The optimization process involves constructing a joint objective function and then pruning the solution space based on constraints to obtain the optimal set of waveform parameters that meet both communication and sensing performance requirements.

[0021] Based on the situation assessment results, the parameters of the sensing fusion waveform are designed and optimized. The waveform parameter design requires first determining the carrier center frequency and signal bandwidth. The waveform parameter optimization adopts a multi-objective optimization algorithm, with communication capacity and sensing accuracy as optimization objectives, and hardware limitations, power constraints, and spectrum regulations as optimization constraints. The optimization algorithm adopts a genetic algorithm to find the Pareto optimal combination of waveform parameters through iterative search. To further improve the synergistic optimization effect of communication performance and sensing performance, the joint optimization objective function of communication and sensing is constructed as follows: in, Represents the set of waveform parameters for synesthetic fusion. Represents the communication capacity function. Represents the perception error function. This represents the cross-interference term between communication and sensing. The weighting coefficients are obtained by dynamic normalization to satisfy... ; in: To achieve adaptive adjustment under different application scenarios, the weight coefficients adopt a dynamic update mechanism, and their expression is as follows: Where i = 1, 2, 3 Indicates the first The environmental state quantities corresponding to class performance indicators can be quantified as follows: ; These are adjustment parameters used to control the sensitivity to weight changes.

[0022] Based on the above technical solution, S1, after completing the waveform parameter design, the generation of the fused waveform and the configuration of the radio frequency front end are carried out. First, the communication baseband processor generates the communication data stream to be transmitted. After channel coding and modulation mapping, the baseband communication signal is formed. At the same time, the sensing and detection unit generates a broadband frequency modulation signal as a carrier. Then, the communication signal is modulated onto the sensing carrier through the waveform superposition algorithm to form the final fused waveform. The RF front-end configuration includes transmit power setting, antenna array pointing adjustment, and beamforming weight calculation. The transmit power setting needs to achieve a balance between sensing and detection range and communication coverage. The antenna array adopts massive MIMO technology, and the beamforming weight is calculated through beam management algorithm. After the RF front-end is configured, it can start transmitting the fusion signal and simultaneously receive target echo and communication signal. Synesthesia fusion signal can be represented as: in, Represents the communication sub-signal, Indicates the sensing sub-signal. and For dynamic weighting coefficients, This is a phase adjustment term; in: Based on the above technical solution, S2, the reflected echo signal from the low-altitude target and the direct signal from the communication terminal are collected synchronously by the receiving antenna array. The received signal contains the target reflected echo, the communication signal transmitted by the communication terminal, various interference signals and noise. The collected signal needs to be amplified, denoised and converted. It is pre-amplified by the amplifier to remove out-of-band interference and noise, and the analog signal is converted into a digital signal. A two-stage separation scheme combining code domain separation and time-frequency analysis is adopted to separate the pure target sensing echo signal from the mixed received signal. Code domain separation uses pilot pattern differences for initial separation. Time-frequency analysis extracts the target echo signal from the initially separated signal through short-time Fourier transform. Then, pulse coherent accumulation is used to improve the signal-to-noise ratio and accurately obtain the echo's time delay and Doppler frequency shift.

[0023] Based on the above technical solution, S2, target parameter estimation and detection and identification are performed based on the extracted target echo signal. The target parameter estimation includes three dimensions: distance estimation, velocity estimation, and angle estimation. Distance estimation is achieved by accurately measuring the echo delay, velocity estimation is achieved by measuring the Doppler frequency offset of the echo signal, and angle estimation is achieved by beamforming technology of the receiving antenna array. The target detection uses a constant false alarm rate (CFAR) detection algorithm to detect the presence of targets while maintaining a preset false alarm probability. Detected targets need to be individually identified. For UAV targets, the type and size of the UAV are determined by analyzing the micro-Doppler modulation features generated by its rotor blades. The detection and identification results are output in the form of a target state vector, including the target's position, speed, heading, and identification information.

[0024] To improve the separation accuracy of communication signals and sensing echo signals, a joint sparse optimization model is constructed as follows: in, In order to receive signals, and These are feature dictionaries for communication signals and sensing signals, respectively, which can be constructed using training samples or prior templates. and These are the corresponding sparse representation coefficients; , For sparse regularization coefficients, the solution methods can be OMP (Orthogonal Matching Pursuit), LASSO, or iterative reconstruction algorithms.

[0025] In the signal separation process, a signal decoupling model based on joint sparse representation is constructed, which represents the mixed signal as a sparse combination of communication signal components and sensing echo components. By introducing a dictionary learning mechanism to model different signal features, high-precision separation of communication signals and sensing echoes is achieved. Simultaneously, by combining iterative reconstruction algorithms, the separation results are gradually optimized, thereby effectively suppressing mutual interference between signals and improving the detection capability of weak targets.

[0026] Based on the above technical solution, S3, while completing the separation of sensing signals, the communication signal is separated from the mixed signal and demodulated. Pilot-assisted channel estimation and interference cancellation technology are used to separate the communication signal. The separated communication signal is demodulated after channel estimation. Channel estimation uses pilot symbols to obtain the frequency response estimate of the channel. The channel response within the entire signal bandwidth is obtained through an interpolation algorithm. Channel equalization compensates the received signal based on the channel estimation results to offset the frequency selective fading caused by multipath propagation. Demodulation uses a demodulation algorithm corresponding to the modulation method of the transmitter to map the received symbols back to the original information bits. The demodulated bit stream is then decoded by the channel to correct the errors introduced during transmission and finally recover the original data sent by the transmitter. It is also necessary to evaluate the performance quality of the communication link and optimize the link adaptation based on the evaluation results. The indicators for communication quality evaluation include signal strength, signal-to-noise ratio, error vector amplitude, bit error rate, and throughput. By analyzing the indicators of the signal at the receiving end, the communication performance status under the current channel conditions can be obtained. Based on the communication quality assessment results, adaptive link adaptation is performed to adjust communication parameters to optimize transmission performance. Through the adaptive mechanism, optimal communication performance is maintained at all times under changing channel conditions.

[0027] Based on the above technical solution, in S4, the sensing information from multiple integrated sensing base stations is fused to form a continuous tracking trajectory for low-altitude targets. Each base station first completes target detection and parameter estimation locally to obtain local target state estimation. Then, the local estimation results are uploaded to the fusion center and globally fused through a distributed fusion algorithm. The fusion center integrates observation information and eliminates measurement redundancy. The target tracking uses an algorithm based on filtering theory, most commonly Kalman filtering and its variants. Track management is also required during the tracking process, including track initialization, track maintenance, track termination, and track splitting and merging. The target trajectory information in the fused output includes the current position and velocity estimate. Based on the requirements of sensing tasks and the load of communication services, the system resources are dynamically allocated and tasks are scheduled. The system resources include computing resources, spectrum resources, power resources, and antenna resources. The optimization objective of resource allocation can be set as maximizing system utility and minimizing the weighted sum of sensing and communication performance. Task scheduling is responsible for coordinating the execution sequence of various sensing and communication tasks. Low-altitude airspace sensing and communication collaboration involves multiple types of tasks, including periodic sensing and detection tasks, event-triggered sensing tasks, real-time communication transmission tasks, and batch data download tasks.

[0028] Based on the above technical solution, S4 improves overall performance through distributed cooperation, including two aspects: cooperative sensing and cooperative communication. Cooperative sensing is achieved by multiple base stations jointly observing the same target, using spatial diversity and geometric diversity to improve sensing performance. Cooperative communication is achieved by multiple base stations cooperating in transmission and reception. Cooperative transmission is achieved by multiple base stations simultaneously sending information to the same user, using transmit diversity to improve the receive signal-to-noise ratio. Cooperative reception is achieved by multiple base stations simultaneously receiving information sent by the user, using receive merging to improve signal quality. Simultaneously, based on system performance monitoring results and service change trends, network parameters and working modes are dynamically adjusted to achieve network optimization. The optimization includes cell parameter adjustment, load balancing, and interference coordination. Cell parameter adjustment includes transmission power and handover parameters. Load balancing distributes service traffic to multiple base stations. Interference coordination coordinates resource allocation between adjacent base stations to reduce mutual interference. Network optimization adopts a closed-loop control architecture. The performance monitoring module continuously collects network operation data, the optimization engine analyzes the data and generates optimization suggestions, which are then executed automatically or manually after approval.

[0029] In the process of multi-base station collaboration, an information fusion mechanism based on uncertainty assessment is introduced, which assigns confidence weights to the target state estimates uploaded by each base station, and adaptively fuses multi-source information by constructing a weighted fusion model. Meanwhile, during the fusion process, the observation angle, signal-to-noise ratio, and historical tracking stability of each base station are considered, and the fusion results are dynamically corrected to improve the continuity and accuracy of target tracking. The result of multi-base station fusion can be expressed as: in, Indicates the first Target state estimation results for each base station This indicates the uncertainty of the corresponding estimate.

[0030] Based on the above technical solution, S5 encapsulates the processed perception information into standardized service products and distributes them to various terminals. The perception information service includes various product forms, including real-time target situation products, target detailed information products, historical data analysis products, and early warning products. The real-time target situation product graphically displays the position, trajectory, and characteristic information of all currently detected low-altitude targets. The target details product provides complete attribute information for specific targets, including target type, size estimation, payload status, and historical trajectory analysis. The historical data analysis product stores and retrieves historical sensing data, supporting post-event analysis and evidence collection applications. The early warning and alert product sends alarm information to relevant personnel when abnormal targets and potential threats are detected. The service distribution adopts a service-oriented architecture, where various types of sensing information are provided to the outside world in the form of services. Application terminals subscribe to and obtain sensing services through standard interfaces, and have functions such as user authentication, permission management, and service billing to ensure the secure and controllable distribution of sensing information services. Communication service types include control command transmission, status information feedback, video image feedback, and batch data transmission. Quality assurance is achieved by configuring different QoS parameters for different types of services, including priority, maximum latency, minimum bandwidth, and bit error rate threshold. This enables the scheduler to make scheduling decisions based on the QoS parameters of the services, with high-priority services receiving resources first, ensuring that the performance requirements of critical services are met.

[0031] Based on the above technical solutions, S5 also needs to continuously manage and optimize the performance of the integrated sensing system to ensure long-term stable and efficient operation. This includes equipment status monitoring, fault alarm handling, software upgrades and updates, security policy execution, and performance optimization based on the analysis results of operating data to continuously improve system parameters and working algorithms. Equipment status monitoring collects data on the operating temperature, voltage, current, and radio frequency indicators of the base station hardware to keep track of the equipment's health status in real time. Fault alarm handling automatically triggers an alarm process when an anomaly is detected, notifying maintenance personnel to handle it. Software upgrades and updates adopt a gradual upgrade strategy, first upgrading and verifying on some nodes before being fully rolled out. Security policy execution includes user authentication, access control, and encrypted data transmission.

[0032] To achieve adaptive optimization of system operation, a reinforcement learning update mechanism is introduced, the expression of which is: in, Indicates the system status. Indicates the scheduling strategy. Represents the reward function, For learning rate, Discount factor; in: r= • Communication performance indicators+ •Perception performance indicators- • Resource consumption Communication performance metrics can include link quality, throughput, or signal-to-noise ratio, etc. Sensing performance indicators can include target detection accuracy, Doppler estimation error, or tracking stability, etc. Resource consumption can be measured in terms of power, spectrum usage, or computational resource consumption. These are weighting coefficients used to adjust the importance of each indicator in the reward function; For learning rate, Discount factor; reward function It can dynamically calculate based on the real-time system operating status to guide Q-learning updates, enabling the system to continuously optimize resource allocation and parameter configuration in dynamic low-altitude environments.

[0033] The relevant parameters in the above model can be dynamically updated using historical data and online operational data to adapt to changes in the low-altitude environment and business needs.

[0034] A closed-loop optimization model based on reinforcement learning is constructed, taking the system operating state, communication performance indicators, and perception error as state inputs. By learning the mapping relationship between different scheduling strategies and system performance, intelligent optimization of resource allocation and parameter configuration is achieved. Through online learning and strategy updates, the system continuously approaches the optimal operating state in the dynamically changing low-altitude environment.

[0035] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-data fusion-based integrated sensing and communication coordination method for low-altitude airspace perception, characterized in that: By deeply integrating communication and sensing functions, integrated collaborative applications can be achieved, including the following steps: Step S1, Design and generation of synesthetic fusion signals; Step S2: Acquisition and processing of target perception signals; Step S3: Processing of communication signals and extraction of information; Step S4: Communication and sensing collaborative decision-making and optimization; Step S5: Differentiated service delivery and closed-loop operation and maintenance.

2. The low-altitude airspace perception and communication cooperation method based on multi-data fusion and integrated with sensing and communication according to claim 1, characterized in that: S1 involves comprehensively monitoring electromagnetic signals within the target airspace using a wide-spectrum scanning receiver to obtain key parameters of the area, as well as real-time meteorological information. In addition, it is necessary to collect airspace management information, process the collected wide-spectrum data, identify different types of signal sources, and output the situation assessment results in a standardized data format. Based on the situation assessment results, the parameters of the sensing fusion waveform are designed and optimized. The waveform parameter design requires first determining the carrier center frequency and signal bandwidth. The waveform parameter optimization adopts a multi-objective optimization algorithm, with communication capacity and sensing accuracy as optimization objectives, and hardware limitations, power constraints, and spectrum regulations as optimization constraints. The Pareto optimal combination of waveform parameters is found through iterative search. To further improve the synergistic optimization effect of communication performance and sensing performance, the joint optimization objective function of communication and sensing is constructed as follows: in, Represents the set of waveform parameters for synesthetic fusion. Represents the communication capacity function. Represents the perception error function. This represents the cross-interference term between communication and sensing. These are the weighting coefficients; To achieve adaptive adjustment under different application scenarios, the weight coefficients adopt a dynamic update mechanism, and their expression is as follows: wherein, represents the environmental state quantity corresponding to the is a tuning parameter for controlling the sensitivity of the weight variation.

3. The method of claim 2, wherein the method is characterized in that: In step S1, after completing the waveform parameter design, the generation of the fused waveform and the configuration of the radio frequency front end are carried out. First, the communication baseband processor generates the communication data stream to be transmitted. After channel coding and modulation mapping, the baseband communication signal is formed. At the same time, the sensing and detection unit generates a broadband frequency modulation signal as a carrier. Then, the communication signal is modulated onto the sensing carrier through a waveform superposition algorithm to form the final fused waveform. The RF front-end configuration includes transmit power setting, antenna array pointing adjustment, and beamforming weight calculation. Once configured, the RF front-end can start transmitting inductive fusion signals and simultaneously receive target echoes and communication signals. Synesthesia fusion signal can be represented as: in, Represents the communication sub-signal, Indicates the sensing sub-signal. and For dynamic weighting coefficients, This is the phase adjustment term.

4. The method for integrated sensing and communication collaboration in low-altitude airspace based on multi-data fusion as described in claim 1, characterized in that: S2 synchronously acquires reflected echo signals from low-altitude targets and direct signals from communication terminals through a receiving antenna array. The received signals include target reflected echoes, communication signals transmitted by the communication terminal, various interference signals and noise, and the acquired signals need to be processed. A two-stage separation scheme based on code domain separation and time-frequency analysis is adopted to separate the pure target sensing echo signal from the mixed received signal. Then, the signal-to-noise ratio is improved by pulse coherent accumulation, and the time delay and Doppler frequency shift of the echo are accurately obtained.

5. The multi-data fusion based integrated sensing and communication method for low-altitude airspace awareness according to claim 4, characterized in that: S2, based on the extracted target echo signal, performs target parameter estimation and detection and identification. The target parameter estimation includes three dimensions: distance estimation, velocity estimation, and angle estimation. The target detection uses a constant false alarm rate (CFAR) detection algorithm to detect the presence of targets while maintaining a preset false alarm probability. Detected targets need to be individually identified. For UAV targets, the type and size of the UAV are determined by analyzing the micro-Doppler modulation features generated by its rotor blades. The detection and identification results are output in the form of a target state vector. To improve the separation accuracy of communication signals and sensing echo signals, a joint sparse optimization model is constructed as follows: wherein, is the received signal, and are feature dictionaries for the communication signal and the sensing signal, respectively, and are the corresponding sparse representation coefficients.

6. The multi-data fusion based integrated sensing and communication method for low-altitude airspace awareness according to claim 4, characterized in that: In step S3, while completing the separation of sensing signals, communication signals are separated from the mixed signals and demodulated. Pilot-assisted channel estimation and interference cancellation techniques are used to separate the communication signals. The separated communication signals are then demodulated after channel estimation. It is also necessary to evaluate the performance quality of the communication link and optimize the link adaptation based on the evaluation results. The communication performance status under the current channel conditions can be obtained by analyzing the signal indicators at the receiving end.

7. The multi-data fusion based integrated sensing and communication method for low-altitude airspace awareness according to claim 1, characterized in that: In step S4, the sensing information from multiple integrated sensing base stations is fused to form a continuous tracking trajectory for low-altitude targets. Each base station first completes target detection and parameter estimation locally to obtain a local target state estimate. Then, the local estimation results are uploaded to the fusion center and globally fused through a distributed fusion algorithm. The fusion center integrates the observation information and eliminates measurement redundancy.

8. The multi-data fusion based integrated sensing and communication method for low-altitude airspace awareness according to claim 7, characterized in that: The S4 improves overall performance through distributed cooperation, including two aspects: cooperative sensing and cooperative communication. Cooperative sensing improves sensing performance by having multiple base stations jointly observe the same target and by utilizing spatial diversity and geometric diversity. Cooperative communication improves sensing performance by having multiple base stations cooperate in transmission and reception. Simultaneously, based on performance monitoring results and service change trends, network parameters and working modes are dynamically adjusted to achieve network optimization. Optimization includes cell parameter adjustment, load balancing, and interference coordination. The result of multi-base station fusion can be expressed as: wherein, denotes the target state estimate of the denotes the uncertainty of the corresponding estimate.​ 9. The multi-data fusion based integrated sensing and communication method for low-altitude airspace awareness according to claim 1, characterized in that: S5 encapsulates the processed perception information into standardized service products and distributes them to various terminals. The perception information service includes various product forms, including real-time target situation products, target detailed information products, historical data analysis products, and early warning products. The service distribution adopts a service-oriented architecture, and various types of sensing information are provided to the outside world in the form of services. Application terminals subscribe to and obtain sensing services through standard interfaces, and have functions such as user authentication, permission management, and service billing. Quality assurance is achieved by configuring different QoS parameters for different types of services, including priority, maximum latency, minimum bandwidth, and bit error rate threshold. This enables the scheduler to make scheduling decisions based on the QoS parameters of the services, allowing high-priority services to obtain resources first.

10. The method for integrated sensing and communication collaboration in low-altitude airspace based on multi-data fusion according to claim 9, characterized in that: The S5 also requires continuous operation and maintenance management and performance optimization of the integrated sensing system, including equipment status monitoring, fault alarm handling, software upgrades and updates, and security policy execution. Performance optimization is based on the analysis results of operating data, and continuous improvement of system parameters and working algorithms. To achieve adaptive optimization of system operation, a reinforcement learning update mechanism is introduced, the expression of which is: in, Indicates the system status. Indicates the scheduling strategy. Represents the reward function, For learning rate, Discount factor; The relevant parameters in the above model can be dynamically updated using historical data and online operational data to adapt to changes in the low-altitude environment and business needs.

Images