Drone Swarm Cooperative Communication and Navigation Method Based on 5G Red Cap Module

Abstract

The present invention belongs to the field of communication technologies, and particularly relates to a method for collaborative communication and navigation of a drone swarm based on a 5G Red Cap module. The method includes: Step 1: Using a 5G Red Cap module to perform high-resolution sampling on the multipath spatio-temporal channels among the drones in the drone group, and constructing a link channel matrix; Step 2: Based on each link channel matrix, for each drone, constructing a joint channel matrix in combination with its neighbor set, and extracting an optimal beamforming vector; Step 3: According to the optimal beamforming vector and the link channel matrix, extracting the time delay, phase, and Doppler frequency shift, and estimating the position information of each drone; Step 4: According to the obtained position information and optimal beamforming vector of each drone, ensuring the collaborative optimization of navigation and communication. The present invention can effectively suppress interference, reduce energy consumption, and ensure the stability of the communication link and the reliability of navigation.

Description

Technical Field

[0001] The present invention belongs to the field of communication technologies, and particularly relates to a method for cooperative communication and navigation of a drone swarm based on a 5G Red Cap module. Background Art

[0002] With the rapid development and wide application of drone technologies, application scenarios such as cooperative combat of drone swarms, environmental monitoring, logistics transportation, and emergency rescue have attracted increasing attention. Traditional drone systems mostly rely on single drones to independently execute tasks and lack effective cooperative communication and navigation mechanisms. However, cooperative operation of drone swarms can achieve resource sharing, information interconnection, and task division of labor, thereby greatly improving the overall operation efficiency and flexibility. In recent years, with the rise of the fifth-generation mobile communication technology (5G), its characteristics of low latency, high data rate, and large connection number have provided new technical support for drone swarm communication. In particular, the 5G Red Cap (Reduced Capability) module has become an ideal choice for many low-cost drones due to its advantages of low power consumption and low complexity. However, although certain progress has been made in 5G communication technologies and drone navigation and positioning in the prior art, there are still many problems and deficiencies in practical applications.

[0003] Most existing drone swarm communication systems adopt traditional wireless communication technologies such as Wi-Fi and dedicated radio frequency band communication. Their transmission rate and anti-interference ability are limited, making it difficult to meet the requirements for real-time and stable data transmission in high-density and multi-node cooperative operations. In addition, since drones will encounter phenomena such as multipath effects, signal fading, and Doppler frequency shift during flight, traditional communication systems usually can only adopt omnidirectional antennas or fixed beamforming methods, making it difficult to achieve efficient focusing of signal energy and interference suppression in a dynamic environment. This causes problems such as communication interruption, data loss, or increased bit error rate in drone swarms during high-speed movement, formation transformation, and complex terrain environments, thereby affecting the task execution effect. Summary of the Invention

[0004] In view of this, the main objective of the present invention is to provide a method for cooperative communication and navigation of a drone swarm based on a 5G Red Cap module. By constructing a joint channel matrix through high-resolution channel sampling, and then adopting an optimal beamforming technology and a multi-dimensional wireless channel characteristic fusion positioning method, high-precision positioning and efficient communication of the drone swarm in a dynamic and multi-path environment are achieved. At the same time, the present invention organically combines flight trajectory planning with dynamic allocation of communication resources, enabling the drone swarm to adaptively adjust the beam direction, transmit power, and flight path, effectively suppressing interference, reducing energy consumption, and ensuring the stability of the communication link and the reliability of navigation, thereby greatly improving the task cooperation efficiency and the overall performance of the system, and having broad application prospects.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A method for collaborative communication and navigation of a drone swarm based on a 5G Red Cap module, the method comprising:

[0007] Step 1: Use a 5G Red Cap module to perform high-resolution sampling on the multipath spatio-temporal channels between each drone in the drone group, discretize the spatio-temporal channel data of each link, and construct a link channel matrix;

[0008] Step 2: Based on each link channel matrix, for each drone, construct a joint channel matrix in combination with its neighbor set, and extract the optimal beamforming vector;

[0009] Step 3: According to the optimal beamforming vector and the link channel matrix, extract the time delay, phase, and Doppler frequency shift, and in combination with the initial relative position relationship between the drones, estimate the position information of each drone by minimizing the residual function of the positioning error;

[0010] Step 4: According to the obtained position information and optimal beamforming vector of each drone, predict the effective channel gain of each link, construct an optimization problem, and simultaneously plan the flight trajectories and dynamic communication resource allocations of the drones to ensure the collaborative optimization of navigation and communication.

[0011] Further, in step 1, the drones in the drone group and the drone The wireless link between them is affected by multiple paths, each path having different attenuation degrees, phases, time delay Doppler frequency shifts, and the channel impulse response is represented by the following formula:

[0012] ;

[0013] Wherein, represents the number of effective paths between the drone and the drone , and each path corresponds to a signal component; is the amplitude fading coefficient of the th path, reflecting the attenuation degree of the signal when propagating on this path; is the time delay of the th path, indicating the time taken for the signal to pass through this path from the transmitter to the receiver; is the Doppler frequency shift of the th path; is the phase of the th path; is an integer subscript index; is the standard deviation of the time delays of all paths; is the sampling time for high-resolution sampling; is the delay variable; is the imaginary symbol.

[0014] Furthermore, in step 1, for each path, let the sampling time set be and the delay sampling set be , and through the following formula, discretize the spatio-temporal channel data of each link to construct the link channel matrix between the UAV and the UAV :

[0015] ;

[0016] where and are both integer subscript indices; is the number of sampling times; is the number of delay sampling points.

[0017] Furthermore, in step 2, by solving the following generalized eigenvalue problem, based on each link channel matrix, for each UAV, construct a joint channel matrix in combination with its neighbor set, and extract the optimal beamforming vector:

[0018] ;

[0019] where represents the self-link channel matrix of the UAV ; is the Hermitian conjugate operation of the matrix; is the identity matrix; is the eigenvalue in the generalized eigenvalue problem. When solving this generalized eigenvalue problem, select the eigenvector corresponding to the largest eigenvalue as the optimal beamforming vector; represents the neighbor set of the UAV ; the optimal beamforming vector of the UAV satisfies the following constraints: ; represents the received noise power of the UAV .

[0020] Furthermore, in step 3, according to the optimal beamforming vector and the link channel matrix, let the extracted delay be , the phase be , the Doppler frequency shift be , and in combination with the relative position relationship between UAVs, use the following formula to define the residual function that minimizes the positioning error :

[0021] ;

[0022] Among them, is the initial position of the UAV ; is the initial position of the UAV ; is the Doppler frequency shift variance; is the time delay variance; is the phase variance; is the velocity vector of the UAV ; is the velocity vector of the UAV ; is the transpose operation; is the Y-axis coordinate of the initial position of the UAV ; is the Y-axis coordinate of the initial position of the UAV ; is the X-axis coordinate of the initial position of the UAV ; is the X-axis coordinate of the initial position of the UAV ; is the carrier wavelength.

[0023] Furthermore, in step 3, through the following formula, using the residual function , the position information of each UAV is estimated:

[0024] ;

[0025] Among them, is the estimated position of the UAV ; is the estimated velocity vector of the UAV ;

[0026] Furthermore, in step 4, the effective link gain between the UAV and the UAV is predicted to be ; Among them, is the optimal beamforming vector of the UAV ;

[0027] Furthermore, in step 4, through the following formula, an optimization problem is constructed to simultaneously plan the flight trajectory of the UAV and the dynamic communication resource allocation, ensuring the collaborative optimization of navigation and communication:

[0028] ;

[0029] Among them, is the transmission power of the UAV ; is the target position of the UAV. By solving this optimization problem, it is calculated to obtain the corresponding to the minimum and ; is the L2 norm operation; The interference attenuation coefficient ranges from 2 to 5.

[0030] Furthermore, the transmission power of the UAV satisfies the following constraint conditions:

[0031] ;

[0032] ;

[0033] where is the preset signal-to-noise ratio threshold.

[0034] Adopting the above technical solutions, the present invention has the following beneficial effects: The present invention adopts a 5G Red Cap module, and through high-resolution sampling and fine discretization of the wireless channels among the UAV swarm, it realizes the comprehensive capture and accurate modeling of channel characteristics such as multipath propagation, time delay, phase, and Doppler frequency shift. This precise channel modeling method is not only difficult to achieve in traditional communication technologies, but also greatly improves the communication reliability and navigation accuracy of the UAV swarm in complex dynamic environments. Compared with the prior art, the present invention utilizes advanced joint channel matrix construction and beamforming algorithms, enabling each UAV to adaptively adjust the signal transmission and reception directions during flight, thereby achieving focused transmission of signal energy and interference suppression. Such a design effectively solves the problems of signal fading and link instability faced by traditional omnidirectional or fixed beam transmission methods in scenarios of high-speed flight, formation transformation, and severe environmental interference, providing a higher data transmission rate and lower latency for UAV swarm communication. In addition, the present invention constructs a residual function based on multiple wireless channel characteristics, combines multi-dimensional information such as signal propagation time delay, phase change, and Doppler frequency shift with the actual geometric positions and motion states among the UAVs, and realizes the minimization optimization of the relative positioning error of the UAV swarm. After adopting this method, even in the case of GPS signal occlusion or large complex environmental interference, the UAVs can accurately calculate the distances and motion states between each other through the high-precision wireless measurement data obtained by themselves, and thus complete high-precision relative positioning and navigation control. This not only significantly improves the cooperative operation ability during UAV formation flight, but also provides a solid data foundation and algorithm support for realizing UAV autonomous navigation, intelligent obstacle avoidance, and task cooperation. In terms of communication resource scheduling, the present invention designs a joint optimization scheme that closely combines the flight trajectory planning of the UAVs with dynamic communication resource allocation. By adjusting the target position and transmission power in real time during flight, each UAV can always maintain the best communication link state while ensuring the completion of tasks. This scheme not only fully considers the smoothness of the flight trajectory and the continuity of the path to prevent excessive energy consumption caused by sharp turns or frequent adjustments, but also takes into account the maximization of link gain and the minimization of adjacent interference, thereby realizing the collaborative optimization of communication and navigation within the swarm. The results show that this method can significantly reduce the overall energy consumption of the system, improve the stability of the communication link, and the reliability of data transmission. Brief Description of the Drawings

[0035] Figure 1 It is a schematic flowchart of the method for UAV swarm cooperative communication and navigation based on a 5G Red Cap module provided by an embodiment of the present invention. Detailed Embodiments

[0036] All features disclosed in this specification, or steps in all methods or processes disclosed, can be combined in any way, except for mutually exclusive features and / or steps.

[0037] Any feature disclosed in this specification (including any additional claims, abstract) can be replaced by other equivalent or similar-purpose alternative features, unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only an example of a series of equivalent or similar features.

[0038] Example 1: Refer to Figure 1 , a method for collaborative communication and navigation of a drone swarm based on a 5G Red Cap module, the method comprising:

[0039] Step 1: Use a 5G Red Cap module to perform high-resolution sampling on the multipath spatio-temporal channels between drones in a drone group, discretize the spatio-temporal channel data of each link, and construct a link channel matrix;

[0040] The 5G Red Cap module has an efficient channel state information (CSI) acquisition ability and can complete the high-resolution reconstruction of channel characteristics under the conditions of low power consumption and low complexity. In the communication environment of a drone swarm, due to the high-speed movement of drones and the complex spatial structure, the signal propagation is inevitably affected by the multipath effect, which causes the received signal to exhibit highly variable characteristics in time, space, and frequency domain. Therefore, it is difficult to accurately capture the dynamic evolution characteristics of the channel only relying on traditional single-channel modeling methods. The application of the 5G Red Cap module provides a channel measurement method that can adapt to the high-speed changing environment of the drone swarm. Based on advanced channel sounding and estimation algorithms, the system can perform fine-grained sampling of the multipath propagation characteristics between drones without adding too much computational burden. In practical applications, the high-resolution sampling of the multipath spatio-temporal channel by the 5G Red Cap module mainly relies on its built-in channel sounding mechanism, which includes the channel state information acquisition technology based on pilot signals. By transmitting specific pilot sequences, the receiving end can resolve the delay, fading characteristics, and spatial direction information of the channel. Since drones move in three-dimensional space, their channel models usually involve not only classical Rayleigh fading or Rice fading effects but also are affected by complex factors such as dynamic occlusion, Doppler frequency shift, and asymmetric path loss. Therefore, the simple static channel estimation method cannot meet the requirements of high-precision sampling. For this reason, the present invention utilizes the high-resolution sampling ability provided by the 5G Red Cap module, continuously samples the channel state in multiple time windows, and combines an adaptive filtering method to improve the accuracy of channel estimation, thereby obtaining more accurate channel state information. On this basis, in order to realize the mathematical modeling of the link channel between drones, it is necessary to discretize the sampled data, that is, numerically approximate the continuous change characteristics of the channel so that it can be expressed in the form of a matrix. Here, the discretization process is not just simply sampling and storing the data, but rather, on the premise of ensuring the integrity of channel information, reasonably reducing the dimension of the data to reduce the computational complexity and improve the stability of subsequent channel processing.

[0041] Step 2: Based on each link channel matrix, for each drone, construct a joint channel matrix in combination with its neighbor set, and extract the optimal beamforming vector;

[0042] In this step, each drone constructs a joint channel matrix based on the link channel matrix obtained in the previous step and in combination with its neighbor set, thereby comprehensively considering the characteristics of multiple links instead of only optimizing a single link. The advantage of this method is that traditional single-link optimization methods often fail to fully utilize the cooperation of neighboring drones, easily leading to beam conflicts, signal interference, or inefficient power allocation. In the design of the present invention, by constructing a joint channel matrix, the beamforming strategy of each drone can not only be optimized for its own best communication direction but also avoid interfering with the signals of neighboring drones, improving the communication efficiency and cooperation of the entire drone swarm. During the construction of the joint channel matrix, the system comprehensively considers the channel state information between all drones, including parameters such as the channel gain, phase shift, time delay information of each link, and the direction angles of neighboring drones, thereby establishing a complete spatial channel feature representation. In this process, the low power consumption and low complexity characteristics of the 5G Red Cap module enable real-time calculation, ensuring that even in the case of high-speed flight or a large-scale drone swarm environment, channel estimation and matrix operations can still be completed with a relatively low communication delay. Based on the constructed joint channel matrix, each drone needs to extract the optimal beamforming vector, and the core goal of this process is to find the best antenna weight configuration to maximize the signal energy in the desired direction and minimize it in the interference direction. The intelligent beam management capability of the 5G Red Cap module plays a key role in this process. By using the beam training and feedback mechanism in 5GNR (New Radio), the drone can dynamically adjust its beam direction and beam shape to adapt to the changing channel environment. During the dynamic formation process of the drone swarm, beamforming needs to be able to quickly respond to environmental changes. Therefore, the present invention combines a beam search algorithm based on machine learning. By analyzing historical channel data and real-time measurement data, it predicts the best beam direction at future moments, thereby reducing the computational overhead caused by exhaustive search in traditional methods and improving the real-time performance of beam adjustment. In addition, due to the low power consumption characteristics of the Red Cap module, these calculations can be efficiently executed on the embedded platform of the drone without significantly increasing the energy consumption of the system.

[0043] To further optimize the beamforming process, the present invention adopts a beam adaptive mechanism based on feedback closed-loop. That is, after initially calculating the beamforming vector, each drone will make fine adjustments to the beam direction according to the received signal quality indicators (such as signal strength, bit error rate, signal-to-interference-plus-noise ratio, etc.), thereby further optimizing the communication quality. This adaptive mechanism can effectively cope with channel changes in complex environments. For example, in an urban environment, the reflection and diffraction effects of buildings will cause the channel characteristics to change continuously, and when flying at high altitude, wind speed and attitude changes will also affect the accuracy of beam pointing. Through closed-loop feedback adjustment, it is possible to improve the accuracy and robustness of beamforming without increasing excessive computational complexity. In addition, since drone swarms often need to switch between multiple frequency bands and multiple antenna modes, the optimization of beamforming also needs to combine multi-carrier and multi-antenna information to ensure that the communication performance in different frequency bands can be maintained at the optimal state. At the final stage of this step, all drones will share their optimal beamforming vectors so that the entire drone swarm can form a consistent communication topology structure, thereby ensuring that any node within the group can efficiently exchange data. This method not only improves the overall communication ability of the drone swarm but also enhances the navigation accuracy because, based on accurate channel modeling and beamforming, each drone can obtain a more stable signal, thereby improving the accuracy of subsequent positioning algorithms. Compared with the traditional fixed beam configuration scheme, the beamforming optimization method of the present invention can more flexibly adapt to different task requirements. For example, when performing a target tracking task, the beam direction can be adjusted to enhance the signal coverage of the target area, and when performing a formation flight task, the beamforming can be optimized to reduce interference between drones. This efficient beamforming optimization scheme enables the present invention to maintain excellent communication performance and navigation accuracy in various complex environments.

[0044] Step 3: According to the optimal beamforming vector and the link channel matrix, extract the time delay, phase, and Doppler frequency shift, and combine the initial relative position relationship between the drones. By minimizing the residual function of the positioning error, estimate the position information of each drone;

[0045] After beamforming is completed, each drone will utilize the radio frequency signal processing capabilities of the Red Cap module to conduct a detailed analysis of the communication channels between itself and neighboring drones, and extract the delay characteristics in a multipath environment. The delay information reflects the propagation time experienced by the wireless signal from the transmitting drone to the receiving drone. Since the millimeter wave and ultra-wideband technologies used in 5G communication have extremely high time resolution, it is possible to obtain sub-nanosecond delay information through precise time measurements, thereby calculating the relative distance between drones. In addition, in a high-speed moving environment, the relative movement of drones will cause a frequency shift of the wireless signal, namely the Doppler frequency shift, which is directly related to the relative speed of the drones. Therefore, it can be used as an important parameter for estimating the dynamic displacement of drones. In the 5G Red Cap module, with advanced frequency estimation algorithms, it is possible to maintain extremely high frequency offset measurement accuracy even under low signal-to-noise ratio conditions, which enables drones in a dynamic environment to accurately perceive their relative motion states. In addition to delay and Doppler frequency shift, the phase information of the signal is also fully utilized. The extraction of phase information requires combining the coherent reception characteristics of the 5G Red Cap module. By analyzing the cumulative change of the signal phase, the accuracy of distance measurement can be further improved. The key challenge in phase measurement is that the multipath effect may cause unstable changes in the phase. Therefore, in the present invention, by utilizing the channel characteristics optimized by beamforming to maximize the energy of the main path signal, the stability of phase measurement is greatly improved. In addition, by modeling the time series of phase information, it is possible to effectively filter out phase mutations caused by instantaneous interference or channel fading, thereby obtaining a more stable phase measurement result.

[0046] After obtaining the delay, phase, and Doppler frequency shift information, the system constructs a residual function of the positioning error in combination with the initial relative position relationship of the UAVs. The goal of this residual function is to minimize the error between the measurement data and the actual geometric position. Since the propagation of wireless signals is affected by environmental factors such as reflection, diffraction, and occlusion, a single measurement value may be biased. Therefore, using the method of minimizing the error to fuse multiple measurement data can effectively improve the stability of positioning. Specifically, the method of the present invention uses an optimization algorithm to adjust the relative position estimation of the UAVs during continuous iteration, so that it gradually converges to the true position state. During the optimization process, the low-power consumption characteristics of the 5G RedCap module are particularly considered, and a lightweight computing framework is adopted, enabling the algorithm to run efficiently on the embedded computing unit of the UAV without additional reliance on high-performance ground computing resources. Through the above method, the present invention can not only achieve high-precision relative positioning of UAV swarms in an environment where GPS signals are limited, but also update the positions of UAVs in real time in a dynamic environment, improving the navigation accuracy of the entire UAV swarm. Compared with the traditional GPS / IMU fusion positioning method, the method of the present invention has stronger robustness in a dynamic environment and can provide more stable position information in a complex environment. In addition, since this method relies on the channel information extraction ability of the 5G Red Cap module and can make full use of the existing 5G infrastructure without additional dedicated positioning hardware, this solution has higher economy and feasibility in actual deployment.

[0047] Step 4: According to the obtained position information and optimal beamforming vectors of each UAV, predict the effective channel gains of each link, construct an optimization problem, and simultaneously plan the flight trajectories of the UAVs and dynamic communication resource allocation to ensure the collaborative optimization of navigation and communication.

[0048] In the first stage of this step, the system predicts the channel gains of each link by using the UAV position information and the optimal beamforming vector obtained in the previous step. Channel gain is an important indicator for measuring the quality of wireless communication. It determines the received signal strength, signal-to-noise ratio, and data transmission rate. Since the flight state of the UAV is constantly changing and the channel gain also shows dynamic changes, it is necessary to predict the channel gain at future moments by combining the wireless propagation model with historical channel measurement data. In the implementation of the present invention, the 5G Red Cap module provides high-precision channel state information, including characteristics such as path loss, multipath distribution, and shadow fading, enabling the channel gain prediction to be dynamically updated based on real measurement data rather than relying solely on static calculations of theoretical models. Through machine learning algorithms, the system can analyze the motion patterns of the UAV and the channel change trends, predict the channel gain conditions of each link in a short period of time, and thus provide accurate data support for subsequent optimization decisions. Based on the channel gain prediction results, the system constructs an optimization problem with the goal of maximizing communication quality and minimizing energy consumption, while taking into account the task requirements and environmental constraints of the UAV. The core variables of the optimization problem include the flight trajectory of the UAV and the allocation strategy of communication resources. Among them, the flight trajectory optimization aims to ensure that the UAV can always maintain good communication conditions during task execution and minimize power consumption as much as possible. For example, when the UAV swarm executes a cruise mission, the system will give priority to planning trajectories that can reduce frequent channel switching to reduce the risk of communication interruption caused by channel hopping. When executing a target tracking mission, it will give priority to considering paths that can ensure the stability of the link channel gain to ensure high-quality video or sensing data transmission. In addition, considering the energy-limited characteristics of the UAV, the optimization problem also needs to balance the relationship between communication quality and energy consumption to ensure that the UAV will not cause rapid power consumption due to frequent path adjustments. Therefore, the optimization goal not only focuses on maximizing communication performance but also takes into account the endurance ability of the UAV.

[0049] Another crucial aspect in this step is dynamic communication resource allocation. Since the UAV swarm shares limited wireless spectrum resources and the communication requirements and channel states of different UAVs vary, an intelligent resource management strategy must be adopted to ensure that each UAV can obtain sufficient bandwidth and power support. With the support of 5G Red Cap modules, the system can perceive the communication requirements of UAVs in real time and dynamically adjust the resource allocation scheme. For example, when the communication load is light, the system can reduce the transmission power to reduce energy consumption. During peak communication demand periods, it can intelligently adjust the beamforming strategy to improve spatial multiplexing efficiency and ensure that multiple UAVs can communicate effectively within the same frequency band. In addition, by jointly optimizing the trajectories and resource allocation of UAVs, the system can avoid a decline in communication quality caused by channel interference or resource conflicts. For example, when the communication interference is strong in a certain area, the system can adjust the flight path of the UAV to avoid the interference source, thus ensuring the stability of data transmission. Finally, the entire optimization process forms a closed loop. That is, during the continuous update of the UAV's position, the system will adjust the channel gain prediction, optimization calculation, and trajectory planning scheme in real time, enabling the UAV swarm to adapt to complex environments and achieve efficient communication and navigation co-optimization.

[0050] Embodiment 2: In step 1, the UAVs in the UAV group and the UAV The wireless link between them is affected by multiple paths, and each path has different attenuation levels, phases, time delays, and Doppler frequency shifts. The channel impulse response is expressed by the following formula:

[0051] ;

[0052] where represents the number of effective paths between UAV and UAV , and each path corresponds to a signal component; is the amplitude fading coefficient of the th path, reflecting the attenuation degree of the signal when propagating on this path; is the time delay of the th path, indicating the time taken for the signal to travel from the transmitter to the receiver through this path; is the Doppler frequency shift of the th path; is the phase of the th path; is an integer subscript index; is the standard deviation of the time delays of all paths; is the sampling moment of high-resolution sampling; is a delay variable; is the imaginary unit symbol.

[0053] Specifically, in this model, the wireless channels between drones are affected by effective paths, and the signal strength of each path is weighted by to characterize the attenuation degree of different paths. Since wireless signals encounter various obstacles during propagation in the air, such as buildings, trees, or other drones, these obstacles cause the signals to propagate through multiple reflection or scattering paths. Therefore, the fading on each path is different, affecting the synthesis of the final signal. Compared with the traditional free-space propagation model, the channel modeling method of the present invention takes into account the multipath effect in a complex environment, making the channel estimation more accurate. In addition, during the flight of the drones, their relative positions are constantly changing, and the phase of the signal propagation path will also shift. Therefore, the phase shift of each path is adjusted by To compensate for this change, the present invention utilizes the high-precision coherent detection ability of the 5G Red Cap module to be able to monitor and adjust the beam direction in real time, ensuring that the signals are combined within an appropriate phase range, thereby reducing the signal loss caused by phase mismatch.

[0054] In terms of time delay, the channel propagation time between drones is characterized by the time delay of each path and due to the different propagation distances of different paths, the signals do not arrive synchronously at the receiving end. To address this issue, the present invention introduces a Gaussian distribution weight in the time delay model to characterize the statistical characteristics of the path time delay. In this way, the system can effectively filter out the time delay errors with short-term mutations, improving the time synchronization accuracy of the signals. At the same time, the ultra-wideband characteristic of the 5G Red Cap module can provide extremely high time resolution, enabling the accuracy of time delay estimation to reach the nanosecond level, thereby improving the relative positioning accuracy of the drone swarm. In terms of Doppler frequency shift, since drones usually fly at a relatively high speed, relative motion will cause frequency shift of the wireless signals, and the Doppler frequency shift of each path is characterized by The Doppler effect will affect the communication quality of the drone swarm. Especially when the drones are highly maneuverable or the formation changes, the rapid change of the channel state will cause the frequency shift to intensify, thus affecting the stability of data transmission. The present invention adopts a high-precision frequency tracking technology based on the 5G Red Cap module, enabling the system to accurately estimate the Doppler frequency shift under low signal-to-noise ratio conditions and adjust the carrier frequency in real time to ensure the stability of the communication link.

[0055] Embodiment 3: In step 1, for each path, let the set of sampling times be and the set of delay samples be , discretize the spatio-temporal channel data of each link through the following formula to construct the UAV and the UAV The link channel matrix between them is as follows:

[0056] ;

[0057] where and are both integer subscript indices; is the number of sampling times; is the number of delay sampling points.

[0058] Specifically, first set the sampling time set , where represents the number of sampling times, and each time corresponds to a high-precision channel measurement. Since the UAV has a high flight speed, its channel characteristics may change drastically in a short time. Therefore, the present invention adopts an adaptive time sampling strategy, dynamically adjusts the sampling interval according to the channel change rate, so as to increase the sampling density when the channel changes rapidly, and reduce the sampling frequency when the channel is relatively stable, thus taking into account both computational efficiency and data accuracy. Compared with the traditional fixed sampling interval method, this adaptive sampling strategy can effectively reduce redundant data while ensuring that key channel features are not lost. At the same time, in order to accurately capture the delay characteristics of the channel, define the delay sampling set , where represents the number of delay sampling points, and each delay point corresponds to a discrete delay value. In a multipath propagation environment, the propagation times of different paths are different. Therefore, the signal will experience different delay distributions at the receiving end. In order to accurately characterize these propagation characteristics, the present invention utilizes the high-resolution delay measurement ability of the 5G Red Cap module to construct a sampling set covering all possible channel delays and calculates the channel impulse response at discrete time points. In this way, sufficient resolution can be obtained in the delay domain to improve the accuracy of subsequent beamforming and navigation estimation. Based on the above spatio-temporal sampling strategy, the channel information between the UAV and the UAV is organized into a two-dimensional matrix , where each element represents the sampling time and the delay The channel response values below. Each row of this matrix corresponds to different time samplings, and each column corresponds to different delay samplings, thus forming a complete spatio-temporal channel representation. This matrix-based channel modeling method has multiple advantages. First, it can provide high-dimensional channel information, enabling the system to simultaneously analyze the time evolution characteristics and delay characteristics, improving the accuracy of channel estimation. Second, the matrix structure helps to utilize modern signal processing techniques. For example, for channel prediction methods based on deep learning, the modeling accuracy of the UAV channel can be improved through matrix feature learning. Finally, this matrix can be directly used for subsequent optimal beamforming calculations, enabling the system to dynamically adjust the beam direction according to the current channel state, thereby improving communication quality and navigation accuracy.

[0059] Embodiment 4: In step 2, by solving the following generalized eigenvalue problem, based on each link channel matrix, for each UAV, a joint channel matrix is constructed in combination with its neighbor set, and the optimal beamforming vector is extracted:

[0060] ;

[0061] where represents the self-link channel matrix of UAV ; is the Hermitian conjugate operation of the matrix; is the identity matrix; is the eigenvalue in the generalized eigenvalue problem. When solving this generalized eigenvalue problem, the eigenvector corresponding to the largest eigenvalue is selected as the optimal beamforming vector; represents the neighbor set of UAV ; The optimal beamforming vector of UAV ; represents the received noise power of UAV .

[0062] Specifically, during the communication process of the UAV swarm, each UAV needs to ensure its own communication quality and, at the same time, minimize the interference to neighboring UAVs . Therefore, an optimization problem that can balance gain and interference needs to be constructed. For this purpose, the system first obtains the self-link channel matrix of each UAV. This matrix represents the propagation characteristics of the signal transmitted by UAV after passing through the wireless channel. At the same time, since the wireless communication of the UAV swarm is usually multipath propagation, the channel matrix includes factors such as multipath effects, phase offsets, and fading. When calculating the beamforming vector, in addition to paying attention to its own channel matrix , it is also necessary to combine all neighboring UAVs The link information is used to construct a joint channel matrix, so that the final beamforming strategy can enhance its own signal while reducing the interference to other UAVs. To achieve this optimization goal, the system needs to solve a generalized eigenvalue problem, where the goal is to maximize the beamforming vector under the action of the channel gain, while suppressing the neighboring UAV signal interference. In the optimization expression, the left term represents the channel gain of the UAV , while the right term contains the sum of the channel matrices of neighboring UAVs , and a noise impact term . Among them, represents the received noise power of the UAV , and the identity matrix is used to ensure numerical stability and avoid the problem of ill-conditioned matrices when solving eigenvalues. Under this optimization framework, by solving the generalized eigenvalue decomposition, the system can find the optimal beamforming vector , which corresponds to the largest eigenvalue, so that the signal gain of the UAV reaches the maximum, while suppressing the interference of neighboring UAVs.

[0063] The core advantage of this method lies in that through joint channel matrix modeling, beamforming optimization not only considers the channel characteristics of a single drone but also combines the cooperative effects of the drone swarm. Therefore, it can significantly improve the communication efficiency of the entire network. Compared with traditional beamforming methods, the method of the present invention uses the solution of the generalized eigenvalue problem, making the calculation of the beamforming vector more robust. Even in a complex channel environment, it can adaptively adjust the beam direction to ensure that the drone swarm always maintains the best communication quality during flight. In addition, due to the low-power computing ability of the 5G Red Cap module, this optimization method can operate efficiently on the embedded computing platform of the drone without introducing excessive computational overhead, thus ensuring the real-time communication optimization requirements of the drone swarm. In the dynamic environment of the drone, the optimal beamforming vector needs to be updated continuously over time. Therefore, the system needs to recalculate the generalized eigenvalue regularly and adjust the beam direction. In the method of the present invention, an intelligent prediction mechanism is adopted, enabling beam optimization to anticipate channel changes in advance and make pre-adjustments according to the movement trajectory of the drone. This prediction method combines historical channel data with real-time measurement data, enabling beamforming to be adjusted before significant channel changes occur, thereby reducing signal quality fluctuations and improving communication stability. In addition, to further reduce the computational complexity, the present invention uses a matrix sparsity analysis method, enabling only the main influencing factors to be concerned and ignoring the secondary channel components during the solution of the generalized eigenvalue, thus accelerating the calculation speed and enhancing real-time performance. Another key technical point is that this optimization method can adapt to different task requirements. For example, in a formation flight task, the system can optimize the beam direction to keep the communication links between all drones stable, while in an environmental monitoring task, it can adjust the beamforming strategy to enable the drone to establish an optimal communication link with the ground station. In addition, in a high-density drone formation, traditional fixed beamforming methods are prone to signal interference problems, while the method of the present invention enables the beamforming of each drone to be adaptively adjusted through joint channel matrix optimization, thereby reducing signal conflicts and increasing the overall network capacity. In addition, this optimization method can also effectively reduce the power consumption of the drone. Since the optimal beamforming vector can maximize the channel gain, the drone can use a lower transmit power under the same signal quality requirements, thereby reducing energy consumption and extending the battery life. This technical feature makes the method of the present invention particularly suitable for long-endurance drone tasks, such as large-scale area monitoring, air logistics networks, and remote disaster relief applications. Compared with traditional full-power transmission schemes, the method of the present invention can optimize the power control strategy while ensuring communication quality, significantly improving the energy utilization efficiency of the entire drone swarm.

[0064] Embodiment 5: In step 3, according to the optimal beamforming vector and the link channel matrix, assume the extracted time delay is , the phase is , and the Doppler shift is . Combining the relative position relationship between the UAVs, the following formula is used to define the residual function that minimizes the positioning error :

[0065] ;

[0066] Among them, is the initial position of UAV ; is the initial position of UAV ; is the Doppler shift variance; is the time delay variance; is the phase variance; is the velocity vector of UAV ; is the velocity vector of UAV ; is the transpose operation; is the Y-axis coordinate of the initial position of UAV ; is the Y-axis coordinate of the initial position of UAV ; is the X-axis coordinate of the initial position of UAV ; is the X-axis coordinate of the initial position of UAV ; is the carrier wavelength.

[0067] Specifically, in the definition of the residual function , the first term represents the ranging error based on the propagation time of the wireless signal. Since the propagation speed of the wireless signal is constant, the time delay can be used to calculate the distance estimate between UAV and UAV , where is the speed of light. However, the actually measured signal propagation time will be affected by multipath effects, channel fading, and noise. Therefore, the calculated distance may deviate from the true physical distance . To optimize this ranging error, the variance normalization method is adopted, that is, using the time delay variance As weights, the ranging errors under different channel conditions can be reasonably balanced, thereby reducing the impact of extreme errors. This optimization process ensures that the measurement values with smaller ranging errors contribute more to the final positioning result, while the measurement values with larger noise have lower weights, thus improving the overall positioning accuracy. The second term of the residual function represents the angular error based on phase measurement. Since the 5G Red Cap module has high-precision phase estimation ability, it can measure the and the drone relative phase shift between . According to the geometric relationship, the direction of the line connecting the two drones can be calculated through the coordinate relationship, that is , indicating the azimuth angle from to . However, in actual measurement, the phase information will be affected by channel phase noise, environmental interference, and device errors. Therefore, the error term needs to be normalized by the phase variance , so that the phase information with higher measurement accuracy has a greater impact on the positioning result, while the phase measurement with more serious interference has a lower weight. This optimization strategy can effectively reduce the impact of phase measurement errors on the final positioning result and improve the positioning robustness. The third term of the residual function is based on Doppler frequency shift measurement and is used to estimate the relative velocity between drones. Since during the flight of the drone, its velocity vectors and directly affect the frequency shift of the signal. According to the Doppler effect, the frequency shift is proportional to the projection of the relative velocity of the drones in the direction of their connection line. Specifically, calculates the theoretical Doppler frequency shift value based on position and velocity information and compares it with the actual measured value to calculate the square of the error. Since the Doppler frequency shift measurement is also affected by noise interference, the frequency shift variance is used for normalization, so that the path with smaller Doppler measurement error has a higher weight, while the measurement value with larger error has less impact. This method can use Doppler information to optimize the motion estimation of drones and improve the positioning accuracy in dynamic environments. Residual function Combines three independent but complementary wireless channel information of time delay, phase and Doppler frequency shift, and optimizes the positioning error through variance normalization, making the final UAV position estimation more accurate and stable. Compared with the traditional GPS positioning method, the method of the present invention can provide high-precision relative positioning in an environment where GPS signals are limited or severely interfered, and at the same time combines the real-time channel estimation ability of the 5G Red Cap module, enabling the system to continuously optimize the UAV position estimation in a dynamic environment and improving the cooperative navigation ability of the UAV swarm. In addition, since this method uses a variety of wireless channel characteristics for comprehensive optimization, compared with a single measurement method, it has higher positioning accuracy and stronger anti-interference ability, and is especially suitable for the autonomous flight, formation control and mission execution of UAV swarms in complex environments.

[0068] Embodiment 6: In step 3, through the following formula, using the residual function , estimate the position information of each UAV:

[0069] ;

[0070] Where is the estimated position of UAV ; is the estimated velocity vector of UAV .

[0071] Specifically, the wireless signal features extracted from the optimal beamforming vector and the link channel matrix are combined with the geometric and motion models between the UAVs for the actually measured channel parameters, thereby constructing an error evaluation index. Here, the residual function incorporates errors in three main aspects: First, the time-delay-based error term estimates the distance between UAVs by measuring the propagation delay of the wireless signal, and compares it with the distance obtained by multiplying the speed of light by the measured time delay; Second, the phase-based error term utilizes the geometric relationship of the connection line between UAVs, calculates the theoretical phase angle through the function, and then compares it with the actually measured phase offset; Finally, the Doppler frequency shift error term calculates the theoretical Doppler frequency shift using the velocity information of the UAVs and combining their relative positions, and then compares it with the actually measured frequency shift value. Each term is passed through the corresponding variance parameter (such as time-delay variance , phase variance and Doppler frequency shift variance )(Perform normalization processing so that each error term can automatically adjust the weight according to the measurement accuracy during the optimization process, avoiding excessive influence of a single noise source or extreme measurement error on the overall positioning result. The residual function constructed in this way can not only fully reflect the internal relationship between wireless channel measurement data and physical position and speed, but also effectively suppress random errors caused by factors such as multipath effect, signal fading, and environmental interference. During the mission execution of the UAV, due to high-speed flight and environmental changes, traditional GPS or inertial navigation systems often struggle to maintain continuous and stable positioning accuracy. However, this method utilizes the high-resolution channel sampling and beamforming technology of the 5G Red Cap module to obtain more detailed time delay, phase, and Doppler information, thus providing a solid data foundation for positioning optimization.)

[0072] (During the actual solution process, the system inputs the wireless signal characteristics collected in real time into the residual function) (and uses numerical optimization algorithms (such as gradient descent, Newton's method, or extended Kalman filter, etc.) to search for the optimal solution in the state space of the UAV, that is, the and (that minimize the error function. The key to this step lies in constructing a smooth and globally optimal objective function to ensure rapid convergence to the global optimal solution in a multi-variable non-linear optimization problem. The obtained optimal position and velocity vector represent the UAV state that best conforms to the measurement data and physical motion laws in the current wireless channel environment, enabling the entire positioning system to adapt to complex and changing environments and providing real-time and accurate position references during the collaborative communication and navigation of UAV swarms. It should be noted that this optimization positioning method based on the residual function is not only applicable to the positioning problem of a single UAV, but more importantly, it can utilize the collaborative information between UAV swarms, that is, through the relative measurement data between neighboring UAVs, to form a distributed positioning network, further improving the overall robustness and positioning accuracy of the system. The low latency and high bandwidth characteristics of the 5G Red Cap module ensure that these calculations can be completed in real time on the UAV embedded processing unit without affecting the real-time nature of the flight mission. At the same time, through multi-source information fusion, this method can effectively resist the risk of single measurement data failure or extreme noise interference, enabling the UAV swarm to maintain stable and continuous positioning services in complex urban environments, indoors, or scenarios with GPS signal occlusion.)

[0073] Example 7: In step 4, the effective link gain between the predicted UAV and the UAV is ; where is the UAV The optimal beamforming vector.

[0074] Specifically, during flight, drones are faced with complex factors such as multipath effects, signal fading, environmental interference, and Doppler frequency shift caused by relative motion. These factors make it difficult for traditional omnidirectional antennas or fixed-beam communication methods to meet the requirements of high-precision communication and navigation. With its high-resolution spatio-temporal sampling ability and low-latency advantage, the 5G Red Cap module can measure and estimate the complex state of the wireless channel in real time, and use the obtained channel matrix as a refined description of the channel characteristics between drones, while the beamforming vector and are the optimal transmit and receive beam directions optimized through algorithms such as joint channel matrix construction and generalized eigenvalue decomposition in the early stage. This beamforming technology can maximize the gain of the signal in the desired direction while suppressing interference from other directions. The conjugate transpose operation in the formula is used to ensure that the receiving end can accurately correct the phase information of the signal during signal synthesis, so as to achieve coherent superposition of signals from each path. Its absolute value operation converts the complex result into the actual signal amplitude, intuitively reflecting the effective gain of the link. In other words, by multiplying the pre-designed transmit beam with the comprehensive channel characteristics such as multipath, fading, and noise included in and combining the beamforming at the receiving end, the system can obtain a quantization index that reflects the actual signal transmission quality between drones. This index not only directly reflects the role of beam directivity in signal enhancement, but also effectively reduces the adverse effects caused by interference and noise. At the same time, this formula also has important significance for the collaborative operation of drone swarms, because in actual tasks, the communication link conditions between different drones directly affect the connectivity and data transmission efficiency of the overall network. Using this link gain prediction model, the system can update the link state information between each drone in real time during flight, so as to reasonably schedule communication resources, and dynamically adjust the flight trajectory and beamforming strategy according to the change of link gain, so that the entire drone swarm can still maintain a high-quality and low-latency communication link in the face of a complex dynamic environment. Especially in scenarios where drones are moving at high speed, the formation is changing, or interference is suddenly encountered, the traditional fixed-beam mode often cannot quickly adapt to the drastic fluctuations of the channel state, while the method of the present invention is to use the channel matrix Combined with the finely optimized beamforming vectors, it realizes the adaptive adjustment of channel gain. This method not only provides a valid link gain evaluation model mathematically, but also offers a strong technical support for cooperative navigation and communication within the UAV swarm, enabling the system to maintain stable and reliable communication performance in high-dynamic and complex interference environments. On the other hand, leveraging the low-power consumption and high processing capacity advantages of the 5G Red Cap module, this link gain prediction method can run in real time on the embedded processing unit of the UAV without relying on a large central processor, thus greatly reducing the overall power consumption and latency of the system. Further, through the dynamic monitoring of the link gain the system can also anticipate potential future channel fading or interference regions in advance, and then adopt avoidance or channel switching strategies in flight control and mission planning to ensure that the UAV swarm always maintains the best communication state when performing tasks such as monitoring, search and rescue, and logistics transportation.

[0075] Example 8: In step 4, construct an optimization problem through the following formula, and simultaneously plan the flight trajectory of the UAV and dynamic communication resource allocation to ensure the collaborative optimization of navigation and communication:

[0076] ;

[0077] where, is the transmission power of the UAV ; is the target position of the UAV. By solving this optimization problem, the corresponding is calculated when is minimized, and ; is the L2 norm operation; is the interference attenuation coefficient, and its value range is from 2 to 5.

[0078] Specifically, the first term is used to measure the distance difference between the target position of the UAV and the current estimated position , that is, it reflects the deviation degree of the flight trajectory. This term plays a role in smoothing trajectory adjustment and restricting large-amplitude position jumps during the optimization process, ensuring that the UAV can meet the mission objectives when adjusting the trajectory, and at the same time will not increase the flight risk or energy consumption due to frequent or drastic trajectory changes. Especially in complex environments, the UAV may need to adjust its flight path due to obstacle avoidance, interference, or mission switching, and this term emphasizes the priority of small-scale adjustments in the form of the sum of squares, thus minimizing the stability problems caused by path changes while achieving navigation accuracy. Then, the second term is directly related to the dynamic allocation of communication resources, where represents the transmission power of the UAV, while is the link gain between the UAV and the target UAV , and is the noise power in the system. The design intention of this term is to balance the relationship between the transmission power and the channel conditions. When the link gain is high, it indicates that the channel condition is good, and the signal can be fully amplified during transmission. At this time, the denominator is large, and the overall penalty term is small, so low-power communication can be achieved; conversely, if the link gain is low, the denominator decreases, and the penalty term increases accordingly, forcing the system to increase the power or adjust the beamforming strategy to improve the communication quality. In this way, the optimization problem can dynamically adapt to environmental changes, achieve energy-saving management of power on the premise of ensuring the stability of the communication link, and at the same time reduce the impact of the uncertainty of the wireless channel on the communication performance. The third term focuses on the interference problem and spatial cooperation effect within the UAV swarm. In a network composed of multiple UAVs, the UAVs will affect each other due to signal overlap and interference problems. This term accumulates the influence of neighboring UAVs and uses the -norm of the distance to the power (where is the path loss coefficient, usually taking values between 2 and 5) as the attenuation function to reflect the natural suppression effect of distance on the interference intensity.

[0079] When two UAVs are far apart, the interference is naturally small; when the distance is close, the interference effect will increase significantly, so that the penalty term increases rapidly, prompting the system to adjust the flight trajectory or transmission power to reduce the interference between each other. At the same time, the part in the numerator emphasizes the importance of signal gain in the communication link again. Only under high link gain can the signal be effectively transmitted, and this term also implies the dependence on beamforming technology, because efficient beamforming can maximize the signal energy in the desired direction and suppress unnecessary side lobe interference. After the entire objective function is accumulated and summed, it is then integrated over the time Perform cumulative optimization on the time scale, which means that the system continuously minimizes this objective function throughout the flight process to achieve real-time and dynamic collaborative optimization. The optimization formula in Embodiment 8 not only unifies the navigation and communication problems of the UAVs into an integrated objective mathematically, but also fully utilizes the advantages of high-precision channel measurement, beamforming technology, and low-latency data processing of the 5G Red Cap module in engineering implementation. When the UAV swarm executes tasks, the fine-tuning of its flight trajectory is directly related to the quality of the communication link. This formula realizes the dynamic balance between the two by jointly considering position changes, transmission power, and adjacent interference. Especially in a complex and changing environment, the UAVs need to continuously adjust their flight routes to avoid obstacles while ensuring stable communication with other UAVs and the ground control station. This optimization problem ensures that the UAVs can efficiently complete tasks and reduce energy consumption and interference risks under the condition of limited communication resources through the penalty mechanism for various indicators. In addition, by adjusting the value of the path attenuation coefficient the system can be flexibly configured for different application scenarios and environmental conditions, enabling this method to achieve the best performance in complex scenarios such as urban environments, mountainous areas, or indoors.

[0080] Embodiment 9: The transmission power of the UAV satisfies the following constraint conditions:

[0081] ;

[0082] ;

[0083] wherein, is a preset signal-to-noise ratio threshold.

[0084] Specifically, this constraint system is mainly achieved through two aspects. The first aspect is to limit the transmission power of the UAV not to exceed the upper limit determined by the channel gain and path attenuation characteristics, that is, the part of the expression . Here, the actual distance between the UAVs should be reflected in the denominator to reflect the path attenuation effect suffered by the wireless signal during spatial propagation. The essence of this constraint is that the signal will experience more significant attenuation as the distance increases during transmission. If the UAV is at a relatively long distance or the channel state is poor, the power required for its transmission should be strictly limited, so as to not only ensure the transmission efficiency but also avoid excessive interference caused by too high transmission power, thereby saving energy and maintaining the stability of the entire UAV swarm communication network. The second aspect introduces a signal-to-noise ratio (SNR) constraint condition, and the ratio term in the expression is used to measure the UAV The effectiveness of its own signal transmission in relation to interference from other drones and system noise. Specifically, this part of the constraint constructs a ratio, where the numerator is the effective signal gain obtained after beamforming optimization in its own link and the change in the current position estimate The ratio between them, while the denominator is the sum of the interference caused by all other drones to the current drone plus the receiver noise power . This construction method ingeniously takes into account the state of its own channel, changes in spatial position, and the impact of neighboring drones on interference, ensuring that the entire ratio must not be lower than a preset SNR threshold to meet the communication requirements. In other words, if the effective signal of the drone is weak relative to interference and noise, this constraint will force the system to readjust its transmit power, beamforming strategy, or flight position to ensure that the link quality is always maintained above a safe and reliable level. Such a design is particularly important in practical applications because drone swarms usually operate in a dynamically changing environment, where the channel state, relative position, and interference situation can change rapidly. Only by satisfying similar constraints in real time can the communication links between drones be ensured to have sufficient robustness, providing stable data support for navigation and mission coordination. Furthermore, based on the high-resolution channel measurement and real-time computing capabilities of 5G Red Cap modules, such power constraints and SNR guarantees can be quickly solved in the drone embedded system, enabling dynamic adjustment without additional hardware support. This method can not only prevent a single drone from causing local interference due to excessive transmit power but also optimize the allocation of wireless resources at the group level, balancing the energy consumption and performance of each drone between communication and navigation.

[0085] Although the specific implementation manners of the present invention are described above, those skilled in the art should understand that these specific implementation manners are only examples. Without departing from the principles and essence of the present invention, those skilled in the art can make various omissions, substitutions, and changes to the details of the above methods and systems. For example, combining the above method steps so as to perform substantially the same function in a substantially the same manner to achieve substantially the same result falls within the scope of the present invention. Therefore, the scope of the present invention is only defined by the appended claims.

Claims

1. A drone swarm collaborative communication and navigation method based on a 5G Red Cap module, characterized in that: The method comprises: Step 1: Use the 5G Red Cap module to perform high-resolution sampling of the multipath spatiotemporal channels between each drone in the drone group, discretize the spatiotemporal channel data of each link, and construct a link channel matrix; Step 2: Based on the channel matrix of each link, for each UAV, a joint channel matrix is ​​constructed in combination with its neighbor set to extract the optimal beamforming vector; Step 3: According to the optimal beamforming vector and link channel matrix, the delay, phase and Doppler frequency shift are extracted, and the position information of each UAV is estimated by minimizing the residual function of the positioning error in combination with the initial relative position relationship between the UAVs; Step 4: Based on the obtained location information of each UAV and the optimal beamforming vector, predict the effective channel gain of each link, construct an optimization problem, and simultaneously plan the flight trajectory of the UAV and the dynamic communication resource allocation to ensure the coordinated optimization of navigation and communication; In step 1, the drones in the drone group With drones The wireless link between the two is affected by multiple paths, each of which has different attenuation, phase, delay, and Doppler shift. The channel impulse response is expressed using the following formula: ； in, Indicates drone With drones The number of valid paths between them, each path corresponds to a signal component; For the The amplitude attenuation coefficient of a path reflects the attenuation degree of the signal when it propagates on the path; For the The delay of a path represents the time it takes for the signal to pass through the path from the transmitter to the receiver. For the The Doppler shift of the path; For the The phase of each path; is an integer subscript index; is the standard deviation of the delay of all paths; is the sampling time of high-resolution sampling; is the delay variable; is the imaginary number symbol; In step 2, by solving the following generalized eigenvalue problem, based on the channel matrix of each link, for each drone, a joint channel matrix is ​​constructed in combination with its neighbor set to extract the optimal beamforming vector: ； in, Indicates drone The self-link channel matrix of is the Hermitian conjugation operation of the matrix; is the identity matrix; is the eigenvalue in the generalized eigenvalue problem. When solving the generalized eigenvalue problem, select the eigenvector corresponding to the maximum eigenvalue as the optimal beamforming vector; Indicates drone Neighborhood collection; drone The optimal beamforming vector The following constraints are met: ; Indicates drone The received noise power; For drones With drones The link channel matrix between ; In step 3, according to the optimal beamforming vector and link channel matrix, the extracted delay is set to , the phase is , the Doppler shift is , combined with the relative position relationship between drones, the following formula is used to define the residual function that minimizes the positioning error : ； in, For drones The initial position of For drones The initial position of is the Doppler shift variance; is the delay variance; is the phase variance; For drones The velocity vector of For drones The velocity vector of is the transpose operation; For drones The Y-axis coordinate of the initial position; For drones The Y-axis coordinate of the initial position; For drones The X-axis coordinate of the initial position; For drones The X-axis coordinate of the initial position; is the carrier wavelength; It is the speed of light; In step 4, the optimization problem is constructed through the following formula, and the flight trajectory of the drone and the dynamic communication resource allocation are planned at the same time to ensure the coordinated optimization of navigation and communication: ; in, For drones The transmission power; is the target position of the UAV. By solving the optimization problem, we can calculate The minimum corresponding and ; is the L2 norm operation; Interference attenuation coefficient, ranging from 2 to 5; For the estimated drone location; Drones for prediction With drones The effective link gain between 2. The drone group collaborative communication and navigation method based on the 5G Red Cap module as claimed in claim 1, characterized in that: In step 1, for each path, let the sampling time set be And the delayed sampling set is , the spatiotemporal channel data of each link is discretized through the following formula to construct the UAV With drones The link channel matrix between: ； in, and All are integer subscript indices; is the number of sampling moments; is the number of delayed sampling points.

3. The drone group collaborative communication and navigation method based on the 5G Red Cap module as claimed in claim 2, characterized in that: In step 3, the residual function is used by the following formula , estimate the position information of each drone: ； in, For the estimated drone The velocity vector.

4. The drone group collaborative communication and navigation method based on the 5G Red Cap module as claimed in claim 3, characterized in that: In step 4, predict the drone With drones The effective link gain between ;in, For drones The optimal beamforming vector of .

5. The drone group collaborative communication and navigation method based on the 5G Red Cap module as claimed in claim 4, characterized in that: Drones Transmit power The following constraints are met: ； ； in, is the preset signal-to-noise ratio threshold.

Images