A user clustering, beamforming and power allocation method for a NOMA-based UAV synaesthesia integrated system

Abstract

The present invention relates to a user clustering, beamforming and power allocation method for a NOMA-based unmanned aerial vehicle (UAV) synaesthesia integrated system, and belongs to the field of wireless communication technology. The present invention divides users in a mission area into several clusters according to location information through a clustering algorithm, and then decomposes the original non-convex problem into two sub-problems through a block coordinate descent method: based on a fixed power allocation ratio of the UAV, the beamforming vector of the UAV is optimized; under a fixed transmission beam vector, the power allocation ratio is optimized and adjusted, and finally an approximate optimal solution for the optimization variable is obtained through an alternating iterative optimization algorithm. The present invention optimizes the beamforming vector and power allocation of the UAV under the conditions of dynamic clustering of communication users and transmission rate restrictions, UAV power restrictions, and perception target signal-to-noise ratio and effective perception power restrictions, and maximizes the sum of the minimum user transmission rate and effective perception power while ensuring the fairness of the synaesthesia function.

Description

Technical Field

[0001] The present invention relates to a user clustering, beamforming vector and power allocation method of a NOMA-based unmanned aerial vehicle (UAV) synaesthesia integrated system, and belongs to the technical field of wireless communications. Background Art

[0002] In recent years, the development of next-generation wireless networks has garnered widespread attention, and the application scenarios of wireless communication systems are expanding. Furthermore, many visions for B5G / 6G networks share a common consensus that sensing will play a more crucial role, particularly in common location and environmental awareness scenarios. Therefore, B5G / 6G networks require both communication and perception capabilities. To address this trend, integrated communication and perception technologies have been proposed, integrating radar sensing and wireless communications, sharing the same spectrum and infrastructure. Current designs for integrated communication and perception based on ground base stations face inherent sensing limitations, limiting their suitability for providing perception and communication services within a fixed range.

[0003] Drones have the advantages of high maneuverability and flexibility. As aerial base stations, they can quickly and effectively establish emergency communications and provide users with low-cost and high-efficiency wireless communication connections. At the same time, they can obtain more diversity gain by carrying multiple antennas. However, while the UAV is equipped with multiple antennas, it will also cause channel interference between communication users while increasing the system gain, which will reduce the communication performance and fail to meet the communication needs. In addition, the existing UAV telepathy integrated design is mostly aimed at scenarios with a small number of users. The service range of the UAV is relatively small, and the advantages of high maneuverability and high flexibility of the UAV cannot be fully utilized. In order to solve the problem of interference between the above-mentioned communication users, non-orthogonal multiple access (NOMA) technology can be introduced for processing. NOMA technology actively introduces interference information and gradually subtracts the interference of the user with the maximum signal power at the receiving end through serial interference cancellation technology to achieve correct demodulation. The simultaneous use of clustering algorithms and NOMA technology can realize multi-cluster and multi-user services, thereby expanding the scope of telepathy functions. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the NOMA-based UAV synesthesia integrated system combined with dynamic user clustering to enhance coverage and quality of service (QoS) has not been fully explored. A user clustering, beamforming and power allocation method for the NOMA-based UAV synesthesia integrated system is provided to improve system performance and ensure the fairness of synesthesia performance.

[0005] The technical solution of the present invention is: a user clustering, beamforming and power allocation method for a NOMA-based UAV synaesthesia integrated system, the specific steps of which are:

[0006] Step 1: Divide all users into several clusters based on their random distribution on the ground, and obtain the cluster center location information of each cluster;

[0007] Step 2: Based on the clusters obtained by dividing users and the NOMA technology, continuous interference cancellation technology is used for users in each cluster in turn to reduce interference between users and establish a corresponding communication model. Based on the cluster center position and the randomly generated sensing target position, the spatial distance is calculated to obtain the cluster center position closest to each sensing target and establish a corresponding sensing model.

[0008] Step 3: Based on the communication model and perception model, consider the power allocation of the UAV to each communication user in the cluster and the maximum energy constraint of the UAV, and establish a corresponding power model;

[0009] Step 4: Based on the communication model, perception model, and power model, under the constraints of the minimum transmission rate, minimum perception signal-to-noise ratio, minimum perception power, and user clustering of the UAV, the corresponding non-convex optimization problem is established by jointly optimizing the beamforming vector and power allocation to maximize the sum of the minimum user transmission rate and effective perception power.

[0010] Step 5: Decompose the above non-convex optimization problem using the block coordinate descent method, introduce auxiliary variables, and solve it using Taylor expansion and convex optimization tools to obtain the optimal beamforming vector and power allocation scheme for the UAV.

[0011] The Step 1 is specifically as follows:

[0012] Step 1.1: Given the user coordinates, set the number of clusters N and initialize N cluster centers;

[0013] Step 1.2: Calculate the Euclidean distance from each user coordinate position to the N cluster centers;

[0014] Step 1.3: Assign users to the nearest cluster center to form N clusters;

[0015] Step 1.4: Calculate the centroid of each cluster and use the centroid as the new cluster center;

[0016] Step 1.5: Repeat Step 1.2-Step 1.4 to continuously update the new cluster center position;

[0017] Step 1.6: When the cluster center position no longer changes, output the clustering results.

[0018] When establishing the communication model and perception model in Step 2, the coordinates of the jth target are recorded as g j =[x j ,y j ]T ; Let the nth cluster be C n , n∈{1,2,···,N}, the corresponding cluster center position is represented by g n =[x n ,y n ] T , the i-th user in the n-th cluster is represented as Its horizontal coordinate position is expressed as i∈{1,2,···,I n}, where I n is the number of communication users in the nth cluster;

[0019] Since the UAV uses a uniform linear array, the antenna array response vector between the UAV and the i-th user in the n-th cluster is:

[0020]

[0021] Where d represents the distance between adjacent antennas, λ represents the wavelength of the carrier center frequency, are the azimuth angles from the UAV to the i-th user in the n-th cluster.

[0022] The channel gain between the UAV and the i-th user in the n-th cluster can be expressed as:

[0023]

[0024] Where β0 is the channel gain when the distance between the UAV and the user is 1 m.

[0025] Since NOMA technology is used to eliminate the interference between users, assuming that the channel gain of users in the cluster for UAV is sorted in descending order, for the i-th user in the n-th cluster The signal to interference and noise ratio can be expressed as:

[0026]

[0027] Then the transmission rate of the i-th user in the n-th cluster is:

[0028]

[0029] The UAV perceives J targets randomly distributed in the working range. The selection criteria of the perceived targets are: when the jth target is adjacent to the nth cluster center g n When the distance is closest, the UAV reaches g n The jth target is sensed at this time, so the perception signal of the jth target received by the UAV is:

[0030]

[0031] in: is the channel gain from UAV to the perception target, specifically expressed as:

[0032]

[0033] in: a rec (θ) represents the receive beam vector, a tra (θ) is the transmit beam vector. Assuming that the receiving antenna and the transmitting antenna are the same antenna array, we have:

[0034]

[0035] SIC is used to eliminate the communication signal and obtain the interference-free radar echo signal. Therefore, the signal-to-noise ratio of the radar echo signal obtained by the UAV from the jth perceived target is:

[0036]

[0037] Where: η is the power spectral density constant; B is the channel bandwidth; is the variance of the predicted radar echo; is the corresponding beamforming vector; is Gaussian white noise.

[0038] The effective perception power of the jth target received at the UAV can be expressed as:

[0039]

[0040] Where: R w Represents the corresponding covariance matrix, which is specifically expressed as The total effective sensing power is further obtained as:

[0041]

[0042] In Step 3, the power model is specifically:

[0043] user The received signal is represented as:

[0044]

[0045] in, represents the power allocated to the i-th user in the n-th cluster; l∈C n ,l≠i represents the other users in the nth cluster except the i-th user; n i Obey the normal distribution (0,σ 2 ), represents additive white Gaussian noise.

[0046] Assume that the maximum transmission power of the UAV is pmax , then the following constraints exist:

[0047]

[0048] The Step 4 is specifically as follows:

[0049] The user clustering, beamforming and power allocation problem of the NOMA-based UAV interawareness integrated system to maximize the sum of the user's minimum transmission rate and limited perception power is modeled as follows:

[0050]

[0051] Among them, ψ c and ψ s is a normalization parameter, C1 represents the constraint of satisfying the user's minimum communication performance, γ com is the user's minimum transmission rate threshold; C2 is the constraint to ensure the minimum perceived performance, γ sen is the minimum transmission signal-to-noise ratio threshold for sensing the target; C3 is to ensure that the target meets the minimum sensing power P min ; C4 is the constraint of transmission power, p max is the maximum transmission power of the UAV; C5 is to limit a user to belong to only one cluster, is a binary variable, when When , it means that user i belongs to the nth cluster. When , it means that user i does not belong to the nth cluster; C6 is the power allocation constraint for users within the cluster.

[0052] The Step 5 is specifically as follows:

[0053] Step 5.1: Initialize the drone’s transmit beamforming and power distribution ratio P (0) , set the convergence accuracy ε, the maximum number of iterations Z max ; Iteration number z = 0;

[0054] Step 5.2: Based on the known w l (0) and P (0) Calculate the initial objective function value;

[0055] Step 5.3: According to the given w1 (z-1) ,w2 (z-1) ,P (z-1) ,w1 (0) ,w2 (0) ,P (0) , by solving the sub-problem about the beamforming vector, we get the communication beamforming solution w1 (z) ;

[0056] Step 5.4: According to the given w2 (z-1) ,P (z-1) ,w1 (z) , by solving the sub-problem about the beamforming vector, we get the perceptual beamforming solution w2 (z) ;

[0057] Step 5.5: According to the given w1 (z) ,w2 (z) , through the power allocation sub-problem, we get the power allocation solution P (z) ;

[0058] Step 5.6: Update z = z + 1, when the target value is greater than the convergence accuracy or z ≥ Z max End and output the results.

[0059] The beneficial effects of the present invention are as follows: the present invention clusters users according to the characteristics of their random distribution, thereby expanding the service range of the drone and being able to perceive known targets within the mission range. By jointly optimizing dynamic user clustering, beamforming vectors, and power allocation, the present invention communicates with users in each cluster in turn and performs perception tasks at the same time, thereby improving the system's minimum user transmission rate and effective perception power. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flow chart of the user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system of the present invention;

[0061] Figure 2 This is a model diagram of the UAV synaesthesia integrated system of the present invention;

[0062] Figure 3 The user clusters and cluster center position diagram obtained by using the clustering algorithm of the present invention;

[0063] Figure 4 It is a flight trajectory diagram of the UAV after the present invention determines each cluster;

[0064] Figure 5 It is the beam transmission gain diagram for sensing targets at two different positions in the present invention.

[0065] Figure 6 This is a graph showing the relationship between the minimum user transmission rate and power of the system under the comparison between the joint optimization scheme proposed in the present invention and the benchmark scheme. DETAILED DESCRIPTION

[0066] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0067] Example 1: Figure 1As shown in FIG, a user clustering, beamforming and power allocation method for a NOMA-based UAV synaesthesia integrated system is specifically as follows:

[0068] Step 1: Divide all users into several clusters based on their random distribution on the ground, and obtain the cluster center location information of each cluster;

[0069] Step 2: Based on the clusters obtained by dividing users and the NOMA technology, continuous interference cancellation technology is used for users in each cluster in turn to reduce interference between users and establish a corresponding communication model. Based on the cluster center position and the randomly generated sensing target position, the spatial distance is calculated to obtain the cluster center position closest to each sensing target and establish a corresponding sensing model.

[0070] Step 3: Based on the communication model and perception model, consider the power allocation of the UAV to each communication user in the cluster and the maximum energy constraint of the UAV, and establish a corresponding power model;

[0071] Step 4: Based on the communication model, perception model, and power model, under the constraints of the minimum transmission rate, minimum perception signal-to-noise ratio, minimum perception power, and user clustering of the UAV, the corresponding non-convex optimization problem is established by jointly optimizing the beamforming vector and power allocation to maximize the sum of the minimum user transmission rate and effective perception power.

[0072] Step 5: Decompose the above non-convex optimization problem using the block coordinate descent method, introduce auxiliary variables, and solve it using Taylor expansion and convex optimization tools to obtain the optimal beamforming vector and power allocation scheme for the UAV.

[0073] like Figure 2 As shown, in this embodiment, a UAV interawareness system model is established. This model provides communication services to users within each cluster when the UAV flies to the cluster center position, and simultaneously performs the perception task at the cluster center position closest to the perception target. Consider a UAV interawareness system consisting of a fixed-wing UAV, J perception targets, and K communication users. The UAV acts as an aerial base station equipped with a uniform linear array (ULA) with M antennas, operating at a fixed altitude H; the perception target j∈{1,…,J}, J≥1; the K communication users are unevenly distributed on the ground and divided into N clusters.

[0074] The Step 1 is specifically as follows:

[0075] In order to conveniently describe the location of the sensing target and the communication user, the coordinates of the jth target are recorded as g j =[x j ,y j ] T ; Let the nth cluster be Cn , n∈{1,2,···,N}, the corresponding cluster center position is represented by g n =[x n ,y n ] T The i-th user in the n-th cluster is represented as Its horizontal coordinate position is expressed as i∈{1,2,···,I n}, where I n is the number of communicating users in the nth cluster.

[0076] Assume that the communication signal sent by UAV to the i-th user in the n-th cluster is represents the corresponding transmit beamforming vector, represents the transmitted sensing signal. Therefore, the signal sent by the UAV to the i-th user in the n-th cluster.

[0077]

[0078] Since the UAV uses a uniform linear array, the antenna array response vector between the UAV and the i-th user in the n-th cluster is:

[0079]

[0080] The channel gain between the UAV and the i-th user in the n-th cluster can be expressed as:

[0081]

[0082] Where β0 is the channel gain when the distance between the UAV and the user is 1 m.

[0083] Since NOMA technology is used to eliminate the interference between users, assuming that the channel gain of users in the cluster for UAV is sorted in descending order, for the i-th user in the n-th cluster The signal to interference and noise ratio can be expressed as:

[0084]

[0085] Then the transmission rate of the i-th user in the n-th cluster is:

[0086]

[0087] The UAV perceives J targets randomly distributed in the working range, so the perception signal of the j-th target received by the UAV is:

[0088]

[0089] in: is the channel gain from UAV to the perception target, specifically expressed as:

[0090]

[0091] in: a rec (θ) represents the receive beam vector, a tra (θ) is the transmit beam vector.

[0092] SIC is used to eliminate the communication signal and obtain the interference-free radar echo signal. Therefore, the signal-to-noise ratio of the radar echo signal obtained by the UAV from the jth perceived target is:

[0093]

[0094] Where: η is the power spectral density constant; B is the channel bandwidth; is the variance of the predicted radar echo; is the corresponding beamforming vector; is Gaussian white noise.

[0095] The effective perception power of the jth target received at the UAV can be expressed as:

[0096]

[0097] Where: R w Represents the corresponding covariance matrix, which is specifically expressed as The total effective sensing power is further obtained as:

[0098]

[0099] Therefore, the problem of user clustering, beamforming, and power allocation in the NOMA-based UAV interawareness integrated system to maximize the sum of the user's minimum transmission rate and limited perception power is modeled as follows:

[0100]

[0101] Among them, ψ c and ψ s is a normalization parameter, C1 represents the constraint of satisfying the user's minimum communication performance, γ com is the user's minimum transmission rate threshold; C2 is the constraint to ensure the minimum perceived performance, γ sen is the minimum transmission signal-to-noise ratio threshold for sensing the target; C3 is to ensure that the target meets the minimum sensing power P min ; C4 is the transmission power constraint p max is the maximum transmission power of the UAV; C5 is to limit a user to belong to only one cluster, is a binary variable, when When , it means that user i belongs to the nth cluster. When , it means that user i does not belong to the nth cluster; C6 is the power allocation constraint for users within the cluster.

[0102] To solve problem P1, an iterative optimization algorithm based on BCD is proposed. The variables in each subproblem are optimized sequentially using a fixed variable method, and the optimization iterations are performed alternately until convergence. The joint optimization problem P1 can be decomposed into two subproblems: the beamforming optimization subproblem and the power optimization subproblem. The original problem is converted into an approximate convex problem by using a first-order Taylor expansion, continuous convex approximation, and the introduction of auxiliary variables. An alternating iterative joint optimization algorithm is designed, and the resulting convex subproblems are alternately optimized to obtain an approximate optimal solution. The non-convex optimization problem with highly coupled variables is decomposed into three subproblems using block coordinate descent. Semi-positive relaxation methods and convex optimization tools are employed to solve the problem, resulting in a suboptimal solution to the original non-convex problem.

[0103] For ease of processing, define At the same time, the auxiliary variable R is introduced tr Indicates the minimum user transmission rate.

[0104] The communication beamforming vector problem is expressed as:

[0105]

[0106] In order to solve the non-convex problem, the continuous convex approximation technique is used to solve the non-convex objective function and non-convex constraints, and the approximate solution of the convex problem is obtained. k The approximate lower bound of , the constraint C1 of subproblem P2 can be equivalently transformed into:

[0107]

[0108] Since R2 is relative to Concave function, according to the property that the first-order Taylor expansion of the concave function at any point is its global upper bound, the global upper bound of R2 at the zth iteration is:

[0109]

[0110] in: is the beamforming vector given in the zth iteration. ξ is specifically expressed as:

[0111]

[0112] It's about The convex function of , according to the properties of convex functions, the expansion of a convex function at any point is its lower bound. Therefore, The first-order Taylor expansion can be obtained as The lower bound.

[0113] With fixed transmit power P and communication beamforming vector w1, the sub-problem about sensing beamforming vector w2 is:

[0114]

[0115] Similarly, we can get:

[0116]

[0117] And simplifying C2 we can get:

[0118]

[0119] Fixing the beamforming vectors w1 and w2, and solving the transmit power P, the original problem is simplified to:

[0120]

[0121] Due to the influence of SIC technology, the user with the largest signal gain in the cluster is not affected by other users with smaller channel gains, so its corresponding signal to interference and noise ratio is:

[0122]

[0123] The user obtains the maximum channel gain in the nth cluster, namely:

[0124]

[0125] To meet the QOS of communication users, the minimum power of the i-th user in the n-th cluster can be obtained as:

[0126]

[0127] According to the above The equivalent convex forms of R1 and R2 are obtained by first-order Taylor expansion:

[0128]

[0129] The final problem-solving process is expressed as:

[0130] Step 5.1: Initialize the drone’s transmit beamforming and power distribution ratio P (0) , set the convergence accuracy ε, the maximum number of iterations Z max ; Iteration number z = 0;

[0131] Step 5.2: Based on the known and P (0) Calculate the initial objective function value;

[0132] Step 5.3: According to the given P (z-1) , P (0) , by solving the sub-problem about the beamforming vector, we get the communication beamforming solution

[0133] Step 5.4: According to the given P (z-1) , By solving the subproblem about the beamforming vector, we get the perceptual beamforming solution

[0134] Step 5.5: According to the given Through the power allocation sub-problem, we get the power allocation solution P (z) ;

[0135] Step 5.6: Update z = z + 1, when the target value is greater than the convergence accuracy or z ≥ Z max End and output the results.

[0136] In this implementation, 20 communication users and two sensing targets are randomly distributed in a 1000m x 1000m scenario, with the 20 users divided into eight clusters. The drone's starting position is (0, 0) and its final arrival position is (1000, 1000). The flight altitude is H = 50m, the maximum transmit power is 30dBm, and the system noise power is -110dBm.

[0137] like Figure 3 and Figure 4 As shown in Figure 2, after being divided into 8 clusters by users, the UAV traverses each cluster and provides communication services in turn, and performs the perception task at the cluster center closest to the perception target.

[0138] like Figure 5 As shown in the figure, the proposed algorithm can achieve correct perception of the target and realize beam alignment, while the transmit beam gain is slightly lower than the beam gain of using radar to only realize the perception function.

[0139] like Figure 6 As shown in the figure, the minimum user rate increases with the increase of total transmission power. At the same time, the proposed scheme outperforms the other three comparison schemes, which illustrates the superiority of the proposed algorithm.

[0140] The above describes the specific embodiments of the present invention in detail with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the knowledge of ordinary technicians in this field without departing from the scope of the present invention.

Claims

1. A user clustering, beamforming, and power allocation method for a NOMA-based UAV synaesthesia integrated system, characterized by: Step 1: Divide all users into several clusters based on their random distribution on the ground, and obtain the cluster center location information of each cluster; Step 2: Based on the clusters obtained by dividing users and the NOMA technology, continuous interference cancellation technology is used on the users in each cluster in turn to reduce interference between users and establish a corresponding communication model. Based on the cluster center position and the randomly generated sensing target position, the spatial distance is calculated to obtain the cluster center position closest to each sensing target and establish a corresponding sensing model. Step 3: Based on the communication model and perception model, consider the power allocation of the UAV to each communication user in the cluster and the maximum energy constraint of the UAV, and establish a corresponding power model; Step 4: Based on the communication model, perception model, and power model, under the constraints of the UAV's minimum transmission rate, minimum perceived signal-to-noise ratio, minimum perceived power, and user clustering, the corresponding non-convex optimization problem is established by jointly optimizing the beamforming vector and power allocation to maximize the sum of the minimum user transmission rate and effective perceived power. Step 5: Decompose the above non-convex optimization problem using the block coordinate descent method, introduce auxiliary variables, and solve it using Taylor expansion and convex optimization tools to obtain the optimal beamforming vector and power allocation scheme for the UAV.

2. The user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system according to claim 1 is characterized in that: The Step 1 specifically includes: Step 1.1: Given the user coordinates, set the number of clusters N and initialize N cluster centers; Step 1.2: Calculate the Euclidean distance from each user coordinate position to the N cluster centers; Step 1.3: Assign users to the nearest cluster center to form N clusters; Step 1.4: Calculate the centroid of each cluster and use the centroid as the new cluster center; Step 1.5: Repeat Step 1.2-Step 1.4 to continuously update the new cluster center position; Step 1.6: When the cluster center position no longer changes, output the clustering results.

3. The user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system according to claim 1 is characterized in that: When establishing the communication model and perception model in Step 2, the coordinates of the jth target are recorded as g j =[x j ,y j ] T ; Let the nth cluster be C n , n∈{1,2,···,N}, the corresponding cluster center position is represented by g n =[x n ,y n ] T , the i-th user in the n-th cluster is represented as Its horizontal coordinate position is expressed as Among them I n is the number of communication users in the nth cluster; The UAV uses a uniform linear array, and the antenna array response vector between the UAV and the i-th user in the n-th cluster is: Where d represents the distance between adjacent antennas, λ represents the wavelength of the carrier center frequency, are the azimuth angles from the UAV to the i-th user in the n-th cluster; The channel gain between the UAV and the i-th user in the n-th cluster is expressed as: Where β0 is the channel gain when the distance between UAV and user is 1m; Assume that the channel gain of the users in the cluster to the UAV is sorted in descending order, for the i-th user in the n-th cluster The signal to interference and noise ratio is expressed as: Then the transmission rate of the i-th user in the n-th cluster is: The UAV perceives J targets randomly distributed in the working range. The selection criteria of the perceived targets are: when the jth target is adjacent to the nth cluster center g n When the distance is closest, the UAV reaches g n The jth target is sensed at this time, so the perception signal of the jth target received by the UAV is: in: is the channel gain from UAV to the perception target, specifically expressed as: in: a rec (θ) represents the receive beam vector, a tra (θ) is the transmit beam vector. Assuming that the receiving antenna and the transmitting antenna are the same antenna array, we have: SIC is used to eliminate the communication signal and obtain the interference-free radar echo signal. Therefore, the signal-to-noise ratio of the radar echo signal obtained by the UAV from the jth perceived target is: Where: η is the power spectral density constant; B is the channel bandwidth; is the variance of the predicted radar echo; is the corresponding beamforming vector; is Gaussian white noise; The effective perception power of the jth target received at the UAV is expressed as: Where: R w Represents the corresponding covariance matrix, which is specifically expressed as The total effective sensing power is further obtained as:

4. According to the user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system according to claim 1, the power model in Step 3 is specifically: user The received signal is represented as: in, represents the power allocated to the i-th user in the n-th cluster; l∈C n ,l≠i represents the other users in the nth cluster except the i-th user; n i Obey the normal distribution (0,σ 2 ), represents additive white Gaussian noise; Assume that the maximum transmission power of the UAV is p max , then the following constraints exist:

5. According to the user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system according to claim 1, Step 4 specifically includes: The user clustering, beamforming and power allocation problem of the NOMA-based UAV interawareness integrated system to maximize the sum of the user's minimum transmission rate and limited perception power is modeled as follows: P1: s.t.C1: C2: C3: C4: C5: C6: Among them, ψ c and ψ s is a normalization parameter, C1 represents the constraint of satisfying the user's minimum communication performance, γ com is the user's minimum transmission rate threshold; C2 is the constraint to ensure the minimum perceived performance, γ sen is the minimum transmission signal-to-noise ratio threshold for sensing the target; C3 is to ensure that the target meets the minimum sensing power P min ; C4 is the constraint of transmission power, p max is the maximum transmission power of the UAV; C5 is to limit a user to belong to only one cluster, is a binary variable, when When , it means that user i belongs to the nth cluster. When , it means that user i does not belong to the nth cluster; C6 is the power allocation constraint for users within the cluster.

6. According to the user clustering, beamforming and power allocation method of the NOMA-based UAV synaesthesia integrated system according to claim 1, Step 5 specifically includes: Step 5.1: Initialize the drone’s transmit beamforming and power distribution ratio P (0) , set the convergence accuracy ε, the maximum number of iterations Z max ; Iteration number z = 0; Step 5.2: Based on the known and P (0) Calculate the initial objective function value; Step 5.3: According to the given P (z-1) , P (0) , by solving the sub-problem about the beamforming vector, we get the communication beamforming solution Step 5.4: According to the given P (z-1) , By solving the subproblem about the beamforming vector, we get the perceptual beamforming solution Step 5.5: According to the given Through the power allocation sub-problem, we get the power allocation solution P (z) ; Step 5.6: Update z = z + 1, when the target value is greater than the convergence accuracy or z ≥ Z max End and output the results.

Images