Unmanned aerial vehicle communication and sensing integrated trajectory and beam forming joint optimization method

Abstract

The invention relates to an unmanned aerial vehicle communication and sensing integrated trajectory and beam forming joint optimization method. According to the method, firstly, an unmanned aerial vehicle communication sensing environment is constructed, in the environment, an unmanned aerial vehicle sends superposed signals by using a non-orthogonal multiple access technology, then a bit rate model and a sensing power model are established, and a target function is constructed according to the bit rate model and the sensing power model; and finally, unmanned aerial vehicle flight parameter constraints are set, and under the constraints, the optimal solution of the target function is solved by adopting the improved particle swarm optimization. Compared with the prior art, the method has the advantages that weighted sum maximization of communication user throughput and effective target sensing power is achieved, the spectrum utilization rate is improved, the system application scene is expanded, and the accuracy of unmanned aerial vehicle communication and sensing integrated trajectory and beam forming joint optimization solution is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar communication integration, and in particular to a joint optimization method for trajectory and beamforming of unmanned aerial vehicle (UAV) synaesthesia integration. Background Art

[0002] As the cost of drones decreases and their ability to perform various complex tasks increases, drones that can move freely at a certain height can be used as a new type of aerial platform to realize the Integrated Sensing And Communications (ISAC) function, and can provide downlink data transmission for communication users through transmitting end beamforming, while realizing the perception of targets in the area. However, when the channels of multiple users are highly correlated, the receiving end may receive mixed signals from multiple users, resulting in decoding difficulties, thereby causing interference between users. This interference will lead to a decrease in signal quality, which in turn affects the reliability and efficiency of communication.

[0003] The invention patent with publication number CN115314135A discloses a communication perception integrated waveform design method for UAV collaboration. This method calculates the channel gain value between the UAV and each communication user based on the line-of-sight wireless transmission model; establishes a waveform design optimization problem with the goal of maximizing the communication rate and effective sensing power, and performs beamforming on the information symbols of K users on the UAV side, and sends a common waveform to all users and M targets to be sensed through N antennas. It is possible to directly calculate and obtain the optimal UAV integrated waveform in the current network topology through a mature optimization method without exceeding the maximum transmission power limit of the UAV. However, this method cannot maximize the weighted sum of the communication user throughput and the effective target perception power. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a joint optimization method of trajectory and beamforming of unmanned aerial vehicle integrated with synaesthesia.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] According to one aspect of the present invention, a joint optimization method of trajectory and beamforming for integrated synaesthesia of unmanned aerial vehicles is provided, and the method comprises the following steps:

[0007] S1. Constructing a UAV communication perception environment, in which the UAV uses non-orthogonal multiple access technology to send superimposed signals;

[0008] S2. establishing a bit rate model and a perceived power model according to the superimposed signal, and constructing an objective function by the bit rate model and the perceived power model;

[0009] S3. Set the UAV flight parameter constraints. Under these constraints, the improved particle swarm algorithm is used to solve the optimal solution of the objective function.

[0010] As a preferred technical solution, the UAV communication perception environment in S1 includes a NOMA-based UAV, K communication users, and M perception targets.

[0011] As a preferred technical solution, in S1, the UAV uses non-orthogonal multiple access technology to transmit the superimposed signal, including: the UAV sends the superimposed signal to the communication user and the UAV sends the superimposed signal to the perception target.

[0012] As a preferred technical solution, the superimposed signal sent by the drone to the communication user in S1 is:

[0013]

[0014] Among them, s i [n] is the signal sent to the i-th communication user in time slot n, is the corresponding transmit beamforming vector.

[0015] As a preferred technical solution, the superposition signal sent by the drone to the sensing target in S1 is:

[0016]

[0017] Among them, R w [n] is the covariance matrix, is the corresponding transmit beamforming vector, and H is the conjugate transposed identifier.

[0018] As a preferred technical solution, the bit rate model, that is, the communication throughput of k in time slot n, is specifically formulated as follows:

[0019] R k [n] = min{R k→k [n],R k→k+1 [n],…,R K [n]},

[0020] Among them, R k [n] is the bit rate of user k in time slot n.

[0021] As a preferred technical solution, the specific formula of the perceived power model is:

[0022] P(θ m [n]) = a H (θ m [n])R w [n]a(θ m [n]),

[0023] Among them, R w [n] is the covariance matrix, θ m [n] represents the azimuth angle from the UAV to the mth perceived target in time slot n, a(θ m [n]) is the guidance vector of the UAV in the direction of the target at time slot n.

[0024] As a preferred technical solution, the specific formula of the objective function constructed in S2 is:

[0025]

[0026] Among them, R k [n] is the bit rate of user k in time slot n, w k [n] is the corresponding transmit beamforming vector, θ m [n] is the pitch angle from the UAV to the th sensing target, ρr is the regularization parameter of the sensing power, and ρc is the regularization parameter of the communication rate.

[0027] As a preferred technical solution, the UAV flight parameters in S3 include the flight starting point, end point, maximum flight speed of the UAV and maximum transmission power of the UAV;

[0028] The specific formula for setting the UAV flight parameter constraints is:

[0029] s.tq[1]=q0

[0030] q[N]=q T

[0031]

[0032] Among them, the constraint q[1] = q0 fixes the starting point of the UAV flight, and the constraint q[N] = q T Fixed the end position of the drone flight, constraints Ensure that the average transmission power of the drone in time slot n does not exceed the maximum transmission power of the drone, and constrain The maximum flight distance of the drone in each time slot is limited. t is the time slot length; V max is the maximum flight speed of the drone.

[0033] As a preferred technical solution, the logarithmic function is introduced into the improved particle swarm algorithm in S3 to construct a nonlinear asynchronous learning factor. The specific steps of using the improved particle swarm algorithm to solve the optimal solution of the objective function include:

[0034] S31, initializing the position of the drone particle swarm, the beamforming particle swarm value, the flight speed of the drone position particle swarm, and the speed of the beamforming value particle swarm;

[0035] S32, using the position of the drone particle swarm to calculate the individual values ​​of the drone particle swarm and the beamforming particle swarm, and defining them as individual optimal solutions;

[0036] S33, introduce logarithmic function to construct nonlinear asynchronous learning factor;

[0037] S34, using nonlinear asynchronous learning factors to obtain the fitness values ​​of the drone position population and the beamforming population;

[0038] S35, based on the fitness values ​​of the drone position population and the beamforming population obtained in S34, the individual optimal solutions of the two populations are updated respectively, and the individual optimal solutions are compared one by one in the particle group to obtain the group optimal solution;

[0039] S36, determine whether the maximum number of iterations has been reached, if so, end the solution; if not, return to S32.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention first constructs a UAV communication perception environment, in which the UAV uses non-orthogonal multiple access technology to send superimposed signals, then establishes a bit rate model and a perception power model, and constructs an objective function based on the bit rate model and the perception power model; finally, the UAV flight parameter constraints are set, and under the constraints, an improved particle swarm algorithm is used to solve the optimal solution of the objective function. For the NOMA-based UAV ISAC system, the communication signal and the perception signal are weighed by jointly optimizing the design of the UAV flight trajectory and the transmit beamforming vector, so as to maximize the weighted sum of the communication user throughput and the effective target perception power.

[0042] 2. The present invention uses non-orthogonal multiple access technology to transmit superimposed signals, including: the UAV sends the superimposed signal to the communication user and the UAV sends the superimposed signal to the perception target. By introducing non-orthogonal multiple access technology, the present invention effectively reduces multi-user interference in the UAV ISAC system. The data of multiple users can be transmitted simultaneously in the same frequency band. This non-orthogonal resource reuse method greatly improves the utilization efficiency of the spectrum.

[0043] 3. The present invention uses an improved particle swarm optimization algorithm to solve the objective function. The improved particle swarm optimization algorithm introduces a logarithmic function to construct a nonlinear asynchronous learning factor, which solves the problem that the particle swarm optimization algorithm is prone to fall into local optimality and cannot obtain a global optimal solution. It increases the accuracy of the solution and provides strong support for the wide application of the ISAC system.

[0044] 4. In the present invention, the UAV communication perception environment includes a NOMA-based UAV, K communication users, and M perception targets. NOMA-based UAVs significantly improve spectrum efficiency through non-orthogonal resource reuse, and can accommodate more users within the same bandwidth, thereby meeting the needs of high-density user scenarios. The present invention can provide effective solutions in crowded shopping malls, hospitals, stadiums and other scenarios. At the same time, with the continuous development of wireless communication technology, the present invention can also be combined with other advanced technologies to bring more possibilities to the future communication and perception fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of the steps of a joint optimization method for trajectory and beamforming of a UAV with integrated synaesthesia in the present invention;

[0046] Figure 2 This is a diagram of the architecture of a UAV synaesthesia integrated communication system in the embodiment;

[0047] Figure 3 This is the flight trajectory diagram of the drone after the IPSO algorithm is optimized in the embodiment;

[0048] Figure 4 It is an iterative convergence curve diagram of PSO and IPSO algorithms in the embodiment;

[0049] Figure 5 A schematic diagram of the trade-off between the communication rate and the sensing power of the PSO and IPSO algorithms in the embodiment;

[0050] Figure 6 1 is a transmission beam diagram obtained under the synaesthesia integration system in the embodiment. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0052] Example

[0053] Non-Orthogonal Multiple Access (NOMA) is a new multiple access technology that allows multiple users to transmit data simultaneously on the same spectrum channel, which can effectively reduce interference between multiple users and achieve higher spectrum efficiency than traditional multiple access technologies.

[0054] This scheme is designed to jointly optimize the flight trajectory and power allocation of NOMA-based drones in order to maximize the user's minimum achievable average rate; in addition, the base station is designed to use NOMA technology to provide services to multiple users and perceive the target through superimposed NOMA signals, which is beneficial to improving the performance of the ISAC system.

[0055] Therefore, this scheme proposes a joint optimization method of trajectory and beamforming for UAV synaesthesia integration. The specific steps of the method are as follows: Figure 1 As shown, including:

[0056] S1. Constructing a UAV communication perception environment, in which the UAV uses non-orthogonal multiple access technology to send superimposed signals;

[0057] S2. establishing a bit rate model and a perceived power model according to the superimposed signal, and constructing an objective function by the bit rate model and the perceived power model;

[0058] S3. Set the UAV flight parameter constraints. Under these constraints, the improved particle swarm algorithm is used to solve the optimal solution of the objective function.

[0059] In step S1, a synaesthesia integrated system is first established, which includes a NOMA-based drone, K ground users, and M ground targets. The drone is equipped with U antennas as an aerial dual-function access point, equipped with a vertically placed uniform linear array (ULA), K ground users constitute the communication system, and M ground targets constitute the perception system.

[0060] In this embodiment, the established synaesthesia integrated system structure is as follows: Figure 2 As shown, in the system, the UAV sends superimposed signals to perception target 1 and perception target 2 respectively to complete the perception behavior; sends superimposed signals to communication user 1, communication user 2 and communication user 3 to complete the communication behavior; communication user 1, communication user 2 and communication user 3 interfere with each other.

[0061] In the synaesthesia integration system, the NOMA-based drone is dynamic. Considering the total mission time T of a drone in the ISA C scenario, it is discretized into N time slots, and the drone flies during the total time. By jointly optimizing the design of the drone flight trajectory and the transmit beamforming vector, while ensuring the constraints of the flight starting point, end point, maximum flight rate and transmit power, the weighted sum of the communication user throughput and the effective target perception power is maximized. Combined with the high maneuverability and high flexibility of the drone itself, the PSO algorithm is used for solution. In this embodiment, considering that the original particle swarm optimization algorithm is prone to fall into local optimality, and there are nonlinear problems in the objective function and constraints, the IPSO algorithm is designed to solve the optimization problem.

[0062] Based on the established interawareness integrated system, we further constructed the flight scenarios of UAVs under communication and perception tasks, and used NOMA technology to send superimposed signals to achieve multi-user communication and ground target perception;

[0063] This solution introduces NOMA technology, which enables data of multiple users to be transmitted simultaneously in the same frequency band, thereby effectively reducing multi-user interference in the drone ISAC system. This non-orthogonal resource multiplexing method greatly improves the utilization efficiency of the spectrum.

[0064] Since NOMA can significantly improve spectrum efficiency through non-orthogonal resource reuse, it means that more users can be accommodated within the same bandwidth, thus meeting the needs of high-density user scenarios. It can provide effective solutions in crowded shopping malls, hospitals, stadiums and other scenarios. At the same time, with the continuous development of wireless communication technology, it can also be combined with other advanced technologies to bring more possibilities to the future communication and perception fields.

[0065] Thus, the superimposed signal sent by the drone includes sending a superimposed signal to the user and sending a superimposed signal to the perceived target;

[0066] The superimposed signal sent to the user communication is specifically:

[0067]

[0068] In the formula, s i [n] is the signal sent by the i-th communication user in time slot n. is the corresponding transmit beamforming vector.

[0069] The superposition signal sent to the sensing target is specifically:

[0070]

[0071] In the formula, R w [n] is the covariance matrix, is the corresponding transmit beamforming vector, and H is the conjugate transposed identifier.

[0072] In step S2, take the signal of user k transmitted in the channel of j as an example. At this time, the signal greater than k is interference. At time slot n, the bit rate of the communication signal at user k after SIC technology is as follows:

[0073]

[0074] The signal of user k is transmitted in the channel of user j (j>k), and decoding is also required. After SIC technology, the bit rate at user j is as follows:

[0075]

[0076] At user K, the interference from all other users is cancelled by SIC, so the bit rate at user K is:

[0077]

[0078] The specific formula for the communication throughput of user k in time slot n is as follows:

[0079] R k [n] = min{R k→k [n],R k→k+1 [n],…,R K [n]},

[0080] The specific formula for the total communication throughput of K users in the nth time slot is:

[0081]

[0082] Among them, h k (q[n],u k ) is the channel gain from the UAV to the kth user in this time slot.

[0083] Assuming that the link between the UAV and the ground user is mainly line-of-sight transmission, its channel gain follows the free space model. The specific formula for the channel gain between the UAV and user k in time slot n is:

[0084]

[0085] Where β is the link channel gain at a distance of 1 m from the drone, and is d(q[n], u k ) is the distance from the UAV to user k, q[n] represents the position of the UAV in the nth time slot, u k Represents the ground user location.

[0086] Define a(q[n], u k ) represents the array steering vector from the UAV to user k in time slot n, and its specific formula is:

[0087]

[0088] Where λ and d represent the carrier wavelength and the distance between two adjacent antennas respectively; U is the number of antennas equipped on the UAV transmitter; θ(q[n], u k ) is the pitch angle between the UAV and user k in time slot n, and H is the flight altitude of the UAV.

[0089] In addition, the covariance matrix of the transmitted signal is designed to meet the perception requirements. The goal of the perception function is to maximize the effective perception power, that is, the strength or energy of the signal emitted by the drone in the direction of the target when detecting the ground target in time slot n.

[0090] P(θ m [n]) = a H (θ m [n])R w [n]a(θ m [n]),

[0091]

[0092] Among them, R w [n] is the covariance matrix, θ m [n] represents the azimuth angle from the UAV to the mth perceived target in time slot n. At the same time, a(θ m [n]) is the guidance vector of the UAV in the direction of the target at time slot n.

[0093] In order to maximize the weighted sum of communication throughput and effective sensing power, the flight trajectory and beamforming design of the UAV are jointly optimized, while ensuring that the constraints of the flight starting point, end point, maximum flight speed of the UAV and maximum transmission power of the UAV are met. The optimization problem is expressed as follows:

[0094]

[0095] st q[1]=q0

[0096] q[N]=q T

[0097]

[0098] Among them, in time slot n, R k [n] is the bit rate at user k, w k [n] is the corresponding transmit beamforming vector; θ m [n] is the pitch angle from the UAV to the first perceived target; V max is the maximum flight speed of the UAV; ρr is the regularization parameter of the sensing power; ρc is the regularization parameter of the communication rate.

[0099] In step S3, the PSO algorithm is used to solve the above optimization problem, initialize the relevant parameters of the particle swarm algorithm, construct the fitness function, that is, the objective function, and update the speed of the particle swarm according to the following formula;

[0100] v=weight×v+c1×rand( )×(pbest_position-positions)+c2×rand( )×(gbest_position-positions),

[0101] Among them, v is the velocity of the ith particle, c1 is the individual cognitive factor, c2 is the group cognitive factor, rand(*) is a randomly generated number (0,1), pbest_position is the optimal position of the ith particle, gbest_position is the global optimal position of the particle swarm, and positions is the current position of the particle.

[0102] During the solution process, the constraints are determined. If the particle swarm meets the constraints, the latest position of the particle swarm is substituted to solve the fitness value. The individual optimal solution and the group optimal solution are updated. Determine whether the maximum number of iterations has been reached. If so, exit. If not, return to the calculation.

[0103] This scheme takes into account that the weighted sum of maximizing the communication user throughput and the effective perceived power of the ground target is a nonlinear function and a non-convex function, and the original particle swarm optimization algorithm is prone to fall into the local optimum, and the obtained objective function value is small. Therefore, an improved PSO is designed and the IPSO based on the logarithmic function is used to solve the problem in order to obtain the global optimal solution. The solution steps are as follows:

[0104] S31, initialize the initial position q0 of the drone particle swarm, the initial value w0 of the beamforming particle swarm, the initial flight speed v0 of the particle swarm representing the drone position, and the initial speed vw0 of the particle swarm representing the beamforming value.

[0105] S32, substitute the initial position of the particle swarm into the objective function, the specific formula is as follows:

[0106]

[0107] In this embodiment, the initial individual values ​​of the two particle swarms are calculated, which are defined as individual optimal values, and then the individual initial optimal solutions pbest-value are compared one by one in the particle swarm to obtain the group initial optimal solution gbest-value.

[0108] S33. Introducing logarithmic functions to construct nonlinear asynchronous learning factors

[0109]

[0110] Among them, iter is the current number of iterations, iter_num is the maximum number of iterations, individual learning factor is c1, and group learning factor is c2.

[0111] In this embodiment, the speeds of the drone position particle swarm and the beamforming value particle swarm are updated respectively; and the positions of the drone position particle swarm and the beamforming value particle swarm are updated respectively.

[0112] S34, determine the following constraints and solve the fitness value.

[0113] s.tq[1]=q0

[0114] q[N]=q T

[0115]

[0116] Among them, the constraint q[1] = q0 fixes the starting position of the drone flight; the constraint q[N] = q T Fixed the end position of the drone flight; constraints Ensure that the average transmission power of the drone during the time slot does not exceed the maximum transmission power of the drone; Constraint The maximum flight distance of the drone in each time slot is limited. t Indicates the time slot length, V max Indicates the maximum flight speed of the drone.

[0117] S35. Based on the fitness values ​​of the drone position population and the beamforming population obtained in S34, the individual optimal solutions and the group optimal solutions of the two populations are updated respectively.

[0118] S36, determine whether the maximum number of iterations has been reached, if so, exit the calculation, if not, return to S32.

[0119] The flight trajectory of the UAV after the IPSO algorithm is optimized in this embodiment is shown in the figure Figure 3 As shown in the figure, the cross symbol represents the flight trajectory of the UAV after optimization by the IPSO algorithm, the circle symbol represents the position of the communication user, and the star symbol represents the position of the sensed target. When the total flight time T = 15s, the UAV flies an optimal trajectory from the starting point to the end point.

[0120] The iterative convergence curves of PSO and IPSO algorithms are shown in Figure 4 As shown in the figure, the line connected by the circle symbol represents the iterative convergence curve of the PSO algorithm, and the line connected by the diamond symbol represents the iterative convergence curve of the IPSO algorithm. By comparison, it can be seen that the objective function value of the IPSO optimization algorithm is significantly improved compared with the objective function value of the PSO algorithm. After 13 iterations, the objective function value can converge to a stable value.

[0121] The trade-off between the communication rate and the sensing power of the PSO and IPSO algorithms is as follows: Figure 5As shown in the figure, the line connected by the star symbol represents the trade-off between the communication rate and the perceived power of the IPSO algorithm, and the line connected by the circle symbol represents the trade-off between the communication rate and the perceived power of the PSO algorithm. By comparison, it can be seen that as the communication rate increases, the perceived power decreases, and vice versa, that is, each solution on the Pareto front in the figure is a trade-off between the communication rate and the perceived power.

[0122] The final transmission beam pattern obtained under the synaesthesia integrated system is as follows: Figure 6 As shown, the angle corresponding to the main peak is the pitch angle between the UAV and the ground target.

[0123] In summary, this scheme targets the NOMA-based UAV ISAC system. By jointly optimizing the design of the UAV flight trajectory and the transmit beamforming vector, the communication signal and the perception signal are weighed to maximize the weighted sum of the communication user throughput and the effective target perception power.

[0124] Under limited resource conditions, the efficient use of system resources is achieved by jointly optimizing the design of UAV flight trajectory and transmit beamforming. This optimization strategy can reduce system power consumption and cost, improve resource utilization efficiency, and provide strong support for the widespread application of ISAC systems, while ensuring that the UAV sends superimposed signals to ground users and ground targets.

[0125] By jointly optimizing the flight trajectory of the UAV and the transmit beamforming, the weighted sum of the throughput of the communication user and the effective target perception power can be maximized. By introducing the improved particle swarm optimization algorithm for solution, the disadvantage of the particle swarm optimization algorithm that it is easy to fall into the local optimum and cannot obtain the global optimal solution is solved. The simulation results show that the objective function value of the improved particle swarm algorithm is significantly improved compared with the objective function value of the original particle swarm algorithm. In addition, the performance of the UAV synaesthesia integration system using NOMA technology is better than that of the UAV synaesthesia integration system using OMA technology. By balancing the communication throughput and perception power, the weighted sum of the communication user throughput and the effective target perception power, as well as the optimal flight trajectory and transmit beamforming vector of the UAV can be obtained.

[0126] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A joint optimization method of trajectory and beamforming for unmanned aerial vehicle synaesthesia integration, characterized in that: The method steps include: S1. Constructing a UAV communication perception environment, in which the UAV uses non-orthogonal multiple access technology to send superimposed signals; S2. establishing a bit rate model and a perceived power model according to the superimposed signal, and constructing an objective function by the bit rate model and the perceived power model; S3. Set the UAV flight parameter constraints. Under the constraints, the improved particle swarm algorithm is used to solve the optimal solution of the objective function to maximize the weighted sum of communication user throughput and effective target perception power.

2. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1 is characterized in that: The UAV communication perception environment in S1 includes a NOMA-based UAV, K communication users, and M perception targets.

3. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 2 is characterized in that: In S1, the UAV sends the superimposed signal using the non-orthogonal multiple access technology, including: the UAV sends the superimposed signal to the communication user and the UAV sends the superimposed signal to the perception target.

4. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 3 is characterized in that: The superimposed signal sent by the drone to the communication user in S1 is: Among them, s i [n] is the signal sent to the i-th communication user in time slot n, is the corresponding transmit beamforming vector.

5. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 3 is characterized in that: The superposition signal sent by the drone to the sensing target in S1 is: Among them, R w ([n] is the covariance matrix, is the corresponding transmit beamforming vector, and H is the conjugate transposed identifier.

6. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1 is characterized in that: The bit rate model is the communication throughput of k in time slot n, and the specific formula is: R k [n]=min{R k→k [n], R k→k+1 [n],…,R K [n]}, Among them, R k [n] is the bit rate of user k in time slot n.

7. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1 is characterized in that: The specific formula of the perceived power model is: P(θ m [n])=a H (i m [n])R w [n]a(θ m [n]), Among them, R w [n] is the covariance matrix, θ m [n] represents the azimuth angle from the UAV to the mth perceived target in time slot n, a(θ m [n]) is the guidance vector of the UAV in the direction of the target at time slot n.

8. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1 is characterized in that: The specific formula of the objective function constructed in S2 is: Among them, R k [n] is the bit rate of user k in time slot n, w k [n] is the corresponding transmit beamforming vector, θ m [n] is the pitch angle from the UAV to the first perceived target, ρ r is the regularization parameter of the perceptual power, ρ c is the regularization parameter of the communication rate.

9. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1, characterized in that: The UAV flight parameters in S3 include the flight starting point, the end point, the maximum flight speed of the UAV and the maximum transmission power of the UAV; The specific formula for setting the UAV flight parameter constraints is: st q[1]=q0 q[N]=q T Among them, the constraint q[1] = q0 fixes the starting point of the UAV flight, and the constraint q[N] = q T Fixed the end position of the drone flight, constraints Ensure that the average transmission power of the drone in time slot n does not exceed the maximum transmission power of the drone, and constrain The maximum flight distance of the drone in each time slot is limited. t is the time slot length; V max is the maximum flight speed of the drone.

10. The method for joint optimization of trajectory and beamforming of unmanned aerial vehicle synaesthesia according to claim 1, characterized in that: The improved particle swarm algorithm in S3 introduces a logarithmic function to construct a nonlinear asynchronous learning factor, and the specific steps of using the improved particle swarm algorithm to solve the optimal solution of the objective function include: S31, initializing the position of the drone particle swarm, the beamforming particle swarm value, the flight speed of the drone position particle swarm, and the speed of the beamforming value particle swarm; S32, using the position of the drone particle swarm to calculate the individual values ​​of the drone particle swarm and the beamforming particle swarm, and defining them as individual optimal solutions; S33, introduce logarithmic function to construct nonlinear asynchronous learning factor; S34, using nonlinear asynchronous learning factors to obtain the fitness values ​​of the drone position population and the beamforming population; S35, based on the fitness values ​​of the drone position population and the beamforming population obtained in S34, the individual optimal solutions of the two populations are updated respectively, and the individual optimal solutions are compared one by one in the particle group to obtain the group optimal solution; S36, determine whether the maximum number of iterations has been reached, if so, end the solution; if not, return to S32.

Images