Perception-assisted unmanned aerial vehicle reliable communication and tracking integrated system and optimization method

Abstract

The invention belongs to the technical field of wireless communication, and provides a perception-assisted unmanned aerial vehicle reliable communication and tracking integrated system and optimization method, and the system comprises a base station and an unmanned aerial vehicle. The base station comprises a uniform linear array of Nt transmitting antennas and Nr receiving antennas; the base station predicts the motion state of the unmanned aerial vehicle to obtain a prediction state vector; transmitting a communication sensing integrated signal to the unmanned aerial vehicle through two-stage predictive beam forming, and tracking the unmanned aerial vehicle while communicating with the unmanned aerial vehicle to obtain an estimated state vector; and planning and optimizing a predicted flight path according to the predicted state vector and the estimated state vector, and controlling the flight of the unmanned aerial vehicle. According to the method, through approximate calculation of the outage probability and alternate optimization of the predicted flight path of the unmanned aerial vehicle, the communication outage capacity of the unmanned aerial vehicle can be maximized under the condition that the maximum outage probability of communication of the unmanned aerial vehicle is met, the reliability of communication service of the unmanned aerial vehicle is improved, performance analysis of the system is facilitated, and the needed calculation complexity is reduced.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wireless communication, and specifically relates to a perception-assisted UAV reliable communication and tracking integrated system and optimization method, in particular, in the scene of applying the communication and perception integrated technology to UAV, the perception-assisted UAV reliable communication and tracking integrated system and optimization method are used. BACKGROUND

[0002] Due to the superior mobility and flexibility, UAVs have been widely used in logistics, remote sensing, environmental detection, industrial monitoring, border patrol and emergency communication and many other fields. In recent years, due to the rapid development of the UAV industry and the planning of the low-altitude economy, it can be predicted that the number of UAVs will continue to grow in the future. However, with the explosive growth of the number of UAVs, signal interference and network congestion problems will be exacerbated. In this case, ensuring the communication and tracking performance of UAVs has become one of the key challenges to support UAV applications.

[0003] With the development of wireless communication technology, communication and perception integration as one of the key technologies of the upcoming sixth generation (6G) network can provide more accurate and reliable wireless coverage for UAVs, thereby reducing signal interference between multiple UAVs and enabling more UAVs to access the network simultaneously. Therefore, how to use the communication and perception integration technology to provide high-quality communication and tracking services for UAVs has become a research hotspot today.

[0004] In the existing communication and perception integrated signal processing and architecture design, perception-assisted predictive beamforming has attracted much attention due to its ability to simultaneously improve target tracking accuracy and user communication link effectiveness. This technology can significantly improve system performance in various application scenarios such as vehicle networking and UAV networking. However, existing research on perception-assisted predictive beamforming mainly focuses on improving instantaneous spectral efficiency, and lacks adequate research on how to ensure reliable communication service quality under random channel changes. In addition, the trajectory of network-connected UAVs can be partially optimized to improve overall system performance.

[0005] The patent document "UAV air computing system based on full-duplex relay and trajectory and power optimization method" (CN114499626A) uses an iterative optimization algorithm to optimize the UAV trajectory and sensor power through a full-duplex relay UAV air computing system, solves the problem of direct communication between the sensor and the base station, realizes efficient information transmission, and reduces the mean square error of the system, but cannot alternately optimize the outage probability and flight trajectory, and has high computational complexity.

[0006] The patent document "Cooperative detection method and system in unmanned aerial vehicle sensing integrated network" (CN119135250A) proposes a cooperative detection method in an unmanned aerial vehicle sensing integrated network. By constructing an unmanned aerial vehicle sensing integrated system, the EKF algorithm is used for perception data fusion, and combined with joint optimization of unmanned aerial vehicle trajectory and beamforming algorithm, the communication performance is enhanced, the multi-beam signal design is realized to simultaneously perform communication and sensing functions, but it cannot maximize the unmanned aerial vehicle communication interruption capacity under the condition of meeting the maximum interruption probability of unmanned aerial vehicle communication, and the reliability is low.

[0007] Therefore, how to optimize the trajectory of the networked unmanned aerial vehicle under the framework of perception-aided predictive beamforming to maximize the reliable communication performance (such as interruption capacity) of the system has become an important problem to be solved. SUMMARY

[0008] In view of the defects in the prior art, the purpose of the present application is to provide a perception-aided unmanned aerial vehicle reliable communication and tracking integrated system and optimization method.

[0009] The perception-aided unmanned aerial vehicle reliable communication and tracking integrated system provided by the present application comprises a base station and an unmanned aerial vehicle.

[0010] The base station comprises a uniform linear array of N t transmitting antennas and N r receiving antennas.

[0011] The unmanned aerial vehicle comprises a single receiving antenna.

[0012] The base station predicts the motion state of the unmanned aerial vehicle to obtain a predicted state vector.

[0013] The base station transmits a communication and sensing integrated signal to the unmanned aerial vehicle through two-stage predictive beamforming, communicates with the unmanned aerial vehicle while tracking the unmanned aerial vehicle to obtain an estimated state vector.

[0014] According to the predicted state vector and the estimated state vector, a predicted flight trajectory is planned and optimized, and the unmanned aerial vehicle is controlled to fly. Preferably, the flight trajectory x n of the unmanned aerial vehicle at the nth time slot is:

[0015] x n =Gx n-1 +u n +z p,n ,

[0016]

[0017] The motion control input of the base station is:

[0018]

[0019] wherein G represents a state transition matrix;

[0020] u n represents a motion control input of the base station;

[0021] z p,n represents process noise caused by a control error, with a covariance matrix of Q p ;

[0022] Q p represents a covariance matrix;

[0023] I2represents a 2x1 all-one vector;

[0024] represents a process noise intensity;

[0025] ΔTrepresents a set time slot second;

[0026] respectively represent a predicted state vector and an estimated state vector of the UAV;

[0027] x n-1 represents a flight trajectory of the n-1th time slot.

[0028] Preferably, the two-stage predictive beamforming includes a prediction stage and an estimation stage.

[0029] The prediction stage includes a first w n proportional part.

[0030] The base station generates a predicted state vector and a predicted beamforming vector by receiving echo signals reflected by the UAV, measuring an azimuth angle θ n and a distance d n , and obtaining an actual motion state of the UAV:

[0031]

[0032] wherein, represents a measurement result;

[0033] represents a measured azimuth angle;

[0034] represents a measured distance;

[0035] w n , respectively represent a perception duration ratio and a predicted azimuth angle;

[0036] respectively represent the x-axis coordinate, the y-axis coordinate of the predicted state vector;

[0037] x n represents the flight trajectory of the UAV at the nth time slot,

[0038] respectively represent the x-axis coordinate, the speed along the x-axis, the y-axis coordinate, the speed along the y-axis;

[0039] h(x n ) represents the measurement function about x n ;

[0040] a(·) represents the transmission antenna array response vector of the base station;

[0041] H represents the fixed flight height of the UAV;

[0042] z m,n represents the measurement noise vector;

[0043] z i,n , respectively represent the azimuth angle, the distance measurement noise.

[0044] The expressions of x

[0045]

[0046] wherein ρ r represents the sensing power gain coefficient;

[0047] P A represents the transmission base station power;

[0048] N sym represents the matching filter gain within the entire time slot;

[0049] σ 2 represents the received end additive Gaussian white noise power;

[0050] σ RCS represents the target radar cross section;

[0051] λ represents the carrier wavelength;

[0052] a1, a2 respectively represent the measurement capability coefficients of the base station in azimuth angle and distance.

[0053] Preferably, the estimation stage includes the 1-w n proportional part of the nth time slot.

[0054] The base station completes the measurement to obtain the predicted state vector Linearized measurement model:

[0055]

[0056] Calculate the prediction mean square error matrix:

[0057] M p,n = GM n-1 G T + Q p

[0058] Wherein, Q p Indicates the process noise covariance matrix;

[0059] M n-1 Indicates the estimated mean square error matrix at the n-1 moment.

[0060] Calculate the Kalman gain matrix:

[0061]

[0062] Wherein, Indicates the measurement noise covariance matrix.

[0063] Get the estimated state vector

[0064]

[0065] Calculate the estimated mean square error matrix:

[0066]

[0067] Wherein, I indicates the unit matrix.

[0068] The base station obtains the estimated beamforming vector based on the estimated state vector:

[0069]

[0070] Wherein, Respectively indicate the x-axis coordinate of the estimated state vector, the velocity along the x-axis, the y-axis coordinate, and the velocity along the y-axis.

[0071] Preferably, the reachable speed of the unmanned aerial vehicle in the prediction and estimation stage of the n time slot is respectively:

[0072]

[0073] β0=(λ / 4π) 2

[0074] Wherein, Indicates the reachable speed coefficient;

[0075] The superscript c indicates the conjugate transpose;

[0076] β0 represents the channel power gain at the set reference distance of 1 meter;

[0077] γ p,n , γ e,n respectively represent the signal-to-noise ratios in the prediction and estimation stages.

[0078] The outage probabilities of the UAV in the prediction and estimation stages in the nth time slot are respectively represented as:

[0079] ζ p,n = P(ξ p,n < 0), ζ e,n = P(ξ e,n < 0)

[0080] ζ p,n = ε out

[0081] ζ e,n = ε out

[0082] wherein, respectively represent the target constant signal-to-noise ratios of the UAV in the prediction and estimation stages in the nth time slot;

[0083] ξ p,n , ξ e,n respectively represent the outage probability intermediate variables in the prediction and estimation stages.

[0084] ε out represents the outage probability threshold value.

[0085] The outage capacities of the UAV in the prediction and estimation stages in the nth time slot are respectively:

[0086]

[0087] C n = w n C p,n +(1-w n )C e,n

[0088] wherein, C n represents the total outage capacity.

[0089] The perception-assisted UAV reliable communication and tracking integrated system optimization method provided by the application comprises: a base station predicts the motion state of a UAV to obtain a predicted state vector;

[0090] The communication-aware integrated signal is transmitted to the UAV through two-stage predictive beamforming, the UAV is tracked while communicating with the UAV, and an estimated state vector is obtained;

[0091] The base station plans a predicted flight trajectory according to the predicted state vector and the estimated state vector;

[0092] The predicted flight trajectory of the UAV in the nth time slot is jointly optimized Until the maximum of the overall outage capacity of the UAV in each time slot is satisfied and a set threshold constraint is met;

[0093] The two-stage predictive beamforming includes a prediction stage and an estimation stage;

[0094] respectively represent the x-axis coordinate and the y-axis coordinate of the predicted state vector.

[0095] Preferably, the base station solves the optimization problem P1 in the prediction stage of the nth time slot:

[0096]

[0097] wherein, represents the estimated flight trajectory of the UAV in the (n-1)th time slot;

[0098] respectively represent the x-axis coordinate and the y-axis coordinate of the estimated flight trajectory of the UAV in the (n-1)th time slot;

[0099] w n represents the awareness time ratio;

[0100] γ n represents the target constant signal-to-noise ratio vector;

[0101] y min represents the minimum coordinate of the y-axis of the flight area;

[0102] w min , w max represent the minimum and maximum awareness time ratios;

[0103] represents the difference between the maximum outage probability of the nth time slot and the outage probability threshold ε out ;

[0104] ζ p,n , ζ e,n respectively represent the outage probability of the UAV in the prediction stage and the estimation stage of the nth time slot;

[0105] represents the maximum target constant signal-to-noise ratio brought by the maximum beamforming gain and the minimum path loss;

[0106] denotes the achievable rate coefficient;

[0107] N t denotes the number of transmitting antennas;

[0108] H denotes the fixed flight altitude of the UAV.

[0109] Preferably, the solving optimization comprises:

[0110] the interruption probability ζ p,n and ζ e,n are approximated as:

[0111]

[0112] wherein, denote the approximated interruption probabilities of the prediction, estimation phase, respectively;

[0113] denotes the difference between the actual x-axis coordinate of the UAV in the nth time slot and the x-axis coordinate of the predicted flight trajectory;

[0114] denotes the difference between the actual x-axis coordinate of the UAV in the nth time slot and the x-axis coordinate of the estimated flight trajectory.

[0115] denote the x-axis coordinate values of the predicted state vector, the estimated state vector of the UAV, respectively;

[0116] x n denotes the x-axis coordinate value of the flight trajectory in the nth time slot;

[0117] denote the mathematical expectation of

[0118] denote the approximated interruption complementary region boundary functions of the prediction, estimation phase, respectively;

[0119] the constraints are approximated to obtain the approximated problem P2:

[0120]

[0121] wherein, C n denotes the total interruption capacity.

[0122] Preferably, the approximated problem P2 is solved by alternately optimizing sub-problems within a limited maximum number of iterations, the sub-problems comprising:

[0123]

[0124] wherein,​ Let P2 and P2.2 represent the solutions to subproblems P2.1 and P2.2 obtained in the i-th iteration, respectively.

[0125] Preferably, solving the subproblems P2.1 and P2.2 includes:

[0126] P2.1 Optimization steps: In the i-th iteration, given the solution to subproblem P2.2 w is obtained through the golden ratio search. n The search value for γ n Perform a binary search to obtain a given With w n Given the search value, determine the optimal solution to the subproblem of problem P2.1, and identify the w that satisfies the convergence condition. n and the corresponding γ n As the solution to problem P2.1 in the i-th iteration, it is denoted as and

[0127] P2.2 Optimization steps: In the i-th iteration, given the solution to subproblem P2.1 Using the continuous convex approximation method, in the m-th iteration, the objective function of subproblem P2.2 is replaced with an approximation function based on the second-order Taylor expansion:

[0128]

[0129] in, This represents the Taylor expansion point in the (m-1)th iteration;

[0130] Q represents a given positive real number that ensures convexity;

[0131] Indicates that in a given and hour, about At point gradient at;

[0132] The optimal solution was obtained using the CVX tool.

[0133] Compared with the prior art, the present invention has the following beneficial effects:

[0134] 1. This invention is aimed at the application scenarios of drones using integrated communication and sensing technology. It adopts a predictive sensing-assisted beamforming system, which can maximize the drone communication interruption capacity while meeting the maximum probability of drone communication interruption, thereby improving the reliability of drone communication services.

[0135] 2、The system optimization method provided by the application can realize approximate calculation of the outage probability and outage capacity of the unmanned aerial vehicle communication which is difficult to calculate in the system, provides an effective theoretical tool for system design, and is beneficial to performance analysis and evaluation of the system.

[0136] 3、The application can effectively and quickly complete optimization of the predicted flight trajectory of the unmanned aerial vehicle, the perception time length ratio and the target constant signal-to-noise ratio vector through approximate calculation of the outage probability and alternating optimization of the predicted flight trajectory of the unmanned aerial vehicle, and reduces the required calculation complexity. BRIEF DESCRIPTION OF DRAWINGS

[0137] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments with reference to the attached drawings:

[0138] Figure 1 The figure is a schematic diagram of an unmanned aerial vehicle reliable communication and tracking integrated system architecture;

[0139] Figure 2 The figure is a precision diagram of outage probability approximate calculation in the unmanned aerial vehicle reliable communication and tracking integrated system optimization method;

[0140] Figure 3 The figure is a convergence speed schematic diagram of alternating optimization of the approximate problem in the unmanned aerial vehicle reliable communication and tracking integrated system;

[0141] Figure 4 The figure is a schematic diagram of outage capacity comparison between the unmanned aerial vehicle reliable communication and tracking integrated system optimization method and the prior art method. DETAILED DESCRIPTION

[0142] The application will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be pointed out that those skilled in the art can make several changes and improvements without departing from the concept of the application. These all belong to the protection scope of the application.

[0143] According to the application, a perception-assisted unmanned aerial vehicle reliable communication and tracking integrated system is provided, which comprises a ground base station and a network-connected unmanned aerial vehicle. Figure 1 For example, the system comprises a ground base station and a network-connected unmanned aerial vehicle.

[0144] The base station is equipped with a uniform linear array of N t transmitting antennas and N r receiving antennas. The unmanned aerial vehicle is equipped with a single receiving antenna, and the flight height is fixed at H meters.

[0145] The base station predicts the motion state of the network-connected unmanned aerial vehicle, and controls the unmanned aerial vehicle to fly according to the predicted motion state through a wireless control link.

[0146] Specifically, a time ΔT seconds is defined as a time slot, which is short enough. In each time slot, the motion state of the UAV keeps unchanged, and the base station updates the trajectory planning and tracking results of the UAV.

[0147] The base station is located at the origin in a three-dimensional Cartesian coordinate system, and the motion state vector of the UAV in the nth time slot can be expressed as The components respectively represent the x-axis coordinate, the velocity along the x-axis, the y-axis coordinate, and the velocity along the y-axis. The flight trajectory of the UAV in the nth time slot, i.e., the motion state vector, is expressed as:

[0148] x n =Gx n-1 +u n +z p,n ,

[0149] wherein G represents a state transition matrix, u n represents the motion control input of the base station, z p,n represents the process noise caused by the control error, and is subject to a Gaussian distribution with a mean vector of 0 and a covariance matrix of Q p .

[0150] The expressions of G and Q p are respectively:

[0151]

[0152] wherein I2 represents a full 1 vector with a size of 2x1, represents the process noise intensity.

[0153] In more preferred examples, the base station realizes the trajectory optimization of the UAV by optimizing the predicted flight trajectory of the UAV. In the nth time slot, the predicted state vector and the estimated state vector of the UAV are respectively expressed as and Therefore, in the nth time slot, the motion control input of the base station is:

[0154]

[0155] x n and The relationship between x n and x can be expressed as:

[0156]

[0157] Due to factors such as control errors, there is a discrepancy between the predicted motion state and the actual motion state of the UAV. Therefore, the base station also uses perception-assisted predictive beamforming technology to transmit integrated communication and sensing signals to the UAV, tracking the UAV while communicating with it to obtain its estimated motion state. By employing predictive perception-assisted beamforming, the system can maximize the UAV communication interruption capacity while meeting the maximum probability of UAV communication interruption, thereby improving the reliability of UAV communication services.

[0158] Specifically, in the nth time slot, the base station transmits integrated communication and sensing signals to the UAV through two-stage predictive beamforming, including a prediction stage and an estimation stage.

[0159] The prediction phase includes the first w of the nth time slot. n In the proportional part, the base station generates a predicted state vector. and beamforming vector Among them, w n , These represent the sensing duration ratio and the predicted azimuth angle, respectively, and a(·) represents the base station's transmission antenna array response vector, such as... It can be represented as e represents the natural base, j represents the imaginary unit, π is the mathematical constant pi, cos(·) is the cosine function, [·] T This indicates the transpose of a vector or matrix.

[0160] Predictive beamforming vector Sufficient precision is required to ensure the drone is within the main lobe of the beam's illumination range and that the base station can receive the echo signal reflected from the drone. Using the received echo signal, the base station measures the drone's azimuth angle θ. n and distance d n The azimuth angle θ of the drone n and distance d n The relationship with the actual motion state of the drone is as follows:

[0161]

[0162] in, Indicates the measurement result, Indicates the measurement of azimuth. z represents the distance being measured. m,n Represents the measurement noise vector; z i,n , The measurement noises for azimuth and distance are respectively, and both have a mean of 0 and a variance of . The Gaussian distribution. The expressions are as follows:

[0163]

[0164] where ρ r is the perceived power gain coefficient, P A is the transmit base station power, N sym is the matched filter gain over the entire time slot, σ 2 is the received end additive white Gaussian noise power, σ RCS represents the target radar cross section, λ is the carrier wavelength, a1 and a2 represent the base station's measurement capability coefficients for azimuth angle and distance, respectively, h(x n ) represents the measurement function with respect to x n .

[0165] The estimation stage includes the remaining (1-w n ) proportion of the n th time slot, the base station completes the measurement and obtains the estimated state vector by the extended Kalman filtering method.

[0166] The predicted state vector is obtained by:

[0167] Linearize the measurement model:

[0168]

[0169] Calculate the predicted mean square error matrix:

[0170] M p,n = GM n-1 G T +Q p

[0171] where Q p represents the process noise covariance matrix, M n-1 represents the estimated mean square error matrix at the n-1 th time.

[0172] Calculate the Kalman gain matrix:

[0173]

[0174] where K represents the measurement noise covariance matrix;

[0175] Obtain the estimated state vector:

[0176]

[0177] Calculate the estimated mean square error matrix:

[0178]

[0179] where I represents the unit matrix.

[0180] Based on the estimated state vector, the base station designs its beamforming vector as:

[0181]

[0182] In more preferred examples, the achievable rates of the UAV in the prediction, estimation phases of the n-th time slot under the two-stage predictive beamforming framework are:

[0183]

[0184] where the coefficient is defined as β0=(λ / 4π) 2 denotes the channel power gain at the reference distance of 1 meter; γ p,n and γ e,n are the signal-to-noise ratios in the prediction and estimation phases, respectively, with the superscript c denoting the conjugate transpose. Accordingly, the outage probabilities of the UAV in the prediction and estimation phases of the n-th time slot are ζ p,n P(ξ p,n <0), ξ e,n P(ξ e,n <0), respectively, where ξ p,n and ξ e,n are defined as and and are the target constant signal-to-noise ratios of the UAV in the prediction and estimation phases of the n-th time slot, respectively. and satisfy ζ p,n = ε out and ζ e,n = ε out , respectively, where ε out is the outage probability threshold. Given and the outage capacities of the UAV in the prediction and estimation phases of the n-th time slot are:

[0185]

[0186] and the overall outage capacity is:

[0187] C n = w n C p,n +(1-w n )C e,n

[0188] Due to measurement errors and other factors, the estimated motion state of the UAV is different from the actual motion state. The wireless channel between the base station and the UAV is in line-of-sight conditions, and the propagation of wireless signals on the channel experiences free space path loss.

[0189] To realize reliable UAV communication, an awareness-assisted UAV reliable communication and tracking integrated system optimization method is provided according to the present application, which jointly optimizes the predicted flight trajectory of the UAV in the nth time slot The perception duration ratio w n and the target constant signal-to-noise ratio vector The constraint that the overall outage capacity of the UAV in each time slot is maximized and the outage probability is not higher than the threshold ε out provides an effective theoretical tool for system design, which is conducive to the performance analysis and evaluation of the system. Specifically, it includes:

[0190] The base station solves the following optimization problem at the beginning of the prediction stage in the nth time slot:

[0191]

[0192] wherein, represents the estimated flight trajectory of the UAV in the (n-1)th time slot, y min represents the minimum coordinate of the y-axis of the flight area, w min represents the minimum perception duration ratio, w max represents the maximum perception duration ratio, represents the difference between the maximum outage probability in the nth time slot and the outage probability threshold ε out . represents the maximum target constant signal-to-noise ratio brought by the maximum beamforming gain and the minimum path loss.

[0193] The constraint conditions (1a)-(1e) in problem P1 represent the constraints on the maximum speed of the UAV, the minimum coordinate of the y-axis of the flight area, the perception duration ratio, the maximum outage probability, and the target constant signal-to-noise ratio, respectively.

[0194] To solve problem P1, the outage probabilities ζ p,n and ζ e,n are approximated, and then solved by an algorithm based on alternating optimization, which includes the following steps:

[0195] Outage probability approximation: the outage probabilities ζ p,n and ζ e,n in the nth time slot are approximated according to the following formulas, respectively:

[0196]

[0197] wherein, and is the approximate outage probability, denotes the difference between the actual x-axis coordinate of the UAV and the x-axis coordinate of the predicted flight trajectory at the nth time slot, denotes the difference between the actual x-axis coordinate of the UAV and the x-axis coordinate of the estimated flight trajectory at the nth time slot. and denote the derivative of and denote the mathematical expectation, the function and are the approximate outage complementary region boundary functions for the prediction and estimation stages, respectively, and are expressed as:

[0198]

[0199] where erf(·) is the error function, and are the approximate outage complementary region boundary functions for the prediction and estimation stages, respectively, and are expressed as:

[0200]

[0201] where det(·) is the determinant of a matrix, and the matrix and are expressed as:

[0202]

[0203] where [·] kl denotes the (k, l)th element of a matrix.

[0204] Approximate problem alternating optimization: the constraint (1d) in the problem P1 is approximated as:

[0205]

[0206] and the approximate problem P2 is obtained, which is expressed as:

[0207]

[0208] The problem P2 is solved by alternating optimization of the following two sub-problems within a limited maximum number of iterations:

[0209]

[0210] where and represent the solutions of the sub-problems P2.1 and P2.2 obtained in the ith iteration, respectively.

[0211] Specifically, by approximating the interruption probability and alternately optimizing the predicted flight trajectory of the UAV, the optimization of the predicted flight trajectory, perception duration ratio, and target constant signal-to-noise ratio vector can be completed effectively and quickly, reducing the required computational complexity. The solution methods for subproblems P2.1 and P2.2 include:

[0212] Given Optimize (w) n ,γ n ): In the i-th iteration, given the solution to subproblem P2.2 First, w is obtained through the golden ratio search. n The search value, and then through γ n Perform a binary search to obtain a given With w n Given the search value, the optimal solution to the subproblem of problem P2.1 is obtained, ultimately yielding w that satisfies the convergence condition. n and the corresponding γ n Let be the solution to problem P2.1 in the i-th iteration, denoted as . and

[0213] Given (w) i ,γ i )optimization In the i-th iteration, given the solution to subproblem P2.1 The solution to problem P2.2 is obtained using a continuous convex approximation method. Specifically, in the m-th iteration, the objective function of subproblem P2.2 is replaced with the following approximation function based on a second-order Taylor expansion:

[0214]

[0215] in, Let Q represent the Taylor expansion point in the (m-1)th iteration, and let Q be a given positive real number that ensures the convexity of the above expression. Indicates that in a given and hour, about At point The gradient at point P2.2. After replacing the objective function with the above equation, P2.2 becomes a convex optimization problem, and the optimal solution is obtained using the CVX tool (Convex Optimization Toolbox).

[0216] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A perception-assisted unmanned aerial vehicle (UAV) reliable communication and tracking integrated system, characterized in that, include: Base stations and drones; The base station includes N t One transmitting antenna and N r A uniform linear array of receiving antennas; The drone includes a single receiving antenna; The base station predicts the motion state of the drone and obtains the predicted state vector; By transmitting integrated communication and sensing signals to the UAV through two-stage predictive beamforming, the UAV is tracked while communicating with it, and the estimated state vector is obtained. The flight trajectory is planned and optimized based on the predicted state vector and the estimated state vector to control the flight of the UAV.

2. The perception-assisted UAV reliable communication and tracking integrated system according to claim 1, characterized in that, The flight trajectory x of the UAV in the nth time slot n for: x n =Gx n-1 +u n +z p,n , The motion control input for the base station is: Where G represents the state transition matrix; u n This indicates the motion control input of the base station; z p,n The covariance matrix Q represents the variance caused by control error. p Process noise; Q p Represent the covariance matrix; I2 represents a vector of size 2×1 consisting entirely of 1s; Indicates the intensity of process noise; ΔT represents the set number of seconds in the time slot; These represent the predicted state vector and the estimated state vector of the UAV, respectively. x n-1 This represents the flight trajectory in the (n-1)th time slot.

3. The perception-assisted UAV reliable communication and tracking integrated system according to claim 1, characterized in that, The two-stage predictive beamforming includes a prediction stage and an estimation stage; The prediction phase, the first w of the nth time slot n Proportional section; The base station generates a predicted state vector. and predicted beamforming vector The azimuth angle θ is measured by receiving the echo signal reflected from the drone. n and distance d n To obtain the actual motion state of the drone: in, Indicates the measurement result; Indicates the measured azimuth angle; Indicates the distance being measured; w n , These represent the sensing duration ratio and the predicted azimuth angle, respectively. These represent the x-axis coordinates and y-axis coordinates of the predicted state vector, respectively. x n This represents the flight trajectory of the drone in the nth time slot. These represent the x-axis coordinate, velocity along the x-axis, y-axis coordinate, and velocity along the y-axis, respectively. h(x n ) represents x n The measurement function; a(·) represents the transmission antenna array response vector of the base station; H represents the fixed flight altitude of the drone; z m,n Represents the measurement noise vector; z i,n , These represent the measurement noise for azimuth and distance, respectively. The expressions for i = 1 and 2 are as follows: Where, ρ r Indicates the perceived power gain coefficient; P A Indicates the power of the transmitting base station; N sym This represents the matched filter gain over the entire time slot; σ 2 This represents the additive white Gaussian noise power at the receiving end; σ RCS Indicates the target's radar cross-section; λ represents the carrier wavelength; a1 and a2 represent the measurement capability coefficients of the base station for azimuth and distance, respectively.

4. The perception-assisted UAV reliable communication and tracking integrated system according to claim 3, characterized in that, The estimation stage includes the 1-w of the nth time slot. n Proportional section; The base station completes the measurement and obtains the predicted state vector. Linearized measurement model: Calculate the prediction mean square error matrix: M p,n =GM n-1 G T +Q p Among them, Q p Represents the process noise covariance matrix; M n-1 This represents the estimated mean square error matrix at time n-1; Calculate the Kalman gain matrix: in, Represents the measurement noise covariance matrix; Obtain the estimated state vector Calculate the estimated mean square error matrix: Where I represents the identity matrix; The base station obtains the estimated beamforming vector based on the estimated state vector: in, These represent the x-axis coordinate, velocity along the x-axis, y-axis coordinate, and velocity along the y-axis of the estimated state vector, respectively.

5. The perception-assisted UAV reliable communication and tracking integrated system according to claim 4, characterized in that, The achievable rates of the UAV in the prediction and estimation phases of the nth time slot are as follows: β0=(λ / 4π) 2 in, Represents the reachability rate coefficient; The superscript 'c' indicates the conjugate transpose; β0 represents the channel power gain at a reference distance of 1 meter; γ p,n γ e,n These represent the signal-to-noise ratios in the prediction and estimation stages, respectively. The interruption probabilities of the UAV in the prediction and estimation stages of the nth time slot are expressed as follows: g p,n =P(ξ p,n <0),ζ e,n =P(ξ e,n <0) in, These represent the constant signal-to-noise ratio of the target during the prediction and estimation stages of the UAV in the nth time slot, respectively. ξ p,n ξ e,n These represent intermediate variables representing the probability of interruption during the prediction and estimation stages, respectively. ε out Indicates the interruption probability threshold; The interruption capacities of the UAV in the prediction and estimation phases of the nth time slot are as follows: C n =w n C p,n +(1-w n )C e,n Among them, C n This indicates the total interruption capacity.

6. A method for optimizing a perception-assisted unmanned aerial vehicle (UAV) reliable communication and tracking integrated system, used to optimize the perception-assisted UAV reliable communication and tracking integrated system as described in any one of claims 1-5, characterized in that, include: The base station predicts the motion state of the drone and obtains the predicted state vector; By transmitting integrated communication and sensing signals to the UAV through two-stage predictive beamforming, the UAV is tracked while communicating with it, and the estimated state vector is obtained. The base station plans and predicts the flight trajectory based on the predicted state vector and the estimated state vector; Jointly optimize the predicted flight trajectory of the UAV in the nth time slot Until the overall interruption capacity of the drone is maximized in each time slot and the set threshold constraint is met; The two-stage predictive beamforming includes a prediction stage and an estimation stage; These represent the x-axis coordinates and y-axis coordinates of the predicted state vector, respectively.

7. The optimization method for a perception-assisted UAV reliable communication and tracking integrated system according to claim 6, characterized in that, The base station solves optimization problem P1 during the prediction phase of the nth time slot: in, This represents the estimated flight trajectory of the UAV within the (n-1)th time slot; These represent the x-axis and y-axis coordinates of the estimated flight trajectory of the UAV within the (n-1)th time slot, respectively. w n Indicates the ratio of perceived duration; γ n Represents the target constant signal-to-noise ratio vector; y min This represents the minimum y-coordinate of the flight area; w min w max This represents the ratio of minimum to maximum perception duration; This represents the maximum interruption probability and the interruption probability threshold ε for the nth time slot. out difference; ζ p,n ,ζ e,n These represent the interruption probabilities of the UAV during the prediction and estimation phases in the nth time slot, respectively. This represents the maximum target constant signal-to-noise ratio resulting from the maximum beamforming gain and minimum path loss; Represents the reachability rate coefficient; N t Indicates the number of transmitting antennas; H indicates the fixed flight altitude of the drone.

8. The optimization method for a perception-assisted UAV reliable communication and tracking integrated system according to claim 7, characterized in that, The solution optimization includes: For the interruption probability ζ p,n and ζ e,n Perform approximate calculations: in, These represent the interruption probabilities in the approximate prediction and estimation stages, respectively. This represents the difference between the actual x-axis coordinate and the predicted x-axis coordinate of the UAV in the nth time slot; This represents the difference between the actual x-axis coordinate and the estimated x-axis coordinate of the UAV in the nth time slot. These represent the x-axis coordinates of the predicted state vector and the estimated state vector of the UAV, respectively. x n The x-axis coordinate value represents the flight trajectory in the nth time slot; They represent respectively to Find the expected value; These represent the approximate boundary functions of the interrupted complementary regions during the prediction and estimation stages, respectively. Approximating the constraints, we obtain approximate problem P2: Among them, C n This indicates the total interruption capacity.

9. The optimization method for a perception-assisted UAV reliable communication and tracking integrated system according to claim 8, characterized in that, The approximation problem P2 is solved by alternately optimizing subproblems within a finite maximum number of iterations. These subproblems include: in, Let P2 and P2.2 represent the solutions to subproblems P2.1 and P2.2 obtained in the i-th iteration, respectively.

10. The optimization method for a perception-assisted UAV reliable communication and tracking integrated system according to claim 9, characterized in that, Solving the subproblems P2.1 and P2.2 includes: P2.1 Optimization steps: In the i-th iteration, given the solution to subproblem P2.2 w is obtained through the golden ratio search. n The search value for γ n Perform a binary search to obtain a given with w n Given the search value, determine the optimal solution to the subproblem of problem P2.1, and identify the w that satisfies the convergence condition. n and the corresponding γ n As the solution to problem P2.1 in the i-th iteration, it is denoted as and γ i * ; P2.2 Optimization steps: In the i-th iteration, given the solution (w) to subproblem P2.1 i * ,γ i * By using the continuous convex approximation method, in the m-th iteration, the objective function of subproblem P2.2 is replaced with an approximation function based on second-order Taylor expansion: in, This represents the Taylor expansion point in the (m-1)th iteration; Q represents a given positive real number that ensures convexity; Indicates that given w i * and γ i * hour, about At point gradient at; The optimal solution was obtained using the CVX tool.

Images