Energy consumption minimization method and device in IRS-assisted multi-antenna UAV video transmission system

Abstract

The present application relates to an energy consumption minimization method and device for a multi-antenna unmanned aerial vehicles (UAVs) video transmission system assisted by intelligent reflecting surfaces (IRS), which utilizes a network model, a communication model, and an energy consumption model constructed by an aerial video surveillance system based on a UAV; a reconfigurable UAV-IRS-GU link is constructed by deploying an IRS to enhance the quality of air-to-ground communication; a model (P1) for minimizing the total energy consumption of the UAV is constructed based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle; the model (P1) for minimizing the total energy consumption of the UAV is optimized through a first-stage algorithm and a second-stage algorithm to obtain a suboptimal solution, wherein the first-stage algorithm adopts a path discretization method, an alternating optimization method, and a continuous convex approximation algorithm (SCA) to minimize the propulsion energy consumption of the UAV; and the second-stage algorithm adopts a double-loop iterative algorithm based on a penalty-based block coordinate descent technique (P-BCD) to minimize the communication energy consumption of the UAV.

Description

Technical Field

[0001] The present application relates to the field of wireless communications, and in particular to a method and apparatus for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system. Background Art

[0002] Unmanned aerial vehicles (UAVs) have gained widespread popularity across various fields due to their high flexibility and low cost. Camera-equipped UAVs are widely applicable to a wide range of aerial surveillance applications, such as traffic monitoring, disaster management, and infrastructure inspection. Compared to traditional ground-based video surveillance systems with fixed cameras, aerial surveillance can expand the monitoring range, continuously track moving targets, and adjust positions for flexible obstacle avoidance. Unlike other data types, video surveillance requires higher quality and smoother video streaming, and has strict requirements on latency and data rate. However, during UAV flight, the quality of the air-to-ground (A2G) channel can fluctuate dramatically with changes in the dynamic environment, thereby degrading network performance and increasing latency.

[0003] Furthermore, the Line-of-Sight (LoS) link between UAVs and ground users (GUs) is easily blocked by obstacles in urban environments, resulting in a significant degradation in video transmission performance. Therefore, ensuring smooth video transmission for urban users while maintaining quality of service (QoS) becomes a key challenge. Summary of the Invention

[0004] This application provides a method and apparatus for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system, maximizing the minimum computational throughput of all ground equipment. The technical solution of this application is as follows:

[0005] According to a first aspect of an embodiment of the present application, a method for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system is provided, the method comprising:

[0006] S1: Use the UAV-based aerial video surveillance system to build a network model, communication model, and energy consumption model;

[0007] The UAV-based aerial video surveillance system includes K single-antenna ground users GUs, a multi-antenna rotor UAV, and a building with An intelligent reflecting surface IRS with 1 reflecting unit is used to establish a UAV-IRS-GU link;

[0008] S2: Based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle, a model (P1) is constructed to minimize the total energy consumption of the UAV, where the total energy consumption of the UAV includes propulsion energy consumption and communication energy consumption;

[0009] S3: The minimization of the total energy consumption model of the UAV (P1) is optimized through the first-stage algorithm and the second-stage algorithm, wherein the first-stage algorithm adopts the path discretization method, the alternating optimization method and the continuous convex approximation algorithm (SCA) to minimize the propulsion energy consumption of the UAV; the second-stage algorithm adopts the double-loop iterative algorithm based on the penalty-based block coordinate descent (P-BCD) technology to minimize the communication energy consumption of the UAV.

[0010] Optionally, a network model is constructed based on a UAV aerial video surveillance system, specifically including:

[0011] The UAV is set to fly at a fixed altitude H0, and the starting and ending positions of the UAV are determined to be q I and q F ; Among them, the UAV flight time is represented by T, and at time t∈[0,T], the three-dimensional position and velocity of the UAV can be represented by and express;

[0012] The IRS is deployed on the top of the building to assist UAV in serving GUs. For each reflection unit n∈{1,…,N} of the IRS, the reflection amplitude β n ∈[0,1] and reflection angle θ n ∈[0,2π) are independently adjusted by the intelligent controller of the IRS to achieve passive beamforming gain.

[0013] Optionally, a UAV-based aerial video surveillance system is used to build a communication model, specifically including:

[0014] Assuming that the channel between GU k and UAV follows Rayleigh fading, the corresponding channel coefficient can be expressed as Where α1 represents the path loss index of the channel, in represents the carrier wavelength, c is the speed of light, f c is the carrier frequency;

[0015] Setting variables It follows an independent and identically distributed circularly symmetric complex Gaussian distribution with a mean of zero and a variance of 1, which is expressed as

[0016] Optionally, an energy consumption model is constructed using a UAV-based aerial video surveillance system, specifically including:

[0017] Let the total energy consumption of the UAV include communication energy consumption and propulsion energy consumption, then we have Among them, P com (t) and P fly (t) represent the communication power and propulsion power of the UAV at time t, respectively.

[0018] Optionally, it is characterized in that

[0019] The communication power of the UAV at time t is expressed as:

[0020]

[0021] The propulsion power of the UAV at time t is expressed as:

[0022]

[0023] Among them, k represents the kth ground user, w k (t) represents the beamforming vector of the kth ground user, P0 represents the blade power of the UAV in the hovering state; Ω represents the angular velocity of the UAV rotor blade; r represents the UAV rotor radius; P i represents the induced power of the drone in the hovering state; v0 represents the average rotor induced speed of the drone in the hovering state; d0 represents the fuselage drag; ρ represents the air density; s represents the rotor firmness of the drone; A r Characterizing the UAV rotor disk area.

[0024] Optionally, the method of constructing a model (P1) for minimizing the total energy consumption of the UAV based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle specifically includes:

[0025] Jointly optimize the UAV's 3D trajectory {q(t)}, flight time T, beamforming vector {w k (t)} and IRS reflection angle {u(t)}, the optimization problem model is described as:

[0026] (P1):

[0027] Constraints include:

[0028]

[0029]

[0030] q(0)=q I,q(T)=q F ,

[0031] ||v(t)||≤V max ,t∈[0,T].

[0032] Where u(t) represents the reflection angle vector of IRS at time t; u n (t) represents the nth element in u(t); B is the bandwidth (MHz); SINR k (t) represents the SINR of the kth terrestrial user at time t; β k represents the maximum tolerable delay of the kth ground user; v(t) represents the speed of the UAV at time t (m / s); V max Indicates the maximum speed of the UAV (m / s).

[0033] Optionally, the For limiting the range of reflection angle, the For giving the QoS requirement of video transmission, the ||v(t)||≤V max ,t∈[0,T] is used to limit the speed of the drone.

[0034] Optionally, the minimization of the total energy consumption model of the drone (P1) is optimized by a first-stage algorithm and a second-stage algorithm, wherein the first-stage algorithm specifically includes:

[0035] Divide the drone path into L line segments connected by L+1 path points;

[0036] make and Based on the path discretization method, the problem of minimizing the propulsion energy of L connecting line segments is expressed as model (P2);

[0037] The optimization variables are divided into and

[0038] While fixing the variables in one block, the variables in the other block are optimized alternately until convergence, or until the reduction ratio of the propulsion energy consumption is lower than the threshold ∈ 1.

[0039] According to a second aspect of an embodiment of the present application, a device for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system is provided, comprising:

[0040] processor;

[0041] a memory for storing instructions executable by the processor;

[0042] The processor is configured to execute the instructions to implement the energy consumption minimization method in the IRS-assisted multi-antenna UAV video transmission system described in any one of the methods mentioned in the first aspect above.

[0043] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0044] The present application relates to a method and apparatus for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system. A network model, a communication model, and an energy consumption model are constructed using a UAV-based aerial video surveillance system. An IRS is deployed to construct a reconfigurable UAV-IRS-GU link to enhance air-to-ground communication quality. A model (P1) for minimizing the total energy consumption of the UAV is constructed based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle. This problem is modeled as a non-convex optimization problem with tightly coupled variables. The model (P1) for minimizing the total energy consumption of the UAV is then optimized using a first-stage algorithm and a second-stage algorithm to obtain a suboptimal solution. The first-stage algorithm employs a path discretization method, an alternating optimization method, and a continuous convex approximation algorithm to minimize the propulsion energy consumption of the UAV. The second-stage algorithm employs a double-loop iterative algorithm based on a penalty-based block coordinate descent technique to minimize the communication energy consumption of the UAV.

[0045] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0047] Figure 1 1 is a schematic diagram illustrating an architecture for establishing communication between a drone and a ground device according to an exemplary embodiment;

[0048] Figure 2 1 is a flow chart illustrating a method for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system according to an exemplary embodiment;

[0049] Figure 3 The figure is a schematic diagram of a UAV trajectory according to an exemplary embodiment.

[0050] Symbol definition:

[0051] T: UAV working time; t: tth moment;

[0052] q(t): 3D position of the UAV at time t;

[0053] K: the number of ground users; k: the kth ground user;

[0054] w k (t): beamforming vector of the kth terrestrial user;

[0055] u(t): reflection angle vector of IRS at time t;

[0056] u n (t): the nth element in u(t);

[0057] P com (t): UAV communication power at time t (W); P fly (t): UAV propulsion power at time t (W);

[0058] B: bandwidth (MHz);

[0059] SINR k (t): SINR of the kth terrestrial user at time t;

[0060] β k : the maximum tolerable delay of the kth ground user;

[0061] q I : starting position of UAV; q F : The final position of the UAV;

[0062] v(t): UAV velocity at time t (m / s); V max : UAV maximum speed (m / s). DETAILED DESCRIPTION

[0063] In order to enable ordinary people in the art to better understand the technical solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0064] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or precedence. Therefore, the terms used in this manner are interchangeable where appropriate, and the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Instead, they are merely examples of apparatus and methods consistent with certain aspects of this application, as detailed in the appended claims.

[0065] The application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Ordinary technicians in this field can know that with the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0066] Unmanned aerial vehicles (UAVs) provide powerful technical support for various aerial surveillance applications, such as traffic monitoring, disaster management, and infrastructure inspection. Related technologies use dynamic adaptive streaming over HTTP (DASH) and adaptive bitrate streaming (ABR) for UAV video streaming, selecting the appropriate bitrate and resolution for the video based on channel quality. However, most existing UAV video streaming work uses a single-antenna UAV, which can only provide video transmission services to a single ground user (GU) at a time, resulting in limited system throughput and high latency.

[0067] To further improve transmission performance, multi-antenna UAVs are introduced, and beamforming technology is used to enable the UAV to simultaneously provide transmission services to a group of GUs. Specifically, precise beamforming design can achieve spatial selectivity in signal transmission, thereby eliminating inter-user interference for multiple GUs using the same time / frequency resources.

[0068] However, the Line-of-Sight (LoS) link between UAVs and GUs is easily blocked by obstacles in urban environments, resulting in a serious degradation of video transmission performance. Therefore, how to provide smooth video transmission for urban users while meeting the Quality of Service (QoS) becomes a key challenge.

[0069] This application deploys intelligent reflecting surfaces (IRS) to build a reconfigurable UAV-IRS-GU link to enhance the quality of air-to-ground (A2G) communications.

[0070] In some embodiments, the IRS, a digitally controlled metasurface, consists of an intelligent controller and a large number of reconfigurable reflective units. The intelligent controller independently adjusts the reflection angle and amplitude of each reflective unit to the incident signal, dynamically reshaping the signal propagation environment based on specific requirements, such as enhancing or suppressing the incident signal. Placing the IRS in a highly visible location can establish a new UAV-IRS-GU link, further improving the quality of A2G communications.

[0071] For ease of explanation, the symbols involved in this application are first explained. This application adopts standard mathematical symbol conventions, and scalars, matrices and vectors are represented by italics, bold uppercase and bold lowercase letters respectively. and denote N×M real-valued and complex-valued matrix spaces respectively. For the time-varying function x(t), represents its first-order derivative. For complex-valued vector y, y H represents its conjugate transpose, diag(y) represents its diagonal matrix, where each element on the diagonal is an element in y. For a general matrix M, M(i,j) represents its element in the i-th row and j-th column. represents the Kronecker product of matrices M and N. |·| and ||·|| represent the absolute value and vector norm, respectively. Re{·} represents the real part of a complex number, [·] T Represents matrix transpose.

[0072] like Figure 2 As shown, the present application provides an energy consumption minimization method in an IRS-assisted multi-antenna UAV video transmission system, specifically including:

[0073] S1: Use the UAV-based aerial video surveillance system to build network models, communication models, and energy consumption models.

[0074] Network Model

[0075] In some embodiments of the present application, a system model is constructed for a UAV mobile edge computing network. Figure 1 As shown, the present application provides a UAV aerial video surveillance system with IRS assistance. The system consists of K single antenna GUs and a The UAV is equipped with a camera and flies from the starting point to the end point, capturing video during the flight and transmitting it to all GUs. The starting point / end point can correspond to the pre-mission / post-mission location or charging station. Since the LoS link between the UAV and the GU may be blocked by ground obstacles in a complex urban environment, a camera is deployed on the buildings near the GU. The IRS of the UAV is connected to the GU, thereby establishing a UAV-IRS-GU link to enhance the A2G communication by reconstructing the propagation environment.

[0076] To avoid the energy consumption caused by frequent takeoff and landing of the aircraft, it is assumed that the UAV flies at a fixed altitude H0. The UAV flight time is represented by T. At time t∈[0,T], the three-dimensional position and velocity of the UAV can be expressed as and The starting and ending positions of the UAV are predetermined and are represented by q I and q F express.

[0077] In some embodiments, to avoid signal blocking, IRS is deployed on the top of the building to help UAV better serve GUs. For each reflection unit n∈{1,…,N}, the reflection amplitude β n ∈[0,1] and reflection angle θ n ∈[0,2π) can be independently adjusted by the intelligent controller of the IRS to achieve passive beamforming gain.

[0078] Since maximizing the reflection amplitude is very helpful to improve the received signal strength at the user, this application sets β = 1 and only optimizes θ = [θ1,…,θ N The reflection angle vector of IRS at time t is represented by u(t)=[u1(t),…,u N (t)] H Indicates that The reflection angle range can be converted into an equivalent unit modulus form, that is, In addition, the distance variation between different IRS reflective elements and UAV or GU in this application can be ignored. Assume that all IRS reflective elements are at the same position, that is, the center coordinate of IRS I = [x r ,y r ,z r ] T Then the horizontal and three-dimensional distances between UAV and IRS at time t are respectively given by and d u,k (t)=||q(t)-I||.

[0079] Assume that the position of GUk∈{1,…,K} is g k =[x k ,y k ,0] T , can be obtained by and d u,k (t)=||q(t)-g k|| Calculate the horizontal and three-dimensional distances between UAV and GUk at time t. Similarly, we can also use and d r,k =||Ig k ||, calculate the horizontal and three-dimensional distances between IRS and GUk respectively.

[0080] communication model

[0081] In some embodiments of the present application, a communication model is also constructed for the drone mobile edge computing network.

[0082] In some embodiments, IRS reflection angle optimization requires accurate channel state information (CSI) between the IRS, UAV, and GU. It is assumed that the UAV is fully aware of the corresponding CSI and sends the resource allocation and reflection angle optimization results to the IRS controller and GUs after executing the algorithm. Denotes the equivalent channel coefficient between UAV and IRS, using and are the equivalent channel coefficients in the IRS-GU and UAV-GU links corresponding to GU k. Since the path attenuation is very significant, the signals reflected multiple times by the IRS are ignored.

[0083] In urban environments, the LoS link between GU and UAV is always blocked by obstacles such as buildings and trees. Therefore, assuming that the channel between GU k and UAV follows Rayleigh fading, the corresponding channel coefficient can be expressed as Where α1 represents the path loss index of the channel, in represents the carrier wavelength, c is the speed of light, f c is the carrier frequency. Assume that the variable Following the independent and identically distributed circularly symmetric complex Gaussian (CSCG) distribution, its mean is zero and its variance is 1, it can be expressed as

[0084] Since drones usually fly at high altitudes and IRSs are always deployed on top of tall buildings, it is assumed that there is a pure LoS channel between the drone and the IRS. Represents, where α2 represents the path loss exponent of the UAV-IRS channel,

[0085]

[0086] In formula (1), d rrepresents the IRS unit spacing, d u represents the UAV antenna spacing, φ r,u and θ r,u are the horizontal and vertical angles of arrival (AoA) between the UAV and the IRS, respectively. Similarly, φ u,r and θ u,r Respectively represent the horizontal and vertical angle of departure (AoD) from the UAV to the IRS. Specifically,

[0087] In some embodiments, the channel between IRS and GU includes two parts: LoS and NLoS. The Rician channel model is used to describe h r,k Modeling, that is

[0088]

[0089] Where α3 represents the loss index of the channel, is the determined LoS component. is the NLos component of random scattering, and G is the Rician factor. Specifically, It can be expressed as

[0090]

[0091] where φ r,k and θ r,k denote the horizontal and vertical AoD from IRS to GU k, respectively, on the other hand,

[0092] The effective channel between the UAV and GU k consists of two parts, namely the direct UAV-GU link and the reflected UAV-IRS-GU link. Therefore, the equivalent channel coefficient between the UAV and GU k at time t can be expressed as in Linear precoding is used on the UAV to assign a dedicated beamforming vector to each GU. The complex baseband signal transmitted by the UAV at time t can be expressed as in is the precoding vector, is the transmitted signal of user k. Then, at time t, the total signal received by user k from the two links can be expressed as

[0093]

[0094] in, Indicates that the power at the receiving end GUk is Therefore, the signal interference plus noise ratio of GUk at time t can be given by the following formula:

[0095]

[0096] Therefore, the achievable rate of user k at time t can be calculated by Blog2(1+SINR k (t)) is calculated, where B represents the bandwidth. Accordingly, the delay of transmitting 1 bit of data from the UAV to user k can be expressed as To ensure the QoS requirements of video transmission, the transmission delay of each GU is subject to the maximum tolerable delay β k The limitation, that is

[0097]

[0098] Table II Physical meaning and values ​​of UAV propulsion parameters

[0099] parameter Physical meaning Simulation values <![CDATA[P0]]> Blade power in hovering state 79.86 watts Ω Rotor blade angular velocity 300 rad / s r rotor radius 0.4 meters <![CDATA[P i ]]> Induced power in hovering state 88.63 watts <![CDATA[v0]]> Average rotor induced speed in hovering state 4.03 rad / s <![CDATA[d0]]> fuselage resistance 0.3 ρ Air density 1.225 kg / m3 s Rotor strength 0.05 cubic meters <![CDATA[A r ]]> Rotor disc area 0.503 square meters

[0100] Energy consumption model

[0101] Due to the limited onboard energy, energy consumption is a key factor limiting UAV communication. UAV energy consumption is mainly used for communication and flight propulsion, which can be modeled as Among them, P com (t) and P fly (t) represent the communication and propulsion power of the UAV at time t. Specifically, the communication power of the UAV at time t can be expressed as:

[0102]

[0103] The propulsion power of the rotorcraft at time t can be calculated by the following formula:

[0104]

[0105] Among them, Equation (8) is the sum of blade drag power, waste drag power and induced power. The physical meaning and simulation values ​​of each parameter are summarized in Table II.

[0106] Step 2: Based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle, a model (P1) for minimizing the total energy consumption of the UAV is constructed, where the total energy consumption of the UAV includes propulsion energy consumption and communication energy consumption.

[0107] To ensure that from the initial point qI To the end point q F Energy-saving aerial video transmission, this application aims to meet the QoS constraints of video transmission while minimizing the total energy consumption of the drone. To this end, the drone trajectory {q(t)}, flight time T, beamforming vector {w k (t)} and IRS reflection angle {u(t)}. The optimization problem model can be described as:

[0108] (P1):

[0109]

[0110]

[0111] q(0)=q I ,q(T)=q F , (11)

[0112] ||v(t)||≤V max ,t∈[0,T]. (12)

[0113] In some embodiments, constraint (9) limits the reflection angle range, constraint (10) gives the QoS requirements for video transmission, and constraint (12) limits the speed of the drone.

[0114] This application proposes an IRS-assisted UAV aerial video surveillance system. A multi-antenna UAV can simultaneously transmit video to multiple ground users with the help of an IRS. To ensure sustainable and low-latency aerial surveillance, the UAV trajectory, flight time, transmit beamforming vector, and IRS reflection angle are jointly optimized under the QoS requirements of video streaming, with the goal of minimizing the total UAV energy consumption. This problem is modeled as a non-convex optimization problem with tightly coupled variables.

[0115] S3, the minimization of the total energy consumption model of the UAV (P1) is optimized through the first-stage algorithm and the second-stage algorithm, wherein the first-stage algorithm adopts the path discretization algorithm, the alternating optimization algorithm and the SCA algorithm to minimize the propulsion energy consumption of the UAV; the second-stage algorithm adopts the double-loop iterative algorithm of the P-BCD technology to minimize the communication energy consumption of the UAV.

[0116] Minimizing the total energy consumption model of the UAV (P1) is a non-convex optimization problem. Specifically, first, (P1) contains countless optimization variables and an unknown integral upper limit T; second, constraints (9) and (10) are non-convex constraints, and the objective function is a non-convex function; third, the UAV trajectory, IRS reflection angle, and transmission beamforming vector are all in the SINR. k(t) is tightly coupled. To solve the above problem and effectively find the suboptimal solution of the non-convex problem (P1), this application proposes a two-stage algorithm, such as Figure 3 shown.

[0117] Typically, drone propulsion consumes the vast majority of its total energy consumption. For example, drone propulsion energy is typically measured in kilojoules, while communication energy is measured only in joules. Furthermore, the deployment of an IRS can further reduce communication-related energy consumption.

[0118] Therefore, in the first stage, the UAV trajectory {q(t)} and the UAV working time T are optimized to minimize the propulsion energy consumption. Given {q(t)} and T obtained in the first stage, the second stage optimizes the beamforming vector {w k (t)} and reflection angle {u(t)} to minimize the energy consumption associated with UAV communication.

[0119] Phase 1: Optimizing drone trajectories and operating times

[0120] 1) Problem Reconstruction: The first stage mainly deals with the continuous variable {q(t)} and the unknown flight time T. This paper adopts the path discretization method to discretize the UAV trajectory into a limited number of line segments with different durations. Specifically, the UAV path is divided into L line segments connected by L+1 path points, and represents the coordinates of the drone at the beginning of the lth line segment. Assume ||q l+1 -q l ||≤d max , where d max Set to be small enough so that the position and velocity of the UAV can be considered to remain unchanged within the lth line segment. max Next, choose a large enough L so that in Is the upper limit of the total flight distance of the UAV. l represents the length of time the drone spends on line segment l. The total flight time of the drone can be calculated by Calculation, the speed of the UAV on the lth line segment can be approximately expressed as Accordingly, the propulsion energy consumption of the UAV on the lth line segment can be approximated as

[0121]

[0122] make and Based on the path discretization method, the propulsion energy minimization problem model of the UAV on l line segments can be expressed as

[0123] (P2):

[0124]

[0125] q1=q I ,q L+1 =q F , (15)

[0126]

[0127] ||q l+1 -q l ||≤d max ,l=1,…,L. (17)

[0128] Since the objective function is a non-convex problem, it is difficult to obtain the global optimal solution of (P2). Therefore, based on the BCD method technology, an effective algorithm is designed to find a high-quality suboptimal solution for (P2).

[0129] 2) Solve (P2): Due to the existence of (14), by Any two groups of variables can determine the value of the remaining group of variables. Therefore, the BCD method can be used to divide the optimization variables into two blocks, namely and Subsequently, the variables of one block can be fixed while the variables of the other block are alternately optimized until convergence, thereby minimizing the propulsion energy consumption in (P2).

[0130] For the first sub-problem model, in a given time Next optimization path point and speed To solve the non-convex objective function, auxiliary variables are introduced in And deduce and Therefore, the first sub-problem model of (P2) can be expressed as

[0131] (P3):

[0132] st(14)-(17),

[0133]

[0134] In this application, the first subproblem model (P3) always obtains the optimal solution when the equation in constraint (18) holds, otherwise it can always be reduced without affecting the target value and other constraints. l To satisfy their equality. Therefore, given Under the condition of , the first subproblem model (P3) is equivalent to the boost energy minimization problem model (P2).

[0135] Since the left-hand-side (LHS) of constraint (18) is non-concave, (18) is still a non-convex constraint. l || 2 Using the first-order Taylor approximation, we can get the linear lower bound of LHS, that is, use Replacing the LHS of (18), the propulsion energy consumption minimization problem model can be approximated as the first subproblem optimization model:

[0136] (P4):

[0137] st(14)-(17),

[0138]

[0139] Since the LHS in (18) is lower bounded by By substitution, any feasible solution of (P4) is also feasible for (P3). The objective function and constraints of (P4) satisfy the standard convex optimization form and can be efficiently solved by solvers such as CVX.

[0140] The second sub-problem model is given a speed In the case of and waypoints It can be expressed as:

[0141] (P5):

[0142] st(14)-(17).

[0143] (P5) is a standard convex optimization problem that can be solved efficiently using existing solvers. By combining BCD and SCA techniques, (P2) can be solved by alternately iteratively optimizing (P4) and (P5).

[0144] 2) Overall algorithm, convergence and complexity, where ∈1 is the given convergence accuracy:

[0145]

[0146] use Indicates that at a given time speed and auxiliary variables Under this condition, the target value of the energy minimization problem model (P2) at the rth iteration can be obtained as follows:

[0147]

[0148] Since the objective value of the first subproblem optimization model (P4) is monotonically decreasing in step 3 based on SCA, (a) holds; since the optimal solution of (P5) can be obtained in step 4, (b) holds. Since the objective value of (P2) is monotonically decreasing during the iteration process, Algorithm 1 is guaranteed to converge to the KKT (Karush-Kuhn-Tucker, KKT) point of problem (P2). In addition, the main steps of Algorithm 1 are to solve the first subproblem optimization model (P4) and the second subproblem model (P5) of the standard convex optimization problem. The computational complexity depends on the number of optimization variables and can be expressed as Phase 2: Beamforming vector and reflection angle optimization

[0149] Using Algorithm 1 of the first stage, the optimized UAV trajectory {q l} and total working time Based on the path discretization method in Section IV-A, w k (t), u(t), P com (t), SINR k (t), h k The corresponding discretization forms of (t) are Since the displacement of the UAV in each time period is relatively small, it is assumed that the coordinates of the UAV in the lth time period are The communication energy minimization problem model can be expressed as

[0150] (P6):

[0151]

[0152]

[0153] It should be noted that (P6) is a non-convex optimization problem due to the existence of the unit mode constraint (21) and the constraint containing tightly coupled variables (22). By contradiction, it can be shown that the minimum communication energy consumption of the UAV in the total time T is equal to the sum of its minimum communication energy consumption on each line segment l (1≤l≤L). Based on this, an algorithm is designed to minimize the communication energy consumption of the UAV in a certain period of time. By applying the proposed algorithm to the line segments l=1,…,L respectively, all the beamforming vectors and IRS reflection angles of the UAV are obtained, thereby minimizing the total communication energy consumption.

[0154] 1) Problem Reconstruction: To facilitate processing, the optimization problem is first transformed into a problem of minimizing the UAV communication energy consumption over a period of time. Definition The transformed problem can be expressed as

[0155] (P7):

[0156]

[0157]

[0158] The key idea to solve the QoS constraint (24) is to introduce auxiliary variables Decoupling the reflection angle u and beamforming vector from the SINR expression Specifically, we first introduce a new equality constraint, which reformulates constraint (24) as the intersection of the following regions:

[0159]

[0160]

[0161] Then, convert (P7) into the following equivalent form

[0162] (P8):

[0163] st(23),(25),(26).

[0164] Although there are still coupled variables in the equality constraint (25), the above transformation helps to update the optimization variables in parallel in the next step. Then, the equality constraint is integrated into the objective function of (P8) using the penalty method. Specifically, the quadratic function |h based on the equality constraint is transformed into k w i -y k,i | 2 Added as a penalty term to the objective function, (P8) is converted to

[0165] (P9):

[0166] st(23),(26).

[0167] Among them, ρ is a positive parameter used to penalize violations of the equality constraint (25), and when ρ→0, (P9) and (P8) are equivalent. At the same time, choosing a very small ρ in practical applications will reduce the weight of power consumption in the objective function, making the solution invalid. To solve this problem, this application proposes a double-loop iterative algorithm based on P-BCD technology, which appropriately updates ρ in each iteration. In the outer loop, ρ is gradually reduced (c<1) by ρ=cρ to amplify the penalty effect of violating the equality constraint until the target threshold is met. In the inner loop, BCD technology is used to divide the optimization variables into different blocks to solve the non-convex optimization problem (P9). Specifically, (P9) is first divided into two sub-problem models: 1) Beamforming vector and auxiliary variables Optimization; 2) IRS reflection angle u optimization, and then iteratively optimize these two sub-problems until convergence. 2) Beamforming vector and auxiliary variable optimization: Under a given reflection angle u, optimize the beamforming vector and the subproblem with auxiliary variable y can be expressed as

[0168] (P10): To handle the non-convex constraint (26), it is converted into the following equivalent form

[0169]

[0170] Using SCA technology, at a given local point Applying the first-order Taylor approximation, we can get |y k,k | 2 The lower bound of

[0171]

[0172] in It's about y k,k Linear function of . k,k | 2 Replaced with the lower bound (P9) can be approximated as the following convex optimization problem

[0173] (P11):

[0174]

[0175] This problem can be solved efficiently using standard convex optimization tools such as CVX.

[0176] 3) Reflection angle optimization: In this sub-problem model, the given beamforming vector and auxiliary variables To optimize the reflection angle u. Since the beamforming vector is fixed, the constant term related to the communication power in the objective function can be ignored and only the penalty term can be minimized. and Can get The sub-problem can be expressed as

[0177] (P12):

[0178] st(23).

[0179] Although (23) is a non-convex unit norm constraint, each reflection angle u nIn (23) they are all independent of each other and are only coupled with other reflection angles in the objective function. This application proposes an algorithm based on BCD technology that can be used to calculate the reflection angles of the given reflection angles. Iteratively optimize the reflection angle u n Specifically, by ignoring the constant term, given Optimize u n The problem can be written as (P13)

[0180] (P13):

[0181] st|u n | 2 =1. (30)

[0182] In question (P13), Therefore, the objective function of (P13) is linear and its optimal solution is It can be obtained from the following closed expression

[0183]

[0184] Based on (31), each reflection angle will be optimized alternately while fixing the other N-1 reflection angles until the convergence condition is met.

[0185] The detailed process of reflection angle optimization is summarized in Algorithm 2.

[0186]

[0187] In some embodiments, Algorithm 2 can be extended to the scenario of discrete reflection angles by slight modification. Assuming that the optional discrete reflection angles are uniformly distributed on [0, 2π), the angle range constraint can be rewritten as 2 of them b Indicates the number of optional reflection angles. First, based on Algorithm 2, the optimized continuous reflection angles can be obtained. Then the Obtain the solution for discrete reflection angles.

[0188]

[0189]

[0190] 4) Overall algorithm, convergence and complexity: Based on the above discussion, the detailed process of solving the problem model (P9) is summarized in Algorithm 3. In the outer loop, the indicator The degree to which the obtained solution violates Equation (25) is evaluated. As the penalty coefficient ρ decreases, the degree of violation will gradually decrease. When φ is lower than the threshold ∈2, the proposed double-loop P-BCD algorithm terminates. By applying Algorithm 3 to each path segment and adding the resulting communication energy consumption, the total communication energy consumption of the UAV during the entire flight can be obtained. It is worth noting that because the coordinate parameters can be adjusted within each time segment, the solution of this application can also be applied to scenarios where the GU is movable.

[0191] Next, we will first analyze the convergence of the inner loop. Given the IRS reflection angle at the rth iteration and auxiliary variables The target value of problem (P9) can be expressed as And have the following relationship

[0192]

[0193] Since the first-order Taylor expansion in step 5 of Algorithm 3 is compact, inequality (a) holds; since Algorithm 2 based on BCD technology has monotonic convergence, inequality (b) holds. Therefore, with a given penalty coefficient ρ It is non-increasing during the iteration process, which ensures the convergence of the inner loop in Algorithm 3.

[0194] The key step of the inner loop is to solve the optimization problem model (P10). The computational complexity depends on the number of optimization variables and can be expressed as The complexity of Algorithm 2 depends on the calculation of u using the closed form (31) n * The number of times can be expressed as Where I0 is the number of iterations required for convergence. Therefore, the overall complexity of Algorithm 3 can be expressed as Among them L inn and L out Represents the number of iterations required for the inner and outer loops respectively.

[0195] This application is based on an IRS-assisted multi-antenna UAV network and uses a UAV-based aerial video surveillance system to build a network model, a communication model, and an energy consumption model; a reconfigurable UAV-IRS-GU link can be built by deploying IRS to enhance the quality of air-to-ground communications. In order to minimize the total energy consumption of the UAV while meeting the latency requirements, this application proposes a two-stage algorithm to jointly optimize the UAV flight time, UAV trajectory, transmit beamforming vector, and IRS reflection angle. Specifically, this application uses path discretization, alternating optimization, and SCA technology to solve the propulsion energy minimization subproblem, and proposes a dual-loop iterative algorithm using P-BCD technology to solve the communication energy minimization subproblem. Specifically, auxiliary variables are introduced to decouple the reflection angle and beamforming vector design, and a quadratic penalty method is used to attach equality constraints to the objective function. The SCA technology is used to solve the subproblems of beamforming and auxiliary variable optimization, and the BCD technology is used to solve the subproblem of reflection angle optimization.

[0196] An embodiment of the present invention further provides an energy consumption minimization device in an IRS-assisted multi-antenna UAV video transmission system, comprising:

[0197] processor;

[0198] a memory for storing instructions executable by the processor;

[0199] The processor executes the computer-executable instructions stored in the memory to implement the solution provided by any of the above method embodiments. The specific functions and technical effects that can be achieved are not described in detail here.

[0200] Those skilled in the art will appreciate that various aspects of the present application can be implemented as systems, methods, or program products. Therefore, various aspects of the present application can be specifically implemented in the following forms: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation that combines hardware and software aspects, which may be collectively referred to herein as a "circuit," "module," or "system."

[0201] In some possible implementations, an electronic device according to the present application includes at least one processor and at least one memory. The memory stores program code that, when executed by the processor, causes the processor to perform the operational data management methods according to various exemplary embodiments of the present application described above. For example, the processor may perform the steps described in the operational data management method.

[0202] Furthermore, the energy consumption minimization device in the IRS-assisted multi-antenna UAV video transmission system implemented in the present application can execute the steps of the energy consumption minimization method in the IRS-assisted multi-antenna UAV video transmission system mentioned in the above embodiment.

[0203] In exemplary embodiments, various aspects of the method and apparatus for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system provided herein may also be implemented as a program product, including program code, etc. When the program product is executed on a computer device, the program code is used to cause the computer device to perform the steps described in this specification, which are included in the method for maximizing the quality of experience in a multi-antenna UAV video transmission system according to various exemplary embodiments of the present application.

[0204] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the application, the features and functions of two or more units described above can be embodied in a single unit. Conversely, the features and functions of a single unit described above can be further divided into multiple units for embodied.

[0205] Furthermore, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. During execution, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0206] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0207] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable image scaling device to produce a machine, so that the instructions executed by the processor of the computer or other programmable image scaling device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0208] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable image scaling device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, the instruction device being implemented in the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0209] These computer program instructions can also be loaded onto a computer or other programmable image scaling device so that a series of operating steps are performed on the computer or other programmable device to produce a computer-implemented process, thereby executing the instructions on the computer or other programmable device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A functional step specified in one or more boxes.

[0210] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims will be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0211] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. If these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application shall also include these modifications and variations.

Claims

1. A method for minimizing energy consumption in an IRS-assisted multi-antenna UAV video transmission system, characterized in that: The method comprises: S1: Use the UAV-based aerial video surveillance system to build a network model, communication model, and energy consumption model; The UAV-based aerial video surveillance system includes K single-antenna ground users GUs, a multi-antenna rotor UAV, and an intelligent reflective surface IRS with N reflective units on a building near the GUs, thereby establishing a UAV-IRS-GUs link; S2: Based on the joint optimization of the UAV trajectory, flight time, beamforming vector and IRS reflection angle, a model P1 for minimizing the total energy consumption of the UAV is constructed, where the total energy consumption of the UAV includes propulsion energy consumption and communication energy consumption; S3: The minimization model P1 of the total energy consumption of the UAV is optimized through the first-stage algorithm and the second-stage algorithm, wherein the first-stage algorithm adopts the path discretization method, the alternating optimization method and the continuous convex approximation algorithm to minimize the propulsion energy consumption of the UAV; the second-stage algorithm adopts the double-loop iterative algorithm based on the penalty-based block coordinate descent technology to minimize the communication energy consumption of the UAV.

2. The method according to claim 1, characterized in that The network model is constructed based on the UAV aerial video surveillance system, including: The UAV is set to fly at a fixed altitude H0, and the starting and ending positions of the UAV are determined to be q I and q F ;Wherein, the UAV flight time is represented by T, the time t∈[0,T], and the three-dimensional position and velocity of the UAV are represented by q(t) and v(t) respectively; The IRS is deployed on the top of the building to assist UAV in serving GUs. For each reflection unit n∈{1,…,N} of the IRS, the reflection amplitude β n ∈[0,1] and reflection angle θ n ∈[0,2π) are independently adjusted by the intelligent controller of the IRS to achieve passive beamforming gain.

3. The method according to claim 2, characterized in that A communication model is constructed using a UAV-based aerial video surveillance system, specifically including: Assuming that the channel between the kth ground user and the UAV follows Rayleigh fading, the corresponding channel coefficient can be expressed as Where α1 represents the path loss index of the channel, in represents the carrier wavelength, c is the speed of light, f c is the carrier frequency; d u,k (t) represents the three-dimensional distance between the UAV and the kth ground user at time t; Setting variables It follows an independent and identically distributed circularly symmetric complex Gaussian distribution with a mean of zero and a variance of 1, which is expressed as 4. The method according to claim 3, characterized in that An energy consumption model is constructed using a UAV-based aerial video surveillance system, specifically including: Let the total energy consumption of the UAV include communication energy consumption and propulsion energy consumption, then we have Among them, P com (t) and P fly (t) represent the communication power and propulsion power of the UAV at time t, respectively.

5. The method according to claim 4, characterized in that The communication power of the UAV at time t is expressed as: The propulsion power of the UAV at time t is expressed as: Among them, k represents the kth ground user, w k (t) represents the beamforming vector of the kth ground user, P0 represents the blade power of the UAV in the hovering state; Ω represents the angular velocity of the UAV rotor blade; r represents the UAV rotor radius; P i represents the induced power of the drone in the hovering state; v0 represents the average rotor induced speed of the drone in the hovering state; d0 represents the fuselage drag; ρ represents the air density; s represents the rotor firmness of the drone; A r Characterizing the UAV rotor disk area.

6. The method according to claim 5, characterized in that The above method constructs a model P1 that minimizes the total energy consumption of the UAV based on the joint optimization of the UAV trajectory, flight time, beamforming vector, and IRS reflection angle, which specifically includes: Jointly optimize the UAV's 3D trajectory {q(t)}, flight time T, beamforming vector {w k (t)} and IRS reflection angle {u(t)}, the optimization problem model is described as: Constraints include: q(0)=q I ,q(T)=q F , ‖v(t)‖≤V max ,t∈[0,T]. Where u(t) represents the reflection angle vector of IRS at time t; u n (t) represents the nth element in u(t); B is the bandwidth; SINR k (t) represents the SINR of the kth terrestrial user at time t; β k represents the maximum tolerable delay of the kth ground user; v(t) represents the speed of the UAV at time t; V max Indicates the maximum speed of the UAV.

7. The method according to claim 6, characterized in that described For limiting the range of reflection angle, the For the QoS requirement of video transmission, the ‖v(t)‖≤V max ,t∈[0,T] is used to limit the speed of the drone.

8. An energy consumption minimization device in an IRS-assisted multi-antenna UAV video transmission system, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the energy consumption minimization method in the IRS-assisted multi-antenna UAV video transmission system according to any one of claims 1 to 7.

Images