Minimum mission time resource management method for UAV-assisted backscatter communication system
By optimizing the drone transmission power, flight trajectory and backscattering equipment scheduling in the drone-assisted backscattering communication system, the problem of limited mission flight time is solved, and the minimum communication requirements and tasks of each reflective device are achieved. Minimize completion time and improve system performance.
Patent Information
- Application Number
- CN202310630208.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-30
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2043-05-30
AI Technical Summary
The existing drone-assisted backscatter communication system cannot meet the minimum data transmission requirements of each reflective device under limited mission flight time, resulting in system performance being affected.
A resource management method for the UAV-assisted backscattering communication system based on minimizing the task completion time is proposed. By establishing equivalent problems and using block coordinate descent and binary search algorithms, the UAV transmission power, flight trajectory and reverse are optimized. Scattering equipment scheduling, meets the minimum communication needs of each reflection node and minimizes task completion time.
It effectively reduces the completion time and energy consumption of drone missions, improves system performance, meets the minimum communication needs of each reflective device at any time, and is better than the algorithm performance of other cases.
Smart Images

Figure CN116707686B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of UAV-assisted backscatter communication system resource management, and specifically focuses on research related to UAV-assisted backscatter communication system resource management under task completion time constraints. Background Art
[0002] With the rapid development of Internet of Things technology, important application fields such as digital RMB hard wallets and smart logistics have put forward higher requirements for information collection, which relies on the deployment of wireless sensors, but due to the mobility of ground users or the inaccurate deployment of sensor nodes. UAVs have the advantages of high maneuverability, dynamic deployment and high probability line of sight links. Combining UAVs with backscatter communication systems has high predictability.
[0003] Based on the UAV-assisted communication system for information transmission and reception, the UAV can be equipped with backscattering equipment for information transmission and reception. Wang Zhengqiang et al. combined UAV communication with backscattering communication in the document "Max-min rate optimization algorithm for UAV-assisted non-orthogonal multiple access backscattering communication system" [J / OL]. Journal of Electronics and Information Technology, 2022[2023-02-24]. Based on the NOMA technology, a system model for maximizing the minimum rate was proposed under the constraints of reflection coefficient, UAV transmission power, minimum information rate requirements and SIC decoding order. The user fairness of the proposed algorithm was verified; DU Y, CHEN Z, HAO J studied a multi-UAV-assisted backscattering communication network in the document "Joint optimization of trajectory and communication in multi-UAVassisted backscatter communication networks[J]" IEEE Access, 2022, 10: 40861-40871. The network uses multiple UAVs to transmit RF carriers to multiple backscattering sensor nodes deployed on the ground and collect data. Under the constraints of backscattering device scheduling, UAV transmission power and flight trajectory, a minimum rate maximization model of the backscattering communication system was established; YEH SH, WANG YS et al. studied a backscattering-assisted wireless sensor network for data collection in the paper "UAV trajectory optimization for data-gathering from backscattering sensor networks[C]." ICC IEEE International Conference on Communications, Dublin, Ireland, 2020:1-6. Under the strict flight time constraint, the number of time slots available for data collection is limited, and the sensor must carefully determine its activation time slot based on the position of the UAV at a given time. NIE Y, ZHAO J et al. proposed a backscattering communication network based on a deep reinforcement learning algorithm in the paper "Energy-efficient UAV trajectory design for backscatter communication: a deep reinforcement learning approach[J]. China communications, 2020, 17(10):129-141." Considering the transmission power, BD scheduling, UAV flight trajectory and reflection coefficient constraints, an energy efficiency maximization model was established.
[0004] In actual application scenarios, each reflection device has the need for data transmission. Considering that the battery capacity carried by the drone is limited, the flight time of the drone mission is strictly limited. The above paper does not consider the information collection needs of each device and the minimization of the mission time, resulting in the failure to meet the minimum data transmission of each reflection device, which affects the performance of the drone-assisted backscatter communication system. Therefore, the present invention considers the ratio of the amount of data that can be achieved by each backscatter device to the amount of data that needs to be transmitted, in order to meet the minimum communication requirements of each reflection node at any time. At the same time, in order to achieve the optimization goal of minimizing the mission completion time, the initial flight trajectory of the drone based on TSP / TSPN is adopted to achieve the optimization goal of minimizing the mission completion time. Summary of the invention
[0005] The present invention aims to solve the above problems of the prior art. A method for managing minimum mission time resources of a drone-assisted backscatter communication system is proposed. The technical solution of the present invention is as follows:
[0006] A method for managing minimum mission time resources of a UAV-assisted backscatter communication system comprises the following steps:
[0007] 101. Establish a UAV-assisted backscatter communication model based on minimizing the task completion time and transform it into an equivalent problem that is easy to handle;
[0008] 102. Initialize the UAV transmission power; iterative algorithm convergence threshold; binary search algorithm parameters T1 = 0, T2 is large enough and convergence threshold;
[0009] 103. Update T=(T1+T2) / 2, initialize the UAV flight trajectory based on the traveling salesman TSP / traveling salesman with neighborhood TSPN, and solve the initial backscattering device scheduling according to the initial flight trajectory;
[0010] 104. Solve the backscatter coefficient using the given UAV transmission power, UAV flight trajectory and backscatter device scheduling, and update it;
[0011] 105. Using the given UAV transmission power and backscatter coefficient, solve the UAV flight trajectory and backscatter device scheduling, and update them;
[0012] 106. Target value update convergence judgment, calculate the updated target value, if the difference between the updated target value and the previous target value is not greater than the iterative algorithm convergence threshold, then the target value is judged to be converged, and the target value iteration is terminated; if the difference between the updated target value and the previous target value is greater than the iterative algorithm convergence threshold, then the newly calculated target value is saved as the target value at this time, and go to step 104 to update the system parameters until the target value meets the convergence condition, and the target value is given;
[0013] 107. The binary search algorithm searches for the minimum task completion time. If the target value is not less than 1, set T2=T, and update the iteratively calculated backscatter coefficient, UAV flight trajectory, backscatter device scheduling and flight time optimization value at the same time; if the target value is less than 1, set T1=T; the binary search algorithm loop ends and judges, if T2-T1 is not greater than the convergence threshold, output the backscatter coefficient, UAV flight trajectory, backscatter device scheduling and flight time optimization value; if T2-T1 is greater than the convergence value, go to step 103 until convergence is met.
[0014] Furthermore, in step 101, a UAV-assisted backscatter communication model based on minimizing the task completion time is established as follows:
[0015]
[0016]
[0017]
[0018]
[0019]
[0020]
[0021] C1f:q(0)=q ini ,q(T f )=q ini ,
[0022]
[0023]
[0024] Among them, T f is the UAV mission completion time, a k (t) represents the scheduling of device BDk at any time t, B is the bandwidth, P(t) represents the transmission power of the drone at any time t, β k (t) is the channel power gain between BDk and the UAV at any time t, b k (t) is the reflection coefficient of BDk at any time t, σ 2 is the system noise, C k is the amount of data collected from BDk, η k Indicates BDk energy efficiency conversion coefficient, E min represents the minimum energy collected, q(t) represents the trajectory of the drone projected on the ground at any time t, V maxIndicates the maximum flight speed of the drone, q ini represents the initial position of the UAV, P max The optimization variable of problem P1 is the maximum transmission power of the UAV. Reflection coefficient Drone flight path and backscatter device scheduling The objective function of problem P1 is the time T for the drone to complete its mission. f , C1a and C1b represent the time division multiple access scheduling constraints of all BDs and the binary constraints of the scheduling indicator variables, C1c represents the constraint on the amount of data that the backscattering device needs to transmit, C1d represents the energy causality constraint, that is, the constraint on energy collection at the kth backscattering device, C1e represents the UAV mobility constraint, and C1f represents the constraint on the mobility of each UAV in each cycle T f It returns to its initial position at the end. C1g represents the UAV transmission power constraint, and C1h represents the reflection coefficient constraint.
[0025] Furthermore, the problem P1 is about time T f Continuous problem, T f As the upper bound of the integral interval is difficult to solve directly, the problem is to jointly optimize the continuous function P f , A f , B f , Q f To get the solution, under a given cycle time T, maximize the ratio θ of the amount of data that can be achieved by each backscattering device to the amount of data that needs to be transmitted. The target solution is expressed as θ * (T), if and only if θ * When (T)≥1, the target rate requirement can be achieved; further discretize the continuous time T, and the given time range [0,T] is discretized into N time slots, that is, T=Nδ t , where δ t is the time slot step, transforming problem P1 into problem P2:
[0026]
[0027]
[0028]
[0029]
[0030]
[0031] C2e:||q[n+1]-q[n]||≤V max δ t,1≤n≤N-1,
[0032] C2f:q[1]=q ini ,q[N]=q ini ,
[0033]
[0034]
[0035] Using the iterative algorithm of block coordinate descent and successive convex approximation method, the original problem P2 is decomposed into three sub-problems, and then each sub-problem is solved by alternate iterative optimization.
[0036] Furthermore, in step 102, T1=0 and T2 are initialized to be large enough; the convergence threshold value υ of the iterative algorithm is greater than 0; the convergence threshold value λ of the binary search algorithm is greater than 0; the specific steps of the given UAV transmission power P as the maximum transmission power of the UAV are:
[0037] Fix A, B and Q, and optimize P and θ subproblem P3 as:
[0038]
[0039]
[0040]
[0041]
[0042] Using proof by contradiction, we first assume that the optimal solution to problem P3 is P * [n], so we get the conclusion P * [n]<P max Then a new feasible solution ρ is introduced, where ρ=P max ,ο=P max / P * [n], due to P * [n]<P max , so ο>1, next we prove that the new feasible solution ρ satisfies constraints C3a and C3b;
[0043] C3b The optimal solution is P * [n], set ο = P max / P * Substituting [n] into:
[0044]
[0045] Similarly, substituting into C3a yields:
[0046]
[0047] It can be seen that the new feasible solution ρ satisfies C3a and C3b and the ratio obtained by ρ is greater than P * [n] The ratio obtained shows that the new feasible solution is the optimal solution to problem P3, which contradicts the conclusion of the hypothesis. * [n] = P max .
[0048] Furthermore, the step 103 updates T=(T1+T2) / 2, uses T to initialize the UAV flight trajectory by using TSP / TSPN, and solves the initial backscatter device scheduling according to the initial flight trajectory.
[0049] Furthermore, the step 104 uses the given UAV transmission power, UAV flight trajectory and backscattering device scheduling to solve the backscattering coefficient and update it, and the specific steps are:
[0050] Fix P, A and Q, and optimize B and θ sub-problem 2 P4 as:
[0051]
[0052]
[0053]
[0054]
[0055] According to the definition of concave function, the left side of C4a is concave with respect to the reflection coefficient B, and other constraints are convex. P3 is a convex optimization problem, which can be effectively solved by convex optimization tools.
[0056] Furthermore, the step 105 uses the given UAV transmission power and backscatter coefficient to solve the UAV flight trajectory and backscatter device scheduling and update, and the specific steps include:
[0057] Fix P and B, and optimize A, Q and θ subproblem three P5 as:
[0058]
[0059]
[0060]
[0061]
[0062]
[0063] C5e:||q[n+1]-q[n]||≤V max δ t ,1≤n≤N-1,
[0064] C5f:q[1]=q ini ,q[N]=q ini .
[0065] Introducing the slack variable ψ k [n], definition C5c can be restated as follows:
[0066]
[0067] Introducing auxiliary variables definition The above formula can be re-expressed as:
[0068]
[0069] The left side of C5d is non-convex with respect to q[n], so a slack variable μ is introduced k [n], definition And rewrite C5d as follows:
[0070]
[0071] Using continuous convex approximation technology, the non-convex constraints are transformed into convex constraints, and the subproblem three is equivalent to P6:
[0072]
[0073]
[0074]
[0075]
[0076]
[0077]
[0078]
[0079]
[0080] C6h:||q[n+1]-q[n]||≤V max δ t ,1≤n≤N-1,
[0081] C6i:q[1]=q ini,q[N]=q ini .
[0082] Among them, in the first iteration of C6c, at a given point and Place (α k [n]+ψ k [n]) 2 Performing a first-order Taylor expansion, we can obtain the following inequality:
[0083] We can get α k [n]ψ k [n] Lower bound
[0084]
[0085] In C6d, take the second derivative of the left side of C5c-3 and prove that is convex, and is converted to a convex constraint as follows:
[0086] is a feasible point. Similarly, for C5c-4,
[0087]
[0088] For the left side of (C5d-1), we can still use the first-order Taylor expansion method to find The first-order Taylor expansion is performed at the point as follows:
[0089] in is a fixed value, P6 is a convex optimization problem and is solved using convex optimization tools.
[0090] Furthermore, in step 106, the target value update convergence judgment is performed to calculate the updated target value θ(T). If θ l+1 (T)-θ l (T)≤υ, where θ l+1 (T) is the target value of the l+1th iteration, θ l (T) is the target value of the first iteration, then the target value is judged to be converged, and the target value θ is given * (T) The iteration ends; otherwise, the newly calculated θ(T) is saved as the target value at this time, and the system parameters are updated in step 104 until the target value meets the convergence condition, giving the target value θ * (T).
[0091] Furthermore, the binary search algorithm in step 107 searches for the minimum task completion time if θ * (T)≥1, let T2=T, update each optimization value A *=A l+1 , Q * =Q l+1 , B * =B l+1 , T * =T; otherwise, let T1 = T; if (T2-T1)≤λ, output the optimization results of each variable {A * , B * , Q *}, T f * =T * , otherwise go to step 103 until convergence is met.
[0092] The advantages and beneficial effects of the present invention are as follows:
[0093] The present invention aims to minimize the UAV task completion time under the constraints of UAV transmission power, reflection coefficient, equipment scheduling and UAV flight trajectory. Since the proposed non-convex problem is difficult to solve directly, the present invention proposes a two-layer iterative algorithm based on block coordinate descent and binary search algorithm for solution. Its innovation lies in that in step 101, the reflection node rate requirements and the shortest path requirements to reach each reflection node are considered to meet the minimum communication requirements of each reflection node at any time and achieve the optimization goal of minimizing the task completion time. In step 103, in order to reduce the task completion time and reduce energy consumption problems at the same time, the UAV initial flight trajectory based on TSP / TSPN is used to improve the performance of the UAV-assisted backscatter communication system. Compared with other cases, the method provided by the present invention has superior algorithm performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0094] Figure 1 The present invention provides a preferred embodiment of a UAV-assisted backscatter communication model based on minimizing the task completion time;
[0095] Figure 2 The ratio of the amount of data that can be achieved in each backscattering device to the amount of data that needs to be transmitted θ and the amount of data collected by each device C K Convergence diagrams for different situations;
[0096] Figure 3 For the present invention C K =Optimization trajectory based on TSP when 200 bits;
[0097] Figure 4 For the present invention C K = 100bit based on TSPN optimized trajectory;
[0098] Figure 5 The backscattering device scheduling is performed when the flight time of the UAV of the present invention is T=79S;
[0099] Figure 6 The task completion time and C under different schemes of the present invention K relation;
[0100] Figure 7 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION
[0101] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.
[0102] The technical solution of the present invention to solve the above technical problems is:
[0103] Figure 7 A method for managing resources of a UAV-assisted backscatter communication system based on minimizing the task completion time under the condition of satisfying the minimum communication demand of each reflection node at any time is disclosed. The method comprises the following steps:
[0104] Step 1: Establish a UAV-assisted backscatter communication model based on minimizing the mission completion time and transform it into an equivalent problem that is easy to handle.
[0105] Step 2: Initialize the UAV transmission power; iterate the algorithm convergence threshold; binary search algorithm parameters T1 = 0, T2 is large enough and the convergence threshold.
[0106] Step 3: Update T = (T1 + T2) / 2, use T to initialize the UAV flight trajectory using TSP / TSPN, and solve the initial backscatter device scheduling based on the initial flight trajectory.
[0107] Step 4: Solve and update the backscatter coefficient using the given UAV transmit power, UAV flight trajectory, and backscatter device scheduling.
[0108] Step 5: Use the given UAV transmit power and backscatter coefficient to solve the UAV flight trajectory and backscatter device scheduling and update.
[0109] Step 6: Target value update convergence judgment, calculate the updated target value, if the difference between the updated target value and the previous target value is not greater than the iterative algorithm convergence threshold, then the target value is judged to have converged, and the target value iteration ends; if the difference between the updated target value and the previous target value is greater than the iterative algorithm convergence threshold, then the newly calculated target value is saved as the target value at this time, and go to step 4 to update the system parameters until the target value meets the convergence conditions and the target value is given.
[0110] Step 7: The binary search algorithm searches for the minimum task completion time. If the target value is not less than 1, set T2 = T, and update the iteratively calculated backscatter coefficient, UAV flight trajectory, backscatter device scheduling, and flight time optimization value; if the target value is less than 1, set T1 = T. The binary search algorithm loop ends the judgment. If T2-T1 is not greater than the convergence threshold, the backscatter coefficient, UAV flight trajectory, backscatter device scheduling, and flight time optimization value are output; if T2-T1 is greater than the convergence value, go to step 3 until convergence is met.
[0111] Furthermore, the step 1 of determining whether there is a feasible solution specifically includes the steps of establishing a UAV-assisted backscatter communication model based on minimizing the task completion time:
[0112]
[0113]
[0114]
[0115]
[0116]
[0117]
[0118] C1f:q(0)=q ini ,q(T f )=q ini ,
[0119]
[0120]
[0121] Among them, T f is the UAV mission completion time, a k (t) represents the scheduling of device BDk at any time t, B is the bandwidth, P(t) represents the transmission power of the drone at any time t, β k (t) is the channel power gain between BDk and the UAV at any time t, b k (t) is the reflection coefficient of BDk at any time t, σ 2 is the system noise, C k is the amount of data collected from BDk, η k Indicates BDk energy efficiency conversion coefficient, E min represents the minimum energy collected, q(t) represents the trajectory of the drone projected on the ground at any time t, and V maxIndicates the maximum flight speed of the drone, q ini represents the initial position of the UAV, P max The optimization variable of problem P1 is the maximum power of the UAV transmission. Reflection coefficient Drone flight path and backscatter device scheduling The objective function of problem P1 is the time T for the drone to complete its mission. f , C1a and C1b represent the time division multiple access scheduling constraints of all BDs and the binary constraints of the scheduling indicator variables, C1c represents the constraint on the amount of data that the backscattering device needs to transmit, C1d represents the energy causality constraint, that is, the constraint on energy collection at the kth backscattering device, C1e represents the UAV mobility constraint, and C1f represents the constraint on the mobility of each UAV in each cycle T f It returns to its initial position at the end. C1g represents the UAV transmission power constraint, and C1h represents the reflection coefficient constraint.
[0122] Question P1 About time T f Continuous problem, T f As the upper bound of the integral interval is difficult to solve directly, the problem is to jointly optimize the continuous function P f , A f , B f , Q f To solve this problem, given a cycle time T, the ratio θ of the amount of data that can be achieved by each backscattering device to the amount of data that needs to be transmitted is maximized. The target value is expressed as θ * (T), if and only if θ * When (T)≥1, the target rate requirement can be achieved. Further discretize the continuous time T, and the given time range [0,T] is discretized into N time slots, that is, T=Nδ t , where δ t is the time slot step, transforming problem P1 into problem P2:
[0123]
[0124]
[0125]
[0126]
[0127]
[0128] C2e:||q[n+1]-q[n]||≤V max δ t,1≤n≤N-1,
[0129] C2f:q[1]=q ini ,q[N]=q ini ,
[0130]
[0131]
[0132] Using the iterative algorithm of block coordinate descent and successive convex approximation method, the original problem P2 is decomposed into three sub-problems, and then each sub-problem is solved by alternate iterative optimization.
[0133] Furthermore, in step 2, T1=0 and T2 are initialized to be large enough; the convergence threshold value υ of the iterative algorithm is greater than 0; the convergence threshold value λ of the binary search algorithm is greater than 0; the UAV transmission power P is given as the maximum transmission power of the UAV. The specific steps are:
[0134] Fix A, B and Q, and optimize P and θ subproblem P3 as:
[0135]
[0136]
[0137]
[0138]
[0139] Using proof by contradiction, we first assume that the optimal solution to problem P3 is P * [n], so we get the conclusion P * [n]<P max Then a new feasible solution ρ is introduced, where ρ=P max ,ο=P max / P * [n], due to P * [n]<P max , so ο>1, next we prove that the new feasible solution ρ satisfies constraints C3a and C3b.
[0140] C3b The optimal solution is P * [n], set ο = P max / P * Substituting [n] into:
[0141]
[0142] Similarly, substituting into C3a yields:
[0143]
[0144] It can be seen that the new feasible solution ρ satisfies C3a and C3b and the ratio obtained by ρ is greater than P * [n] The ratio obtained shows that the new feasible solution is the optimal solution to problem P3, which contradicts the conclusion of the hypothesis. * [n] = P max .
[0145] Furthermore, the step 3 updates T=(T1+T2) / 2, uses T to initialize the UAV flight trajectory by using TSP / TSPN, and solves the initial backscattering device scheduling according to the initial flight trajectory.
[0146] Furthermore, step 4 uses the given UAV transmission power, UAV flight trajectory and backscattering device scheduling to solve and update the backscattering coefficient. The specific steps are:
[0147] Fix P, A and Q, and optimize B and θ sub-problem 2 P4 as:
[0148]
[0149]
[0150]
[0151]
[0152] According to the definition of concave function, it can be deduced that the left side of C4a is concave with respect to the reflection coefficient B, and other constraints are convex. P3 is a convex optimization problem, which can be effectively solved by convex optimization tools.
[0153] Furthermore, the step 5 uses the given UAV transmission power and backscatter coefficient to solve the UAV flight trajectory and backscatter device scheduling and update, and the specific steps include:
[0154] Fix P and B, and optimize A, Q and θ subproblem three P5 as:
[0155]
[0156]
[0157]
[0158]
[0159]
[0160] C5e:||q[n+1]-q[n]||≤Vmax δ t ,1≤n≤N-1,
[0161] C5f:q[1]=q ini ,q[N]=q ini .
[0162] Introducing the slack variable ψ k [n], definition C5c can be restated as follows:
[0163]
[0164] Introducing auxiliary variables definition The above formula can be re-expressed as:
[0165]
[0166] The left side of C5d is non-convex with respect to q[n], so a slack variable μ is introduced k [n], definition And rewrite C5d as follows:
[0167]
[0168] Using continuous convex approximation technology, the non-convex constraints are transformed into convex constraints, and the subproblem three is equivalent to P6:
[0169]
[0170]
[0171]
[0172]
[0173]
[0174]
[0175]
[0176]
[0177] C6h:||q[n+1]-q[n]||≤V max δ t ,1≤n≤N-1,
[0178] C6i:q[1]=q ini ,q[N]=q ini .
[0179] Among them, in the first iteration of C6c, at a given point and Place (α k [n]+ψ k [n]) 2 Performing a first-order Taylor expansion, we can obtain the following inequality:
[0180] We can get α k [n]ψ k [n] Lower bound
[0181]
[0182] In C6d, taking the second derivative of the left side of C5c-3, we can prove that is convex, and is converted to a convex constraint as follows:
[0183] is a feasible point. Similarly, for C5c-4,
[0184]
[0185] For the left side of (C5d-1), we can still use the first-order Taylor expansion method to find The first-order Taylor expansion is performed at the point as follows:
[0186] in is a fixed value. P6 is a convex optimization problem and can be solved using convex optimization tools.
[0187] Furthermore, in step 6, the target value is updated and converged, and the updated target value θ(T) is calculated. If θ l+1 (T)-θ l (T)≤υ, where θ l+1 (T) is the target value of the l+1th iteration, θ l (T) is the target value of the first iteration, then the target value is judged to be converged, and the target value θ is given * (T) The iteration ends; otherwise, the newly calculated θ(T) is saved as the target value at this time, and the system parameters are updated in step 4 until the target value meets the convergence condition, giving the target value θ * (T).
[0188] Further, the binary search algorithm in step 7 searches for the minimum task completion time if θ * (T)≥1, let T2=T, update each optimization value A * =A l+1 , Q* =Q l+1 , B * =B l+1 , T * =T; otherwise, let T1 = T; if (T2-T1)≤λ, output the optimization results of each variable {A * , B * , Q *}, T f * =T * , otherwise go to step 103 until convergence is met.
[0189] The present invention aims to minimize the UAV task completion time under the constraints of UAV transmission power, reflection coefficient, equipment scheduling and UAV flight trajectory. Since the proposed non-convex problem is difficult to solve directly, the present invention proposes a two-layer iterative algorithm based on block coordinate descent and binary search algorithm for solution. Its innovation lies in that the ratio of the amount of data that can be achieved by each backscattering device to the amount of data that needs to be transmitted is considered to meet the minimum communication requirements of each reflection node at any time. At the same time, in order to achieve the optimization goal of minimizing the task completion time, the initial flight trajectory of the UAV based on TSP / TSPN is adopted. Compared with other cases, the method provided by the present invention has better algorithm performance.
[0190] This embodiment is a method for minimizing the completion time of a UAV-assisted backscatter communication task. The backscatter devices are evenly distributed in a 450×400m 2 The number of devices K is set to 6, and the starting position of the drone is set to (-200, 0) m. After collecting data in hovering or continuous flight mode, it returns to the starting point. The A2G model, that is, the free space path loss model, is used between the drone and the BD. The communication channel from the drone to the BD is mainly controlled by LoS propagation. The drone's flight altitude is: H = 10m, and the drone's maximum flight speed is: V max =15m / s, UAV maximum transmission power: P max =3W, reference channel power gain: β0 = -20dBm, noise power: σ 2 =-90dBm, BD energy collection efficiency: η k =0.8, BD minimum collection energy: E min =0.2510 -6 J (Joule), system bandwidth: B = 1 Hz.
[0191] In this example, Figure 1The present invention provides a preferred embodiment of a backscatter communication system assisted by a drone. In the figure, there are K backscatter devices on the ground. A drone is dispatched to the area to serve the backscatter devices on the ground. In this network system, the drone acts as both a signal transmitter and a receiver of the reflected signal of the ground device. In the downlink, the drone broadcasts the signal, and in the uplink, the drone uses the TDMA protocol to communicate with the BD to collect information. Figure 2 The ratio of the amount of data that can be achieved in each backscattering device to the amount of data that needs to be transmitted θ and the amount of data collected by each device C K Convergence diagrams for different situations; Figure 3 For the present invention C K =Optimization trajectory based on TSP when 200 bits; Figure 4 For the present invention C K = 100bit based on TSPN optimized trajectory; Figure 5 The backscattering device scheduling is performed when the flight time of the UAV of the present invention is T=79S; Figure 6 The task completion time and C under different schemes of the present invention K relation; Figure 2 Visible in C k =100 bits and C k =200 bits, the algorithm converges in 8 iterations, and the target value θ increases rapidly with the increase of iteration number, among which the target value θ of the first 3 iterations increases significantly, which verifies the fast convergence of the algorithm; Figure 3 It can be seen that the amount of data required to be transmitted C k =200 bits, the optimized trajectory drone based on TSP uses its high maneuverability to obtain the best communication link quality on top of each BD node in the order of access, and hovers for a period of time to collect data, which is shorter than the initialization trajectory algorithm based on the ellipse; Figure 4 It can be seen that the amount of data required to be transmitted C k =100 bits, the optimized trajectory drone based on TSPN does not need to fly above each BD for communication, but adopts a continuous flight mode for data collection, and the algorithm task completion time is shorter; Figure 5 It can be seen that each BD is in a non-scheduled state in most cases. Only when the drone hovers directly above the BD or flies close enough to the BD, the BD will be scheduled. This is because only by allocating time resources to better channels can higher transmission efficiency be achieved. And under the TDMA protocol, the drone only schedules one backscatter device in each time slot; Figure 6 It can be seen that the task completion time of all algorithms increases with C k increases with the increase of , and the performance of the algorithm proposed in this chapter is better than the two compared algorithms.
[0192] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.
[0193] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0194] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.
Claims
1. A method for managing minimum mission time resources in a UAV-assisted backscatter communication system, characterized in that: The following steps are involved:
101. Establish a UAV-assisted backscatter communication model based on minimizing the task completion time and transform it into an equivalent problem that is easy to handle; 102. Initialize the UAV transmission power; iterative algorithm convergence threshold; binary search algorithm parameters T1 = 0, T2 is large enough and convergence threshold; 103. Update T=(T1+T2) / 2, initialize the UAV flight trajectory based on the traveling salesman TSP / traveling salesman with neighborhood TSPN, and solve the initial backscattering device scheduling according to the initial flight trajectory; 104. Solve the backscatter coefficient using the given UAV transmission power, UAV flight trajectory and backscatter device scheduling, and update it; 105. Using the given UAV transmission power and backscatter coefficient, solve the UAV flight trajectory and backscatter device scheduling, and update them; 106. Target value update convergence judgment, calculate the updated target value, if the difference between the updated target value and the previous target value is not greater than the iterative algorithm convergence threshold, then the target value is judged to be converged, and the target value iteration is terminated; if the difference between the updated target value and the previous target value is greater than the iterative algorithm convergence threshold, then the newly calculated target value is saved as the target value at this time, and go to step 104 to update the system parameters until the target value meets the convergence condition, and the target value is given; 107. The binary search algorithm searches for the minimum task completion time. If the target value is not less than 1, set T2 = T, and update the iteratively calculated backscatter coefficient, UAV flight trajectory, backscatter device scheduling, and flight time optimization value at the same time; If the target value is less than 1, set T1 = T; the binary search algorithm loop ends and judges that if T2-T1 is not greater than the convergence threshold, output the backscatter coefficient, the UAV flight trajectory, the backscatter device scheduling and the flight time optimization value; if T2-T1 is greater than the convergence value, go to step 103 until convergence is met; The step 101 is to establish a UAV-assisted backscatter communication model based on minimizing the task completion time: Among them, T f is the UAV mission completion time, a k (t) represents the scheduling of device BDk at any time t, B is the bandwidth, P(t) represents the transmission power of the drone at any time t, β k (t) is the channel power gain between BDk and the UAV at any time t, b k (t) is the reflection coefficient of BDk at any time t, σ 2 is the system noise, C k is the amount of data collected from BDk, η k Indicates BDk energy efficiency conversion coefficient, E min represents the minimum energy collected, q(t) represents the trajectory of the drone projected on the ground at any time t, V max Indicates the maximum flight speed of the drone, q ini represents the initial position of the UAV, P max The optimization variable of problem P1 is the maximum power of the UAV transmission. Reflection coefficient Drone flight path and backscatter device scheduling The objective function of problem P1 is the time T for the drone to complete its mission. f , C1a and C1b represent the time division multiple access scheduling constraints of all BDs and the binary constraints of the scheduling indicator variables, C1c represents the constraint on the amount of data that the backscattering device needs to transmit, C1d represents the energy causality constraint, that is, the constraint on energy collection at the kth backscattering device, C1e represents the UAV mobility constraint, and C1f represents the constraint on the mobility of each UAV in each cycle T f When it ends, it returns to its initial position. C1g represents the transmission power constraint of the UAV, and C1h represents the reflection coefficient constraint; The problem P1 is about time T f Continuous problem, T f As the upper bound of the integral interval is difficult to solve directly, the problem is to jointly optimize the continuous function P f , A f , B f , Q f To get the solution, under a given cycle time T, maximize the ratio θ of the amount of data that can be achieved by each backscattering device to the amount of data that needs to be transmitted. The target solution is expressed as θ * (T), if and only if θ * When (T)≥1, the target rate requirement can be achieved; further discretize the continuous time T, and the given time range [0,T] is discretized into N time slots, that is, T=Nδ t , where δ t is the time slot step, transforming problem P1 into problem P2: Using the iterative algorithm of block coordinate descent and successive convex approximation method, the original problem P2 is decomposed into three sub-problems, and then each sub-problem is optimized and solved by alternating iterations. In step 102, T1 is initialized to 0, T2 is large enough; the convergence threshold value υ of the iterative algorithm is greater than 0; the convergence threshold value λ of the binary search algorithm is greater than 0; the UAV transmission power P is given as the maximum transmission power of the UAV. The specific steps are: Fix A, B and Q, and optimize P and θ subproblem P3 as: Using proof by contradiction, we first assume that the optimal solution to problem P3 is P * [n], so we get the conclusion P * [n]<P max ; Then introduce a new feasible solution ρ, where ρ=P max ,ο=P max / P * [n], due to P * [n]<P max , so ο>1, next we prove that the new feasible solution ρ satisfies constraints C3a and C3b; C3b The optimal solution is P * [n], set ο = P max / P * Substituting [n] into: Similarly, substituting into C3a yields: It can be seen that the new feasible solution ρ satisfies C3a and C3b and the ratio obtained by ρ is greater than P * [n] The ratio obtained shows that the new feasible solution is the optimal solution to problem P3, which contradicts the conclusion of the hypothesis. * [n] = P max ; The step 103 updates T=(T1+T2) / 2, uses T to initialize the UAV flight trajectory by using TSP / TSPN, and solves the initial backscatter device scheduling according to the initial flight trajectory; The step 104 uses the given UAV transmission power, UAV flight trajectory and backscattering device scheduling to solve the backscattering coefficient and update it. The specific steps are: Fix P, A and Q, and optimize B and θ sub-problem 2 P4 as: According to the definition of concave function, the left side of C4a is concave with respect to the reflection coefficient B, and other constraints are convex. P3 is a convex optimization problem, which can be effectively solved by convex optimization tools. The step 105 uses the given UAV transmission power and backscatter coefficient to solve the UAV flight trajectory and backscatter device scheduling and update, and the specific steps include: Fix P and B, and optimize A, Q and θ subproblem three P5 as: Introducing the slack variable ψ k [n], definition C5c can be restated as follows: Introducing auxiliary variables definition The above formula can be re-expressed as: The left side of C5d is non-convex with respect to q[n], so a slack variable μ is introduced k [n], definition And rewrite C5d as follows: H 2 +‖q[n]-u k ‖ 2 ≤μ k [n](C5d-2)? Using continuous convex approximation technology, the non-convex constraints are transformed into convex constraints. Subproblem 3 P5 is equivalent to P6: Among them, in the first iteration of C6c, at a given point and Place (α k [n]+ψ k [n]) 2 Performing a first-order Taylor expansion, we can obtain the following inequality: We can get α k [n]ψ k [n] Lower bound In C6d, take the second derivative of the left side of (C5c-3) and prove that is convex, and is converted to a convex constraint as follows: is a feasible point given; similarly for (C5c-4), For the left side of (C5d-1), we can still use the first-order Taylor expansion method to find The first-order Taylor expansion is performed at the point as follows: in is a fixed value, P6 is a convex optimization problem and is solved using convex optimization tools.
2. The method for managing minimum mission time resources of a UAV-assisted backscatter communication system according to claim 1, characterized in that: In step 106, the target value is updated and converged, and the updated target value θ(T) is calculated. If θ l+1 (T)-θ l (T)≤υ, where θ l+1 (T) is the target value of the l+1th iteration, θ l (T) is the target value of the first iteration, then the target value is judged to be converged, and the target value θ is given * (T) The iteration ends; Otherwise, the newly calculated θ(T) is saved as the target value at this time, and the system parameters are updated in step 104 until the target value meets the convergence condition, giving the target value θ * (T).
3. The method for managing minimum mission time resources of a UAV-assisted backscatter communication system according to claim 2, characterized in that: In step 107, the binary search algorithm searches for the minimum task completion time. * (T)≥1, let T2=T, update each optimization value A * =A l+1 , Q * =Q l+1 , B * =B l+1 , T * =T; otherwise, let T1 = T; if (T2-T1)≤λ, output the optimization results of each variable {A * , B * , Q * }, T f * =T * , otherwise go to step 103 until convergence is met.
Citation Information
Patent Citations
Resource allocation method based on max-min fairness for unmanned aerial vehicle assisted backscatter communication system
CN114124705A
Unmanned aerial vehicle assisted NOMA backscatter communication system and rate maximization method
CN115002800A
Unmanned aerial vehicle assisted air-ground backscatter communication resource optimization method
CN116032351A
A device, computer program and method for monitoring an UAV
WO2021198634A1
