An unmanned aerial vehicle wireless power supply aerial computing auxiliary internet of things data acquisition method

Abstract

The application discloses a kind of unmanned aerial vehicle wireless energy supply aerial computing auxiliary internet of things data acquisition method, comprising: S1: establishing the three-dimensional coordinate internet of things transmission system of unmanned aerial vehicle base station, the optimization target of three-dimensional coordinate internet of things transmission system of unmanned aerial vehicle base station is established by jointly optimizing the flight trajectory of unmanned aerial vehicle, wireless energy supply time slot allocation, sensor transmission power;S2: establishing resource allocation optimization model P1;S3: establishing unmanned aerial vehicle trajectory optimization model P2;S4: resource allocation optimization model P1, unmanned aerial vehicle trajectory optimization model P2, the optimization target of three-dimensional coordinate internet of things transmission system of unmanned aerial vehicle base station is alternately solved, the flight trajectory of unmanned aerial vehicle, wireless energy supply time slot allocation, sensor transmission power when the uplink data acquisition rate of three-dimensional coordinate internet of things transmission system is maximized are obtained;The application improves the endurance of internet of things equipment, makes data collection simple, improves transmission rate.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wireless energy supply Internet of Things communication transmission, and particularly relates to a method for unmanned aerial vehicle wireless energy supply air computing assisted Internet of Things data collection. BACKGROUND

[0002] In recent years, under the background of interconnection of all things, the development of physical network technology has attracted widespread attention from researchers at home and abroad, and the number of various types of sensor devices has also increased significantly, which puts forward higher requirements for data collection of Internet of Things devices. Considering that some Internet of Things devices are deployed in remote areas or dangerous areas with harsh environment, using manual method to collect data of these Internet of Things devices is not only low in efficiency, but also is not conducive to detection. In order to overcome the bottleneck of communication data transmission, the emergence of unmanned aerial vehicle assisted communication technology brings new opportunities for flexible deployment of large-scale Internet of Things devices. Compared with the traditional fixed position base station deployment, by dynamically optimizing and adjusting the flight trajectory of the unmanned aerial vehicle, efficient transmission of Internet of Things devices can be realized in various complex scenes such as deserts.

[0003] In addition, most of the Internet of Things devices on the market are small in size and poor in endurance, which makes it impossible to realize long-distance uninterrupted communication. Using unmanned aerial vehicles with high mobility and reliability as energy suppliers for ground sensors can effectively improve the endurance time of ground sensor nodes, and further promote the large-area deployment of sensor devices.

[0004] Because there is often a doubly near-fare effect in the wireless downlink energy supply communication scene, that is, the devices far away from the base station will cause the reduction of the uplink transmission rate due to too little energy received, resulting in serious unfairness phenomenon of the system. Using unmanned aerial vehicle technology can provide a line-of-sight channel between the base station and the receiving Internet of Things device, and by adjusting the flight trajectory of the unmanned aerial vehicle in real time, the transmission distance between the base station and the receiving device can be reduced, so as to improve the energy and data transmission efficiency in a more flexible way.

[0005] The core idea of air computing is to use the waveform superposition characteristics of wireless multiple access channel and the concurrent transmission of multiple devices to realize efficient data aggregation. Compared with the traditional "transmission first and then computing" scheme, the air computing scheme has higher spectrum utilization, and the access delay of the air computing scheme will not increase significantly with the increase of the scale of the sensor network. Considering that the number of future ground sensor devices will increase significantly, the idea of air computing is used to collect data of all ground sensor devices.

[0006] Existing technologies propose a wireless communication method that transmits energy in the downlink and data in the uplink, minimizing data aggregation error by jointly optimizing three variables: energy beamforming, data aggregation beamforming, and power control. However, this scheme does not consider using UAV technology to dynamically adjust the transmission path at the transceiver end, resulting in low data transmission rates for edge devices. Another proposed method is UAV-assisted wireless power transmission, which maximizes uplink data transmission rate by jointly optimizing the UAV's flight trajectory and time slot allocation. However, this scheme does not consider employing an aerial computing solution that can effectively reduce potential transmission latency. Summary of the Invention

[0007] To address the problems of poor battery life, difficult data collection, and low transmission rate of existing IoT devices, this invention proposes a method for wirelessly powered aerial computing-assisted IoT data collection using unmanned aerial vehicles (UAVs), which improves the battery life of IoT devices, simplifies data collection, and increases transmission rate.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0009] A method for wirelessly powered, computing-assisted aerial IoT data acquisition for unmanned aerial vehicles (UAVs) includes the following steps:

[0010] S1: Establish a three-dimensional coordinate IoT transmission system for UAV base stations. By jointly optimizing the flight trajectory of UAVs, the allocation of wireless power supply time slots, and the sensor transmission power, the optimization target of the three-dimensional coordinate IoT transmission system for UAV base stations is established.

[0011] S2: Establish a resource allocation optimization model P1 that jointly optimizes wireless power supply time slot allocation and sensor transmission power;

[0012] S3: Establish a drone trajectory optimization model P2 to optimize the drone flight trajectory of the three-dimensional coordinate Internet of Things system;

[0013] S4: Alternately solve the optimization objectives of the resource allocation optimization model P1, the UAV trajectory optimization model P2, and the UAV base station's three-dimensional coordinate IoT transmission system to obtain the UAV's flight trajectory, wireless power supply time slot allocation, and sensor transmission power when the uplink data acquisition rate of the three-dimensional coordinate IoT transmission system is maximized.

[0014] The working principle of this invention is as follows:

[0015] This invention obtains the drone's flight trajectory, wireless power supply time slot allocation, and sensor transmission power when the uplink data acquisition rate of the three-dimensional coordinate IoT transmission system is maximized by alternately solving the constructed resource allocation optimization model P1, the UAV trajectory optimization model P2, and the UAV base station's three-dimensional coordinate IoT transmission system.

[0016] Preferably, the three-dimensional coordinate IoT transmission system of the UAV base station includes a UAV base station and K ground sensor nodes; where K is an integer; the UAV supplies power to the K ground sensors via wireless power transmission and collects data from all ground sensors via data aggregation.

[0017] Furthermore, when establishing a three-dimensional coordinate IoT transmission system for drone base stations, the parameters are initialized.

[0018] The initialization setup method is as follows: the position coordinates of the kth sensor are x k =[x k ,y k ,0] T Where k∈K={1,…,K}; the UAV's transmit power is P. U The drone's flight altitude is H; the time required for the drone to complete one flight mission is T; T is divided into N time slots, each time slot having a length of... The position coordinates of the UAV in each time slot are x[n] = [x[n], y[n], H]. T , n∈N=={1,…,N}; Each flight time slot of the UAV is divided into two sub-time slots, including the first sub-time slot length τ0[n]δ used for downlink energy transmission. t The second sub-slot length τ1[n]δ used for uplink data collection t The drone's flight speed does not exceed V. max .

[0019] Furthermore, the method for optimizing the three-dimensional coordinate IoT transmission system of the drone base station is as follows:

[0020] In the nth time slot, a line-of-sight channel fading model is established based on the air-to-ground wireless channel model and the geographical locations of the sensor nodes and the UAV.

[0021] The line-of-sight channel fading model is as follows:

[0022]

[0023] Where β0 represents the channel gain when the distance between the ground sensor and the UAV is 1m;

[0024] At the τ0[n]δt When the UAV transmits energy to the ground sensor in the downlink for the length of the first sub-time slot, the energy collected by the k-th sensor in the first sub-time slot is:

[0025] E k [n]=ητ0[n]δ t P U h k [n]

[0026] Where η∈(0,1] represents the energy receiving efficiency of each ground sensor;

[0027] At the τ1[n]δ t During the second sub-time slot length, all ground sensors transmit data to the UAV via the uplink. The data acquisition rate of the UAV within the second sub-time slot is:

[0028]

[0029] Among them, [a] + Represents max(a,0); P k [n] represents the transmit power of the k-th sensor for uplink data transmission in the n-th time slot; σ 2 This indicates the noise power received by the drone;

[0030] To ensure sustainable data acquisition, the power of each ground sensor must meet a transmit power constraint, which is as follows:

[0031]

[0032] The mobility constraints for drones are as follows:

[0033] ||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}

[0034] Where, δ d = max δ t ;

[0035] make Indicates drone trajectory, and Indicates wireless power supply time slot allocation, Let represent the transmit power of the k-th sensor; the objective function for optimizing the three-dimensional coordinate IoT transmission system of the UAV base station is the first objective function, which is expressed as:

[0036]

[0037] The constraints of the first objective function include:

[0038] A1. This indicates that the total energy transmitted by the ground sensors is no higher than the total energy received from the drone.

[0039] A2.||x[n+1]-x[n]||≤V max δ t ,n∈N - This indicates that the drone's flight speed does not exceed its maximum flight speed.

[0040] A3.τ0[n]+τ1[n]≤1 means that the sum of the sub-slot allocation coefficients within each slot is not higher than 1.

[0041] A4. This indicates that the allocation coefficient for each time slot is not less than 0.

[0042] A5. This indicates that the data transmission power of each sensor used for the uplink is not less than 0.

[0043] Furthermore, the objective function of the resource allocation optimization model P1 is a second objective function, which is expressed as:

[0044]

[0045] The constraints of the second objective function include:

[0046] B1.

[0047] B2.τ0[n]+τ1[n]≤1.

[0048] B3.

[0049] B4.

[0050] Furthermore, for the second objective function, we introduce a first slack variable γ[n] and let... The second objective function is transformed into a convex perspective function, and a convex constraint is introduced at the same time. Subsequently ordered

[0051] Represented as: The third objective function is obtained, and the third objective function is expressed as:

[0052]

[0053] The constraints of the third objective function include:

[0054] C1.

[0055] C2.

[0056] C3.

[0057] C4.

[0058] C5.

[0059] C6.

[0060] in,

[0061] Furthermore, the objective function of the UAV trajectory optimization model P2 is a fourth objective function, which is expressed as:

[0062]

[0063] The constraints of the fourth objective function include:

[0064] D1.

[0065] D2.||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}.

[0066] Furthermore, for the fourth objective function, a second slack variable is introduced. And a first-order Taylor expansion is used to transform h k The lower bound of [n] is represented as:

[0067]

[0068] Where, x (t) [n] represents the optimal value of x[n] in the t-th iteration; the fifth objective function is obtained, which is expressed as:

[0069]

[0070] The constraints of the fifth objective function include:

[0071] E1.

[0072] E2.

[0073] E3.

[0074] E4.||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}.

[0075] in,

[0076] Furthermore, the alternating solution steps in S4 are as follows:

[0077] S41: Initialize the drone base station's flight path to X (0) Wireless power supply time slot allocation is T0 (0) and T1 (0) The sensor's transmission power is P (0) And calculate the initial uplink data acquisition rate R. (0) The error threshold η is 10 -3 The number of alternating iterations i = 0;

[0078] S42: Set the flight path X of the drone base station (i) Substituting these values ​​into the resource allocation optimization model P1, and using the CVX solver, the optimal solution T0 for wireless power supply time slot allocation and sensor transmission power in the i-th iteration is obtained. (i+1) T1 (i+1) and P k (i+1) ;

[0079] S43: The optimal solution T0 obtained in S42 (i+1) T1 (i+1) and P k (i+1) And flight trajectory X (i) Substituting into the UAV flight trajectory optimization model P2, we obtain the optimal solution X for the UAV flight trajectory in the i-th iteration. (i+1) And calculate the uplink data acquisition rate R. (i+1) ;

[0080] S44: If |R (i+1) -R (i) If |≤η, the optimal UAV flight trajectory X is obtained. (i+1) Time slot allocation T0 (i+1) and T1 (i +1) Sensor transmit power P k (i+1) Otherwise, let i = i + 1 and repeat S42 and S43.

[0081] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it performs the steps of the method described above.

[0082] Compared with the prior art, the beneficial effects of the present invention are:

[0083] 1. By wirelessly powering sensors with short battery life using drones, the battery life of the sensors can be improved.

[0084] 2. By alternately solving for the flight trajectory of the UAV, the wireless power supply time slot allocation, and the sensor transmission power when the data acquisition rate is maximized, the data collection process is simplified and the data acquisition rate of the Internet of Things is effectively improved. Attached Figure Description

[0085] Figure 1 This is a flowchart of a method for wirelessly powered aerial computing-assisted Internet of Things (IoT) data acquisition for unmanned aerial vehicles (UAVs).

[0086] Figure 2 This is a schematic diagram illustrating how a drone transfers energy and collects data from sensors.

[0087] Figure 3 This is a diagram showing the positional relationship between the sensor and the drone.

[0088] Figure 4 This is a diagram showing the relationship between the three-dimensional flight trajectory of the UAV and the wireless power supply time slot allocation.

[0089] Figure 5 A comparison chart showing the energy consumption and data collected by different sensors under different flight trajectory schemes for drones.

[0090] Figure 6 This graph shows the relationship between the sum of system data acquisition rates of the UAV at different maximum flight speeds and the number of algorithm iterations.

[0091] Figure 7 This graph shows the relationship between the sum of system data acquisition rates under different flight trajectory schemes of the UAV and the different transmission powers of the UAV. Detailed Implementation

[0092] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0093] Example 1

[0094] In this embodiment, as Figure 1 As shown, a method for wirelessly powered aerial computing-assisted Internet of Things (IoT) data acquisition for unmanned aerial vehicles (UAVs) includes the following steps:

[0095] S1: Establish a three-dimensional coordinate IoT transmission system for UAV base stations. By jointly optimizing the flight trajectory of UAVs, the allocation of wireless power supply time slots, and the sensor transmission power, the optimization target of the three-dimensional coordinate IoT transmission system for UAV base stations is established.

[0096] S2: Establish a resource allocation optimization model P1 that jointly optimizes wireless power supply time slot allocation and sensor transmission power;

[0097] S3: Establish a drone trajectory optimization model P2 to optimize the drone flight trajectory of the three-dimensional coordinate Internet of Things system;

[0098] S4: Alternately solve the optimization objectives of the resource allocation optimization model P1, the UAV trajectory optimization model P2, and the UAV base station's three-dimensional coordinate IoT transmission system to obtain the UAV's flight trajectory, wireless power supply time slot allocation, and sensor transmission power when the uplink data acquisition rate of the three-dimensional coordinate IoT transmission system is maximized.

[0099] Example 2

[0100] In this embodiment, a method for wirelessly powered aerial computing-assisted Internet of Things (IoT) data acquisition by a drone includes the following steps:

[0101] S1: Establish a three-dimensional coordinate IoT transmission system for UAV base stations. By jointly optimizing the flight trajectory of UAVs, the allocation of wireless power supply time slots, and the sensor transmission power, the optimization target of the three-dimensional coordinate IoT transmission system for UAV base stations is established.

[0102] S2: Establish a resource allocation optimization model P1 that jointly optimizes wireless power supply time slot allocation and sensor transmission power;

[0103] S3: Establish a drone trajectory optimization model P2 to optimize the drone flight trajectory of the three-dimensional coordinate Internet of Things system;

[0104] S4: Alternately solve the optimization objectives of the resource allocation optimization model P1, the UAV trajectory optimization model P2, and the UAV base station's three-dimensional coordinate IoT transmission system to obtain the UAV's flight trajectory, wireless power supply time slot allocation, and sensor transmission power when the uplink data acquisition rate of the three-dimensional coordinate IoT transmission system is maximized.

[0105] In this embodiment, as Figure 2 As shown, the three-dimensional coordinate IoT transmission system of the UAV base station includes a UAV base station and K ground sensor nodes; where K is an integer; the UAV supplies power to the K sensors on the ground via wireless power transmission and collects data from all the sensors on the ground via data aggregation.

[0106] More specifically, when establishing a three-dimensional coordinate IoT transmission system for drone base stations, the parameters are initialized.

[0107] The initialization setup method is as follows: the position coordinates of the kth sensor are x k =[x k ,y k ,0] T Where k∈K={1,…,K}; the UAV's transmit power is P. U The drone's flight altitude is H; the time required for the drone to complete one flight mission is T; T is divided into N time slots, each time slot having a length of... The position coordinates of the UAV in each time slot are x[n] = [x[n], y[n], H]. T , n∈N=={1,…,N}; Each flight time slot of the UAV is divided into two sub-time slots, including the first sub-time slot length τ0[n]δ used for downlink energy transmission. t The second sub-slot length τ1[n]δ used for uplink data collection t The drone's flight speed does not exceed V. max .

[0108] More specifically, the method for optimizing the three-dimensional coordinate IoT transmission system of drone base stations is as follows:

[0109] In the nth time slot, a line-of-sight channel fading model is established based on the air-to-ground wireless channel model and the geographical locations of the sensor nodes and the UAV.

[0110] The line-of-sight channel fading model is as follows:

[0111]

[0112] Where β0 represents the channel gain when the distance between the ground sensor and the UAV is 1m.

[0113] At the τ0[n]δ t When the UAV transmits energy to the ground sensor in the downlink for the length of the first sub-time slot, the energy collected by the k-th sensor in the first sub-time slot is:

[0114] E k [n]=ητ0[n]δ t P U h k [n]

[0115] Where η∈(0,1] represents the energy receiving efficiency of each ground sensor.

[0116] At the τ1[n]δ tDuring the second sub-time slot length, all ground sensors transmit data to the UAV via the uplink. The data acquisition rate of the UAV within the second sub-time slot is:

[0117]

[0118] Among them, [a] + Represents max(a,0); P k [n] represents the transmit power of the k-th sensor for uplink data transmission in the n-th time slot; σ 2 This indicates the noise power received by the drone.

[0119] To ensure sustainable data acquisition, the power of each ground sensor must meet a transmit power constraint, which is as follows:

[0120]

[0121] The mobility constraints for drones are as follows:

[0122] ||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}

[0123] Where, δ d = max δ t .

[0124] make Indicates drone trajectory, and Indicates wireless power supply time slot allocation, Let represent the transmit power of the k-th sensor; the objective function for optimizing the three-dimensional coordinate IoT transmission system of the UAV base station is the first objective function, which is expressed as:

[0125]

[0126] The constraints of the first objective function include:

[0127] A1. This indicates that the total energy transmitted by the ground sensors is no higher than the total energy received from the drone.

[0128] A2.||x[n+1]-x[n]||≤V max δ t ,n∈N - This indicates that the drone's flight speed does not exceed its maximum flight speed.

[0129] A3.τ0[n]+τ1[n]≤1 means that the sum of the sub-slot allocation coefficients within each slot is not higher than 1.

[0130] A4. This indicates that the allocation coefficient for each time slot is not less than 0.

[0131] A5. This indicates that the data transmission power of each sensor used for the uplink is not less than 0.

[0132] More specifically, the objective function of the resource allocation optimization model P1 is the second objective function, which is expressed as:

[0133]

[0134] The constraints of the second objective function include:

[0135] B1.

[0136] B2.τ0[n]+τ1[n]≤1.

[0137] B3.

[0138] B4.

[0139] More specifically, for the second objective function, we introduce a first slack variable γ[n] and let The second objective function is transformed into a convex perspective function, and a convex constraint is introduced at the same time. Subsequently ordered

[0140] Represented as: The third objective function is obtained, and the third objective function is expressed as:

[0141]

[0142] The constraints of the third objective function include:

[0143] C1.

[0144] C2.

[0145] C3.

[0146] C4.

[0147] C5.

[0148] C6.

[0149] in,

[0150] More specifically, the objective function of the UAV trajectory optimization model P2 is the fourth objective function, which is expressed as:

[0151]

[0152] The constraints of the fourth objective function include:

[0153] D1.

[0154] D2.||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}.

[0155] More specifically, for the fourth objective function, a second slack variable is introduced. And a first-order Taylor expansion is used to transform h k The lower bound of [n] is represented as:

[0156]

[0157] Where, x (t) [n] represents the optimal value of x[n] in the t-th iteration; the fifth objective function is obtained, which is expressed as:

[0158]

[0159] The constraints of the fifth objective function include:

[0160] E1.

[0161] E2.

[0162] E3.

[0163] E4.||x[n+1]-x[n]||≤δ d ,n∈N - ={1,…,N-1}.

[0164] in,

[0165] More specifically, the steps for alternating solutions in S4 are as follows:

[0166] S41: Initialize the drone base station's flight path to X (0) Wireless power supply time slot allocation is T0(0) and T1 (0) The sensor's transmission power is P (0) And calculate the initial uplink data acquisition rate R. (0) The error threshold η is 10 -3 The number of alternating iterations i = 0;

[0167] S42: Set the flight path X of the drone base station (i) Substituting these values ​​into the resource allocation optimization model P1, and using the CVX solver, the optimal solution T0 for wireless power supply time slot allocation and sensor transmission power in the i-th iteration is obtained. (i+1) T1 (i+1) and P k (i+1) ;

[0168] S43: The optimal solution T0 obtained in S42 (i+1) T1 (i+1) and P k (i+1) And flight trajectory X (i) Substituting into the UAV flight trajectory optimization model P2, we obtain the optimal solution X for the UAV flight trajectory in the i-th iteration. (i+1) And calculate the uplink data acquisition rate R. (i+1) ;

[0169] S44: If |R (i+1) -R (i) If |≤η, the optimal UAV flight trajectory X is obtained. (i+1) Time slot allocation T0 (i+1) and T1 (i +1) Sensor transmit power P k (i+1) Otherwise, let i = i + 1 and repeat S42 and S43.

[0170] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it performs the steps of the method described above.

[0171] Example 3

[0172] In this embodiment, MATLAB is used to simulate the aforementioned UAV wireless-powered aerial computing-assisted IoT data acquisition method; the simulation parameters are set as follows:

[0173] All UAVs and ground sensor nodes are equipped with only one antenna; the number of ground sensor nodes is 5; all sensors are uniformly and randomly distributed within a 200m * 200m rectangular area; the total flight time for a UAV to complete one mission is T = 20s, and this total flight time is divided into N = 40 time slots; the UAV's flight altitude is H = 50m; the UAV's maximum flight speed is V. max =40m / s; the reference channel power gain of the link between the UAV and the ground sensor node is β0 = -30dBm; the noise variance received at the UAV is σ. 2 = -104dBm; the drone's transmit power is P U =40dBm; Algorithm convergence accuracy is 10 -3 .

[0174] In this embodiment, the "hovering", "equal time slot allocation hovering", and "geometric center hovering" schemes are the position strategy schemes adopted by existing UAVs, and "cruising" is the UAV's mobile flight trajectory calculated by this invention.

[0175] In this embodiment, as Figure 3 , Figure 4 As shown, the sensor positions are represented by "○", and the numbers in "○" represent the sensor serial numbers; "Hover" indicates the optimal position obtained by the UAV using a static hovering strategy; "Cruise" indicates the UAV's flight trajectory obtained using this invention; "Initial Trajectory" indicates the trajectory centered on the optimal position obtained by the UAV using a static hovering strategy. A circle with radius is used as the initial point for alternating optimization of the "cruise" scheme; when the drone can only hover statically, the optimal hovering position is the center of the smallest circumcircle of the convex hull of all sensor node positions; when the drone is in flight, it will hover at point B near sensors 1 and 2 and point A near sensors 3, 4, and 5. Combined with... Figure 4 When approaching points A and B, more time will be allocated to downlink wireless power transmission, while when between points A and B, more time will be allocated to uplink in-flight computing, thereby achieving a balance between energy harvesting and energy consumption among the various sensors; in addition, the optimal flight trajectory always lies within the convex hull of all sensor locations.

[0176] In this embodiment, as Figure 5 As shown, the "equal time slot allocation hovering" scheme means that the UAV hovers at the optimal position and distributes the total flight time T evenly to the downlink energy transmission and the uplink in-flight computing; the "geometric center hovering" scheme means that the UAV always hovers at the geometric center of all sensor positions: that is, Because static hovering of UAVs results in severe long-distance path attenuation, the three methods of "hovering," "equal-slot hovering," and "geometric center hovering" exhibit a significant dual near-far effect. For example, in the "hovering" and "equal-slot hovering" methods, sensors 2 and 3, which are furthest from the optimal hovering position, collect the least energy but consume the most. When sensors 2 and 3 run out of energy, even if other sensors have energy remaining, the system still cannot perform in-flight computation, resulting in low energy efficiency. In the "cruising" method proposed in this invention, the dual near-far problem is effectively improved. Not only do all sensors exhaust their energy simultaneously, but all sensors can also collect more energy to support in-flight computation in the uplink. Therefore, the optimized design proposed in this invention can better achieve efficient data aggregation for large-scale ground-based IoT devices.

[0177] In this embodiment, as Figure 6 As shown, the sum of data acquisition rates is defined as Maximum flight speed V of the drone max The values ​​are 10, 20, and 40 m / s, respectively. As can be seen from the figure, the sum of the system data acquisition rates increases as the UAV's flight speed increases. Furthermore, the alternating optimization method proposed in this invention converges in no more than 5 iterations at different UAV flight speeds, and the number of iterations required for algorithm convergence does not increase significantly with the increase of the UAV's speed.

[0178] In this embodiment, as Figure 7 As shown, with the continuous increase of the UAV's transmission power, the sum of data acquisition rates also increases, and the performance of the "cruising" scheme proposed in this invention is better than the other three schemes. When the UAV's flight speed increases, the performance of the "cruising" scheme improves to a certain extent. When the UAV's flight speed is low, the performance of the "cruising" scheme is similar to that of the "hovering" scheme. When the UAV's transmission power is low, the calculation rate of the "equal time slot allocation hovering" scheme is zero, which reflects the importance of resource allocation in the UAV's wirelessly powered aerial computing system. When the UAV's transmission power is high enough, the performance of the "geometric center hovering" scheme is better than that of the "equal time slot allocation hovering" scheme, verifying the necessity of optimizing the UAV's hovering position.

[0179] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A method for drone wireless powered aerial computing assisted IoT data collection, characterized by the steps of Comprise: S1: establish a three-dimensional coordinate Internet of Things transmission system of unmanned aerial base station, establish an optimization objective of the three-dimensional coordinate Internet of Things transmission system of unmanned aerial base station by jointly optimizing the flight trajectory of the unmanned aerial vehicle, the wireless power supply time slot allocation, and the sensor transmission power; S2: establish a resource allocation optimization model P1 for jointly optimizing the wireless power supply time slot allocation and the sensor transmission power; S3: establish an unmanned aerial vehicle trajectory optimization model P2 for optimizing the flight trajectory of the three-dimensional coordinate Internet of Things transmission system of unmanned aerial base station; S4: alternately solve the optimization objectives of the resource allocation optimization model P1, the UAV trajectory optimization model P2, and the three-dimensional coordinate Internet of Things transmission system of the UAV base station, to obtain the flight trajectory of the UAV, the wireless energy supply time slot allocation, and the sensor transmission power when the uplink data collection rate of the three-dimensional coordinate Internet of Things transmission system is maximized; wherein the flight trajectory is always located within the convex hull of all sensor positions; the initial trajectory is a circle with the optimal position obtained by the UAV adopting a static hovering strategy as the center, when the UAV can only be static, the optimal hovering position is the center of the minimum circumscribed circle of the convex hull of all sensor node positions; when the UAV is in flight, hovering at a preset hovering point close to part of the ground sensor nodes, and when close to the preset hovering point, more time is allocated to the wireless energy transmission of the downlink; during the flight between the preset hovering points, more time is allocated to the aerial computing of the uplink; wherein, represents the total flight time, represents the maximum flight speed of the UAV.

2. The unmanned aerial vehicle wireless-powered aerial computing assisted Internet of Things data collection method according to claim 1, wherein, The aforementioned three-dimensional coordinate IoT transmission system for drone base stations includes a drone base station and One ground sensor node; among which Round to the nearest integer; the drone transmits power wirelessly to the ground. Each sensor is powered and collects data from all sensors on the ground in a data aggregation manner.

3. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 2, wherein, When establishing the three-dimensional coordinate Internet of Things transmission system of unmanned aerial base station, the parameters are initialized and set; The initialization setup method is as follows: The position coordinates of the sensors are ,in The drone's transmission power is The drone's flight altitude is The time required for an unmanned aircraft to complete a single flight mission is ;Will Average score N There are 1 time slot, and the length of each time slot is 1. ; The position coordinates of the UAV at each time slot are , ; Each flight time slot of the UAV is divided into two sub-time slots, including a first sub-time slot length for energy transmission of downlink , a second sub-time slot length for data collection of uplink ; the flight speed of the UAV does not exceed .

4. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 3, wherein, The method for optimizing the three-dimensional coordinate Internet of Things transmission system of unmanned aerial base station is as follows: At the first n In the first time slot, a line-of-sight channel fading model is established according to the air-to-ground wireless channel model and the geographical positions of the sensor node and the unmanned aerial vehicle. The line-of-sight channel fading model is as follows: ； wherein, G1represents the channel gain when the ground sensor is spaced 1 m from the drone; In the When the first sub-time slot length is reached, the UAV transfers energy to the ground sensors in the downlink. Then the... k The energy collected by the sensors in the first sub-time slot is: ； wherein, represents the energy receiving efficiency of each ground sensor; At the first sub-slot length, all ground sensors transmit data to the uplink drone, and the data acquisition rate of the drone within the first sub-slot is: ； wherein represents ; represents the transmission power of the first sensor for uplink data transmission in the first time slot; represents the noise power received by the UAV; In order to meet the sustainable data collection, the power of each ground sensor meets the transmission power constraint, and the transmission power constraint is as follows: ; The unmanned aerial vehicle mobility constraint is as follows: ； wherein ; Let represent the trajectory of the UAV, and represent the wireless power supply time slot allocation, represent the transmission power of the first sensor; the objective function of the optimization target of the three-dimensional coordinate Internet of Things transmission system of the UAV base station is a first objective function, and the first objective function is represented as: ； The constraint conditions of the first objective function include: A1. , , indicates that the total energy transmitted by the ground sensor is not higher than the total energy received from the drone; A2. , indicates that the flight speed of the UAV is not higher than the maximum flight speed of the UAV; A3. denotes that the sum of the sub-slot allocation factors within each slot is not higher than 1; A4. , indicates that each time slot allocation factor is not lower than 0; A5. , indicates that the data transmit power for uplink of each sensor is not lower than 0.

5. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 4, wherein, The objective function of the resource allocation optimization model P1 is a second objective function, and the second objective function is expressed as: ； The constraint conditions of the second objective function include: B1. ， ； B2. ； B3. ； B4. 。 6. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 5, wherein, For the second objective function, introduce the first slack variable and let , the second objective function is transformed into a convex perspective function, while introducing a convex constraint ; then let , be represented as: ; the third objective function is obtained, which is represented as: ； The constraint conditions of the third objective function include: C1. ； C2. ； C3. ； C4. ； C5. ； C6. ； wherein .

7. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 6, wherein, The objective function of the unmanned aerial vehicle trajectory optimization model P2 is a fourth objective function, and the fourth objective function is expressed as: ； The constraint conditions of the fourth objective function include: D1. ； D2. 。 8. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 7, wherein, For the fourth objective function, introduce second slack variables and represent the lower bound of as ； wherein, represents the optimal value of the iteration ; a fifth objective function is obtained, which is represented as: ； The constraint conditions of the fifth objective function include: E1. ； E2. ； E3. ； E4. ； wherein , , .

9. The unmanned aerial vehicle wireless-powered aerial computing-assisted IoT data collection method of claim 8, wherein, The steps of the alternating solution in S4 are as follows: S41: initialize the flight trajectory of the unmanned aerial base station as , the wireless energy supply time slot allocation as , and , the sensor transmission power as , and calculate the initial uplink data collection rate , the error threshold ε as , and the number of alternative iterations ; S42: Set the flight trajectory of the unmanned aerial base station and substituted into the resource allocation optimization model P1, solved by CVX solver to obtain the optimal solution of the wireless power supply time slot allocation and sensor transmission power in the th iteration 、 and ; S43: The optimal solution obtained in S42 , and and flight path Substituting into the UAV flight trajectory optimization model P2, we obtain the first... The optimal solution of the UAV flight trajectory in the next iteration And calculate the uplink data acquisition rate. ; S44: If , the optimal UAV flight trajectory is obtained , time slot allocation and , sensor transmit power ; Else let S42 and S43 are repeated.

10. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1-9 are realized.

Images