Unmanned aerial vehicle deployment method integrating wireless charging and data transmission in passive sensing network

Abstract

An unmanned aerial vehicle deployment method integrating wireless charging and data transmission in a passive sensing network comprises the following steps: 1) constructing system models including a network model and a wireless charging model; 2) establishing an optimization problem, wherein the process is as follows: 2.1) deducing expected time and energy consumption for sending and receiving a data packet; 2.2) calculating the charging time and the data transmission time of the network; 2.3) calculating a network throughput rate; 2.4) optimizing the problem; and 3) an optimization problem is solved, the deployment position of the unmanned aerial vehicle after joint optimization and node-unmanned aerial vehicle association are output, the unmanned aerial vehicle flies to a specified position according to an output instruction, node-unmanned aerial vehicle association is carried out, then nodes are charged, data transmission is carried out, and thus the network throughput rate is maximized. According to the invention, the position of the unmanned aerial vehicle and node-unmanned aerial vehicle association are jointly optimized, so that the charging time and the data transmission time are reduced, and the throughput rate is improved.

Description

Technical Field

[0001] The present invention belongs to the field of passive sensing networks and relates to a method for deploying unmanned aerial vehicles integrating wireless charging and data transmission in a passive sensing network. Background Art

[0002] A large number of passive sensor nodes are deployed in the BFSN, which obtain energy from radio frequency (RF) signals to support the nodes to realize functions such as environmental perception, task execution and data transmission. In BFSN, drones equipped with RF transceivers have been widely used. Drones hover above BFSN and serve as aerial charging stations to charge nodes. At the same time, they also serve as network repeaters for data broadcasting and data collection.

[0003] In some IoT applications, such as smart farms, BFSN is deployed in the target area, and nodes need to exchange information with drones. For example, drones can broadcast a batch of operation instructions to nodes to instruct nodes to perform corresponding tasks. In addition, drones can also collect perception data from nodes to provide data support for managers' decision-making. In this process, before the node transmits data, the drone must first charge the node. After the node obtains enough energy, it can receive instructions from the drone or upload perception data to the drone.

[0004] As can be seen from the above, throughput is crucial for IoT applications. This is because once the throughput is low, the drone's operating instructions cannot be broadcast to the nodes in time, or the data sensed by the nodes cannot be uploaded to the drone in time, thus missing the best time to perform the task. However, there are two main problems in efficiently broadcasting / collecting data in BFSN: (1) Insufficient RF energy supply: The RF energy acquisition efficiency of the nodes is generally low, resulting in a long charging time. (2) Long data transmission time. Traditional BFSNs usually use a tree structure for data transmission. On the one hand, data needs to be forwarded through multiple hops before it can be received. On the other hand, nodes close to the root node are more likely to run out of energy because these nodes must receive and forward data from their child nodes, affecting the sustainability of data transmission. The above problems easily lead to low throughput for data broadcasting and data collection. Summary of the invention

[0005] In order to overcome the shortcomings of the existing technology, the present invention provides a UAV deployment method integrating wireless charging and data transmission in a passive sensing network, which jointly optimizes the position of the UAV and the node-UAV association to reduce the charging time and data transmission time, thereby improving the throughput.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A method for deploying a drone integrating wireless charging and data transmission in a passive sensing network, the method comprising the following steps:

[0008] 1) Build a system model, including a network model and a wireless charging model. The process is as follows:

[0009] 1.1) Construct a network model, which includes M nodes and Q drones. The indexes of nodes and drones are represented by the sets Ψ = {1, 2, ..., M} and Γ = {1, 2, ..., Q} respectively. The mth node is represented by A m , the qth UAV is represented by U q ,The drone hovers above the node and is responsible for charging the node and broadcasting / collecting data. During the charging process, A m Energy can be obtained from the radio frequency signals transmitted by multiple drones. During the data transmission process, each node needs to be associated with a drone to receive data or send data to the drone. A(q) represents the energy associated with U q A collection of associated nodes, each passive node in the network has only one antenna, so charging and data transmission cannot be performed at the same time;

[0010] 1.2) Build a wireless charging model

[0011] In g m,q (t) represents A m At time t, from U q The received signal strength, when Q = 1, is obtained according to the Friss equation:

[0012]

[0013] Among them, G s and G r Represent the antenna gains of the transmitter and receiver respectively; λ is the wavelength; P s is the transmitting power of the UAV; L p is polarization attenuation; D m,q (t) represents A at time t n with U q The distance between them; let η be the energy conversion efficiency of the node energy capture circuit, then A m The charging power at time t is h m (t) = ηg m,q (t); When Q>1, for node A m For example, the RF signal received from each drone has different attenuation and phase shift, then A m At time t, from U q The received signal is Therefore, at time t, A mThe total charging power is Within a period of time T, A m The energy obtained E m (T) = ∫ T h m (t)dt;

[0014] 2) Establish the optimization problem. The process is as follows:

[0015] 2.1) Derive the expected time and energy consumption of sending and receiving a data packet

[0016] A m ,U q ) indicates connection A m and U q The link is b m,q Indicates link (A m ,U q ) bit error rate, bit error rate Among them, B N is the noise bandwidth, R is the data rate, ψ m,q =P R -P th , P R is the received signal strength, calculated by the Friss equation, P th It is the threshold of the received signal, usually set to -85dBm;

[0017] For a given data packet, at most K transmissions are performed. This is because in wireless networks, the MAC layer usually sets the maximum number of retransmissions, which is p. m,q represents the success probability of transmitting a data packet once, then Where L D is the length of the packet in bits; C and σ T represents the channel access time and the value of the timer set at the sender. If the timer times out and the sender does not receive its ACK, the data packet will be retransmitted. The jth transmission will only occur if all the previous j-1 transmissions have failed, and its probability is (1-p m,q ) j-1 , the time required is (j-1)(T D +σ T ), where T D =σ C +L D / R. The probability of successful transmission in the jth time is (1-p m,q ) j-1 p m,q , the time required is (j-1)(T D+σ T )+(T D +T A ), j = 1, 2, ..., K, where T A =σ A +L A / r+σ IFS , σ A is the interval time required before sending ACK, σ IFS is the inter-frame interval, L A is the length of the ACK packet, and r is the data rate at which the ACK packet is transmitted. If the Kth transmission fails, its probability is The time required is K(T D +σ T ). Therefore, A m Through the link (A m ,U q ) sends a data packet to U q The expected time is:

[0018]

[0019] In addition, A m Through the link (A m ,U q ) sends a data packet to U q The expected energy consumption is:

[0020]

[0021] Among them, θ s and θ r are the energy consumption of sending and receiving 1 bit, θ s and θ r The value of can be obtained through actual measurement;

[0022] Take (U q ,A m ) indicates connection U q and A m The link is b q,m Indicates link (U q ,A m ) bit error rate, then A m Through the link (U q ,A m )Receive from U q The expected time for a packet is:

[0023]

[0024] The corresponding expected energy consumption is:

[0025]

[0026] 2.2) Calculate the network charging time and data transmission time

[0027] The time to charge the nodes is determined by the current remaining energy of each node and the expected energy consumption required to receive / send data packets. Represents the number of data packets broadcasted by the drone to the node, expressed as N m Represents node A m The number of packets sent to the drones associated with it, so the node needs to receive packets from the drones. packets and send N m The expected energy required for a data packet to be delivered to the drone is: by Indicates A m The initial energy of A m The total energy that needs to be harvested from the RF signal from the drone is:

[0028]

[0029] Therefore, A m In order to obtain energy m The required charging time is:

[0030]

[0031] Therefore, the total charging time in the network is:

[0032]

[0033] For each drone, it needs to broadcast In addition, it needs to receive N packets from each node associated with it. m data packets, therefore, UAV U q The data transfer time is:

[0034]

[0035] Therefore, the total data transmission time of the network is determined by the UAV with the longest data transmission time, and we get:

[0036]

[0037] 2.3) Calculate network throughput

[0038] The network throughput is defined as the total number of bits of data broadcast and collected by the drone divided by the sum of the charging time and data transmission time. The network throughput is:

[0039]

[0040] 2.4) Optimization Problem

[0041] From formula (11), we can see that the throughput is closely related to the charging time and data transmission time, while the charging time and data transmission time are determined by the location of the UAV and the node-UAV association. Therefore, the throughput is maximized by jointly optimizing the location of the UAV and the node-UAV association;

[0042] With W = {ω1,ω2,...,ω Q} represents the position of Q drones, where ω q =(x q ,y q ,z q ) is U q The coordinates of Representing the node-drone association, the following optimization problem is obtained:

[0043]

[0044] In the above formula, the variables to be optimized are the position of the drone and the node-drone association. The Ω in the first constraint C1 represents the candidate space, which ensures that the position of the drone is in the candidate space and the positions of any two drones do not overlap. The second constraint C2 ensures that any node can only be associated with one drone.

[0045] 3) To solve the optimization problem, a heuristic algorithm Association is designed to optimize the node-UAV association, and then an algorithm based on particle swarm optimization PSO-UAVdeployment is designed to optimize the position of the UAV; the deployment position of the UAV and the node-UAV association after joint optimization are output. The UAV will fly to the specified location and perform node-UAV association according to the output instructions, and then charge the node and transmit data, thereby maximizing the network throughput.

[0046] Further, the process of step 3) is as follows:

[0047] 3.1) A heuristic algorithm Association is designed to optimize the node-drone association. Input: The position of the drone W = {ω1,ω2,...,ω Q}, the node set A m ,m∈Ψ; Output: A(q),q∈Γ, the steps are as follows:

[0048] 3.1.1: Each node A m Select the UAV with the strongest signal for association, thus obtaining the initial node-UAV association A(q), q∈Γ;

[0049] 3.1.2: Sort the drones in reverse order of the number of nodes associated with them to obtain the drone queue Φ;

[0050] 3.1.3: For the first drone U in the drone queue Φ q =Φ[1], determine the nearest neighboring UAV U q′ ;

[0051] 3.1.4: If |A(q)|>|A(q′)|+1, then find the distance UAV U in the set A(q) q' The nearest node A m ;

[0052] 3.1.5: A(q′)←A(q′)∪A m ; A(q)=A(q)\A m , UAV q Dequeue from queue Φ;

[0053] 3.1.6: Return to step 3.1.3 until the drone queue is empty;

[0054] 3.1.7: Return to step 3.1.2 until the difference in the number of drones associated with adjacent drones does not exceed 1;

[0055] 3.1.8: End.

[0056] Furthermore, the step 3) further comprises the following steps:

[0057] 3.2) Design the particle swarm optimization algorithm PSO-UAVdeployment to optimize the position of the UAV. Input: the set of nodes A m ,m∈Ψ, the set of drones U q ,q∈Γ, the number of particles H, the dimension of each particle Q, the maximum speed of the particle V max , number of iterations G, inertia weight w and learning rates c1 and c2, as well as the candidate position Ω of the drone, etc. Output: The position of the drone W = {ω1,ω2,...,ω Q}, the steps are as follows:

[0058] 3.2.1: Randomly initialize the initial position of each particle in H particles: h i =([x iq ,y iq ,z iq ]) i∈{1,2,...,H},q∈Γ ,[x iq ,y iq ,z iq ]∈Ω;

[0059] 3.2.2: Randomly initialize the positions of Q UAVs in each particle: v i = ([α iq , β iq , γ iq ), α iq , β iq , γ iq ∈[-V max , V max ;

[0060] 3.2.3: Set Θ ← 0;

[0061] 3.2.4: Set g ← 0;

[0062] 3.2.5: Set g ← g + 1;

[0063] 3.2.6: For each of the H particles, call the algorithm Association to obtain the node-UAV association;

[0064] 3.2.7: Calculate the throughput Θ corresponding to each particle using Equation (11), i.e., throughput Θ i , i ∈ {1, 2,..., H}; Determine the coordinates corresponding to the maximum throughput obtained by each particle so far:

[0065] 3.2.8: Θ temp ← max i∈{1,2,...,H} {Θ i};

[0066] 3.2.9: I ← arg{max i∈{1,2,...,H} {Θ i}}; Set the best positions found by all particles so far

[0067] 3.2.10: If Θ temp > Θ, then Θ ← Θ temp ;

[0068] 3.2.11: Update the velocity of each particle according to Equation (13) below;

[0069]

[0070] 3.2.12: Update the position of each particle according to Equation (14) below:

[0071]

[0072] 3.2.13: If g < G, go to step 3.2.5, otherwise, output W and go to step 3.2.14;

[0073] 3.2.14: End.

[0074] The beneficial effects of the present invention are mainly manifested in: reducing charging time and data transmission time, and improving throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a schematic diagram of the wireless charging model.

[0076] Figure 2 It is a schematic diagram of data broadcasting / collection. DETAILED DESCRIPTION

[0077] The present invention will be further described below in conjunction with the accompanying drawings.

[0078] Reference Figure 1 and Figure 2 , a method for deploying a UAV integrating wireless charging and data transmission in a passive sensing network, comprising the following steps:

[0079] 1) Build a system model, including a network model and a wireless charging model. The process is as follows:

[0080] 1.1) Build a network model, such as Figure 1 As shown, Figure 1 Contains M nodes and Q drones. The indexes of nodes and drones are represented by the sets Ψ = {1, 2, ..., M} and Γ = {1, 2, ..., Q} respectively. The mth node is represented by A m , the qth UAV is represented by U q ,The drone hovers above the node and is responsible for charging the node and broadcasting / collecting data. During the charging process, A m Energy can be obtained from radio frequency signals transmitted by multiple drones, such as Figure 1 During data transmission, each node needs to be associated with a drone to receive data or send data to the drone, such as Figure 2 As shown. A(q) represents the q A collection of associated nodes, each passive node in the network has only one antenna, so charging and data transmission cannot be performed at the same time.

[0081] 1.2) Build a wireless charging model

[0082] In g m,q (t) represents A m At time t, from U q Received signal strength (RSS), when Q = 1, according to the Friss equation:

[0083]

[0084] Among them, G s and G r Represent the antenna gains of the transmitter and receiver respectively; λ is the wavelength; P s is the transmitting power of the UAV; L p is polarization attenuation; D m,q (t) represents A at time t m with U q The distance between them; let η be the energy conversion efficiency of the node energy capture circuit, then A m The charging power at time t is h m (t) = ηg m,q (t); When Q>1, for node A m For example, the RF signal received from each drone has different attenuation and phase shift, then A m At time t, from U q The received signal is Therefore, at time t, A m The total charging power is Within a period of time T, A m The energy obtained E m (T) = ∫ T h m (t)dt.

[0085] 2) Establish the optimization problem. The process is as follows:

[0086] 2.1) Derive the expected time and energy consumption of sending and receiving a data packet

[0087] A m ,U q ) indicates connection A m and U q The link is b m,q Indicates link (A m ,U q ) bit error rate, bit error rate Among them, B N is the noise bandwidth, R is the data rate, ψ m,q =P R -P th , P R is the received signal strength, calculated by the Friss equation, P th It is the threshold of the received signal, usually set to -85dBm;

[0088] For a given data packet, at most K transmissions are performed. This is because in wireless networks, the MAC layer usually sets the maximum number of retransmissions, which is p.m,q represents the success probability of transmitting a data packet once, then Where L D is the length of the packet in bits; C and σ T represents the channel access time and the value of the timer set at the sender. If the timer times out and the sender does not receive its ACK, the data packet will be retransmitted. The jth transmission will only occur if all the previous j-1 transmissions have failed, and its probability is (1-p m,q ) j-1 , the time required is (j-1)(T D +σ T ), where T D =σ C +L D / R, the probability of successful transmission in the jth time is (1-p m,q ) j-1 p m,q , the time required is (j-1)(T D +σ T )+(T D +T A ), j = 1, 2, ..., K, where T A =σ A +L A / r+σ IFS , σ A is the interval time required before sending ACK, σ IFS is the inter-frame interval, L A is the length of the ACK packet, r is the data rate at which the ACK packet is transmitted, and if the Kth transmission fails, its probability is The time required is K(T D +σ T ). Therefore, A m Through the link (A m ,U q ) sends a data packet to U q The expected time is:

[0089]

[0090] In addition, A m Through the link (A m ,U q ) sends a data packet to U q The expected energy consumption is:

[0091]

[0092] Among them, θ s and θ rare the energy consumption of sending and receiving one bit, θ s and θ r The value of can be obtained through actual measurement;

[0093] Take (U q ,A m ) indicates connection U q and A m The link is b q,m Indicates link (U q ,A m ) bit error rate, then A m Through the link (U q ,A m )Receive from U q The expected time for a packet is:

[0094]

[0095] The corresponding expected energy consumption is:

[0096]

[0097] 2.2) Calculate the network charging time and data transmission time

[0098] The time to charge the nodes is determined by the current remaining energy of each node and the expected energy consumption required to receive / send data packets. Represents the number of data packets broadcasted by the drone to the node, expressed as N m Represents node A m The number of packets sent to the drones associated with it, so the node needs to receive packets from the drones. packets and send N m The expected energy required for a data packet to be delivered to the drone is: by Indicates A m The initial energy of A m The total energy that needs to be harvested from the RF signal from the drone is:

[0099]

[0100] Therefore, A m In order to obtain energy m The required charging time is:

[0101]

[0102] Therefore, the total charging time in the network is:

[0103]

[0104] For each drone, it needs to broadcast In addition, it needs to receive N packets from each node associated with it. m Therefore, UAV U q The data transfer time is:

[0105]

[0106] Therefore, the total data transmission time of the network is determined by the UAV with the longest data transmission time, and we get:

[0107]

[0108] 2.3) Calculate network throughput

[0109] The network throughput is defined as the total number of bits of data broadcast and collected by the drone divided by the sum of the charging time and data transmission time. The network throughput is:

[0110]

[0111] 2.4) Optimization Problem

[0112] From formula (11), we can see that the throughput is closely related to the charging time and data transmission time, while the charging time and data transmission time are determined by the position of the UAV and the node-UAV association. Therefore, the throughput is maximized by jointly optimizing the position of the UAV and the node-UAV association;

[0113] With W = {ω1,ω2,...,ω Q} represents the position of Q drones, where ω q =(x q ,y q ,z q ) is U q The coordinates of Representing the node-drone association, the following optimization problem is obtained:

[0114]

[0115] In the above formula, the variables to be optimized are the position of the drone and the node-drone association. The Ω in the first constraint C1 represents the candidate space, which ensures that the position of the drone is in the candidate space and the positions of any two drones do not overlap. The second constraint C2 ensures that any node can only be associated with one drone.

[0116] 3) To solve the optimization problem, a heuristic algorithm Association is designed to optimize the node-UAV association, and then an algorithm based on particle swarm optimization PSO-UAVdeployment is designed to optimize the position of the UAV; the deployment position of the UAV and the node-UAV association after joint optimization are output. The UAV will fly to the specified location and perform node-UAV association according to the output instructions, and then charge the node and transmit data, thereby maximizing the network throughput.

[0117] The process of step 3) is as follows:

[0118] 3.1) A heuristic algorithm Association is designed to optimize the node-drone association. Input: The position of the drone W = {ω1,ω2,...,ω Q}, the node set A m ,m∈Ψ; Output: A(q),q∈Γ, the steps are as follows:

[0119] 3.1.1: Each node A m Select the UAV with the strongest signal for association, thus obtaining the initial node-UAV association A(q), q∈Γ;

[0120] 3.1.2: Sort the drones in reverse order of the number of nodes associated with them to obtain the drone queue Φ;

[0121] 3.1.3: For the first drone U in the drone queue Φ q =Φ[1], determine the nearest neighboring UAV U q′ ;

[0122] 3.1.4: If |A(q)|>|A(q′)|+1, then find the distance UAV U in the set A(q) q' The nearest node A m ;

[0123] 3.1.5: A(q′)←A(q′)∪A m ; A(q)=A(q)\A m , UAV q Dequeue from queue Φ;

[0124] 3.1.6: Return to step 3.1.3 until the drone queue is empty;

[0125] 3.1.7: Return to step 3.1.2 until the difference in the number of drones associated with adjacent drones does not exceed 1;

[0126] 3.1.8: End;

[0127] 3.2) Design the particle swarm optimization algorithm PSO-UAVdeployment to optimize the position of the UAV. Input: the set of nodes A m ,m∈Ψ, the set of drones U q ,q∈Γ, the number of particles H, the dimension of each particle Q, the maximum speed of the particle V max , number of iterations G, inertia weight w and learning rates c1 and c2, as well as the candidate position Ω of the drone, etc. Output: The position of the drone W = {ω1,ω2,...,ω Q}, the steps are as follows:

[0128] 3.2.1: Randomly initialize the initial position of each particle in H particles: h i =([x iq ,y iq ,z iq ]) i∈{1,2,...,H},q∈Γ ,[x iq ,y iq ,z iq ]∈Ω;

[0129] 3.2.2: Randomly initialize the positions of Q drones in each particle: v i =([α iq ,β iq ,γ iq ]),α iq ,β iq ,γ iq ∈[-V max ,V max ];

[0130] 3.2.3: Set Θ←0;

[0131] 3.2.4: Set g←0;

[0132] 3.2.5: Set g←g+1;

[0133] 3.2.6: For each of the H particles, call the Association algorithm to obtain the node-drone association;

[0134] 3.2.7: Use formula (11) to calculate the revenue corresponding to each particle, that is, the throughput rate i∈{1,2,...,H}; Determine the coordinates corresponding to the maximum throughput achieved so far for each particle:

[0135] 3.2.8:

[0136] 3.2.9: Set all particles to the best position found so far

[0137] 3.2.10: If Θ temp > Θ, then Θ ← Θ temp ;

[0138] 3.2.11: Update the velocity of each particle according to the following formula (13).

[0139]

[0140] 3.2.12: Update the position of each particle according to the following formula (14):

[0141]

[0142] 3.2.13: If g < G, go to step 3.2.5; otherwise, output W and go to step 3.2.14;

[0143] 3.2.14: End.

[0144] In this embodiment, the algorithm is executed on the edge device of the network, such as a gateway. The main entry of the program is the algorithm PSO-UAVdeployment. In each iteration of executing this algorithm, PSO-UAVdeployment will call the algorithm Association. When the number of iterations reaches G, the algorithm outputs the deployment positions of the drones after joint optimization and the node-drone associations. The output of the algorithm will be sent by the gateway to the drones. Subsequently, the drones will fly to the specified positions according to the instructions of the algorithm, perform node-drone associations, then charge the nodes and perform data transmission, so as to maximize the network throughput rate.

[0145] The content described in the embodiments of this specification is only an enumeration of the implementation forms of the inventive concept and is only for illustrative purposes. The protection scope of the present invention should not be regarded as limited to the specific forms stated in this embodiment. The protection scope of the present invention also extends to equivalent technical means that can be thought of by those of ordinary skill in the art according to the inventive concept of the present invention.

Claims

1. A method for deploying unmanned aerial vehicles (UAVs) that integrates wireless charging and data transmission in a passive sensing network, characterized in that, the method comprises the following steps: 1) Construct a system model, including a network model and a wireless charging model, the process is as follows: 1.1) Construct a network model, which includes M nodes and Q drones. The indexes of nodes and drones are represented by the sets Ψ = {1, 2, ..., M} and Γ = {1, 2, ..., Q} respectively. The mth node is represented by A m , the qth UAV is represented by U q ,The drone hovers above the node and is responsible for charging the node and broadcasting / collecting data. During the charging process, A m Energy can be obtained from the radio frequency signals transmitted by multiple drones. During the data transmission process, each node needs to be associated with a drone to receive data or send data to the drone. A(q) represents the energy associated with U q A collection of associated nodes, each passive node in the network has only one antenna, so charging and data transmission cannot be performed at the same time; 1.2) Construct a wireless charging model In g m,q (t) represents A m At time t, from U q The received signal strength, when Q = 1, is obtained according to the Friss equation: Among them, G s and G r Represent the antenna gains of the transmitter and receiver respectively; λ is the wavelength; P s is the transmitting power of the UAV; L p is polarization attenuation; D m,q (t) represents A at time t m with U q The distance between them; let η be the energy conversion efficiency of the node energy capture circuit, then A m The charging power at time t is h m (t) = ηg m,q (t); When Q>1, for node A m For example, the RF signal received from each drone has different attenuation and phase shift, then A m At time t, from U q The received signal is Therefore, at time t, A m The total charging power is Within a period of time T, A m The energy obtained E m (T) = ∫ T g m (t)dt; 2) Establish an optimization problem, the process is as follows: 2.1) Derive the expected time and energy consumption for sending and receiving a data packet A m ,U q ) indicates connection A m and U q The link is b m,q Indicates link (A m ,U q ) bit error rate, bit error rate Among them, B N is the noise bandwidth, R is the data rate, ψ m,q =P R -P th , P R is the received signal strength, calculated by the Friss equation, P th It is the threshold of the received signal, usually set to -85dBm; For a given data packet, at most K transmissions are performed. This is because in wireless networks, the MAC layer usually sets the maximum number of retransmissions, which is p. m,q represents the success probability of transmitting a data packet once, then Where L D is the length of the packet in bits; C and σ T represents the channel access time and the value of the timer set at the sender. If the timer times out and the sender does not receive its ACK, the data packet will be retransmitted. The jth transmission will only occur if all the previous j-1 transmissions have failed, and its probability is (1-p m,q ) j-1 , the time required is (j-1)(T D +σ T ), where T D =σ C +L D / R, the probability of successful transmission in the jth time is (1-p m,q ) j-1 p m,q , the time required is (j-1)(T D +σ T )+(T D +T A ), j = 1, 2, ..., K, where T A =σ A +L A / r+σ IFS , σ A is the interval time required before sending ACK, σ IFS is the inter-frame interval, L A is the length of the ACK packet, r is the data rate at which the ACK packet is transmitted, and if the Kth transmission fails, its probability is The time required is K(T D +σ T ); therefore, A m Through the link (A m ,U q ) sends a data packet to U q The expected time is: In addition, A m Through the link (A m ,U q ) sends a data packet to U q The expected energy consumption is: Among them, θ s and θ r are the energy consumption of sending and receiving 1 bit, θ s and θ r The value of can be obtained through actual measurement; Take (U q ,A m ) indicates connection U q and A m The link is b q,m Indicates link (U q ,A m ) bit error rate, then A m Through the link (U q ,A m )Receive from U q The expected time for a packet is: The corresponding expected energy consumption is: 2.2) Calculate the charging time and data transmission time of the network The time to charge the nodes is determined by the current remaining energy of each node and the expected energy consumption required to receive / send data packets. Represents the number of data packets broadcasted by the drone to the node, expressed as N m Represents node A m The number of packets sent to the drones associated with it, so the node needs to receive packets from the drones. packets and send N m The expected energy required for a data packet to be delivered to the drone is: by Indicates A m The initial energy of A m The total energy that needs to be harvested from the RF signal from the drone is: Therefore, A m In order to obtain energy m The required charging time is: Thus, the total charging time in the network is: For each drone, it needs to broadcast In addition, it needs to receive N packets from each node associated with it. m data packets, therefore, UAV U q The data transfer time is: Therefore, the total data transmission time of the network is determined by the UAV with the longest data transmission time, and we get: 2.3) Calculate the network throughput Define the network throughput as the total number of bits of data broadcast and collected by the UAVs divided by the sum of the charging time and the data transmission time, and the network throughput is obtained as: 2.4) Optimization problem As can be seen from Equation (11), the throughput is closely related to the charging time and the data transmission time, and the charging time and the data transmission time are determined by the positions of the UAVs and the node-UAV associations. Therefore, maximize the throughput by jointly optimizing the positions of the UAVs and the node-UAV associations; With W = {ω 1 ,ω 2 ,...,ω Q } represents the position of Q drones, where ω q =(x q ,y q ,z q ) is U q The coordinates of Representing the node-drone association, the following optimization problem is obtained: In the above formula, the variables to be optimized are the position of the drone and the node-drone association. The first constraint C 1 The Ω in the equation represents the candidate space, which ensures that the position of the drone is in the candidate space and the positions of any two drones do not overlap. The second constraint C 2 Ensure that any node can only be associated with one drone; 3) Solve the optimization problem, design a heuristic algorithm Association to optimize the node-UAV association, and then design an algorithm PSO-UAVdeployment based on particle swarm optimization to optimize the positions of the UAVs; output the deployment positions of the UAVs and the node-UAV associations after joint optimization. The UAVs will fly to the specified positions according to the output instructions and establish node-UAV associations, and then charge the nodes and perform data transmission, thereby maximizing the network throughput.

2. The method for deploying UAVs that integrates wireless charging and data transmission in a passive sensing network according to claim 1, characterized in that, the process of step 3) is as follows: 3.1) A heuristic algorithm Association is designed to optimize the node-drone association, input: drone position W = {ω 1 ,ω 2 ,...,ω Q }, the node set A m ,m∈Ψ; Output: A(q), q ∈ Γ, the steps are as follows: 3.1.1: Each node A m Select the UAV with the strongest signal for association, thus obtaining the initial node-UAV association A(q), q∈Γ; 3.1.2: Sort the UAVs in reverse order according to the number of nodes associated with each UAV to obtain a UAV queue Φ; 3.1.3: For the first drone U in the drone queue Φ q =Φ[1], determine the nearest neighboring UAV U q′ ; 3.1.4: If |A(q)|>|A(q′)|+1, then find the distance UAV U in the set A(q) q′ The nearest node A m ; 3.1.5: A(q′)←A(q′)∪A m ; A(q)=A(q)\A m , UAV q Dequeue from queue Φ; 3.1.6: Return to step 3.1.3 until the UAV queue is empty; 3.1.7: Return to step 3.1.2 until the difference in the number of associated UAVs between adjacent UAVs does not exceed 1; 3.1.8: End.

3. The method for deploying UAVs that integrates wireless charging and data transmission in a passive sensing network according to claim 2, characterized in that, step 3) further comprises the following steps: 3.2) Design the particle swarm optimization algorithm PSO-UAVdeployment to optimize the position of the UAV. Input: the set of nodes A m ,m∈Ψ, the set of drones U q ,q∈Γ, the number of particles H, the dimension of each particle Q, the maximum speed of the particle V max , number of iterations G, inertia weight w and learning rate c 1 and c 2 , and the candidate position Ω of the drone, etc. Output: The position of the drone W = {ω 1 ,ω 2 ,...,ω Q }, the steps are as follows: 3.2.1: Randomly initialize the initial position of each particle in H particles: h i =([x iq ,y iq ,z iq ]) i∈{1,2,...,H},q∈Γ ,[x iq ,y iq ,z iq ]∈Ω; 3.2.2: Randomly initialize the positions of Q drones in each particle: v i =([α iq ,β iq ,γ iq ]),α iq ,β iq ,γ iq ∈[-V max ,V max ]; 3.2.3: Set Θ ← 0; 3.2.4: Set g ← 0; 3.2.5: Set g ← g + 1; 3.2.6: For each of the H particles, call the algorithm Association to obtain the node-UAV association; 3.2.7: Use formula (11) to calculate the revenue corresponding to each particle, that is, the throughput Θ i , i∈{1,2,...,H}; determine the coordinates corresponding to the maximum throughput obtained by each particle so far: 3.2.8: I temp ←max i∈{1,2,...,H} {I i }; 3.2.9: I←arg{max i∈{1,2,...,H} {Θ i }}; Set all particles to the best position found so far 3.2.10: If Θ temp >Θ, then Θ←Θ temp ; 3.2.11: Update the velocity of each particle according to the following Equation (13); 3.2.12: Update the position of each particle according to the following Equation (14): 3.2.13: If g < G, then go to step 3.2.5, otherwise, output W and go to step 3.2.14; 3.2.14: End.

Images