A resource optimization method and system for a drone-assisted data transmission network

Abstract

The present invention discloses a resource optimization method and system for a drone-assisted data transmission network, including: establishing a joint optimization problem P1 based on the traversal reachable rate #imgabs0##imgabs1# and the bandwidth allocation coefficient η; optimizing and solving the joint optimization problem P1 to obtain the noise compression covariance matrix Ω of the drone * And the corresponding bandwidth allocation coefficient η * ; According to the bandwidth allocation coefficient η * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Design the compression noise of the drone; establish the joint optimization problem P2 based on the traversal reachable rates #imgabs2# and #imgabs3#; use the gradient descent algorithm to solve the joint optimization problem P2 and obtain the power allocation vector p of the ground AP * And the noise compression covariance matrix Q * ; According to the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs; effectively improve the collection efficiency of sensor data and the utilization rate of system spectrum resources.

Description

Technical Field

[0001] The present invention belongs to the field of communication resource management, and in particular relates to a resource optimization method and system for an unmanned aerial vehicle assisted data transmission network. Background Art

[0002] With the rapid development of the Internet of Things, various applications with low latency, high reliability and large-scale device connection requirements have emerged. In order to support these applications, a large number of sensor nodes are deployed in areas of interest to collect sensor data, forming wireless sensor networks. Due to the needs of practical applications, most sensor networks are deployed in remote areas.

[0003] With the advantages of flexible maneuverability and easy establishment of line-of-sight channels, drones play a huge role in assisting wireless sensor networks in reliable and timely data collection. Drones usually act as aerial data collectors to receive sensor data from sensors and forward the received data to remote servers for further processing. Drones can improve services for wireless sensor networks at a closer distance and with higher channel quality. Therefore, drone-assisted sensor data collection has huge advantages in performance, such as improving energy efficiency, reducing latency, and extending the life of wireless sensor networks. However, the entire system may generate a large amount of resource consumption. For remote servers, how to make reasonable resource scheduling and allocation is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a resource optimization method and system for a UAV-assisted data transmission network. The resource allocation scheme obtained by joint optimization divides the bandwidth of the data transmission network, designs the compression noise of the UAV, and designs the power allocation and compression noise of the ground AP. Under the premise of meeting the ground user rate requirements and the transmission power of the ground AP and the UAV, the sensor data collection efficiency and the system spectrum resource utilization can be effectively improved.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A first aspect of the present invention provides a method for optimizing resources of a drone-assisted data transmission network, comprising:

[0007] Using line-of-sight channel estimation to estimate the traversal rate of sensor data acquisition and the traversal achievable rate of the wireless fronthaul link between the drone’s antenna and the central processor The data collected by the sensor is sent to the central processor through the antenna of the drone; the effective range of the bandwidth allocation coefficient η of the data transmission network is determined by using the block coordinate descent algorithm;

[0008] Based on traversal reachability rate Traversal reachable rate The effective range of the bandwidth allocation coefficient η is used to establish the joint optimization problem of the noise compression of the UAV and the bandwidth allocation scheme of the data transmission network, which is recorded as the joint optimization problem P1. The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * ;

[0009] According to the bandwidth allocation coefficient η * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Designing for compression noise of UAVs.

[0010] Furthermore, it also includes: using large-dimensional random matrix theory to derive the downlink traversal rate of ground users and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The ground user is connected to the central processor through the ground AP; the ground user, the ground AP, the wired fronthaul link, the sensor, the antenna of the drone, the wireless fronthaul link and the central processor constitute the data transmission network;

[0011] Given a bandwidth allocation coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the ground AP is jointly optimized, which is recorded as joint optimization problem P2. The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ;

[0012] According to the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs.

[0013] Further, the method for determining the effective range of the bandwidth allocation coefficient η of the data transmission network using the block coordinate descent algorithm includes:

[0014] Under the premise of satisfying the transmission power of ground AP and the service quality of ground users, the effective range η of the bandwidth allocation coefficient η of the data transmission network is solved by the block coordinate descent algorithm. min ≤η≤η max , where η min and η maxare the minimum and maximum values ​​of the bandwidth allocation coefficient η respectively.

[0015] Furthermore, based on the traversal reachability rate Traversal reachable rate The method for establishing the joint optimization problem of noise compression of UAV and bandwidth allocation scheme of data transmission network based on the effective range of bandwidth allocation coefficient η includes:

[0016] Add traversal reachable rate Traversal reachable rate The weighted least square error and Dinckelbach algorithm are used to jointly optimize the noise compression of UAVs and the bandwidth allocation scheme of the data transmission network, and the expression is:

[0017]

[0018]

[0019] Ω≥0,

[0020]

[0021] In the formula, Ω is the compression noise matrix of the wireless fronthaul link, ω is the coefficient of the Dinckelbach algorithm, is the ergodic capacity of the wireless fronthaul link between the UAV and the CPU, δ max is η max The reciprocal of E i is the mean square error of the i-th sensor, W i For E i , Σ3 is the introduced auxiliary variable.

[0022] Furthermore, the line-of-sight channel estimation method is used to estimate the traversal rate of sensor data acquisition. The expression formula is:

[0023]

[0024]

[0025]

[0026]

[0027] In the formula, is the characteristic parameter corresponding to the i-th sensor in the receiving matrix in the minimum mean square error receiving scheme, is the transmission power of the i-th sensor, is the line-of-sight channel between the UAV and the i-th sensor, is the sensor transmit power matrix after removing the elements in column i, is the matrix of the line-of-sight channel matrix between the UAV and the sensor after removing the elements in the i-th column, σ 2 is the intensity of Gaussian white noise received by the drone, N is the number of antennas of the drone, I N Represented as an N*N dimensional identity matrix; (·) H Expressed as transposed conjugate.

[0028] Furthermore, the line-of-sight channel estimation method is used to estimate the traversal rate of the wireless fronthaul link between the sensor and the central processor. The expression formula is:

[0029]

[0030]

[0031] In the formula, P sn is the sensor’s transmit power matrix, is the line-of-sight channel matrix between the UAV and the sensor.

[0032] Furthermore, the joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the drone: * And the corresponding bandwidth allocation coefficient η * The specific process includes:

[0033] Initialize Ω (0) =σ 2 I N ,ω (0) =0, r out =0, r in = 0, set the accuracy requirement to ε, and set the maximum number of iterations to I max ;

[0034] renew

[0035] The auxiliary variables are updated using the WMMSE estimation method.

[0036] Update the mean square error of the i-th sensor, and the calculation formula is:

[0037]

[0038] Mean square error The weight coefficient is updated:

[0039] The joint optimization problem P1 is solved by the gradient descent algorithm, and the solution is obtained The calculation formula is:

[0040]

[0041] Update in =r in +1; when | or in ≥I max When the noise compression covariance matrix of the output drone is Denoted as the noise compression covariance matrix Ω * ;

[0042] calculate

[0043] The weight coefficient ω is updated by the Dinkelbach algorithm:

[0044]

[0045] renew r out =r out +1;

[0046] Repeat until or out ≥I max , calculate and output the bandwidth allocation coefficient η of the data transmission network * ;

[0047] Furthermore, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the established ground AP are jointly optimized, and the expression formula is:

[0048] find{p,Q}

[0049]

[0050]

[0051]

[0052] Where p is the power allocation vector of the ground AP, Q is the compression noise covariance matrix of the ground AP, and C m is the fixed capacity of the wired fronthaul link connected to the ground APm, is the rate requirement of ground users, δ * Expressed as bandwidth allocation coefficient η * The reciprocal of m(p,Q) is the power consumption of ground APm, u m,k represents the mth row and kth column element of the RZF precoding matrix, (Q) mm represents the mth diagonal element of the compressed noise covariance matrix Q, P m,max is the maximum transmit power of the ground APm.

[0053] A second aspect of the present invention provides a resource optimization system for a drone-assisted data transmission network, comprising:

[0054] The estimation module is used to estimate the traversal rate of sensor data acquisition using the line-of-sight channel estimation method. and the traversal achievable rate of the wireless fronthaul link between the drone’s antenna and the central processor The data collected by the sensor is sent to the central processor through the antenna of the drone; the effective range of the bandwidth allocation coefficient η of the data transmission network is determined by using the block coordinate descent algorithm;

[0055] Joint optimization module for traversal-based reachability rate Traversal reachable rate The effective range of the bandwidth allocation coefficient η is used to establish the joint optimization problem of the noise compression of the UAV and the bandwidth allocation scheme of the data transmission network, which is recorded as the joint optimization problem P1. The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * ;

[0056] An execution module is used to allocate a coefficient η according to the bandwidth * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Designing for compression noise of UAVs.

[0057] Furthermore, the method further comprises: the estimation module uses large-dimensional random matrix theory to derive the downlink traversal achievable rate of the ground user and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The ground user is connected to the central processor through the ground AP; the ground user, the ground AP, the wired fronthaul link, the sensor, the antenna of the drone, the wireless fronthaul link and the central processor constitute the data transmission network;

[0058] The joint optimization module allocates a given bandwidth coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the ground AP is jointly optimized, which is recorded as joint optimization problem P2. The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ;

[0059] The execution module allocates power according to the power distribution vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] The present invention is based on traversal reachable rate Traversal reachable rate The effective range of the bandwidth allocation coefficient η is used to establish the joint optimization problem of the noise compression of the UAV and the bandwidth allocation scheme of the data transmission network, which is recorded as the joint optimization problem P1. The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * ; According to the bandwidth allocation coefficient η * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Design the compression noise of the UAV; given the bandwidth allocation coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the ground AP is jointly optimized, which is recorded as joint optimization problem P2. The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ; According to the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs; maximize the sum rate of sensor data acquisition while meeting the ground user rate requirements and the transmission power of ground APs and drones, while improving the utilization rate of system spectrum resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart of a resource optimization method for a drone-assisted data transmission network provided in an embodiment. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0064] The sensor data acquisition network has an unmanned aerial vehicle equipped with N (N≥1) antennas and I (I≥1) sensor nodes; the non-cellular large-scale MIMO system is composed of M (M>>1) ground APs, a CPU, and K (K≥1) ground users, and the ground users are connected to the central processor through the ground AP; the unmanned aerial vehicle collects sensor data and transmits it to a non-cellular large-scale MIMO CPU for further processing, wherein the antenna of the unmanned aerial vehicle to the central processor is set as a wireless fronthaul link, and the ground AP to the central processor is set as a wired fronthaul link; the sensor to the unmanned aerial vehicle is set as an uplink data transmission link, and the ground AP to the user is set as a downlink data transmission link; the data transmission network is composed of ground users, downlink data transmission links, ground APs, wired fronthaul links, unmanned aerial vehicle antennas, uplink data transmission links, sensors, wireless fronthaul links and central processors.

[0065] like Figure 1 As shown, this implementation provides a resource optimization method for a drone-assisted data transmission network, including:

[0066] Using line-of-sight channel estimation to estimate the traversal rate of sensor data acquisition and the traversal achievable rate of the wireless fronthaul link between the drone’s antenna and the central processor The methods include:

[0067] The traversal achievable rate of sensor data acquisition by line-of-sight channel estimation method is calculated when the central processing unit (CPU) adopts the minimum mean square error (MMSE) reception scheme. and the traversal achievable rate of the wireless fronthaul link The estimation is performed to achieve the goal of not relying on small-scale information.

[0068] The method for determining the effective range of the bandwidth allocation coefficient η of the data transmission network using the block coordinate descent algorithm includes:

[0069] Under the premise of satisfying the transmission power of ground AP and the service quality of ground users, the effective range η of the bandwidth allocation coefficient η of the data transmission network is solved by the block coordinate descent algorithm. min ≤η≤η max , where η min and η maxare the minimum and maximum values ​​of the bandwidth allocation coefficient η respectively.

[0070] Using large-dimensional random matrix theory to derive the downlink traversal rate of ground users and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The process includes:

[0071] Under the premise that the ground AP adopts Regularized Zero-Forcing (RZF) precoding and independent compression scheme, the uplink traversal achievable rate of ground users that does not depend on small-scale information is derived using large-dimensional random matrix theory. and the traversal achievable rate of the wired fronthaul link

[0072] Add traversal reachable rate Traversal reachable rate The weighted least square error and Dinckelbach algorithm are used to jointly optimize the noise compression of UAVs and the bandwidth allocation scheme of the data transmission network, which is denoted as the joint optimization problem P1. The expression is:

[0073]

[0074]

[0075] Ω≥0,

[0076]

[0077] In the formula, Ω is the compression noise matrix of the wireless fronthaul link, ω is the coefficient of the Dinckelbach algorithm, is the ergodic capacity of the wireless fronthaul link between the UAV and the CPU, δ max is η max The reciprocal of E i is the mean square error of the i-th sensor, W i For E i , Σ3 is the introduced auxiliary variable.

[0078] Among them, the traversal rate of sensor data collection is The expression formula is:

[0079]

[0080]

[0081]

[0082]

[0083] In the formula, is the characteristic parameter corresponding to the i-th sensor in the receiving matrix in the minimum mean square error receiving scheme, is the transmission power of the i-th sensor, is the line-of-sight channel between the UAV and the i-th sensor, is the sensor transmit power matrix after removing the elements in column i, is the matrix of the line-of-sight channel matrix between the UAV and the sensor after removing the elements in the i-th column, σ 2 is the intensity of Gaussian white noise received by the drone, N is the number of antennas of the drone, I N Represented as an N*N dimensional identity matrix; (·) H Expressed as transposed conjugate.

[0084] Traversal reachable rate The expression formula is:

[0085]

[0086]

[0087] In the formula, P sn is the sensor’s transmit power matrix, is the line-of-sight channel matrix between the UAV and the sensor.

[0088] The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * The specific process includes:

[0089] Initialize Ω (0) =σ 2 I N ,ω (0) =0, r out =0, r in = 0, set the accuracy requirement to ε, and set the maximum number of iterations to I max ;

[0090] renew

[0091] The auxiliary variables are updated using the WMMSE estimation method.

[0092] Update the mean square error of the i-th sensor, and the calculation formula is:

[0093]

[0094] Mean square error The weight coefficient is updated:

[0095] The joint optimization problem P1 is solved by the gradient descent algorithm, and the solution is obtained The calculation formula is:

[0096]

[0097] Update in =r in +1; when | or in ≥I max When the noise compression covariance matrix of the output drone is Denoted as the noise compression covariance matrix Ω * ;

[0098] calculate

[0099] The weight coefficient ω is updated by the Dinkelbach algorithm:

[0100]

[0101] renew r out =r out +1;

[0102] Repeat until or out ≥I max , calculate and output the bandwidth allocation coefficient η of the data transmission network * .

[0103] Given a bandwidth allocation coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the established ground AP is jointly optimized, denoted as joint optimization problem P2; the expression is:

[0104] find{p,Q}

[0105]

[0106]

[0107]

[0108] Where p is the power allocation vector of the ground AP, Q is the compression noise covariance matrix of the ground AP, and C m is the fixed capacity of the wired fronthaul link connected to the ground APm, is the rate requirement of ground users, δ * Expressed as bandwidth allocation coefficient η * The reciprocal of m (p,Q) is the power consumption of ground APm, u m,k represents the mth row and kth column element of the RZF precoding matrix, (Q) mm represents the mth diagonal element of the compressed noise covariance matrix Q, P m,max is the maximum transmit power of the ground APm.

[0109] The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ; According to the bandwidth allocation coefficient η * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Design the compression noise of the UAV; according to the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs.

[0110] This embodiment shares the total system bandwidth in the non-cellular massive MIMO network with the wireless access network and the wireless fronthaul network, and uses the processing power of the CPU to process the sensor data collected by the drone. Combined with the data processing power of the CPU in the non-cellular massive MIMO, the further transmission and processing of the data collected by the drone is considered, which can effectively improve the efficiency of sensor data collection and the utilization rate of system spectrum resources, and reduce communication overhead. The scenario of sharing bandwidth between the wireless access network and the wireless fronthaul link is considered. In addition, the compression forwarding scheme is used to alleviate the problem of limited capacity of wired and wireless fronthaul links, further reduce the limitation of sensor data collection rate, and improve the utilization rate of system spectrum resources. It can maximize the sum rate of sensor data collection while meeting the ground user rate requirements and the transmission power of ground AP and UAV.

[0111] Example 2

[0112] This embodiment provides a resource optimization system for a drone-assisted data transmission network. The system in this embodiment can be applied to the method described in Example 1. The resource optimization system includes:

[0113] The estimation module is used to estimate the traversal rate of sensor data acquisition using the line-of-sight channel estimation method. and the traversal achievable rate of the wireless fronthaul link between the drone’s antenna and the central processor The data collected by the sensor is sent to the central processor through the antenna of the drone; the effective range of the bandwidth allocation coefficient η of the data transmission network is determined by the block coordinate descent algorithm; the downlink traversal reachable rate of the ground user is derived by using the large-dimensional random matrix theory. and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The ground user is connected to the central processor through the ground AP; the data transmission network is composed of the ground user, the ground AP, the wired fronthaul link, the sensor, the antenna of the UAV, the wireless fronthaul link and the central processor;

[0114] Joint optimization module for traversal-based reachability rate Traversal reachable rate The effective range of the bandwidth allocation coefficient η is used to establish the joint optimization problem of the noise compression of the UAV and the bandwidth allocation scheme of the data transmission network, which is recorded as the joint optimization problem P1. The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * ;

[0115] At the same time, the joint optimization module allocates a given bandwidth coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the ground AP is jointly optimized, which is recorded as joint optimization problem P2. The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ;

[0116] An execution module is used to allocate a coefficient η according to the bandwidth * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * The compression noise of the UAV is designed; the execution module is based on the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs.

[0117] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0118] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0121] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A resource optimization method for a drone-assisted data transmission network, characterized in that: include: Using line-of-sight channel estimation to estimate the traversal rate of sensor data acquisition The expression formula is: In the formula, is the characteristic parameter corresponding to the i-th sensor in the receiving matrix in the minimum mean square error receiving scheme, is the transmission power of the i-th sensor, is the line-of-sight channel between the UAV and the i-th sensor, is the sensor transmit power matrix after removing the elements in column i, is the matrix of the line-of-sight channel matrix between the UAV and the sensor after removing the elements in the i-th column, σ 2 is the intensity of Gaussian white noise received by the drone, N is the number of antennas of the drone, I N represents the N*N dimensional identity matrix; (·) H represents transposed conjugate; The ergodic achievable rate of the wireless fronthaul link between the sensor and the central processor is estimated using the line-of-sight channel estimation method The expression formula is: In the formula, P (sn) is the sensor’s transmit power matrix, is the line-of-sight channel matrix between the UAV and the sensor; the data collected by the sensor is sent to the central processor through the antenna of the UAV; The method for determining the effective range of the bandwidth allocation coefficient η of the data transmission network using the block coordinate descent algorithm includes: Under the premise of satisfying the transmission power of ground AP and the service quality of ground users, the effective range η of the bandwidth allocation coefficient η of the data transmission network is solved by the block coordinate descent algorithm. min ≤η≤η max , where η min and η max are the minimum and maximum values ​​of the bandwidth allocation coefficient η respectively; Based on traversal reachability rate Traversal reachable rate The effective range of bandwidth allocation coefficient η is used to establish the joint optimization problem of UAV noise compression and data transmission network bandwidth allocation scheme, which is recorded as joint optimization problem P1. The specific process includes: Add traversal reachable rate Traversal reachable rate The weighted least square error and Dinckelbach algorithm are used to express the joint optimization problem of UAV noise compression and data transmission network bandwidth allocation scheme, and the expression is: In the formula, Ω is the noise compression covariance matrix of the wireless fronthaul link, ω is the coefficient of the Dinckelbach algorithm, is the ergodic capacity of the wireless fronthaul link between the UAV and the CPU, δ max is η max The reciprocal of E i is the mean square error of the i-th sensor, W i For E i The weight of Σ3 is the auxiliary variable introduced; I C represents the number of sensor nodes; The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * The specific process includes: Initialize Ω (0) =σ 2 I N ,ω (0) =0, r out =0, r in = 0, set the accuracy requirement to ε, and set the maximum number of iterations to I max ; renew The auxiliary variables are updated using the WMMSE estimation method. Update the mean square error of the i-th sensor, and the calculation formula is: Mean square error The weight coefficient is updated: The joint optimization problem P1 is solved by the gradient descent algorithm, and the solution is obtained The calculation formula is: Update in =r in +1; when or in ≥I max When the noise compression covariance matrix of the output drone is Denoted as the noise compression covariance matrix Ω * ; calculate The weight coefficient ω is updated by the Dinkelbach algorithm: renew r out =r out +1; Repeat until or out ≥I max , calculate and output the bandwidth allocation coefficient η of the data transmission network * ; According to the bandwidth allocation coefficient η * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Design for compression noise of drones; Using large-dimensional random matrix theory to derive the downlink traversal rate of ground users and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The ground user is connected to the central processor through the ground AP; the ground user, the ground AP, the wired fronthaul link, the sensor, the antenna of the drone, the wireless fronthaul link and the central processor constitute the data transmission network; Given a bandwidth allocation coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the established ground AP is jointly optimized, denoted as joint optimization problem P2; the expression formula is: find{p,Q} Where p is the power allocation vector of the ground AP, Q is the noise compression covariance matrix of the ground AP, and C m is the fixed capacity of the wired fronthaul link connected to the ground APm, is the rate requirement of ground users, δ * represents the bandwidth allocation coefficient η * The reciprocal of m (p,Q) is the power consumption of ground APm, u m,k represents the mth row and kth column element of the RZF precoding matrix, (Q) mm represents the mth diagonal element of the noise compression covariance matrix Q, P m,max is the maximum transmit power of ground APm; M represents the number of ground APs; K represents the number of ground users; The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ; According to the power allocation vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs.

2. A resource optimization system for drone-assisted data transmission networks, characterized in that: include: The estimation module is used to estimate the traversal rate of sensor data acquisition using the line-of-sight channel estimation method. and the traversal achievable rate of the wireless fronthaul link between the drone’s antenna and the central processor The data collected by the sensor is sent to the central processor through the antenna of the drone; the effective range of the bandwidth allocation coefficient η of the data transmission network is determined by using the block coordinate descent algorithm; Joint optimization module for traversal-based reachability rate Traversal reachable rate The effective range of bandwidth allocation coefficient η is used to establish the joint optimization problem of UAV noise compression and data transmission network bandwidth allocation scheme, which is recorded as joint optimization problem P1. The joint optimization problem P1 is optimized and solved with the goal of maximizing the total rate of sensor data acquisition to obtain the noise compression covariance matrix Ω of the UAV. * And the corresponding bandwidth allocation coefficient η * ; An execution module is used to allocate a coefficient η according to the bandwidth * The data transmission network bandwidth is divided according to the noise compression covariance matrix Ω * Design for compression noise of drones; The estimation module uses large-dimensional random matrix theory to derive the downlink traversal rate of ground users. and the traversal achievable rate of the wired fronthaul link between the ground AP and the central processor The ground user is connected to the central processor through the ground AP; the ground user, the ground AP, the wired fronthaul link, the sensor, the antenna of the drone, the wireless fronthaul link and the central processor constitute the data transmission network; The joint optimization module allocates a given bandwidth coefficient η * Under the premise of adding ground AP power and ground user rate constraints, based on the downlink traversal achievable rate and traversal reachability rate The power allocation and noise compression scheme of the ground AP is jointly optimized, which is recorded as joint optimization problem P2. The gradient descent algorithm is used to solve the joint optimization problem P2 to obtain the power allocation vector p of the ground AP. * And the noise compression covariance matrix Q * ; The execution module allocates power according to the power distribution vector p * and noise compression covariance matrix Q * Design the power allocation and compression noise of ground APs; The method for determining the effective range of the bandwidth allocation coefficient η of the data transmission network by the estimation module using the block coordinate descent algorithm includes: Under the premise of satisfying the transmission power of ground AP and the service quality of ground users, the effective range η of the bandwidth allocation coefficient η of the data transmission network is solved by the block coordinate descent algorithm. min ≤η≤η max , where η min and η max are the minimum and maximum values ​​of the bandwidth allocation coefficient η respectively; The joint optimization module is based on the traversal reachable rate Traversal reachable rate The method for establishing the joint optimization problem of noise compression of UAV and bandwidth allocation scheme of data transmission network based on the effective range of bandwidth allocation coefficient η includes: Add traversal reachable rate Traversal reachable rate The weighted least square error and Dinckelbach algorithm are used to express the joint optimization problem of UAV noise compression and data transmission network bandwidth allocation scheme, and the expression is: In the formula, Ω is the noise compression covariance matrix of the wireless fronthaul link, ω is the coefficient of the Dinckelbach algorithm, is the ergodic capacity of the wireless fronthaul link between the UAV and the CPU, δ max is η max The reciprocal of E i is the mean square error of the i-th sensor, W i For E i The weight of Σ3 is the auxiliary variable introduced; I C represents the number of sensor nodes; The estimation module uses the line-of-sight channel estimation method to estimate the traversal rate of sensor data acquisition The expression formula is: In the formula, is the characteristic parameter corresponding to the i-th sensor in the receiving matrix in the minimum mean square error receiving scheme, is the transmission power of the i-th sensor, is the line-of-sight channel between the UAV and the i-th sensor, is the sensor transmit power matrix after removing the elements in column i, is the matrix of the line-of-sight channel matrix between the UAV and the sensor after removing the elements in the i-th column, σ 2 is the intensity of Gaussian white noise received by the drone, N is the number of antennas of the drone, I N represents the N*N dimensional identity matrix; (·) H represents transposed conjugate; The estimation module estimates the traversal achievable rate of the wireless fronthaul link between the sensor and the central processor using the line-of-sight channel estimation method The expression formula is: In the formula, P (sn) is the sensor’s transmit power matrix, is the line-of-sight channel matrix between the UAV and the sensor; The joint optimization module optimizes and solves the joint optimization problem P1 with the goal of maximizing the total rate of sensor data acquisition, and obtains the noise compression covariance matrix Ω of the drone * And the corresponding bandwidth allocation coefficient η * The specific process includes: Initialize Ω (0) =σ 2 I N ,ω (0) =0, r out =0, r in = 0, set the accuracy requirement to ε, and set the maximum number of iterations to I max ; renew The auxiliary variables are updated using the WMMSE estimation method. Update the mean square error of the i-th sensor, and the calculation formula is: Mean square error The weight coefficient is updated: The joint optimization problem P1 is solved by the gradient descent algorithm, and the solution is obtained The calculation formula is: Update in =r in +1; when or in ≥I max When the noise compression covariance matrix of the output drone is Denoted as the noise compression covariance matrix Ω * ; calculate The weight coefficient ω is updated by the Dinkelbach algorithm: renew r out =r out +1; Repeat until or out ≥I max , calculate and output the bandwidth allocation coefficient η of the data transmission network * ; The joint optimization module is based on the downlink traversal reachable rate and traversal reachability rate The power allocation and noise compression scheme of the established ground AP are jointly optimized, and the expression formula is: find{p,Q} Where p is the power allocation vector of the ground AP, Q is the noise compression covariance matrix of the ground AP, and C m is the fixed capacity of the wired fronthaul link connected to the ground APm, is the rate requirement of ground users, δ * represents the bandwidth allocation coefficient η * The reciprocal of m (p,Q) is the power consumption of ground APm, u m,k represents the mth row and kth column element of the RZF precoding matrix, (Q) mm represents the mth diagonal element of the noise compression covariance matrix Q, P m,max is the maximum transmission power of ground APm; M represents the number of ground APs; K represents the number of ground users.

Images