Joint optimization method for relay communication of multiple unmanned aerial vehicles in cognitive radio network and application

Abstract

The invention discloses a joint optimization method for multi-unmanned aerial vehicle relay communication in a cognitive wireless network and application, and relates to the technical field of wireless communication, a joint optimization model in a multi-unmanned aerial vehicle relay communication scene in the cognitive wireless network is constructed, the joint optimization model takes maximization of the minimum throughput of a user as an optimization target, and the optimal throughput of the user is maximized. Solving the joint optimization model by taking an unmanned aerial vehicle distribution constraint, a minimum bandwidth constraint, a bandwidth equivalent constraint, a maximum throughput constraint and a master user interference constraint as constraint conditions to obtain a joint optimization scheme; therefore, unmanned aerial vehicle-slave user matching, unmanned aerial vehicle position deployment and bandwidth allocation in a multi-unmanned aerial vehicle relay communication scene in a cognitive wireless network can be jointly optimized, the minimum throughput of the slave user is improved on the premise of ensuring the communication quality of a main network, and the frequency use fairness of the slave user is ensured.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a joint optimization method and application of drone-slave user matching, drone location deployment and bandwidth allocation for multi-drone relay communication in a cognitive wireless network. Background Art

[0002] In cities, cellular mobile networks are usually implemented through the deployment of ground base stations to provide communication services to users. However, in some areas with many obstructions, the communication between ground base stations and users may be interfered. As a new type of communication platform, drones are gradually applied to various fields due to their strong mobility and low manufacturing cost. Using drones as relays between ground base stations and users can effectively solve the above problems, but a single drone is prone to small coverage and limited computing power during the execution of tasks. Compared with a single drone, a multi-drone system has better mission adaptability and application potential. However, scarce spectrum resources have become a limiting factor for multi-drone relay communication. At the same time, with the increase in the number of users, the existing spectrum resource allocation strategy is difficult to guarantee the communication quality of each user, and the fairness of user frequency use is greatly challenged. Therefore, how to improve the spectrum resource utilization of multi-drone relay communication through cognitive radio technology under the premise of ensuring fair spectrum use is a difficult problem in the study of drone spectrum management.

[0003] Researchers have conducted extensive research on multi-UAV relay communication scenarios. For example, Wu Di et al. published a joint optimization method for user matching and spectrum resources in multi-UAV assisted communication in telecommunications technology, which studied the location deployment and resource optimization methods in multi-UAV relay communication. The method used genetic algorithms and convex optimization to alternately iterate to maximize the user's minimum transmission rate and ensure the fairness of user frequency usage. Arribas E et al. published an Exact Resource Allocation for Fair Wireless Relay in IEEE Communications Letters, which studied the allocation of relay UAV bandwidth by continuously increasing the user's minimum throughput with the goal of maximizing the user's minimum throughput, and ultimately achieved fairness in user communication. However, the above studies mainly focus on the frequency allocation problem of multiple UAVs as relays in a single network, without considering the spectrum reuse problem between multiple networks.

[0004] At present, some researchers have put multi-UAV relay communication scenarios into cognitive wireless networks. For example, Ding R et al. published From External Interaction to Internal Inference: An Intelligent Learning Framework for Spectrum Sharing and UAV Trajectory Optimization in IEEE Transactions on Wireless Communications, which studied the spectrum resource allocation and location optimization problems of UAVs in cognitive wireless networks. They used a hybrid online-offline multi-agent neural network optimization method based on external interaction to obtain the maximum total throughput of system users. However, the above research lacks consideration of the fairness of user frequency usage, and there is unfairness in user transmission. Summary of the invention

[0005] The purpose of this application is to provide a joint optimization method and application of multi-UAV relay communication in a cognitive wireless network, which can jointly optimize UAV-slave user matching, UAV location deployment and bandwidth allocation in a multi-UAV relay communication scenario in a cognitive wireless network, and improve the minimum throughput of slave users and ensure the fairness of frequency use of slave users while ensuring the communication quality of the main network.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a joint optimization method for multi-UAV relay communication in a cognitive wireless network, and the joint optimization method for multi-UAV relay communication in a cognitive wireless network includes:

[0008] A joint optimization model for a multi-UAV relay communication scenario in a cognitive wireless network is constructed; the joint optimization model takes maximizing the minimum throughput of slave users as an optimization goal, and takes UAV allocation constraints, minimum bandwidth constraints, bandwidth equality constraints, maximum throughput constraints, and primary user interference constraints as constraints; the primary user is a ground user provided with communication services by the primary network in the cognitive wireless network, the slave user is a ground user provided with communication services by the slave network in the cognitive wireless network, the slave network includes a temporary node, multiple UAVs, and multiple slave users, and the temporary node provides communication services to the slave users through multiple UAV relays;

[0009] The joint optimization model is solved to obtain a joint optimization solution; the joint optimization solution includes a drone-slave user matching optimization solution, a drone location deployment optimization solution and a bandwidth allocation optimization solution.

[0010] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned joint optimization method for multi-UAV relay communications in the cognitive wireless network.

[0011] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned joint optimization method for multi-UAV relay communications in the cognitive wireless network.

[0012] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned joint optimization method for multi-UAV relay communications in the cognitive wireless network.

[0013] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0014] The present application provides a joint optimization method and application of multi-UAV relay communication in a cognitive wireless network, constructs a joint optimization model for a multi-UAV relay communication scenario in a cognitive wireless network, and the joint optimization model takes maximizing the minimum throughput of slave users as the optimization goal, and takes UAV allocation constraints, minimum bandwidth constraints, bandwidth equality constraints, maximum throughput constraints and main user interference constraints as constraints. The joint optimization model is solved to obtain a joint optimization scheme, which includes a UAV-slave user matching optimization scheme, a UAV position deployment optimization scheme and a bandwidth allocation optimization scheme. Therefore, under the condition of simultaneously considering the throughput of slave users and the interference of main users, UAV-slave user matching, UAV position deployment and bandwidth allocation in a multi-UAV relay communication scenario in a cognitive wireless network can be jointly optimized, and the minimum throughput of slave users can be improved and the frequency fairness of slave users can be guaranteed while ensuring the communication quality of the main network. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A method flow chart of a joint optimization method for multi-UAV relay communications in a cognitive wireless network provided in Example 1 of the present application.

[0017] Figure 2Schematic diagram of the system model for a multi-UAV relay communication scenario in a cognitive wireless network provided in Example 1 of the present application.

[0018] Figure 3 Schematic diagram of the drone-user matching optimization solution and drone location deployment optimization solution provided in Example 1 of the present application.

[0019] Figure 4 A schematic diagram showing the comparison of user throughput before and after the linear precision algorithm optimization provided in Example 1 of the present application; Figure 4 (a) in the figure is the throughput from the user before optimization; Figure 4 (b) in the figure is the throughput from the user after optimization.

[0020] Figure 5 A schematic diagram comparing the bandwidth and throughput of a drone before and after optimization of the linear precision algorithm provided in Example 1 of the present application.

[0021] Figure 6 A schematic diagram of the structure of a computer device provided in Example 2 of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] Example 1

[0024] like Figure 1 As shown, an embodiment of the present application provides a joint optimization method for multi-UAV relay communication in a cognitive wireless network, and the joint optimization method for multi-UAV relay communication in a cognitive wireless network includes:

[0025] S1: Construct a joint optimization model for a multi-UAV relay communication scenario in a cognitive wireless network; the joint optimization model takes maximizing the minimum throughput of slave users as the optimization goal, and takes UAV allocation constraints, minimum bandwidth constraints, bandwidth equality constraints, maximum throughput constraints and primary user interference constraints as constraints; the primary user is a ground user provided with communication services by the primary network in the cognitive wireless network, the slave user is a ground user provided with communication services by the slave network in the cognitive wireless network, the slave network includes a temporary node, multiple UAVs and multiple slave users, and the temporary node provides communication services to the slave users through multiple UAV relays.

[0026] S2: Solving the joint optimization model to obtain a joint optimization solution; the joint optimization solution includes a drone-slave user matching optimization solution, a drone location deployment optimization solution, and a bandwidth allocation optimization solution.

[0027] In an actual communication environment, the uneven distribution of spectrum resources objectively exists. By implementing the above steps S1 to S2, the embodiment of the present application can jointly optimize the drone-slave user matching, drone location deployment and bandwidth allocation in the multi-drone relay communication scenario in the cognitive wireless network, and improve the minimum throughput of the slave user while ensuring the communication quality of the main network, thereby ensuring the fairness of the frequency use of the slave users.

[0028] System model of multi-UAV relay communication scenario in cognitive wireless network Figure 2 As shown, in order to alleviate the obstruction of buildings to communication links in urban communications, it is planned to deploy multiple rotor drones as relays over the target area to transmit the transmission signals of temporary nodes (i.e., temporary base stations) to ground users. This embodiment explores the downlink scenario of multi-drone relay communication in cognitive wireless networks. In this scenario, the network contains a temporary node g and multiple drones (i.e., Figure 2 The relay drone in the network is distributed with multiple ground users, which are recorded as slave users (i.e. Figure 2 The temporary nodes are used for temporary communication. Specifically, the temporary nodes provide communication services for the slave users through multiple drone relays. The drones are represented by M∈{1,2…,M}, where M is the total number of drones. The slave users are represented by K∈{1,2…,K}, where K is the total number of slave users. There is a main network in working state near the slave network. The main network includes a cellular network base station G. There is a ground user distributed in the main network, which is recorded as the main user (i.e. Figure 2 The cellular network base station provides communication services for the main network users in the main network. The main network and the slave network share spectrum resources. The spectrum reuse mode of the main network and the slave network adopts the Underlay mode. Mutual interference between the two is inevitable. To ensure smooth communication of the main network, the spectrum use of the slave network must be limited. Conversely, the communication quality of the slave users in the slave network is also affected by the main network.

[0029] When faced with the problem of matching multiple drones with multiple slave users, the slave users will be divided into multiple clusters (i.e., groups) according to the number of drones. Each cluster is managed by a cluster head drone. The decision variables involved in this process are defined as: w r,u , when w r,u = 1, it means that drone r serves user u. On the contrary, when w r,u = 0, it means that drone r does not serve user u.

[0030] In order to ensure the sustainability of communication, the following constraints must be met: any slave user can only be provided with communication services by one drone, namely:

[0031]

[0032] In formula (1), M is the total number of drones; w r,u is a binary variable indicating whether drone r serves user u.

[0033] In the slave network, the temporary node g allocates all bandwidth to the drone, and the drone allocates its bandwidth to the slave users in its cluster (i.e., the slave users that match the drone). The following constraints must be met for the bandwidth allocation of slave users:

[0034] First, considering the fairness of the frequency usage of the slave network, a minimum limit needs to be set for the bandwidth between the drone and the slave user, namely:

[0035]

[0036] In the above formula, w r is the bandwidth of drone r; is the minimum bandwidth of the drone; w u is the bandwidth from user u; The minimum bandwidth from the user.

[0037] Secondly, the sum of the bandwidths of all drones is equal to the maximum bandwidth that the temporary node can provide, and the sum of the bandwidths of the slave users in each cluster is equal to the bandwidth of the drones in the cluster, that is:

[0038]

[0039] In the above formula, is the maximum bandwidth that the temporary node g can provide; K is the total number of slave users.

[0040] Finally, the sum of the throughputs of the slave users of each cluster is less than or equal to the throughput of the drones in the cluster, and the entire relay network also needs to consider the backhaul bottleneck, that is, the sum of the throughputs provided by the temporary node g to all drones cannot exceed its backhaul bottleneck, that is:

[0041]

[0042] In the above formula, T u is the throughput from user u; T r is the throughput of drone r; τ is the backhaul bottleneck of temporary node g.

[0043] UAVs may interfere with the primary users of the primary network. In order to limit this interference, it is set that the interference of each UAV to the primary user shall not exceed a maximum threshold δ, that is:

[0044]

[0045] In formula (8), P r is the transmission power of UAV r; is the channel gain coefficient when UAV r communicates with the primary user u*; δ is the maximum tolerance threshold of the primary user to single-source interference.

[0046] In actual communication scenarios, air-to-ground communication links are often blocked by obstacles. This embodiment uses the A2G (Air-to-Ground) channel model at 2 GHz carrier frequency to describe the communication link from UAV m to ground user k. This communication link usually consists of two parts: Line of Sight (LoS) propagation link and Non-Line of Sight (NLoS) propagation link. The relevant channel gain coefficient g m,k The calculation formula is:

[0047]

[0048] In formula (9), g m,k is the channel gain coefficient when UAV m communicates with ground user k; P LOS is the transmission probability of the line-of-sight link; d Um,k is the horizontal distance between UAV m and ground user k, which is calculated based on the horizontal coordinates of UAV m and ground user k; α u is the space-to-ground path loss index; P NLOS is the transmission probability of the non-line-of-sight link, P NLOS =1-P LOS ; η is the attenuation factor of the non-line-of-sight propagation link, which reflects the additional loss in non-line-of-sight transmission.

[0049]

[0050] In formula (10), a and b are both constants, and they take different values ​​according to different propagation environments to adapt to specific communication scenarios; θ is the elevation angle.

[0051]

[0052] In formula (11), H is the flight altitude of the UAV.

[0053] This embodiment maximizes the minimum throughput of all slave users by optimizing the drone-slave user matching scheme, drone location deployment scheme, and bandwidth allocation scheme. The joint optimization model is:

[0054]

[0055] In formula (12), T u is the throughput from user u; M is the total number of drones; w r,u is a binary variable indicating whether drone r serves user u, all w r,u Composition of drones - matching solutions from users; w r is the bandwidth of drone r; is the minimum bandwidth of the drone; w u is the bandwidth from user u; is the minimum bandwidth from the user; is the maximum bandwidth that the temporary node g can provide; K is the total number of slave users; T r is the throughput of drone r; τ is the backhaul bottleneck of temporary node g; P r is the transmission power of UAV r; g r,u* is the channel gain coefficient when UAV r communicates with the primary user u*; δ is the maximum tolerance threshold of the primary user to single-source interference.

[0056] In formula (12), C1 is the UAV allocation constraint, which means that any slave user can only have one UAV to provide communication services for it; C2 and C3 are minimum bandwidth constraints, which means that the bandwidth of each UAV and each slave user has a minimum value; C4 means that the sum of the bandwidth of the UAVs is equal to the maximum bandwidth that the temporary node g can provide, C5 means that the sum of the bandwidth of the slave users in the cluster is equal to the bandwidth of the UAVs in the cluster, and C4 and C5 are bandwidth equality constraints; C6 means that the sum of the throughput of all slave users in the cluster is less than or equal to the throughput of the UAVs in the cluster, C7 means that the sum of the throughput of all UAVs is less than or equal to the throughput of the temporary node g (i.e., the backhaul bottleneck), and C6 and C7 are maximum throughput constraints; C8 is the master user interference constraint, which means that the interference of the slave network to the master network must be controlled within the maximum threshold range.

[0057] It can be concluded from formula (12) that the size of the slave user throughput is related to the distance between the slave user and the drone and the bandwidth allocation. The location deployment of the drone not only affects the size of the slave user throughput, but also needs to take its interference to the master user into consideration. This embodiment uses an improved K-Means clustering algorithm to optimize the matching of the drone and the slave user and the location deployment of the drone. For bandwidth allocation optimization, a linear precise algorithm is used to continuously increase the bandwidth of the user with the smallest throughput in each iteration without exceeding the backhaul bottleneck, thereby enhancing the fairness of user communication. That is, the joint optimization algorithm of this embodiment includes an improved K-Means clustering algorithm and a linear precise algorithm.

[0058] In actual communication scenarios, the distance between drones and slave users is a significant factor affecting communication quality. Multiple drones are matched and clustered with multiple slave users. One drone is deployed in each cluster, and the distance to the slave users in the cluster must be minimum. In addition, different positions of the deployed drones will cause different degrees of interference to the main user. If the size of the interference reaches the maximum threshold that affects the communication quality of the main network, the position of the drone needs to be re-corrected. Therefore, the location deployment of the drone must take into account both the distance to the cluster users and the distance to the main user. This embodiment uses an improved K-Means clustering algorithm to achieve matching between drones and slave users and the location deployment of drones. The specific solution steps are as follows:

[0059] (1) When the number of drones and the positions of slave users are known, a slave user is randomly selected as the initial center point for initialization, and M initial center points are obtained.

[0060] (2) During the loop, traverse each slave user and calculate the distance from each slave user to each initial center point, as shown below:

[0061] d u,m =||(X u ,Y u )-(X Um ,Y Um )||2 (13)

[0062] In formula (13), d u,m is the distance from user u to the mth initial center point; || ||2 is the Euclidean distance; (X u ,Y u ) are the x-coordinate and y-coordinate of user u respectively; (X Um ,Y Um ) are the x-coordinate and y-coordinate of the m-th initial center point respectively.

[0063] Each slave user is assigned to the initial center point closest to the slave user. At this time, the process of clustering the slave users is completed, and M clusters are obtained. The center point of each cluster is calculated. The coordinates of the center point are the mean of the coordinates of all slave users in the cluster, and M new center points are obtained.

[0064] (3) If the distance between the new center point and the initial center point is less than the set value, as shown in the following formula (14), the loop is stopped. The coordinates of the M new center points are the positions of the M drones, and the drone position deployment plan is obtained. The M clusters are the matching results between the drones and the slave users, and the drone-slave user matching optimization plan is obtained. Otherwise, the new center point is used as the initial center point of the next iteration and the loop continues.

[0065]

[0066] In formula (14), Δd is the distance between the new center point and the initial center point; are the x-coordinate and y-coordinate of the m-th new center point respectively; ε is the set value.

[0067] (4) After obtaining the UAV position deployment plan, for each UAV, if the interference of the UAV position to the primary user exceeds the maximum threshold, the interference is minimized by adjusting the position of the UAV. When adjusting the position of the UAV whose interference to the primary user exceeds the maximum threshold (i.e., the UAV to be adjusted), the maximum threshold of the interference of the UAV to be adjusted to the primary user is taken as the interference size. The maximum channel gain coefficient when the UAV to be adjusted communicates with the primary user is calculated according to formula (8). Then, the safe distance between the UAV to be adjusted and the primary user is inferred according to formula (9). The position of the UAV to be adjusted is adjusted based on the safe distance to obtain the UAV position deployment optimization plan.

[0068] The calculation formula for the maximum channel gain coefficient is:

[0069] g' r,u* =δ / P r (15)

[0070] In formula (15), g' r,u* is the maximum channel gain coefficient when the UAV r to be adjusted communicates with the main user u*; δ is the maximum tolerance threshold of the main user to single-source interference; P r is the transmission power of the UAV r to be adjusted.

[0071] The calculation formula for the safety distance is:

[0072]

[0073] In formula (16), d Ur,u* is the safe distance from the UAV r to be adjusted to the primary user u*; αu is the space-to-ground path loss index; P LOS is the transmission probability of the line-of-sight link; η is the attenuation factor of the non-line-of-sight link; P NLOS is the transmission probability of the non-line-of-sight link.

[0074] The specific process of the above improved K-Means clustering algorithm is shown in Table 1 below.

[0075] Table 1 UAV-user matching and UAV location deployment based on improved K-Means clustering algorithm

[0076]

[0077]

[0078] In the secondary user bandwidth allocation, the increase of the secondary user bandwidth of each cluster will lead to a corresponding increase in the bandwidth of the drone providing communication services for the secondary users in the cluster. At the same time, the bandwidth of the drone cannot exceed the maximum bandwidth that the temporary node can provide, and the total throughput of the system is also subject to the limit of the backhaul bottleneck. In order to optimize the minimum throughput and achieve fair allocation of secondary user spectrum resources, this embodiment uses a linear precise algorithm to optimize the minimum throughput of the secondary user to complete the bandwidth allocation optimization. The linear precise algorithm specifically includes:

[0079] (1) By analyzing the drone position and the main network interference, the signal-to-noise ratio γ of each slave user is calculated r,u The bandwidth of each drone is initially set to the minimum value In each cluster, the bandwidth of the drone is evenly distributed to each slave user in the cluster to obtain the bandwidth of each slave user. The bandwidth of each slave user constitutes the bandwidth allocation scheme of the previous iteration. For each slave user, the signal-to-noise ratio and bandwidth of the slave user are used as inputs, and the throughput of each slave user can be calculated based on Shannon's theorem.

[0080] The signal-to-noise ratio of the slave user is related to the ground position of the slave user. In addition to the environmental noise, the slave user will also be affected by the interference from the main network. The position of each slave user is different, and the interference is also different. Since the transmission power of the base station G of the main network and the drone of the slave network is constant, the signal-to-noise ratio of each slave user can be obtained. The calculation formula of the signal-to-noise ratio is:

[0081]

[0082] In formula (17), γ r,u is the signal-to-noise ratio when drone r communicates with slave user u; w r,u is a binary variable indicating whether drone r provides communication services to user u; p r is the transmission power of UAV r; gr,u is the channel gain coefficient when UAV r communicates with slave user u, which is calculated using formula (9); I CCI To identify the co-channel interference of a master network on a slave network in a wireless network; is the additive white Gaussian noise power.

[0083] I CCI =p G g1 (18)

[0084] In formula (18), p G is the transmission power of the base station G in the main network; g1 is the channel gain coefficient when the base station G in the main network communicates with the slave user u. It is only necessary to replace the drone in formula (9) with the base station G in the main network to calculate g1 using formula (9).

[0085] (2) Find the minimum throughput T of all slave users m , and find all the users whose throughput is second only to the minimum throughput T M ,as follows:

[0086]

[0087] The throughput of the slave users corresponding to the minimum throughput is gradually increased by using a cyclic improvement method. First, we need to determine in which clusters the slave users corresponding to the minimum throughput are distributed. Then, we calculate the number of slave users corresponding to the minimum throughput in each cluster |u|. The throughput increase caused by the minimum throughput in the cluster is |u|β(T M -T m ), β is the adjustment coefficient, and each minimum throughput is increased to obtain the adjusted throughput of each slave user. For each slave user, the signal-to-noise ratio and the adjusted throughput of the slave user are used as inputs, and based on the Shannon theorem, the bandwidth of each slave user in the current iteration is calculated. Further, based on the matching results between the drone and the slave user, the bandwidths of all slave users provided with communication services by the same drone are summed to obtain the bandwidth of each drone in the current iteration.

[0088] (3) Using Shannon’s formula, the increased throughput is expressed as follows:

[0089]

[0090] In formula (20), A is the number of minimum throughputs, that is, the number of minimum throughputs determined by this cycle; M is the total number of drones; is the additional bandwidth that leads to the increase in the throughput of the drone r; γ g,r is the signal-to-noise ratio when the temporary node g communicates with the drone r.

[0091] Based on formula (20), the calculation formula of β can be derived as follows:

[0092]

[0093] In formula (21), β is the adjustment coefficient; is the maximum bandwidth that the temporary node g can provide; M is the total number of drones; w r is the bandwidth of the current iteration of UAV r; A is the number of minimum throughput; T M is the second smallest throughput; T m is the minimum throughput; γ g,r is the signal-to-noise ratio when the temporary node g communicates with the drone r.

[0094] (4) When β is less than 0 or reaches the maximum number of iterations, it means that the sum of the bandwidths of all drones is greater than the maximum bandwidth that the temporary node can provide. At this time, the optimization is stopped, and the last optimization result is retained and output, that is, the bandwidth allocation plan of the last iteration is output.

[0095] (5) Checking the backhaul bottleneck τ: If the total throughput exceeds the backhaul bottleneck τ, the throughput of the highest user is reduced in sequence. Specifically, the throughput of the slave user with the highest throughput is reduced first until the sum of the total throughput is less than or equal to the backhaul bottleneck. If the total throughput is still greater than the backhaul bottleneck after the throughput of the slave user with the highest throughput is reduced to a preset value, a new slave user with the highest throughput is determined, and the process returns to the step of “reducing the throughput of the slave user with the highest throughput”.

[0096] The specific process of the linear precision algorithm is shown in Table 2.

[0097] Table 2 Bandwidth allocation algorithm based on linear precision algorithm

[0098]

[0099]

[0100] In this embodiment, the joint optimization model is solved to obtain a joint optimization solution, which specifically includes:

[0101] (1) A clustering algorithm is used to divide all slave users into multiple clusters, and each cluster corresponds to a drone. The slave users belonging to the cluster are matched with the drone corresponding to the cluster to obtain the drone location deployment plan and the drone-slave user matching optimization plan.

[0102] (2) Based on the primary user interference constraint, the UAV location deployment plan is adjusted to obtain the optimized UAV location deployment plan.

[0103] (3) Taking the drone-slave user matching optimization plan and the drone location deployment optimization plan as input, the linear exact algorithm is used to iteratively optimize and solve the joint optimization model to obtain the bandwidth allocation optimization plan.

[0104] In this embodiment, the drone position deployment plan is adjusted based on the primary user interference constraint to obtain the drone position deployment optimization plan, which specifically includes:

[0105] (1) Based on the UAV location deployment plan, determine whether each UAV meets the primary user interference constraints.

[0106] (2) If yes, the UAV location deployment plan is used as the UAV location deployment optimization plan.

[0107] (3) If not, the UAVs that do not meet the primary user interference constraint are recorded as UAVs to be adjusted; for each UAV to be adjusted, the maximum channel gain coefficient when the UAV to be adjusted communicates with the primary user is calculated based on the primary user interference constraint, and the safe distance from the UAV to be adjusted to the primary user is calculated based on the maximum channel gain coefficient, and the position of the UAV to be adjusted is adjusted based on the safe distance to obtain the adjusted position of the UAV to be adjusted, wherein the position of the UAV to be adjusted is determined based on the UAV position deployment plan.

[0108] Among them, adjusting the position of the drone to be adjusted based on the safety distance specifically includes: determining an adjustment circle with the main user as the center and the safety distance as the radius; and taking the position on the adjustment circle that is closest to the position of the drone to be adjusted as the adjusted position of the drone to be adjusted.

[0109] (4) The positions of all the drones to be adjusted in the drone position deployment plan are replaced with the adjusted positions of all the drones to be adjusted, so as to adjust the drone position deployment plan and obtain an optimized drone position deployment plan.

[0110] In this embodiment, the UAV-slave user matching optimization solution and the UAV location deployment optimization solution are used as inputs, and the linear precision algorithm is used to iteratively optimize and solve the joint optimization model to obtain the bandwidth allocation optimization solution, which specifically includes:

[0111] (1) Based on the bandwidth allocation scheme of the previous iteration, the Shannon theorem is used to calculate the throughput of each slave user in the current iteration.

[0112] (2) The throughput of each slave user in the current iteration is sorted to obtain the minimum throughput and the second minimum throughput of the current iteration. The minimum throughput is the minimum value of the throughput of each slave user in the current iteration, and the second minimum throughput is the minimum value after removing the minimum throughput from the throughput of each slave user in the current iteration.

[0113] (3) Based on the minimum throughput and the second minimum throughput of the current iteration and the adjustment coefficient of the previous iteration, the minimum throughput of the current iteration is increased to obtain the adjusted throughput of each slave user of the current iteration.

[0114] The minimum throughput of the current iteration is increased based on the minimum throughput and the second minimum throughput of the current iteration and the adjustment coefficient of the previous iteration to obtain the adjusted throughput of each slave user of the current iteration, which specifically includes:

[0115] 1) For each minimum throughput of the current iteration, determine the minimum UAV that matches the slave user corresponding to the minimum throughput, and determine the number of users whose throughput is the minimum throughput among the slave users matched with the minimum UAV; calculate the difference between the sub-minimum throughput of the current iteration and the minimum throughput, and calculate the product of the number of users, the adjustment coefficient of the previous iteration and the difference to obtain the increase; calculate the sum of the minimum throughput and the increase to increase the minimum throughput, and obtain the increased throughput corresponding to the minimum throughput.

[0116] 2) The minimum throughput among the throughputs of each slave user in the current iteration is replaced by the increased throughput corresponding to the minimum throughput, so as to obtain the adjusted throughput of each slave user in the current iteration.

[0117] (4) Based on the adjusted throughput of each slave user in the current iteration, the bandwidth of each slave user in the current iteration is calculated using Shannon's theorem to obtain a bandwidth allocation scheme for the current iteration.

[0118] (5) Based on the bandwidth allocation scheme of the current iteration, calculate the adjustment coefficient of the current iteration.

[0119] (6) Determine whether the iteration termination condition is met; if so, determine the bandwidth allocation scheme of the previous iteration as the initial bandwidth allocation scheme; if not, increase the current iteration number by 1, determine the bandwidth allocation scheme of the current iteration as the bandwidth allocation scheme of the previous iteration, and determine the adjustment coefficient of the current iteration as the adjustment coefficient of the previous iteration.

[0120] (7) Taking the backhaul bottleneck of the temporary node as a constraint, the initial bandwidth allocation plan is adjusted to obtain the bandwidth allocation optimization plan.

[0121] Below, this embodiment verifies the effectiveness of the proposed joint optimization algorithm through simulation. The simulation simulates the scenario of multi-UAV relay communication in a cognitive wireless network, where the network covers an area of ​​1000×1000 square meters and the UAV flies at an altitude of 100 meters. The simulation parameters are shown in Table 3 below.

[0122] Table 3 Simulation parameter settings

[0123]

[0124]

[0125] Figure 3 The matching results of drones and slave users and the location deployment results of drones are shown. The location of slave users is fixed, and the five-pointed star marks the location of the optimized drone. This optimization ensures that the distance between each slave user and its corresponding drone is minimized. At the same time, it also ensures that under the spectrum reuse conditions of Underlay, the interference of the transmission signal of the drone of the slave network to the slave user on the main user is kept within a certain range. Under the premise of ensuring the communication quality of the main network, all slave users are clustered and the location deployment of each drone is completed.

[0126] Figure 4 The comparison chart of the throughput of each slave user before and after the optimization using the linear precise algorithm in the scenario of 5 drones and 100 slave users is shown. The throughput setting method before optimization is: the bandwidth of each drone is set to the minimum bandwidth, and the bandwidth of the drone is evenly distributed to the slave users served by the drone, and the throughput of each slave user before optimization is obtained; the throughput setting method after optimization is: the throughput of each slave user after optimization is calculated using the linear precise algorithm. Due to the large number of slave users, every 4 slave users are sampled once, and the throughput comparison chart of 25 slave users is obtained. It can be seen from the figure that the throughput of users with smaller throughput before optimization has been significantly increased after optimization. In addition, some slave users have higher throughput before optimization. After optimization, the throughput of these slave users has been significantly reduced. This is because during the iteration process, the sum of the throughput of all drones exceeds the limit of the backhaul bottleneck, so the throughput of higher users will be reduced until the sum of the throughput of drones in the slave network is lower than the backhaul bottleneck.

[0127] Figure 5The figure shows the changes in bandwidth and throughput of different drones after optimization by the linear precision algorithm in a scenario with 5 drones and 100 slave users. It can be seen that after optimization, the drones in cluster 3 are allocated more bandwidth and have a larger throughput. The reason is that the slave users in cluster 3 are closer to the base station of the main network. Therefore, each slave user suffers greater interference from the main network, so each slave user has a relatively lower signal-to-noise ratio. Since the bandwidth allocation before optimization is evenly distributed, the pre-optimization throughput of the slave users in cluster 3 is smaller than that of other slave users. During the optimization process, more bandwidth is allocated, so the drones in cluster 3 have higher bandwidth and throughput after optimization. On the contrary, the drones in clusters 4 and 5 are farther away from the base station of the main network, so the slave users in the cluster have a larger pre-optimization throughput. Therefore, no additional bandwidth resources are allocated during the optimization process, so the throughput of their drones does not change much before and after optimization.

[0128] This embodiment mainly studies the position optimization and bandwidth allocation of drones in the slave network in the communication scenario of multiple drones as relays in cognitive wireless networks. In order to ensure the communication quality of the primary user and the frequency fairness of the slave users, the improved K-Means algorithm is used to solve the position deployment problem of drones, and the linear precision algorithm is used to optimize the bandwidth allocation of drones to slave users. The simulation results show that the algorithm of this embodiment can greatly improve the minimum throughput.

[0129] Drones have attracted much attention in cognitive wireless networks due to their low cost and flexible deployment. This embodiment addresses the spectrum allocation problem in which multiple drones act as relays to provide temporary services to users in cognitive wireless networks. The spectrum reuse and mutual interference between the master and slave networks are considered. Under the premise of ensuring the communication quality of the master user, the improved K-Means algorithm and the linear precision algorithm are proposed to maximize the minimum throughput of the slave user, respectively, to optimize the location deployment and bandwidth allocation of drones. The simulation results show that the proposed joint optimization method can effectively improve the throughput of user transmission and ensure the fairness of user communication.

[0130] This embodiment studies the situation in which multiple drones are used as relays in a cognitive wireless network and multiplex spectrum resources with the main network, and proposes a joint optimization method for the location deployment and bandwidth allocation of drones to improve the minimum throughput and achieve fairness in frequency usage for users. Compared with the prior art, the advantages of this embodiment are as follows:

[0131] (1) This embodiment constructs a multi-UAV relay communication scenario based on a cognitive wireless network, with the goal of improving the minimum throughput. To this end, an algorithm for jointly optimizing UAV location deployment and bandwidth allocation is proposed to reduce interference, improve the minimum throughput, and ensure frequency fairness, thereby maximizing network performance and spectrum resource utilization efficiency.

[0132] (2) The algorithm proposed in this embodiment takes into account the spectrum reuse problem between the master network and the slave network, and adopts an improved K-Means clustering algorithm and a linear precision algorithm to effectively reduce the mutual interference between the master network and the slave network, while optimizing the allocation of bandwidth resources and significantly improving the minimum throughput in the slave network.

[0133] (3) The simulation results show that the algorithm proposed in this embodiment can significantly improve the minimum throughput of the slave network users while ensuring the communication quality of the master network. This not only improves the overall performance of the communication network, but also ensures the fairness of communication between users. The simulation results also demonstrate the effectiveness and robustness of the proposed algorithm in improving the minimum throughput under different network conditions and scenarios.

[0134] Example 2

[0135] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a joint optimization method for multi-UAV relay communication in a cognitive wireless network is implemented.

[0136] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] In an exemplary embodiment, a computer device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the joint optimization method for multi-UAV relay communication in a cognitive wireless network as described in Example 1.

[0138] Example 3

[0139] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for jointly optimizing multi-UAV relay communications in a cognitive wireless network described in Example 1 is implemented.

[0140] Example 4

[0141] An embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the joint optimization method for multi-UAV relay communication in a cognitive wireless network described in Example 1.

[0142] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A joint optimization method for multi-UAV relay communication in a cognitive wireless network, characterized in that: The joint optimization method for multi-UAV relay communication in the cognitive wireless network includes: A joint optimization model for a multi-UAV relay communication scenario in a cognitive wireless network is constructed; the joint optimization model takes maximizing the minimum throughput of slave users as an optimization goal, and takes UAV allocation constraints, minimum bandwidth constraints, bandwidth equality constraints, maximum throughput constraints, and primary user interference constraints as constraints; the primary user is a ground user provided with communication services by the primary network in the cognitive wireless network, the slave user is a ground user provided with communication services by the slave network in the cognitive wireless network, the slave network includes a temporary node, multiple UAVs, and multiple slave users, and the temporary node provides communication services to the slave users through multiple UAV relays; The joint optimization model is solved to obtain a joint optimization solution; the joint optimization solution includes a drone-slave user matching optimization solution, a drone location deployment optimization solution and a bandwidth allocation optimization solution.

2. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 1 is characterized in that: The joint optimization model is: maxminT u Among them, T u is the throughput from user u; M is the total number of drones; w r,u is a binary variable indicating whether drone r serves user u; w r is the bandwidth of drone r; is the minimum bandwidth of the drone; w u is the bandwidth from user u; is the minimum bandwidth from the user; is the maximum bandwidth that the temporary node g can provide; K is the total number of slave users; T r is the throughput of drone r; τ is the backhaul bottleneck of temporary node g; P r is the transmission power of UAV r; is the channel gain coefficient when UAV r communicates with the primary user u*; δ is the maximum tolerance threshold of the primary user to single-source interference.

3. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 1, characterized in that: The joint optimization model is solved to obtain a joint optimization solution, which specifically includes: All slave users are divided into multiple clusters using a clustering algorithm, and each cluster corresponds to a drone. The slave users belonging to the cluster are matched with the drones corresponding to the cluster to obtain a drone location deployment plan and a drone-slave user matching optimization plan; Adjusting the UAV position deployment plan based on the primary user interference constraint to obtain an optimized UAV position deployment plan; The UAV-slave user matching optimization solution and the UAV position deployment optimization solution are used as inputs, and a linear precision algorithm is used to iteratively optimize and solve the joint optimization model to obtain a bandwidth allocation optimization solution.

4. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 3 is characterized in that: The UAV position deployment scheme is adjusted based on the primary user interference constraint to obtain an optimized UAV position deployment scheme, which specifically includes: Based on the UAV location deployment plan, determining whether each UAV satisfies the primary user interference constraint; If yes, the UAV location deployment plan is used as the UAV location deployment optimization plan; If not, the drone that does not meet the primary user interference constraint is recorded as a drone to be adjusted; for each drone to be adjusted, the maximum channel gain coefficient when the drone to be adjusted communicates with the primary user is calculated based on the primary user interference constraint, and the safety distance from the drone to be adjusted to the primary user is calculated based on the maximum channel gain coefficient, and the position of the drone to be adjusted is adjusted based on the safety distance to obtain the adjusted position of the drone to be adjusted; wherein the position of the drone to be adjusted is determined based on the drone position deployment plan; The positions of all the drones to be adjusted in the drone position deployment plan are replaced by the adjusted positions of all the drones to be adjusted, so as to adjust the drone position deployment plan and obtain an optimized drone position deployment plan.

5. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 4, characterized in that: The calculation formula for the maximum channel gain coefficient is: Among them, g' r,u* is the maximum channel gain coefficient when the UAV r to be adjusted communicates with the main user u*; δ is the maximum tolerance threshold of the main user to single-source interference; P r is the transmission power of the UAV r to be adjusted; The calculation formula for the safety distance is: in, is the safe distance between the UAV r to be adjusted and the primary user u*; α u is the space-to-ground path loss index; P LOS is the transmission probability of the line-of-sight link; η is the attenuation factor of the non-line-of-sight link; P NLOS is the transmission probability of the non-line-of-sight link; Adjusting the position of the to-be-adjusted drone based on the safety distance specifically includes: Determine an adjustment circle with the primary user as the center and the safety distance as the radius; The position on the adjustment circle that is closest to the position of the UAV to be adjusted is used as the adjusted position of the UAV to be adjusted.

6. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 3, characterized in that: Taking the drone-slave user matching optimization solution and the drone location deployment optimization solution as input, the linear precision algorithm is used to iteratively optimize and solve the joint optimization model to obtain a bandwidth allocation optimization solution, which specifically includes: Based on the bandwidth allocation scheme of the previous iteration, the throughput of each slave user in the current iteration is calculated using Shannon's theorem; The throughput of each slave user in the current iteration is sorted to obtain the minimum throughput and the second minimum throughput of the current iteration; the minimum throughput is the minimum value of the throughput of each slave user in the current iteration; the second minimum throughput is the minimum value after the minimum throughput is removed from the throughput of each slave user in the current iteration; Based on the minimum throughput and the second minimum throughput of the current iteration and the adjustment coefficient of the previous iteration, the minimum throughput of the current iteration is increased to obtain the adjusted throughput of each slave user of the current iteration; Based on the adjusted throughput of each slave user in the current iteration, the bandwidth of each slave user in the current iteration is calculated using Shannon's theorem to obtain a bandwidth allocation scheme for the current iteration; Based on the bandwidth allocation scheme of the current iteration, calculating the adjustment coefficient of the current iteration; Determine whether the iteration termination condition is met; if so, determine the bandwidth allocation scheme of the previous iteration as the initial bandwidth allocation scheme; if not, increase the current iteration number by 1, determine the bandwidth allocation scheme of the current iteration as the bandwidth allocation scheme of the previous iteration, and determine the adjustment coefficient of the current iteration as the adjustment coefficient of the previous iteration; Taking the backhaul bottleneck of the temporary node as a constraint, the initial bandwidth allocation scheme is adjusted to obtain a bandwidth allocation optimization scheme.

7. The joint optimization method for multi-UAV relay communication in a cognitive wireless network according to claim 6, characterized in that: Based on the minimum throughput and the second minimum throughput of the current iteration and the adjustment coefficient of the previous iteration, the minimum throughput of the current iteration is increased to obtain the adjusted throughput of each slave user of the current iteration, which specifically includes: For each minimum throughput of the current iteration, determine the minimum drone that matches the slave user corresponding to the minimum throughput, and determine the number of users whose throughput is the minimum throughput among the slave users matched with the minimum drone; calculate the difference between the sub-minimum throughput of the current iteration and the minimum throughput, and calculate the product of the number of users, the adjustment coefficient of the previous iteration and the difference to obtain an increase; calculate the sum of the minimum throughput and the increase to increase the minimum throughput to obtain an increased throughput corresponding to the minimum throughput; Replacing the minimum throughput among the throughputs of each slave user in the current iteration with the increased throughput corresponding to the minimum throughput, to obtain an adjusted throughput of each slave user in the current iteration; The calculation formula of the adjustment coefficient is: Among them, β is the adjustment coefficient; is the maximum bandwidth that the temporary node g can provide; M is the total number of drones; w r is the bandwidth of the current iteration of UAV r; A is the number of minimum throughput; T M is the second smallest throughput; T m is the minimum throughput; γ g,r is the signal-to-noise ratio when the temporary node g communicates with the drone r.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the joint optimization method for multi-UAV relay communication in a cognitive wireless network according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for joint optimization of multi-UAV relay communication in a cognitive wireless network described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for joint optimization of multi-UAV relay communication in a cognitive wireless network described in any one of claims 1 to 7 is implemented.

Images