Methods and devices for deploying emergency communication networks for unmanned aerial vehicles

Abstract

This invention relates to the field of communication equipment technology, and in particular to a method and apparatus for deploying an emergency communication network for unmanned aerial vehicles (UAVs). First, it acquires information on users to be served and UAV service restriction information. Based on the focus of communication, the UAV deployment is decomposed into three issues: the required UAV-user association method, the UAV deployment location, and the UAV's transmission power and spectrum allocation. Then, corresponding expressions are constructed for each issue. By determining these expressions, the UAV-user association method, the UAV deployment location, and the UAV's transmission power and spectrum are obtained, and the UAV deployment is completed accordingly, thus improving the deployment efficiency of the emergency communication network.

Description

Technical Field

[0001] This invention relates to the field of communication equipment technology, and in particular to a method and apparatus for deploying an emergency communication network for unmanned aerial vehicles (UAVs). Background Technology

[0002] Mobile communication networks, as an essential means for disaster victims to connect with the outside world, are susceptible to congestion and paralysis when damaged during disasters, negatively impacting the timeliness and reliability of rescue efforts. Therefore, emergency rescue operations require efficient and rapid deployment and management plans for emergency communication networks. These plans must provide more stable and reliable communication services to affected users and rescue teams under conditions of limited communication resources, thereby ensuring the smooth progress of rescue work.

[0003] Generally, emergency communication network deployment requires high mobility, rapid deployment, sufficient disaster recovery, and reliable transmission. However, deployment based on terrestrial communication networks faces challenges such as the inability to use existing terrestrial communication base stations due to damage, and road damage preventing emergency communication vehicles from reaching designated locations. Furthermore, emergency communication services typically involve high concurrency, and ensuring efficient and reliable communication service quality under resource constraints is a significant challenge. In recent years, with the continuous development of the drone industry, its high flexibility and low cost enable it to provide mobile communication services to users in the air. Its network deployment is unaffected by ground conditions, and in emergency communication scenarios, it can quickly and dynamically adjust to the changing needs of rescue personnel and disaster victims. In addition, traditional Orthogonal Multiple Access (OMA) technology faces spectrum resource constraints in emergency communication scenarios. In recent years, Non-Orthogonal Multiple Access (NOMA) technology has effectively improved spectrum efficiency by enabling multiple users to transmit messages using the same sub-channel at different powers, thus achieving spectrum resource reuse and offering significant advantages in the rational utilization and scheduling of spectrum resources. Therefore, by utilizing NOMA technology, combined with high-altitude UAV design for emergency communication network deployment schemes and rational allocation strategies for related communication resources, rapid UAV networking and efficient utilization of communication resources can be achieved, further improving network transmission rates to meet the diverse needs of emergency rescue missions for timely and reliable communication services.

[0004] Most existing drone network deployment schemes consider traditional wireless channel models, which are not well-suited for special scenarios such as forest disaster areas with severe tree obstruction, resulting in a significant discrepancy between actual deployment performance and theoretical effects. Furthermore, due to the limited spectrum resources of emergency networks, traditional drone network deployment schemes are prone to network congestion or insufficient service capacity when a large number of users access the network. In addition, because emergency rescue operations have extremely high requirements for timeliness and reliability, rapid and efficient deployment of emergency communication networks is crucial. However, existing deployment schemes based on simulated annealing or greedy algorithms have high algorithmic complexity, making it impossible to achieve rapid network deployment decisions, resulting in low deployment efficiency for emergency communication networks. Summary of the Invention

[0005] This invention provides a method for deploying an emergency communication network for unmanned aerial vehicles (UAVs), which addresses the technical deficiency of low deployment efficiency in existing emergency communication networks.

[0006] On one hand, the present invention provides a method for deploying an emergency communication network for unmanned aerial vehicles (UAVs), comprising:

[0007] Acquire information on users awaiting service and drone service restriction information; the information on users awaiting service includes the number of users awaiting service and the size of the service area, and the information on drone service restriction information includes the drone's transmission power and the maximum number of users the drone can serve.

[0008] Based on the user information to be served and the drone service restriction information, a drone-user association expression is constructed, and the drone-user association expression is determined to obtain the drone-user association method to be deployed.

[0009] Based on the service user information and drone service restriction information, a drone deployment location expression is constructed, and the drone deployment location is obtained by determining the drone deployment location expression;

[0010] Based on the service user information and UAV service restriction information, a user spectrum resource allocation expression is constructed, and the UAV's transmit power and spectrum allocation are obtained by determining the user spectrum resource allocation expression;

[0011] The drones are deployed according to the user association method, deployment location, transmission power, and spectrum allocation of the drones as required.

[0012] On the other hand, the present invention also provides a drone emergency communication network deployment device, comprising:

[0013] The acquisition unit is used to acquire information on users to be served and drone service restriction information; the information on users to be served includes the number of users to be served and the size of the service area, and the drone service restriction information includes the drone's transmission power and the maximum number of users that the drone can serve.

[0014] The first processing unit is used to construct a drone-user association expression based on the user information to be served and drone service restriction information, and determine the drone-user association expression to obtain the drone-user association method to be deployed.

[0015] The second processing unit is used to construct a drone deployment location expression based on the service user information and drone service restriction information, and determine the drone deployment location by determining the drone deployment location expression;

[0016] The third processing unit is used to construct a user spectrum resource allocation expression based on the service user information and UAV service restriction information, and determine the user spectrum resource allocation expression to obtain the UAV's transmit power and spectrum allocation;

[0017] The deployment unit is used to deploy drones according to the user association method, deployment location, transmission power, and spectrum allocation of the drones to be deployed.

[0018] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the UAV emergency communication network deployment method as described in any of the preceding claims.

[0019] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the UAV emergency communication network deployment method as described in any of the preceding claims.

[0020] The UAV emergency communication network deployment method provided by this invention first obtains information on users to be served and UAV service restriction information. Based on the focus of communication, the UAV deployment is decomposed into three issues: the required UAV-user association method, the UAV deployment location, and the UAV transmission power and spectrum allocation. Then, corresponding expressions are constructed for each issue. By determining the corresponding expressions, the UAV-user association method, the UAV deployment location, and the UAV transmission power and spectrum are obtained, and the UAV deployment is completed accordingly, thereby improving the deployment efficiency of the emergency communication network. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the method for deploying an emergency communication network for unmanned aerial vehicles (UAVs) according to an embodiment of the present invention;

[0023] Figure 2 This is a diagram illustrating the deployment scenario of an unmanned aerial vehicle (UAV) emergency communication network provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the objective function optimization process provided in an embodiment of the present invention;

[0025] Figure 4 This is a schematic diagram illustrating the association effect between a drone and a user without NOMA matching, as provided in an embodiment of the present invention.

[0026] Figure 5 This is a schematic diagram illustrating the association effect between a drone and a user under NOMA matching conditions, provided in an embodiment of the present invention.

[0027] Figure 6 This is a schematic diagram comparing the simulation time of the optimization rate calculation by the upper bound under the change of the number of iterations of the particle swarm optimization algorithm provided in the embodiment of the present invention.

[0028] Figure 7 A schematic diagram showing the comparison of rates calculated by the rate calculation function provided in this embodiment of the invention under different input powers;

[0029] Figure 8 This is a comparative diagram showing the effects of three drone deployment schemes provided in the embodiments of the present invention;

[0030] Figure 9 This is a schematic diagram illustrating the change in computation time with the number of users provided in an embodiment of the present invention;

[0031] Figure 10 This is a schematic diagram illustrating the variation of computation time with UAV transmission power in an embodiment of the present invention.

[0032] Figure 11 A schematic diagram comparing the performance of the power search optimization algorithms provided in the embodiments of the present invention;

[0033] Figure 12 This is a schematic diagram showing the network traversal rate and comparison under different maximum single-user transmit power limits for different UAVs, provided in an embodiment of the present invention.

[0034] Figure 13 This is a performance comparison diagram of drone network deployment schemes under different numbers of users provided in an embodiment of the present invention;

[0035] Figure 14 This is a schematic diagram illustrating the overall network speed improvement effect of NOMA technology compared to OMA technology, provided by an embodiment of the present invention.

[0036] Figure 15 This is a schematic diagram of the structure of an unmanned aerial vehicle (UAV) emergency communication network deployment device provided in an embodiment of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0038] This invention assumes that existing communication infrastructure is damaged by disasters and can no longer provide wireless communication services. Simultaneously, damage to ground roads prevents ground-based emergency communication equipment from deploying new ground-based emergency communication networks in disaster areas. Therefore, it considers deploying a drone network over disaster areas to provide communication services for rescue workers and affected residents. To overcome the low deployment efficiency of existing emergency communication networks, this invention provides a drone-based emergency communication network deployment method primarily applicable to forest area emergency communication scenarios. Given the complex channel environment and limited spectrum resources in forest areas, this invention considers a dedicated forest area channel model, takes into account the communication needs of affected users, and aims to maximize the overall throughput of the network system. It proposes a low-complexity, fast drone emergency network deployment scheme combining one-dimensional search, the KM (Kuhn-Munkres) algorithm, and an improved K-PSO (PSO-K-means) algorithm to meet the communication transmission needs of more service users.

[0039] Figure 2 This is a deployment scenario diagram of an unmanned aerial vehicle (UAV) emergency communication network provided by an embodiment of the present invention. In cases where damage to ground roads renders ground emergency communication equipment inoperable, the present invention provides such a deployment scenario. Figure 1 The diagram shown illustrates a scenario for the deployment of an emergency communication network for drones, assuming that in... Figure 1 Within the communication service area of ​​the rectangular region, there exist U randomly distributed disaster area users. The disaster area user set is defined as follows: If all users are on the same horizontal plane, then the user's two-dimensional position S user ={s user,1 ,...,s user,u ,...,suser,U Deploy M drones to ensure communication for all users, defining the drone set as... Drones provide communication to users in disaster areas through three-dimensional positioning deployment. The three-dimensional position of the m-th drone is represented as s. UAV,m ={x m ,y m ,h m The set of drone location deployment schemes is defined as S. UAV ={s UAV,1 ,...,s UAV,m ,...,s UAV,M A single drone can serve multiple users in a disaster area, but each user can only be served by one drone. Therefore, a binary variable α is defined. m,u ∈{0,1} represents the relationship between the drone and users in the disaster area. Where α m,u =1 indicates that drone m provides services to user u in the disaster area, α m,u =0 indicates that the drone m does not provide services to the user u in the disaster area.

[0040] To improve spectrum utilization and overall network throughput, this invention employs NOMA technology for communication between the UAV and users. Since NOMA signal decoding uses Successive Interference Cancellation (SIC), the decoding complexity increases exponentially with the number of users in disaster areas. Therefore, considering the portability and cost requirements of terminal devices for users in disaster areas, and to reduce decoding complexity, this invention sets up a NOMA user pair for every two users in disaster areas. Once paired, NOMA users in disaster areas will be served by the same UAV. Thus, UAV m can serve users to form J user pairs, and the index set of these user pairs can be defined as... To ensure successful decoding of power-domain NOMA, users in the disaster area within a NOMA user pair need to meet a certain channel gain ratio δ0. Users in the disaster area that do not meet the channel gain ratio requirement communicate using OMA (Online Communication). To distinguish them from the NOMA user pairs mentioned above, these are referred to here as OMA user pairs. Definition in, For NOMA user pairs subset, This is a subset of OMA user pairs. Users in disaster areas with lower channel fading (i.e., better channel quality) are called strong users, while those with lower channel quality are called weak users. It is assumed that UAVs will communicate with different disaster area user pairs using different frequencies, and that OMA disaster area user pairs will also communicate using their own separate frequency bands. Therefore, there is no signal interference between disaster area users using different frequency bands. In NOMA user pairs, to ensure the correctness of SIC decoding, weak users will be allocated higher transmit power. During decoding, the received signal first demodulates the signal with the highest power, then subtracts this signal from the overall superimposed signal, and then decodes the signal with the lower decoding power. In this invention, L is defined as... m,j,S and L m,j,W Let p represent the channel gain between the drone m and the strong and weak users in the disaster area user pair j, respectively. m,j,S and p m,j,W and represent the transmission power of the signal transmitted by the drone m to the strong and weak users j in the disaster area, respectively. max This indicates the maximum transmission power of the drone. (B) m,j This represents the communication bandwidth allocated by drone m to user j in the disaster area, σ0 2 This represents the noise power. According to Shannon's formula, the traversal rates of strong and weak users in NOMA user j are expressed as follows:

[0041]

[0042]

[0043] Where E[*] represents the expectation of *.

[0044] The traversal rates of strong and weak users in user j by OMA user are expressed as follows:

[0045]

[0046]

[0047] in,

[0048] Based on the above, the traversal rate of user j in the NOMA disaster area is expressed by equation (5):

[0049]

[0050] The traversal rate of the OMA user on j is expressed by equation (6):

[0051]

[0052] Furthermore, to better adapt to the communication links of emergency networks, this invention considers establishing a channel model for the dedicated channels in forest areas when building the channel model. Specifically, based on the characteristics of the communication link between the UAV and ground users, this invention uses a forest area tilted channel model. The additional attenuation in the forest area expressed as a logarithm between the i-th UAV and the j-th user is expressed as:

[0053]

[0054] Where, d L,i,j The distance of the link blocked by vegetation represents the distance, and θ is the angle between the link and the horizontal plane. A, B, C, E, and G are empirical parameters used to characterize the influence of multiple factors such as vegetation type, density, and actual water content.

[0055] In this invention, channel fading is represented as the sum of logarithmic shadow fading and additional forest area attenuation. The total logarithmic channel fading between the i-th UAV and the j-th disaster area user is expressed as:

[0056]

[0057] Where α is the path loss factor, d i,j The distance of the communication link is expressed as:

[0058]

[0059] X σ This represents shadow fading, which follows a variance of σ. 2 The zero-mean Gaussian distribution has the following probability density function:

[0060]

[0061] This represents the free space path loss at the reference distance d0:

[0062]

[0063] Where f is the carrier frequency and c is the speed of light.

[0064] In this invention, with the goal of maximizing the overall network traversal rate, the objective function and corresponding constraints for maximizing the overall network traversal rate are established by comprehensively optimizing factors such as three-dimensional location deployment, user association, and UAV transmission power and spectrum resource allocation in the network service communication of UAVs deployed in forest areas:

[0065]

[0066] Where P1 represents the objective function, and C1, C2, C3, C4, C5, C6, C7, and C8 represent constraints with different focuses. Specifically, C1 ensures communication for all users within the area, meaning that each user must have at least one drone providing service to them in the disaster area; C2 constrains the drone's transmission power, ensuring that the transmission power for a single user cannot exceed the maximum transmission power p. max Furthermore, since power domain NOMA requires strong and weak users to transmit at different transmit powers, and to provide greater fairness to users with poor channel conditions, the transmit power of weak users must be greater than that of strong users; C3 indicates that the shared transmit power of NOMA users cannot exceed the total transmit power p of the UAV. total C3 represents a constraint on the energy consumption of the drone. C4 represents a constraint on spectrum resources, requiring the communication network to be deployed even with limited spectrum resources. C5 represents a constraint on the three-dimensional deployment range of the drone, ensuring that the drone is deployed within a controllable range. C6 represents a constraint on the minimum distance d between two drones. min To ensure a minimum safe flight distance between drones and prevent collisions, C7 represents the constraint on the channel difference between strong and weak users in a NOMA user pair; C8 represents the constraint on the effective gain of NOMA users, where ε is a constant factor.

[0067] Under the above constraints, the objective function P1 is determined to obtain the required drone-user association method, the deployment location of each drone, and the drone's transmission power and spectrum allocation, thereby completing the deployment of the drone emergency communication network.

[0068] Figure 3 This is a schematic diagram of the objective function optimization process provided in an embodiment of the present invention. Further, in order to facilitate the determination of the objective function in this invention, such as... Figure 3 As shown, the objective function P1 can be decomposed into three problems: the drone disaster area user association problem, the drone location deployment problem, and the internal NOMA pairing and power and spectrum resource allocation problem for drones serving disaster area users. These three problems are determined separately to obtain the drone-user association method, the deployment location of each drone, and the drone's transmission power and spectrum allocation.

[0069] The physical meanings of some of the physical parameters appearing in this application are defined as follows:

[0070] M represents the number of drones, m represents any one drone in M, U represents the number of users, u represents any drone in U, and r m,u Q represents the traversal rate of the drone m serving user u; max This indicates the maximum number of users that the drone can serve.

[0071] The following is combined with Figures 1-14 This invention describes a method for deploying an emergency communication network for unmanned aerial vehicles (UAVs).

[0072] Figure 1 For a flowchart of the UAV emergency communication network deployment method provided in this embodiment of the invention, please refer to... Figure 1 The method includes:

[0073] S101. Obtain information on users to be served and drone service restriction information; the information on users to be served includes the number of users to be served and the size of the service area, and the information on drone service restriction includes the drone's transmission power and the maximum number of users that the drone can serve.

[0074] For example, the system can obtain the size of the area that the communication network needs to serve (i.e., the area size) and the number of users to be served within the area, while also obtaining the drone service limitation information (i.e., drone-related parameter information) of the deployed drones, such as the transmit power of a single drone and the maximum number of users that a single drone can serve.

[0075] S102. Construct a drone-user association expression based on the user information to be served and drone service restriction information, and determine the drone-user association expression to obtain the drone-user association method to be deployed.

[0076] For example, user locations S are randomly distributed within a region. user Number of users U, maximum number of users served by the drone Q max The maximum total transmit power of the UAV is P total Maximum transmit power P for a single user max Drone-User Relationship Matrix Drone Location Deployment S UAV Power allocation scheme P.

[0077] For example, the expression for associating a drone with a user is as follows:

[0078] Simultaneously, the constraint conditions corresponding to the drone-user association expression are constructed as follows:

[0079]

[0080]

[0081] Where U represents the number of users, Let m represent any single drone in the drone swarm, and Q represent the entire drone swarm. max α represents the maximum number of users that the drone can serve. m，uA binary variable representing whether a drone is associated with a user.

[0082] Based on the constraints corresponding to the aforementioned drone-user association expression, an improved K-PSO algorithm is used to determine the drone-user association expression, thus obtaining the required drone-user association method. Specifically, the obtained drone-user association method is the drone-user association matrix A, which is parameter A in the drone-user association expression.

[0083] For example, the objective function Transform into This transforms the problem of finding the optimal association between drones and users (i.e., the association matrix A) into finding the minimum sum of distances between drones and users. This problem can be viewed as a clustering problem that associates geographically proximate users with the same drone. The two-dimensional position of the m-th drone is defined as s. m ={x m y m The set of drone location deployment schemes is defined as S. UAV,2D ={s1,...,s m ,...,s M The two-dimensional location of the drone is considered as the cluster center of a partition-based clustering algorithm, making the goal of the partition-based clustering algorithm to minimize the distance between the cluster center and the data belonging to that cluster. The K-means algorithm is used to determine the relevant problem, first based on the number of users in the disaster area U and the maximum number of users Q that the drone can serve in the disaster area. max A preliminary estimate of the initial number of drones is made:

[0084]

[0085] Among the symbols This indicates rounding up to the nearest integer. That is, equation (17) represents the ratio of the number of users U in the disaster area to the maximum number of users Q that the drone can serve in the disaster area. max The initial estimated number of drones is obtained by rounding up the percentage. Drones are then pre-allocated to users in the disaster area based on this initial estimate. If the pre-allocation does not meet the constraints, the drone-user association method is adjusted until the number of drones can provide services to all users in the disaster area. Specifically, the drone-user association optimization algorithm provided in this embodiment is as follows:

[0086] First, the input information includes: user locations S randomly distributed within the region. user Number of users U, maximum number of users served by the drone Q max And the convergence count N, then proceed with the following algorithm flow:

[0087]

[0088]

[0089] The output information obtained from the above process includes: a drone-user association matrix. The set of horizontal coordinates of the centroid of each cluster obtained through clustering is S0.

[0090] S103. Construct a drone deployment location expression based on the service user information and drone service restriction information, and determine the drone deployment location by determining the drone deployment location expression.

[0091] For example, the expression for constructing the drone deployment location is:

[0092] The constraints corresponding to the expression for the drone deployment location are as follows:

[0093]

[0094]

[0095] Among them, s UAV,m Indicates the location deployment of drone m, s UAV,n This indicates the location deployment of drone n. d represents the set of drone locations. min This represents the minimum safe distance parameter for the drone. Then, based on the constraints corresponding to the drone deployment position expression, an improved K-PSO algorithm is used to determine the drone deployment position expression, thus obtaining the drone's deployment position. The variable representing the drone's position is defined in the above expression P. 12 The channel H appears in the expression for the traversal rate r. Changes in position affect the distance between the UAV and the user, thus affecting the channel and consequently the traversal rate. Therefore, determining the traversal rate r allows us to determine the deployment position of the UAV. For example, all UAV positions are stored in a single particle variable as a whole. The algorithm outputs the optimal particle, which represents the optimal position of all UAVs. Specifically, in this embodiment, after obtaining the association between the UAV and the user, the 3D UAV deployment optimization subproblem is a non-convex problem. The optimal solution is quickly found using the particle swarm optimization algorithm. To further improve the algorithm's efficiency, the centroid position output by the K-means algorithm is set as the initial particle position of the particle swarm. Since the fitness function in the particle swarm optimization algorithm does not handle constraints, problem P... 12 The objective function in the equation cannot be directly used as the fitness evaluation function. Therefore, the evaluation function for the fitness of the k-th particle in the l-th iteration is constructed as follows:

[0096]

[0097] Where x is a particle storing the three-dimensional positions of M drones, Rk (x) represents problem P 12 The objective function is F. λ is the penalty factor, and F... k,m (x) represents the three-dimensional position of the m-th UAV in the k-th particle. d min This is the minimum safe distance parameter for drones. The penalty function in the second term of formula (18) can ensure that drones maintain a safe distance from each other.

[0098] In particular, particles with higher fitness have better positions. J global,l (x) represents the historical optimal fitness and global optimal fitness of particle k after the l-th iteration, respectively. x global,l These represent the corresponding particle positions. The velocity of a particle in the next iteration is determined by its current velocity, its historical best position, and its global best position in the particle swarm.

[0099]

[0100] Where v k l Let x be the velocity of the k-th particle in the l-th iteration. k l-1 Let c1 be the position of the k-th particle in the (l-1)-th iteration. c1 and c2 are the learning factors, i.e., weights, for the particle to learn its own historical best solution and the global best solution, respectively. A random number between 0 and 1.

[0101] The particle moves from its current position based on its velocity, determining the particle's position for the next iteration.

[0102]

[0103] Specifically, this embodiment obtains the deployment location of the drone using the following algorithm:

[0104]

[0105]

[0106] S104. Construct a user spectrum resource allocation expression based on the service user information and UAV service restriction information, and determine the UAV's transmit power and spectrum allocation by determining the user spectrum resource allocation expression.

[0107] For example, the expression for constructing user spectrum resource allocation is as follows:

[0108]

[0109] The constraints corresponding to the user spectrum resource allocation expression are as follows:

[0110]

[0111]

[0112]

[0113]

[0114] Among the above constraints, p represents a collection of drones. m，j,S p m，j，W p represents the transmission power of the signal transmitted by the drone m to the strong and weak users j in the disaster area, respectively. max J represents the maximum transmit power of the drone. m NOMA p represents the set of NOMA user pairs served by drone m. total L represents the total transmit power of the drone to the user. m，j,S L m,j,W Let represent the channel gain between UAV m and the strong and weak users in user pair j in the disaster area, respectively; δ0 represents the channel gain ratio δ0 that NOMA users need to satisfy for strong and weak users; and ε represents the constraint on the effective gain for NOMA users, where ε is a constant factor. Let m represent the traversal rates of the NOMA user without service m to the strong and weak users j, respectively. Let represent the traversal rates of the OMA users serving unmanned aircraft m to the strong and weak users j, respectively. Then, based on the constraints corresponding to the user spectrum resource allocation expression, a user bipartite graph is constructed according to the user's strength and weakness attributes. A clustering algorithm is used to determine the optimal matching of the bipartite graph to obtain the UAV's transmit power and spectrum allocation.

[0115] To utilize NOMA for communication, the subproblem of optimizing the transmission power allocation of UAVs to service users is solved by addressing the combination optimization problem of NOMA pairs among users in the disaster area and the problem of searching for the optimal transmission power of NOMA for users.

[0116] This embodiment employs a bipartite graph optimal matching method. Users served by the UAV are divided into strong and weak users based on their channel gain, forming a bipartite graph. The maximum traversal and rate between user pairs are used as the edge weights for strong and weak user nodes. The edge weights for strong and weak users are determined through an improved one-dimensional search combined with bipartite search, and the optimal matching of the bipartite graph is determined using the KM algorithm.

[0117] For ease of calculation, define

[0118]

[0119] Let L represent the normalized channel gain between the drone m and the user u. m，u Let m be the channel gain between the drone and the user u.

[0120] The ergodic rate for the transmit power P of channel Ω is defined as follows:

[0121] R(P,B,Ω)=E[Blog2(1+PΩ)](22)

[0122] According to formulas (5) and (6), we get

[0123]

[0124] Where Ω m,j，S Ω m，j,W Let represent the normalized channel gain between the UAV m and the strong user pair j, respectively.

[0125] Specifically, for cases where the total number of users serving the disaster area via drones is odd, a virtual user is added to the set of weak users in the bipartite graph, and the rate of this node is set as follows:

[0126]

[0127] Meanwhile, the traversal rate of edges connected to this weak user is calculated using the OMA method by default.

[0128] According to formula (23), the objective function of P1 varies with p m,j,W Monotonically increasing, therefore the solution to the objective function will be at p m,j,W =P max or p m,j,S +p m,j,W =P total This is obtained from [the objective function]. Therefore, the search range of the solution space of the objective function is changed from a two-dimensional plane to a one-dimensional plane, thus greatly reducing the complexity of objective determination.

[0129] The improved one-dimensional search algorithm is as follows:

[0130] Step 1: For p m,j,W =P max p m,j,S =min{p max , (p total -p m，j，W Determine whether the boundary points of the solution space of the two boundary line segments of )} satisfy the condition. The constraints further narrow the solution space down to a single line segment.

[0131] The second step is to determine the specific solution space segment based on the judgment results of the first step, and then use the bisection method to search for the optimal solution by controlling a single variable.

[0132] Step 3: If a feasible solution is found, the user in the disaster area will communicate in NOMA mode and obtain the optimal transmission power allocation scheme; if no feasible NOMA power allocation scheme can be found in the solution space, the user in the disaster area will communicate in OMA mode and calculate using the OMA traversal rate formula.

[0133] After obtaining the edge weights of the bipartite graph, the KM algorithm is used to determine the optimal matching of the bipartite graph. The maximum sum of edge weights is the maximum traversal rate of the drone serving users in the disaster area.

[0134] Specifically, the process of a one-dimensional search algorithm includes:

[0135] First, the input information includes: the maximum total transmit power P of the drone. total and the maximum transmit power P of a single disaster area user max Channel power gain Ω for strong and weak users m,j,S Ω m,j,W The convergence threshold δ for the bisection method. Then, the following process is executed:

[0136]

[0137]

[0138] The output information obtained through the above process includes: feasible power allocation scheme P, communication methods for users in the disaster area, and maximum traversal rate for a single user in the disaster area.

[0139] This invention optimizes the association and location deployment of drones with users in disaster areas, namely P. 11 and P 12 In this paper, an improved K-PSO algorithm is proposed. The algorithm flow is shown in steps 1 and 2 of Algorithm 1. A detailed introduction follows.

[0140] The derivation yields... Therefore, the objective function Transform into The problem is transformed into finding the minimum distance sum. This can be viewed as a clustering problem that associates geographically proximate disaster area users with the same drone. The drone's two-dimensional location is considered as the cluster center in a partition-based clustering algorithm, making the goal of this algorithm to minimize the distance between the cluster center and the data belonging to that cluster. The K-means algorithm is used to define the relevant problem, first based on the number of disaster area users U and the maximum number of disaster area users Q that the drone can serve. max A preliminary estimate of the initial number of drones is made:

[0141]

[0142] The number of users in the disaster area (U) and the maximum number of users served by drones in the disaster area (Q) max The initial estimated number of drones is obtained by rounding up the percentage. Drones are initially pre-allocated to users in the disaster area based on the preliminary estimated number of drones. If the pre-allocation result does not meet the constraints, the association method between drones and users is adjusted until the number of drones can provide services to all users in the disaster area.

[0143] After obtaining the correlation between the drones and users in the disaster area, the subproblem of optimizing the three-dimensional positioning of the drones is a non-convex problem. The optimal solution is quickly found using the particle swarm optimization (PSO) algorithm. To further improve the algorithm's efficiency, the centroid positions output by the K-means algorithm are set as the initial particle positions in the PSO algorithm. Since the fitness function in the PSO algorithm does not handle constraints, problem P... 12 The objective function in the equation cannot be directly used as the fitness evaluation function. Therefore, the evaluation function for the fitness of the k-th particle in the l-th iteration is constructed as follows:

[0144]

[0145] Where x is a particle storing the three-dimensional positions of M drones, R k (x) represents problem P 12 The objective function is F. λ is the penalty factor, and F... k，m (x) represents the three-dimensional position of the m-th UAV in the k-th particle.

[0146] In particular, particles with higher fitness have better positions. J global，l (x) represents the historical optimal fitness and global optimal fitness of particle k after the l-th iteration, respectively. x global,l These represent the corresponding particle positions. The velocity of a particle in the next iteration is determined by its current velocity, its historical best position, and its global best position in the particle swarm.

[0147]

[0148] The particle moves from its current position based on its velocity, determining the particle's position for the next iteration.

[0149]

[0150] Furthermore, this embodiment provides an optimization algorithm for associating drones with users, specifically including: input information being: user locations S randomly distributed within a region. user Number of users U, maximum number of users served by the drone Q max And the number of convergences, N. The algorithm proceeds as follows:

[0151]

[0152] The output information obtained through the above process is: Drone-User Association Matrix The set of horizontal coordinates of the centroid of each cluster obtained through clustering is S0.

[0153] To utilize NOMA for communication, the subproblem of optimizing the transmit power allocation of UAVs to serving users is addressed by solving the pairwise combinatorial optimization problem of NOMA among users in the disaster area and the optimal transmit power search problem for NOMA among users. A bipartite graph optimal matching method is considered for this purpose. Users served by the UAVs are divided into strong and weak users based on their channel gain, forming a bipartite graph. The maximum traversal and rate between user pairs are used as the edge weights of the strong and weak user nodes. The edge weights of strong and weak users are determined by an improved one-dimensional search combined with a bipartite search, and the optimal matching of the bipartite graph is determined using the KM algorithm.

[0154] For ease of calculation, define

[0155]

[0156] This represents the normalized channel gain between the drone m and the user u.

[0157] The ergodic rate for the transmit power P of channel Ω is defined as follows:

[0158] R(P,B,Ω)=E[Blog2(1+PΩ)]

[0159] Based on the derivation of formulas (5) and (6), we obtain

[0160]

[0161]

[0162]

[0163]

[0164] Specifically, for cases where the total number of users serving the disaster area via drones is odd, a virtual user is added to the set of weak users in the bipartite graph, and the rate of this node is set as follows:

[0165]

[0166] Meanwhile, the traversal rate of edges connected to this weak user is calculated using the OMA method by default.

[0167] Derivation (0.2) shows that the objective function of P1 varies with p W Monotonically increasing, therefore the solution to the objective function will be at pW =P max or p S +p W =P total This is obtained from [the objective function]. Therefore, the search range of the solution space of the objective function is changed from a two-dimensional plane to a one-dimensional plane, thus greatly reducing the complexity of objective determination.

[0168] The improved one-dimensional search algorithm is as follows, and the algorithm flow is shown in Algorithm 3:

[0169] Step 1: For p W =P max p S =min{p max ,(p total -p w Determine whether the boundary points of the solution space of the two boundary line segments of )} satisfy the condition. The constraints further narrow the solution space down to a single line segment.

[0170] The second step is to determine the specific solution space segment based on the judgment results of the first step, and then use the bisection method to search for the optimal solution by controlling a single variable.

[0171] Step 3: If a feasible solution is found, the user in the disaster area will communicate in NOMA mode and obtain the optimal transmission power allocation scheme; if no feasible NOMA power allocation scheme can be found in the solution space, the user in the disaster area will communicate in OMA mode and calculate using the OMA traversal rate formula.

[0172] After obtaining the edge weights of the bipartite graph, the KM algorithm is used to determine the optimal matching of the bipartite graph. The maximum sum of edge weights is the maximum traversal rate of the drone serving users in the disaster area.

[0173] The one-dimensional search algorithm process includes:

[0174] Input parameters include: UAV maximum total transmit power P total and the maximum transmit power P of a single disaster area user max Channel power gain Ω for strong and weak users S Ω W The convergence threshold δ for the bisection method. Output parameters include: feasible power allocation scheme P, communication methods for users in the disaster area, and the maximum traversal rate for a single user pair in the disaster area. The specific algorithm flow includes:

[0175]

[0176]

[0177] To further reduce the algorithm complexity, the objective function (formula (10)) is integrated, which has a high time complexity. Therefore, we consider using an approximate function based on the upper bound to optimize the objective function.

[0178] According to Jason's inequality, we get:

[0179] E[log2(1+PX)]≤log2[1+PE(X)] (25)

[0180] Therefore, the upper bound function of the objective function is expressed as:

[0181]

[0182]

[0183] Define upper bound rate

[0184]

[0185] Therefore, the upper bound rate for users in the disaster area can be expressed as:

[0186]

[0187] Using an upper bound rate function instead of directly calculating the objective function can further reduce the algorithm's computation time.

[0188] S105. Deploy drones according to the required drone-user association method, drone deployment location, drone transmission power, and spectrum allocation.

[0189] For example, after obtaining the association method between the drones to be deployed and the users, the deployment location of the drones, and the transmission power and spectrum allocation of the drones, the drones are deployed in the corresponding locations based on this information to form an emergency communication network.

[0190] The UAV emergency communication network deployment method provided by this invention decomposes UAV deployment into three problems based on the focus of communication: the required UAV-user association method, the UAV deployment location, and the UAV transmission power and spectrum allocation. Corresponding expressions are constructed for each problem. By determining the corresponding expressions, the UAV-user association method, the UAV deployment location, and the UAV transmission power and spectrum are obtained, and the UAV deployment is completed accordingly, thereby improving the deployment efficiency of the emergency communication network.

[0191] Regarding the UAV emergency communication network deployment method provided in this embodiment, a simulation method is provided below to verify the technical effect of the UAV emergency communication network deployment method in this embodiment.

[0192] The relevant simulation parameters are shown in Table 1:

[0193] Table 1

[0194]

[0195]

[0196] Figure 4 This is a schematic diagram illustrating the association effect between a drone and a user without NOMA matching, as provided in an embodiment of the present invention. Figure 5 This is a schematic diagram illustrating the association effect between a drone and a user under NOMA matching conditions, provided in an embodiment of the present invention; for example... Figure 4 and Figure 5 As shown, Figure 4 and Figure 5 This demonstrates the deployment effect of drones in three-dimensional space. In this embodiment, users are randomly distributed within a 100m*100m planar area. Figure 4 This embodiment demonstrates the deployment locations of drones and the association effects between drones and users obtained by the proposed method. First, an optimized K-means algorithm is used to determine the association method between drones and users, and to associate drones with users in disaster areas, ensuring that the number of users served by drones in disaster areas does not exceed Q. max Given a threshold of 10, the minimum number of drones and user-connection methods are determined to provide services to users in the disaster area, and the centroid output by the algorithm is used as the initial position for 3D deployment optimization in the particle swarm optimization algorithm. Secondly, Figure 5 This demonstrates the further optimization effect on matching NOMA disaster area users within the disaster area user cluster after completing the optimization of drone location and disaster area user clustering. From Figure 5 As can be seen, the method provided in this embodiment can associate geographically closer disaster area users with a single drone within the drone deployment range and select a better location for the drone to provide service. Furthermore, among disaster area users served by the same drone, users with better channel conditions are matched with those with worse channel conditions to form NOMA disaster area user pairs, providing greater communication fairness for disaster area users with poor channel conditions, thereby improving both the overall network speed and ensuring the communication stability of disaster area users.

[0197] Figure 6 This diagram illustrates a comparison of simulation times for calculating the optimization rate via an upper bound under varying iteration counts in the particle swarm optimization algorithm, as provided in this embodiment of the invention. Figure 6This paper demonstrates the performance improvement in algorithm simulation time achieved by optimizing the calculation function based on an upper bound. Here, `rate-D` represents the rate calculated directly using equation (1.23), and `rate-UB` represents the rate calculated using the rate function optimized by the upper bound of equation (1.29). Taking the computation time of the PSO algorithm in the overall scheme as an example, as the number of iterations of the particle swarm optimization algorithm increases, directly calculating the traversal rate to calculate the objective function significantly increases the simulation time. Under the condition of 10 iterations, the particle swarm optimization algorithm takes more than 220 seconds to compute, which is highly detrimental to the timely deployment of emergency solutions. After optimizing the objective function based on the upper bound rate, the simulation time for iterations within ten is kept within 0.1 seconds, greatly improving the algorithm's running efficiency. This verifies the optimization of the scheme in reducing algorithm complexity.

[0198] Figure 7 This is a schematic diagram comparing the rates calculated by the rate calculation function provided in this embodiment of the invention under different input powers. Figure 7 The results demonstrate that the rates calculated using the rate calculation function based on the upper bound and the directly calculated rates differ very little from those calculated with different input powers, with a difference of only 0.027% in some regions. Furthermore, the monotonicity of the calculated rate functions is in good agreement with the original functions, indicating that the calculated rates can effectively simulate and replace the original functions. Figure 6 The results demonstrate that the upper bound rate optimization function can significantly improve the algorithm efficiency when it closely matches the original function.

[0199] Figure 8 This is a comparative diagram showing the effects of three drone deployment schemes provided in this embodiment of the invention. Figure 8 The overall network traversal rate is demonstrated under different drone transmit power conditions, after deploying drones and associating them with disaster area users using the proposed algorithm. K-2D represents the scheme where disaster area users are first associated using the K-means algorithm, and the centroid position is used as the drone's 2D position for deployment at the same altitude. K-rand-3D represents the scheme where disaster area users are clustered using the K-means algorithm to obtain the drone's 2D position, and then the drones are deployed at random altitudes within a height constraint. K-PSO-3D represents the scheme in this paper where the 2D drone position is obtained and then used as the initial particle position for 3D position optimization using a particle swarm optimization algorithm. Simulation results show that, under the same drone transmit power setting, the proposed drone deployment scheme significantly improves the overall system speed compared to the other two schemes.

[0200] Figure 9 This is a schematic diagram illustrating the change in computation time with the number of users, provided in an embodiment of the present invention. Figure 10 This is a schematic diagram illustrating the variation of computation time with UAV transmission power in an embodiment of the present invention. Figure 9The results show that as the scale of users in the disaster area continues to increase, the computation time of all three schemes increases, but the computation time of all three schemes remains within 0.05s. K-rand-3D is the UAV 3D positioning deployment scheme proposed in this paper, which combines the K-means algorithm and the PSO algorithm. Although it sacrifices some time compared to the 2D positioning deployment using only the K-means algorithm and the K-rand-3D deployment of UAVs at random altitudes, with an overall difference of milliseconds, it does not affect the overall efficiency. Figure 8 The calculation results of the K-PSO scheme were significantly improved, which clearly improved the network traversal speed and verified the effectiveness of the scheme. Figure 10 With varying drone transmission power, all three schemes maintained a computation time of less than 0.02 seconds, demonstrating high decision-making rates.

[0201] Figure 11 This is a schematic diagram comparing the performance of the power search optimization algorithm provided in the embodiments of the present invention. Figure 11 Simulations were conducted on a power optimization scheme for UAVs. The results compared a standard bisection method for searching power between strong and weak users in a NOMA disaster area, an exhaustive search method within the solution space, and a proposed scheme that reduces the dimensionality of the solution space by incorporating the monotonicity between the objective function and variables, using a bisection method in one-dimensional space for optimal solution searching. Under the simulated conditions, with fewer than 8 iterations, neither the one-dimensional search method nor the standard bisection method found a feasible NOMA power allocation scheme, as all users communicated via OMA. With more than 11 iterations, the proposed one-dimensional search scheme found the optimal solution. The bisection search method within the planar solution space failed to find a feasible NOMA power allocation scheme within the tested iteration count. The exhaustive search algorithm exhibited significant randomness, and due to the large search space, the results were unstable with fewer iterations. Furthermore, the found solution was only feasible, not necessarily the optimal solution under the same conditions. Based on the above analysis, the power optimization scheme in this embodiment is superior to the other two algorithms in terms of determining speed and the quality of the output solution.

[0202] Figure 12 This is a schematic diagram illustrating the network traversal rate and comparison under different maximum single-user transmit power limits for various UAVs, provided in this embodiment of the invention. Figure 12Simulations were performed on a network scheme optimized by 3D deployment of UAVs and association with disaster area users using OMA technology. This paper compares this scheme with a scheme optimized by using NOMA technology, which further optimizes the association between disaster area users and UAVs, and UAV location, through NOMA-based matching of disaster area users and power allocation for UAVs. As the maximum transmit power limit for a single UAV user in a disaster area increases, the overall network traversal rate and efficiency improve. Furthermore, under the same power limit, this invention achieves a higher network rate using NOMA technology compared to OMA technology.

[0203] Figure 13 This is a performance comparison diagram of drone network deployment schemes under different user numbers provided in this embodiment of the invention. Figure 14 This is a schematic diagram illustrating the overall network speed improvement effect of NOMA technology compared to OMA technology, provided by an embodiment of the present invention. Figure 13 This study compares the overall network traversal speed of network deployment schemes using OMA and NOMA technologies as the user base increases. As the number of users in the disaster area grows, the difference in speed improvement between the NOMA network and the OMA network also increases. Figure 14 For the same number of users in a disaster area, the percentage increase in network speed achieved by NOMA technology compared to not using NOMA technology remains at 7% to 8% for medium-sized disaster areas with fewer than 100 users. While the effect slightly decreases with larger disaster area user numbers, it still remains above 6%. This demonstrates that using NOMA technology to improve the traversal speed of network deployment schemes is highly effective.

[0204] The simulation results demonstrate that the UAV emergency communication network deployment method provided in this embodiment utilizes NOMA technology to improve the spectral efficiency of UAV communication with disaster-stricken users on the ground, while ensuring fairness in communication quality for all service users. Combining one-dimensional search, the KM algorithm, and an improved K-PSO, a low-complexity, fast UAV emergency network deployment algorithm was designed. This reduces the deployment time of the emergency communication network, supports high-spectral-efficiency UAV network deployment solutions in emergency scenarios, and ensures communication quality for all disaster-stricken users within the service area while achieving higher spectral efficiency and communication rates.

[0205] The UAV emergency communication network deployment device provided by the present invention is described below. The UAV emergency communication network deployment device described below and the UAV emergency communication network deployment method described above can be referred to in correspondence.

[0206] Figure 15 This is a schematic diagram of the structure of the UAV emergency communication network deployment device provided in an embodiment of the present invention, as shown below. Figure 15As shown, the drone emergency communication network deployment device includes:

[0207] The acquisition unit 150 is used to acquire information on users to be served and drone service restriction information; the information on users to be served includes the number of users to be served and the size of the service area, and the information on drone service restriction includes the drone's transmission power and the maximum number of users that the drone can serve.

[0208] The first processing unit 151 is used to construct a drone-user association expression based on the user information to be served and drone service restriction information, and determine the drone-user association expression to obtain the drone-user association method to be deployed.

[0209] The second processing unit 152 is used to construct a drone deployment location expression based on the service user information and drone service restriction information, and determine the drone deployment location by determining the drone deployment location expression.

[0210] The third processing unit 153 is used to construct a user spectrum resource allocation expression based on the service user information and UAV service restriction information, and determine the UAV's transmit power and spectrum allocation by determining the user spectrum resource allocation expression.

[0211] Deployment unit 154 is used to deploy drones according to the user association method, deployment location, transmission power and spectrum allocation of the drones as needed.

[0212] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the UAV emergency communication network deployment method provided by the above methods.

[0213] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the UAV emergency communication network deployment method provided by the above methods.

[0214] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0215] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for deploying an emergency communication network for unmanned aerial vehicles (UAVs), characterized in that, include: Obtain information on users awaiting service and drone service restriction information; The information on users to be served includes the number of users to be served and the size of the service area; the information on drone service limitations includes the drone's transmission power and the maximum number of users that the drone can serve. Based on the user information to be served and the drone service restriction information, a drone-user association expression is constructed, and the drone-user association expression is determined to obtain the drone-user association method to be deployed; Based on the service user information and drone service restriction information, a drone deployment location expression is constructed, and the drone deployment location is obtained by determining the drone deployment location expression; Based on the service user information and UAV service restriction information, a user spectrum resource allocation expression is constructed, and the UAV's transmit power and spectrum allocation are obtained by determining the user spectrum resource allocation expression; The step of constructing a user spectrum resource allocation expression based on the service user information and drone service restriction information includes: The expression for user spectrum resource allocation is constructed as follows: The constraints corresponding to the user spectrum resource allocation expression are as follows: Among the above constraints, p represents a collection of drones. m,j,S p m,j,W p represents the transmission power of the signal transmitted by the drone m to the strong and weak users j in the disaster area, respectively. max J represents the maximum transmit power of the drone. m NOMA p represents the set of NOMA user pairs served by drone m. total L represents the total transmit power of the drone to the user. m,j,S L m,j,W Let represent the channel gain between UAV m and the strong and weak users in user pair j in the disaster area, respectively; δ0 represents the channel gain ratio that the NOMA user pair needs to satisfy for the strong and weak users; and ε represents the constraint on the effective gain for the NOMA user, where ε is a constant factor. Let m represent the traversal rates of the NOMA user without service m to the strong and weak users j, respectively. Let m represent the traversal rates of the unattended OMA user to the strong and weak users in j, respectively. This represents the traversal rate of NOMA users on j served by drone m. This represents the traversal rate of the OMA user serving drone m on j; The process of determining the user spectrum resource allocation expression to obtain the UAV's transmit power and spectrum allocation includes: Based on the constraints corresponding to the user spectrum resource allocation expression, a user bipartite graph is constructed according to the user's strength and weakness attributes. The KM algorithm is used to determine the optimal matching of the bipartite graph to obtain the UAV's transmit power and spectrum allocation. The drones are deployed according to the user association method, deployment location, transmission power, and spectrum allocation of the drones as required.

2. The method for deploying an emergency communication network for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step of constructing a drone-user association expression based on the user information to be served and the drone service restriction information includes: The expression for associating drones with users is constructed as follows: Simultaneously, the constraint conditions corresponding to the drone-user association expression are constructed as follows: Where U represents the number of users, Let m represent a set of drones, and Q represent any single drone in the drone swarm. max α represents the maximum number of users served by the drone. m,u A binary variable, r, representing whether the drone is associated with a user. m,u This represents the traversal rate of the drone m serving user u.

3. The method for deploying an emergency communication network for unmanned aerial vehicles (UAVs) according to claim 2, characterized in that, The step of determining the association expression between the drone and the user to obtain the association method between the drone and the user to be deployed includes: Based on the constraints corresponding to the drone-user association expression, the improved K-PSO algorithm is used to determine the drone-user association expression, thereby obtaining the drone-user association method to be deployed.

4. The method for deploying an emergency communication network for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step of constructing the drone deployment location expression based on the service user information and drone service restriction information includes: The expression for constructing the drone deployment location is as follows: The constraints corresponding to the UAV deployment location expression are as follows: Among them, s UAV,m Indicates the location deployment of drone m, s UAV,n This indicates the location deployment of drone n. d represents the set of drone locations. min r represents the minimum safe distance parameter for drones. m,u This represents the traversal rate of the drone m serving user u.

5. The method for deploying an emergency communication network for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step of determining the drone deployment location expression to obtain the drone deployment location includes: Based on the constraints corresponding to the UAV deployment location expression, the improved K-PSO algorithm is used to determine the UAV deployment location expression, thereby obtaining the UAV deployment location.

6. A device for deploying an emergency communication network for unmanned aerial vehicles (UAVs), characterized in that, include: The acquisition unit is used to acquire information about users to be served and drone service restriction information; The information on users to be served includes the number of users to be served and the size of the service area; the information on drone service limitations includes the drone's transmission power and the maximum number of users that the drone can serve. The first processing unit is used to construct a drone-user association expression based on the user information to be served and drone service restriction information, and determine the drone-user association expression to obtain the drone-user association method to be deployed. The second processing unit is used to construct a drone deployment location expression based on the service user information and drone service restriction information, and determine the drone deployment location by determining the drone deployment location expression; The third processing unit is configured to construct a user spectrum resource allocation expression based on the service user information and UAV service restriction information, and determine the UAV's transmit power and spectrum allocation by determining the user spectrum resource allocation expression; the step of constructing the user spectrum resource allocation expression based on the service user information and UAV service restriction information includes: The expression for user spectrum resource allocation is constructed as follows: The constraints corresponding to the user spectrum resource allocation expression are as follows: Among the above constraints, p represents a collection of drones. m,j,S p m,j,W p represents the transmission power of the signal transmitted by the drone m to the strong and weak users j in the disaster area, respectively. max J represents the maximum transmit power of the drone. m NOMA p represents the set of NOMA user pairs served by drone m. total L represents the total transmit power of the drone to the user. m,j,S L m,j,W Let represent the channel gain between UAV m and the strong and weak users in user pair j in the disaster area, respectively; δ0 represents the channel gain ratio that the NOMA user pair needs to satisfy for the strong and weak users; and ε represents the constraint on the effective gain for the NOMA user, where ε is a constant factor. Let m represent the traversal rates of the NOMA user without service m to the strong and weak users j, respectively. Let m represent the traversal rates of the unattended OMA user to the strong and weak users in j, respectively. This represents the traversal rate of NOMA users on j served by drone m. This represents the traversal rate of the OMA user serving drone m on j; The process of determining the user spectrum resource allocation expression to obtain the UAV's transmit power and spectrum allocation includes: Based on the constraints corresponding to the user spectrum resource allocation expression, a user bipartite graph is constructed according to the user's strength and weakness attributes. The KM algorithm is used to determine the optimal matching of the bipartite graph to obtain the UAV's transmit power and spectrum allocation. The deployment unit is used to deploy drones according to the user association method, deployment location, transmission power, and spectrum allocation of the drones to be deployed.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the UAV emergency communication network deployment method as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the UAV emergency communication network deployment method as described in any one of claims 1 to 5.

Images