Unmanned aerial vehicle network communication and control system assisted by multiple intelligent reflecting surfaces

Abstract

Complete phase shift, transmitting power distribution, decoding and flight control factors are considered, a multi-intelligent-reflection-surface-assisted unmanned aerial vehicle network communication and control system is designed in a combined mode, and the system divides the complex original problem of solving phase shift, transmitting power distribution, decoding and flight control into three sub-problems for alternate optimization. And applying different methods in each sub-problem to finally obtain an optimal result. Therefore, a multivariable joint design problem can be effectively solved, a stable convergence effect under relatively low complexity can be realized, a design objective of maximum rate can be realized, system communication rate maximization under the condition of the lowest communication rate of each ground user can be met, the system can complete communication tasks as efficiently as possible, and the communication efficiency is improved. Therefore, the communication requirements of ground users can be better met; and the achievable rate of the system is further increased by adjusting the multi-IRS reflection phase shift.

Description

Technical Field

[0001] The present invention belongs to the technical field of mobile communications, and in particular relates to a network communication and control system for unmanned aerial vehicles assisted by multiple intelligent reflective surfaces. Background Art

[0002] With the rise of the Internet of Things, the number of different types of terminal devices such as cloud sensors, smartphones, and wearable devices has continued to surge; intelligent applications such as face recognition, interactive games, and virtual reality are also emerging, and the demand for spectrum resources and data transmission rates is growing rapidly. On the other hand, with the acceleration of urbanization, wireless communications are facing a harsh wireless transmission environment, which seriously reduces the communication quality of ground users. In this context, unmanned aerial vehicles (UAVs), non-orthogonal multiple access (NOMA), and intelligent reflective surfaces (IRS) as the latest technologies can effectively solve the problems of spectrum resource shortage and harsh wireless transmission environment.

[0003] With the advancement and maturity of modern industrial technology, the performance of drones has been significantly improved, playing a vital role in the military and civilian fields, performing tasks such as material transportation, rescue and reconnaissance. Especially in the field of communications, drones often play the role of flying base stations (BS), aiming to improve the network data transmission capacity and expand the service coverage. On the one hand, compared with ground communications, drones have the advantage of being easy to build air-to-ground (A2G) links based on line of sight (LoS), which can provide ground users with stable and high-speed communication links on this basis. On the other hand, drones have excellent maneuverability and can be flexibly deployed near ground users, providing better channel conditions and thus improving communication quality. The next generation of communication systems aims to expand service coverage with high speed and low latency. Drones are considered to be one of the promising solutions to meet the needs of the next generation of communication systems with their many advantages and are receiving increasing attention. However, with the surge in the number of communication devices and the emergence of various new applications and services, the demand for spectrum resources and data transmission rates is growing rapidly.

[0004] Non-orthogonal multiple access allows multiple ground users to use the same frequency resources at the same time, which greatly improves spectrum utilization and increases the communication capacity of the system. Therefore, it is considered to be a very promising technical solution and has attracted widespread attention from academia and industry. UAVs and NOMA each have their own advantages. A novel idea is to further integrate the two into the UAV-NOMA structure. On the one hand, UAVs can move flexibly to make up for the lack of coverage of traditional base stations. On the other hand, NOMA can greatly improve spectrum efficiency and system capacity. This joint utilization optimizes the allocation of spectrum resources and improves network coverage and data transmission efficiency, especially in high-density ground user areas and remote areas. This structure also enhances the flexibility and responsiveness of the system, enabling it to adapt to rapidly changing communication needs and environmental conditions, providing strong support for future smart city and Internet of Things (IoT) applications.

[0005] In addition, wireless communications also face harsh transmission environments. The presence of a large number of obstructions in urban environments seriously hinders signal transmission. Fortunately, smart reflective surfaces are considered to be a promising solution to this problem and have been widely studied by academia and industry. Specifically, IRS is a 2D plane composed of many inexpensive passive reflective elements. By adjusting these independent elements, the incident signal can be changed, thereby reshaping the wireless channel. In addition, IRS is an almost passive device that can be quickly installed on the surface of a building to reconfigure the wireless channel at a very low cost, thereby improving the quality of communication. In this sense, the application of IRS to UAV-NOMA networks can inspire new technical applications and advantages. On the other hand, it can support large-scale connections while reducing communication delays and costs, providing a strong impetus for future communication development.

[0006] Although the flexibility and communication rate of traditional communication systems can be further enhanced through the auxiliary design of drones, at present, both academia and industry rarely consider the simultaneous application of IRS and NOMA technologies in drone-assisted communication scenarios, and there is even less work on communication and control methods in multi-IRS-assisted UAV-NOMA networks. However, in reality, the multi-IRS-assisted drone communication system using NOMA technology is more general and more in line with the needs of IoT application scenarios. On the one hand, according to the NOMA protocol, the decoding order of ground users is determined by their channel conditions. Dynamic drones will inevitably bring about dynamic NOMA decoding orders, resulting in a high degree of coupling between multi-IRS phase shifts, drone trajectories, and NOMA decoding order, which makes it very difficult to deal with this problem. On the other hand, for multi-IRS and multi-ground user scenarios, the optimal phase shift is no longer just a coherent combination of the reflected signal and the direct signal, because the reflective element needs to serve multiple ground users at the same time, so the IRS phase shift optimization becomes more difficult to handle. Therefore, it is very important to deeply study the phase shift, power, decoding and control methods of multi-IRS-assisted UAV-NOMA networks. Summary of the invention

[0007] In order to solve the above problems existing in the prior art, the present invention provides a multi-intelligent reflective surface assisted UAV network communication and control system. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0008] A multi-intelligent reflective surface-assisted UAV network communication and control system includes:

[0009] The initialization module is used to design system parameters according to the distribution of UAVs, ground users and IRS in the target area, and initialize them to obtain the initialization system parameters;

[0010] A multi-IRS phase shift control module is used to construct an initial first optimization problem for solving the reflection phase shift by using the initial system parameters, and convert it from a non-convex form to a convex form, and solve the convex form of the first optimization problem to obtain the reflection phase shift of each IRS;

[0011] A power allocation module is used to construct an initial second optimization problem for solving the transmission power by using the reflection phase shift of each IRS, and convert it from a non-convex form to a convex form, and solve the convex form of the second optimization problem to obtain the transmission power of each UAV in each time slot;

[0012] The decoding order and trajectory planning module is used to construct the initial third optimization problem using the transmission power of the UAV and the reflection phase shift of the IRS, and convert it from a non-convex form to a convex form, and solve the convex form of the third optimization problem to obtain the decoding order and trajectory planning results of the NOMA of each UAV;

[0013] The output module is used to determine whether the output condition is met according to the decoding order and trajectory planning result of the NOMA. If not, the execution process of the multi-IRS phase shift control module, the power allocation module, the decoding order and the trajectory planning module is repeated until the output condition is met to obtain the optimal reflection phase shift, the optimal transmission power, the optimal decoding order and the optimal trajectory planning result.

[0014] Beneficial effects:

[0015] The present invention studies the UAV-NOMA communication network assisted by multiple IRSs, and considers multiple factors of multiple IRS phase shifts, UAV transmission power, NOMA decoding order and UAV control party of the multiple IRS assisted UAV-NOMA communication system. By jointly optimizing multiple IRS phase shifts, UAV transmission power, NOMA decoding order and UAV trajectory, the defects of low efficiency and severe channel fading of traditional UAV communications are broken, and the purposes of multiple IRS phase shift control, UAV transmission power allocation, NOMA decoding order and UAV trajectory planning are achieved. A multi-intelligent reflector-assisted UAV network communication and control system with a complete phase shift, transmission power allocation, decoding and flight control joint design is proposed.

[0016] The multi-intelligent reflective surface-assisted UAV network communication and control system provided by the present invention can realize the system, divide the complex original problem into three sub-problems for alternate optimization, and apply different methods in each sub-problem to finally obtain the best result. This can effectively solve the multi-variable joint design problem, and achieve a stable convergence effect under low complexity, and achieve the design goal of maximum rate, so that the system can complete the communication task as efficiently as possible, so as to better meet the communication needs of ground users; and by adjusting the multi-IRS phase shift, the achievable rate of the system is further increased.

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of a multi-intelligent reflective surface-assisted UAV network communication and control system provided by the present invention;

[0019] Figure 2 is a flow chart of a multi-IRS phase shift control module provided by the present invention;

[0020] Figure 3 It is a schematic diagram of the flow of the power distribution module provided by the present invention;

[0021] Figure 4 It is a flowchart of the decoding sequence and trajectory planning module provided by the present invention. DETAILED DESCRIPTION

[0022] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0023] The present invention mainly includes five modules: information initialization, multi-IRS phase shift control, UAV transmission power allocation, NOMA decoding order and UAV trajectory planning, and result output. The present invention adopts a joint design method based on an alternating optimization structure. In each iteration, the multi-IRS phase shift control module is first designed separately, and then the UAV transmission power allocation is jointly designed. Finally, the NOMA decoding order and the UAV trajectory are jointly planned. The final output system can achieve the maximum rate, as well as the optimal multi-IRS reflection phase shift, UAV transmission power, NOMA decoding order and UAV trajectory.

[0024] like Figure 1 As shown, the present invention provides a multi-intelligent reflective surface assisted UAV network communication and control system, comprising:

[0025] The initialization module is used to design system parameters according to the distribution of UAVs, ground users and IRS in the target area, and initialize them to obtain the initialization system parameters;

[0026] Before the system is run, the present invention first designs the system initialization parameters through the information initialization module. The initialization system parameters include: multi-IRS reflection phase shift range, initial transmission power of the UAV, constraint relationship of NOMA decoding order and initial trajectory of the UAV.

[0027] The present invention first defines the relevant parameters of the UAV, ground user and IRS in the target area; the relevant parameters include: represents all terrestrial users, Indicates the number of reflection units of a single IRS, represents all smart reflective surfaces, Θ l [n] represents the reflection phase shift matrix of the lth IRS, w k represents the location of ground user k; w r Indicates the location of the IRS; represents the trajectory of the drone; Represents the total time slot of the system; I k,j [n] indicates the decoding order of NOMA; p k [n] represents the transmission power of the UAV; the system communication bandwidth and white noise power are represented by B and σ respectively 2 ;

[0028] The present invention considers a finite time period T. For analysis, the time period T is divided into N equal time slots, and the interval between each time slot is δ t Therefore, the horizontal trajectory of the drone can be expressed as q[n] = (x[n], y[n]), n = 1, ..., N, and the drone flies at a fixed height H. The starting point and end point of the drone are denoted by q I and q F Therefore, the UAV maneuverability constraint is

[0029] q I =q[1], (1)

[0030] q F =q[N], (2)

[0031] ||q[n+1]-q[n]|| 2 ≤(V max δ t ) 2 ,n=1,...,N-1, (3)

[0032] Among them, V max is the maximum flight speed of the UAV. For the convenience of analysis, the present invention uses h l [n], respectively, represent the channels from UAV to ground user k, UAV to IRS1, and IRS1 to ground user k. In this paper, the channels involved are considered to be quasi-static flat fading, and by applying the channel estimation algorithm, the UAV can fully understand the channel state information.

[0033] Therefore, the channel from the UAV to IRS1 is expressed as

[0034]

[0035] Where β0 represents the average channel power gain when the reference distance is 1m, represents the IRS1 array response, expressed as:

[0036]

[0037] where g y and g z denote the element spacing along the y-axis and z-axis, respectively, λ denotes the carrier wavelength, In addition, θ l [n] and denote the vertical and horizontal arrival angles of the signal from the UAV to IRS l, sinθ l [n]=(h u [n]-h l )||q[n]-wl || -1 ,

[0038] According to the previous analysis, the channel from the UAV to the ground user k is given by

[0039]

[0040] in α u,k and κ1 are the path loss index and Rice factor of the k channel of the UAV ground user, respectively. u,k [n]=||q[n]-w k ||,

[0041] The channel between the drone and IRS1 is expressed as

[0042]

[0043] Among them, γ and κ2 are the path loss exponent and Ricean factor of the channel between IRS1 and ground user k, respectively. Represents the LOS component:

[0044]

[0045] in, In addition, θ l [n] and denote the vertical and horizontal arrival angles of the signal from the UAV to IRS l, sinθ l,k =h l ||w k -w l || -1 ,

[0046] The communication rate function of the system is defined by the related parameters, which is expressed as:

[0047]

[0048] Among them, R k [n] is the receiving task bit of the kth access point in the nth time slot, expressed as:

[0049] R k [n] = log2(1 + γ k [n])(10)

[0050] in, is the direct channel gain between the UAV and the ground user, For the reflection channel between UAV-multiple IRS-ground user, is the channel between IRS1 and ground user k, h l [n] is the channel from the drone to IRS1.

[0051] The signal sent by the UAV can reach the ground user through multiple paths, but due to severe signal attenuation, the present invention assumes that the signal reflected twice or more by the IRS is ignored. Therefore, the combined channel power gain from the UAV to the ground user k is:

[0052] In the information initialization module, according to the location distribution of ground users, a regular trajectory initialization method can be used to initialize the UAV trajectory, such as using a straight line uniform speed flight method to fly from a given starting point to an end point within a total time T. In addition, the initialization obtains the UAV transmission power allocation and NOMA decoding order I k,j [n].

[0053] Using the NOMA decoding order I k,j [n] The constraints that determine the NOMA decoding order are expressed as:

[0054]

[0055] Among them, d u,k [n]=||q[n]-w k || 2 represents the distance from the UAV to the kth ground user, d u,j [n]=||q[n]-w j || 2 represents the distance from the UAV to the jth ground user, w j represents the coordinates of the jth ground user.

[0056] According to the location distribution of ground users, the trajectory of the UAV and the power allocation of the UAV are initialized, and the NOMA decoding order I is obtained according to the initialized UAV trajectory. k,j [n].

[0057] A multi-IRS phase shift control module is used to construct an initial first optimization problem for solving the reflection phase shift by using the initial system parameters, and convert it from a non-convex form to a convex form, and solve the convex form of the first optimization problem to obtain the reflection phase shift of each IRS;

[0058] In the multi-IRS phase shift control module, by giving the UAV's transmit power, NOMA decoding order and UAV trajectory, the problem is transformed into a semi-definite program (SDP) problem using the semidefinite relaxation (Semidefinite Relaxation, SDR) technique, and then the reflection phase shift of the UAV can be obtained by using the convex-concave procedure (CCCP) and successive convex approximation (SCA) method.

[0059] In the target area (such as outdoor scenes), there are There are multiple intelligent reflectors to assist a UAV to communicate with a ground user. Both the ground user and the UAV are equipped with a single antenna. Indicates the total time slot of the system; I k,j [n] indicates the decoding order; the horizontal trajectory of the drone is expressed as The height is fixed as H. The transmission power of the drone in each time slot is expressed as p k [n]; the system communication bandwidth and white noise power are represented by B and σ respectively 2 .

[0060] refer to Figure 2 As shown, the multi-IRS phase shift control module is used to:

[0061] a, using the initial transmission power P of the drone and the NOMA decoding order I k,j [n], UAV trajectory q[n] and reflection phase shift Θ of multiple IRS l [n] Construct the initial first optimization problem, expressed as:

[0062]

[0063] In the formula, R min is the minimum communication quality constraint for ground users, r k [n] is the first slack variable introduced, R k [n] is the receiving task bit of the kth ground user in the nth time slot, θ l,m [n] is the reflected phase shift of the lth IRS in the nth time slot;

[0064] b. Using the semidefinite relaxation SDR technology, the initial first optimization problem is converted into a semidefinite programming SDP problem. During the conversion process, the rank-one constraint is ignored to obtain a convex form of the first optimization problem.

[0065] Since the initial first optimization problem has non-convex constraints (13a) and (13b), the initial first optimization problem (SP1) is non-convex, and the present invention can apply the SDR method to process it. For analysis, the present invention can convert H k [n] The matrix can be rewritten as follows:

[0066]

[0067] in,

[0068] Based on the above transformation, H k [n] is further converted into the form of trace additive constant, namely: in, V[n]f 0,rank(V[n])=1.

[0069] Therefore, (13b) can be rewritten as:

[0070]

[0071] in, α k [n] is another slack variable introduced, satisfying:

[0072]

[0073] The right side of equation (15) is the difference of two concave functions. The present invention can apply the CCCP method. Specifically, by Performing a first-order Taylor expansion, the present invention can obtain The upper bound of can be expressed as

[0074]

[0075] in is the α in the lth SCA iteration k The value of [n].

[0076] In summary, the initial first optimization problem (SP1) can be reformulated as

[0077]

[0078] In the formula, α k [n] is the second slack variable introduced, p j [n] is the power allocated by the UAV to ground user j, U[n], V[n], V m,m [n], O k [n] is an intermediate variable in the process of transforming the initial first optimization problem into a convex problem. is the channel from the UAV to the ground user k, σ 2 is the white noise power of the system, yes The upper bound of L is the Lth IRS, M is the number of reflection units in a single IRS;

[0079] c. Using a convex optimization tool, the first optimization problem is solved to obtain the reflection phase shift of each IRS.

[0080] Due to the rank-one constraint, the problem (SP1-1) is still non-convex. Ignoring the rank-one constraint, the first optimization problem (SP1-1) is a standard semi-definite programming problem that can be solved efficiently by convex optimization tools such as CVX. The solution can be made to satisfy the rank-one constraint later by applying a Gaussian randomization procedure.

[0081] A power allocation module is used to construct an initial second optimization problem for solving the transmission power by using the reflection phase shift of each IRS, and convert it from a non-convex form to a convex form, and solve the convex form of the second optimization problem to obtain the transmission power of each UAV in each time slot;

[0082] In the power allocation module, the optimal power allocation of the UAV can be obtained by using CCCP and SCA methods given the multi-IRS phase shifts, NOMA decoding order and UAV trajectory.

[0083] refer to Figure 3 As shown, the power distribution module is specifically used for:

[0084] a, using NOMA decoding order I k,j [n], UAV trajectory q[n], reflection phase shift Θ l [n] Construct the initial second optimization problem to solve the transmit power, expressed as:

[0085]

[0086] In the formula, r 2k [n] is the third slack variable introduced, p j [n] is the power allocated to ground user j, P max is the maximum transmission power of the UAV, p k [n] is the power allocated to terrestrial user k.

[0087] b. Using the semidefinite relaxation SDR technology, the initial second optimization problem is transformed into a semidefinite programming SDP problem to obtain a convex second optimization problem.

[0088] Due to the non-convex constraint (19b), the initial second optimization problem (SP2) is also non-convex. The structure of constraint (19b) is similar to that of (13b), so the same method can be used for constraint (19b). Therefore, constraint (19b) can be transformed into the following form:

[0089]

[0090] in

[0091] where α 2k [n] is the fourth slack variable introduced, satisfying:

[0092]

[0093] The right side of constraint (20) is the subtraction form of two concave functions, and the present invention uses the CCCP method again. Similarly, we can get The upper bound of is expressed as:

[0094]

[0095] Therefore, the non-convex initial second optimization problem can be transformed into a convex second optimization problem, expressed as:

[0096]

[0097] In the formula, H k [n] is the combined channel power gain from the UAV to the ground user k.

[0098] c. Use the CVX solver to solve the convex form of the second optimization problem and obtain the UAV’s transmission power.

[0099] The decoding order and trajectory planning module is used to construct the initial third optimization problem using the transmission power of the UAV and the reflection phase shift of the IRS, and convert it from a non-convex form to a convex form. The convex form of the third optimization problem is solved to obtain the decoding order and trajectory planning results of the NOMA of each UAV.

[0100] In the decoding order and trajectory planning module, by giving the multi-IRS phase shift and the transmission power of the UAV, the penalty function method and SCA method can be used to plan the optimal NOMA decoding order and UAV trajectory.

[0101] refer to Figure 4 , decoding order and trajectory planning module, specifically used for:

[0102] a, using the reflection phase shift matrix Θ l [n] and the transmission power P construct the initial third optimization problem for solving the UAV trajectory and NOMA decoding order, which is expressed as:

[0103]

[0104] The NOMA decoding order of ground users is determined by the following formula

[0105]

[0106] Among them, d u,k [n]=||q[n]-w k || 2 represents the distance from the UAV to the kth ground user, d u,j [n]=||q[n]-w j || 2 Represents the distance from the UAV to the jth ground user.

[0107] The initial third optimization problem (SP3) is still intractable because the objective function is non-convex and includes integer constraints (11)-(12).

[0108] Can be H k [n] After analysis and transformation, we can get:

[0109]

[0110] in C k,l(l-1) / 2 [n] No specific expression, by k [n] is obtained by expanding the square.

[0111] Next, the present invention introduces the fifth relaxation variable {u k [n]>0} means d u,k [n]Upper bound, {u l [n]>0} means d l The upper bound of [n]. We can get:

[0112]

[0113] Substituting (25) and (26) into (24), the present invention can obtain the lower limit of the combined channel power gain of the UAV to its ground user:

[0114]

[0115] Note that A k [n],B l,k [n],C k,l(l-1) / 2 [n],D l,kThe arrival angle variable of the signal sent by the drone to the IRS in [n] is related to the position of the drone, which makes the problem difficult to handle. Therefore, the following constraints are introduced to approximate the problem:

[0116]

[0117] where d m represents the maximum allowable displacement of the UAV after each iteration of SCA. m is small enough, the arrival angle between the lth iteration and the next iteration can be considered to remain unchanged.

[0118] By the previous approximation and the additional constraint (28), the convexity of the right side of (27) depends only on u k [n] and u l [n], the present invention defines the right side of (27) as h k [n].

[0119] The present invention takes into account the coefficient A k [n],B l,k [n] is always greater than zero, and the coefficient C k,l(l-1) / 2 [n],D l,k [n] is less than zero. h k The convexity of [n] depends on the sign of these coefficients.

[0120] When A k [n],B l,k [n], C k,l(l-1) / 2 [n],D l,k [n],C k,l(l-1) / 2 [n],D l,k When [n] are both greater than zero, h k [n] relative to u k [n] and u l [n] is jointly convex, because h k The Hessian matrix of [n] is greater than zero. In this case, we can k [n] and u l [n] Perform a first-order Taylor expansion to obtain h k The lower bound of [n] is

[0121] When the coefficient C k,l(l-1) / 2 [n],D l,k [n] exists less than zero h k [n] two convex functions are subtracted. At this time, the present invention can obtain by performing a first-order Taylor expansion of the previous convex function

[0122] Therefore, given and The value of can be deduced Therefore, constraint (27) can be transformed into

[0123]

[0124] Next, the sixth slack variable Ψ is introduced k [n], satisfying:

[0125]

[0126] Therefore, the lower bound of the objective function of the initial third optimization problem (SP3) can be expressed as

[0127]

[0128] For the non-convex constraints (25) and (26), the present invention notes that their left side (LHS) is a convex function. Therefore, by and Applying the first-order Taylor expansion at , we can obtain the lower limits respectively as follows:

[0129]

[0130] Next, we analyze the constraint condition (11) related to the NOMA decoding order. k,j [n], relax it to a continuous variable of 0-1, satisfying:

[0131]

[0132] At the same time, the following two equations should be satisfied:

[0133]

[0134] where π j [n] is the seventh slack variable introduced. Combined with equation (11), equations (34), (35), (36) and (37) are equivalent to equations (11) and (12). Adding equation (35) as a penalty term to the objective function, the new calculation rate function is: where λ p represents the penalty factor, when λ p When it approaches infinity, the new calculation capacity function is the same as the original calculation capacity function. The second term of as well as Performing a first-order Taylor expansion, we can The conversion into a convex function is as follows:

[0135]

[0136] where r 3k [n] is the eighth slack variable introduced.

[0137] Then, we analyze equation (37) and express it in the following form:

[0138]

[0139] Using the SCA method, the second term on the left side of equation (39) is calculated at a given point. and By performing a first-order Taylor expansion, the convexity of the left side of (39) can be guaranteed. Similarly, using the SCA method, the right side of (39) is calculated at a given local point q (l) [n] Performing a first-order Taylor expansion can ensure that the right side of (39) is non-convex. In this way, (39) can be transformed into a convex constraint as shown below:

[0140]

[0141] Among them are

[0142] b. Using a convex optimization method, the initial third optimization problem is converted into a convex third optimization problem, which is expressed as:

[0143]

[0144] In the formula, r 3k [n] is the eighth slack variable introduced, q I is the starting point of the drone, q F is the end point of the drone. The horizontal trajectory of the drone is expressed as q[n]=(x[n],y[n]),n=1,...,N,V max is the maximum flight speed of the drone, δ t is the interval between each time slot, d m is the maximum allowed displacement of the drone after each iteration of SCA, Yes k The lower bound of [n], h k [n] is the intermediate parameter used for presentation, u k [n] is the third slack variable introduced, Ψ k [n] is the sixth slack variable introduced, is the lower bound of the combined channel power gain from UAV to ground users, is the u in the rth SCA iteration k The value of [n], is the u in the rth SCA iteration l The value of [n], w l are the coordinates of IRS1, It is an intermediate variable in the process of converting the initial third optimization problem into a convex third optimization problem.

[0145] c. Use convex optimization tools to solve the convex form of the third optimization problem to obtain the decoding order and trajectory planning results of the NOMA of each drone.

[0146] The output module is used to determine whether the output condition is met according to the decoding order and trajectory planning result of the NOMA. If not, the execution process of the multi-IRS phase shift control module, the power allocation module, the decoding order and the trajectory planning module is repeated until the output condition is met to obtain the optimal reflection phase shift, the optimal transmission power, the optimal decoding order and the optimal trajectory planning result.

[0147] In the result output module, after one iteration is completed, the system's communication rate, multi-IRS phase shift, drone's transmission power, NOMA decoding order and drone's trajectory are updated. If the system's communication rate converges to a certain accuracy (the accuracy can be pre-set according to actual conditions), the iteration is terminated and the best result is output; or if the module records that the number of system iterations is greater than the maximum allowed number, the iteration is also terminated; otherwise, the next iteration process is entered.

[0148] The output condition is a preset convergence accuracy or a maximum number of iterations.

[0149] The present invention considers the complete phase shift, transmission power allocation, decoding and flight control factors, and jointly designs a multi-intelligent reflector-assisted UAV network communication and control system. The system divides the complex original problem of solving phase shift, transmission power allocation, decoding and flight control into three sub-problems for alternate optimization, and applies different methods in each sub-problem to finally obtain the best result. In this way, the multi-variable joint design problem can be effectively solved, and a stable convergence effect under low complexity can be achieved, and the design goal of maximum rate can be achieved, and the system communication rate under the condition of the minimum communication rate of each ground user can be maximized, so that the system can complete the communication task as efficiently as possible, so as to better meet the communication needs of ground users; and the achievable rate of the system is further increased by adjusting the multi-IRS reflection phase shift.

[0150] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0151] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other variations of the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality of components or steps.

[0152] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A multi-intelligent reflective surface assisted UAV network communication and control system, characterized in that: include: The initialization module is used to design system parameters according to the distribution of UAVs, ground users and IRS in the target area, and initialize them to obtain the initialization system parameters; A multi-IRS phase shift control module is used to construct an initial first optimization problem for solving the reflection phase shift by using the initial system parameters, and convert it from a non-convex form to a convex form, and solve the convex form of the first optimization problem to obtain the reflection phase shift of each IRS; A power allocation module is used to construct an initial second optimization problem for solving the transmission power by using the reflection phase shift of each IRS, and convert it from a non-convex form to a convex form, and solve the convex form of the second optimization problem to obtain the transmission power of each UAV in each time slot; The decoding order and trajectory planning module is used to construct the initial third optimization problem using the transmission power of the UAV and the reflection phase shift of the IRS, and convert it from a non-convex form to a convex form, and solve the convex form of the third optimization problem to obtain the decoding order and trajectory planning results of the NOMA of each UAV; The output module is used to determine whether the output condition is met according to the decoding order and trajectory planning result of the NOMA. If not, the execution process of the multi-IRS phase shift control module, the power allocation module, the decoding order and the trajectory planning module is repeated until the output condition is met to obtain the optimal reflection phase shift, the optimal transmission power, the optimal decoding order and the optimal trajectory planning result.

2. The multi-intelligent reflective surface-assisted drone network communication and control system according to claim 1 is characterized in that: The initialization system parameters include: the initialization system parameters include: multi-IRS reflection phase shift range, the initial transmission power of the drone, the constraint relationship of the NOMA decoding order and the initial trajectory of the drone.

3. The multi-intelligent reflective surface-assisted drone network communication and control system according to claim 2 is characterized in that: The initialization module is specifically used for: Define the relevant parameters of the UAV, ground users and IRS in the target area; the relevant parameters include: represents all terrestrial users, Indicates the number of reflection units of a single IRS, represents all smart reflective surfaces, Θ l [n] represents the reflection phase shift matrix of the lth IRS, w k represents the location of ground user k; w r Indicates the location of the IRS; represents the trajectory of the drone; Represents the total time slot of the system; I k,j [n] indicates the decoding order of NOMA; p k [n] represents the transmission power of the UAV; the system communication bandwidth and white noise power are represented by B and σ respectively 2 ; The communication rate function of the system is defined by the related parameters, which is expressed as: Among them, R k [n] is the receiving task bit of the kth access point in the nth time slot, expressed as: R k [n]=log2(1+γ k [n]); in, is the direct channel gain between the UAV and the ground user, is the reflection channel gain of UAV-multiple IRS-ground user, is the channel between IRS1 and ground user k, h l [n] is the channel from the drone to the lth IRS; According to the location distribution of ground users, the trajectory of the UAV and the power allocation of the UAV are initialized, and the NOMA decoding order I is obtained according to the initialized UAV trajectory. k,j [n]; The constraints of NOMA decoding order are expressed as: Among them, d u,k [n]=||q[n]-w k || 2 represents the distance from the UAV to the kth ground user, d u,j [n]=||q[n]-w j || 2 represents the distance from the UAV to the jth ground user, w j represents the coordinates of the jth ground user.

4. The multi-intelligent reflective surface-assisted drone network communication and control system according to claim 3 is characterized in that: The multi-IRS phase shift control module is used to: Using the initial transmission power P of the drone and the NOMA decoding order I k,j [n], UAV trajectory q[n] and reflection phase shift Θ of multiple IRS l [n] Construct the initial first optimization problem, expressed as: In the formula, R min is the minimum communication quality constraint for ground users, r k [n] is the first slack variable introduced, R k [n] is the receiving task bit of the kth ground user in the nth time slot, θ l,m [n] is the reflected phase shift of the lth IRS in the nth time slot; The initial first optimization problem is transformed into a semidefinite programming (SDP) problem using the semidefinite relaxation (SDR) technique. During the conversion, the rank-one constraint is ignored to obtain the first optimization problem in a convex form, which is expressed as: In the formula, α k [n] is the second slack variable introduced, p j [n] is the power allocated to ground user j, U[n], V[n], V m,m [n], O k [n] are all intermediate variables introduced in the process of transforming the non-convex first optimization problem into a convex problem. is the channel from the UAV to the ground user k, σ 2 is the white noise power of the system, yes The upper bound of L is the Lth IRS, M is the number of reflection units in a single IRS; The first optimization problem is solved using a convex optimization tool to obtain the reflection phase shift of each IRS.

5. The multi-intelligent reflective surface-assisted drone network communication and control system according to claim 4 is characterized in that: Power distribution module, specifically used for: Using NOMA decoding order I k,j [n], UAV trajectory q[n], reflection phase shift Θ l [n] Construct the initial second optimization problem to solve the transmit power, expressed as: In the formula, r 2k [n] is the third slack variable introduced, P max is the maximum transmission power of the UAV, p k [n] is the power allocated to terrestrial user k; The initial second optimization problem is transformed into a semidefinite programming (SDP) problem using the semidefinite relaxation (SDR) technique to obtain a convex second optimization problem, which is expressed as: In the formula, H k [n] is the combined channel power gain from the UAV to the ground user; The convex form of the second optimization problem is solved using the CVX solver to obtain the UAV's transmission power.

6. The multi-intelligent reflective surface-assisted UAV network communication and control system according to claim 5 is characterized in that: Decoding sequence and trajectory planning module, specifically used for: Using the reflection phase shift matrix Θ l [n] and the initial transmission power P construct the initial third optimization problem for solving the UAV trajectory and NOMA decoding order, which can be expressed as: Using the convex optimization method, the initial third optimization problem is converted into a convex third optimization problem, which is expressed as: st q I =q[1] q F =q[N] ||q[n+1]-q[n]|| 2 ≤(V max δ t ) 2 ,n=1,...,N-1 In the formula, r 3k [n] is the eighth slack variable introduced, q I is the starting point of the drone, q F is the end point of the drone. The horizontal trajectory of the drone is expressed as q[n]=(x[n],y[n]),n=1,...,N,V max is the maximum flight speed of the drone, δ t is the interval between each time slot, d m is the maximum allowed displacement of the drone after each iteration of SCA, Yes k The lower bound of [n], h k [n] is the intermediate parameter used for presentation, u k [n] is the third slack variable introduced, Ψ k [n] is the sixth slack variable introduced, is the lower bound of the combined channel power gain from UAV to ground users, is the given u in the rth iteration k The value of [n], is the given u in the rth iteration l The value of [n], w l are the coordinates of IRSl; It is an intermediate variable in the process of converting the initial third optimization problem into a convex third optimization problem; The convex optimization tool is used to solve the third optimization problem in convex form to obtain the decoding order and trajectory planning results of the NOMA of each drone.

7. The multi-intelligent reflective surface-assisted drone network communication and control system according to claim 1 is characterized in that: The output condition is a preset convergence accuracy or a maximum number of iterations.

Images