A resource allocation method for RIS-assisted multi-UAV communication system

Abstract

The present invention discloses a resource allocation method for a RIS-assisted multi-UAV communication system. The method comprises the following steps: introducing an intelligent reflective surface into a multi-UAV assisted non-orthogonal multiple access system to construct an intelligent reflective surface-assisted multi-UAV communication system; establishing a mathematical model for minimizing the total energy consumption of the intelligent reflective surface-assisted multi-UAV communication system; and jointly optimizing the position deployment of UAVs and resource allocation. The present invention enhances the signal coverage of UAVs with the help of intelligent reflective surfaces; adopts NOMA technology to enable multiple users to share the same spectrum resources under the constraint of ensuring the minimum transmission rate of users, thereby realizing access of massive users and improving spectrum efficiency.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a resource allocation method for a RIS-assisted multi-UAV communication system. Background Art

[0002] The rapid development of the Internet of Things (IoT) has led to an exponential growth in wireless devices, which will support IoT applications such as automated manufacturing, virtual reality, smart homes, and smart cities. To support IoT applications, wireless networks must meet high-performance requirements such as low-latency communication, ultra-high capacity, and large-scale connectivity. Non-orthogonal multiple access (NOMA) is considered to be one of the technologies that meet these stringent requirements. Its principle is to enable multiple users to share the same spectrum resources through superposition coding and continuous interference cancellation to improve the system's spectrum efficiency. However, despite the advantages of the NOMA scheme, the performance gain of NOMA technology is still limited by the propagation environment, especially for users with small channel gain differences.

[0003] Recently, reconfigure intelligent surfaces (RIS) have been recognized as a potential technology for controlling wireless propagation environments. RIS is a planar structure composed of multiple low-cost reflective elements, each of which can intelligently tune the amplitude and phase of the reflected signal via controllable integrated circuits. This enables the RIS to reconfigure the propagation of reflected signals, thereby improving communication quality. Combining NOMA technology with RIS can flexibly adjust user channel conditions by adjusting the RIS reflection coefficient, thereby improving system performance. Because RIS are deployed on building surfaces or walls and can only serve users distributed in the front half of the space, ensuring quality of service for edge users is difficult. Unmanned aerial vehicles (UAVs) offer advantages such as high maneuverability, strong autonomy, and low cost. They can be rapidly deployed in the air to provide reliable service to edge users. By integrating UAVs into RIS to enhance NOMA networks, a "virtual" line-of-sight link can be established between the UAVs and ground users, thereby extending coverage and reducing energy consumption.

[0004] In the existing technology, Liu et al. studied a DQN-based method to minimize the total energy cost of the UAV-RIS network by optimizing the UAV mobile position, transmission power, and NOMA demodulation order (Liu X., Liu Y., Chen Y., Machine Learning Empowered Trajectory and Passive Beamforming Design in UAV-RIS Wireless Networks[J]. IEEE J. Sel. Areas Commun., 2021, 39(7): 2042-2055.). This technology aims to minimize the energy consumption of the RIS-assisted UAV-NOMA system. However, since the above scenario only considers a single UAV scenario, this solution cannot be directly applied to scenarios where users are distributed over a wide area.

[0005] Mu et al. studied the throughput maximization problem of RIS-assisted multi-UAV NOMA system, considering UAV position deployment, power allocation, RIS phase shift matrix, and NOMA demodulation order. This optimization problem was solved using the BCD method (X.Mu, Liu Y., Guo L., et al. Intelligent Reflecting Surface Enhanced Multi-UAV NOMA Network Works [J]. IEEE J. Sel. Areas Commun., 2021, 39 (10): 3051-3066.). This technique focuses on the resource allocation strategy of RIS-assisted multi-UAV NOMA system, assuming that UAVs and users are equipped with single antennas. Therefore, this algorithm cannot meet the user's QoS requirements and reduce system energy consumption, especially when the user channel quality is poor. Summary of the Invention

[0006] The purpose of the present invention is to address the shortcomings of the existing technology and solve the problems of network coverage, user access and energy efficiency. While ensuring the minimum transmission rate of users and the safe distance between drones, a mathematical optimization problem based on minimizing total energy consumption is established. The drone position deployment, smart reflector phase shift matrix, beamforming vector and transmission power, and demodulation order are jointly optimized to provide a resource allocation method based on smart reflector-assisted multi-drone.

[0007] The purpose of the present invention is achieved by at least one of the following technical solutions.

[0008] A resource allocation method for a RIS-assisted multi-UAV communication system comprises the following steps:

[0009] S1. Introducing intelligent reflective surfaces into a multi-UAV assisted non-orthogonal multiple access system to build an intelligent reflective surface assisted multi-UAV communication system;

[0010] S2. Establish a mathematical model for minimizing the total energy consumption of a multi-UAV communication system assisted by an intelligent reflective surface;

[0011] S3. Jointly optimize drone location deployment and resource allocation.

[0012] Furthermore, in step S1, the intelligent reflective surface assisted multi-UAV communication system is specifically as follows:

[0013] K quadrotor drones provide wireless communication services for K user groups, each of which includes multiple users; the kth drone serves the kth user group;

[0014] All drones carry N t antennas, and each user has 1 antenna;

[0015] Since the direct path between the drone and the user is blocked by obstacles, a smart reflective surface device equipped with N reflective units is deployed on the surface of the building to reflect the signal transmitted from the drone to the user;

[0016] The channel g from the kth UAV to the smart reflective surface device k as follows:

[0017]

[0018] Where ρ0 is the channel power gain when the reference distance is 1m, q k is the three-dimensional position of the kth UAV, k = 1~K, w r is the location of the smart reflective surface; represents the array response of the kth UAV generated by the antenna unit; represents the array response of the kth UAV generated by the RIS unit, and Obtained through channel estimation;

[0019] The channel from the kth drone to the i-th user in the k-th user group as follows:

[0020]

[0021] in, is the position of the i-th user in the k-th user group; α ug is the path loss between the UAV and the user; κ ug is the Ricean factor of the channel between the drone and the user; and are the direct path component and the non-line-of-sight path component of the i-th user in the k-th user group served by the k-th UAV, respectively, k = 1~K;

[0022] The channel from the smart reflector device to the i-th user in the k-th user group is as follows:

[0023]

[0024] Among them, α rg is the path loss between the smart reflector and the user; κ rg is the Ricean factor of the channel between the smart reflector and the user; and are the direct path component and the non-direct path component of the i-th user in the k-th user group, respectively;

[0025] Assuming that all drones use non-orthogonal multiple access technology to serve the user group, the signal-to-interference-and-noise ratio (SINR) of the i-th user in the k-th user group is k,i It is expressed as follows:

[0026]

[0027] Among them, f k is the precoding vector of the kth user group; p k,i is the transmission power of the i-th user in the k-th user group; Θ is the phase shift matrix of the smart reflection surface; σ 2 is additive white noise; the achievable rate R of the i-th user in the k-th user group k,i It is expressed as follows:

[0028] R k,i =log2(1+SINR k,i ). (5)

[0029] Furthermore, in step S2, a mathematical model for minimizing the total energy consumption of the intelligent reflective surface-assisted multi-UAV communication system is established, including determining mathematical expressions of optimization variables, objective functions, and constraints;

[0030] The total energy consumption of the intelligent reflector-assisted multi-UAV communication system is expressed as:

[0031] P sum =P+P RIS +P UAV ; (6)

[0032] in, is the total transmission power of all UAVs, M k is the number of users in the kth user group, P RIS is the power consumed by the smart reflector, PUAV is the power consumed by the drone;

[0033] The optimization variables of the mathematical model for minimizing total energy consumption include:

[0034] 1) The three-dimensional position q of the kth UAV k ;

[0035] 2) Transmission power p of the i-th user in the k-th user group k,i ;

[0036] 3) smart reflection surface phase shift matrix Θ;

[0037] 4) Precoding vector f of the kth user group k ;

[0038] 5) The demodulation order u of the user.

[0039] Furthermore, the constraints of the mathematical model for minimizing total energy consumption include:

[0040] a) Transmission power constraint: p k,i ≥0;

[0041] b) Minimum transmission rate constraint: log2(1+SINR k,i )≥R min ; R min is the transmission rate threshold;

[0042] c) Minimum safe distance constraints between drones: Δ min is the minimum safe distance between drones;

[0043] d) Reflection unit phase coefficient constraint: θ n ∈[0,2π);θ n is the phase coefficient of the nth reflective unit in the intelligent reflective surface device, n = 1 to N;

[0044] e) Constraints on the demodulation order of all users: is a feasible set of demodulation orders, which can be obtained by brute force algorithm;

[0045] f) Constraints on the demodulation order of a single user: u k (i)>u k (t), u k (i) and u k (t) are the demodulation order of the i-th user in the k-th user group and the demodulation order of the t-th user in the k-th user group, respectively, and {u k (i),u k(t)∈u};

[0046] g) Constraints on the precoding vector: ||f k || 2 ≤1.

[0047] Furthermore, the mathematical model based on minimizing total energy consumption is as follows:

[0048]

[0049] stp k,i ≥0, (7b)

[0050] log2(1+SINR k,i )≥R min , (7c)

[0051]

[0052] θ n ∈[0,2π), (7e)

[0053]

[0054]

[0055]

[0056] ||f k || 2 ≤1. (7h)

[0057] Furthermore, step S3 includes the following steps:

[0058] S3.1. According to the maximum ratio transmission technique, the optimal precoding vector f for the kth user group is obtained. k ; Using the convex approximation algorithm, the three-dimensional position q of the kth UAV is obtained k , where k = 1, 2, ..., K;

[0059] S3.2. Using algebraic transformation methods and introducing slack variables, the total energy consumption minimization problem is transformed into the differential form of two equivalent convex functions. Furthermore, a Gaussian randomization process is used to derive a closed-loop expression for the smart reflector phase shift matrix Θ.

[0060] S3.3. According to the convex optimization tool, combined with the user demodulation order u, the transmission power p of the i-th user in the k-th user group is obtained k,i , where k = 1, 2, ..., K, i∈M k ;

[0061] S3.4, based on the principles of superposition coding technology and continuous interference elimination technology, combined with the three-dimensional position q of the kth UAV k , the transmission power p of the i-th user in the k-th user group k,i , the precoding vector f of the kth user group k Impact on the total energy consumption of the system, and obtain the demodulation order u of all users k (i), and then the total demodulation order u is obtained.

[0062] Furthermore, in step S3.1, the precoding vector f of the kth user group k The calculation formula is:

[0063]

[0064] in,(.) H Represents conjugate transpose; According to the constraint function formula (7c) and the precoding vector f k The calculation formula (8) is used to solve the optimization problem (9) to obtain the three-dimensional position q of the k-th UAV. k :

[0065]

[0066] stlog2(1+SINR k,i )≥R min (9b)

[0067]

[0068] q j represents the three-dimensional position of the j-th UAV, j = 1~K and j≠k.

[0069] Furthermore, in step S3.2, the precoding vector f of the kth user group is obtained based on formula (8): k And the three-dimensional position q of the kth UAV obtained by formula (9a)-formula (9c) k , and according to the phase coefficient θ of the nth reflection unit n , calculate the smart reflection surface phase shift matrix Θ, the formula is as follows:

[0070]

[0071] Among them, e is the natural base; is a complex vector; is a complex vector The Nth element of ; U is a unitary matrix; ∑ is a diagonal matrix; S is a complex vector with mean 0 and variance 1; is the Nth element of the smart reflection surface phase shift matrix Θ, i.e., the phase coefficient; j is an imaginary number.

[0072] Furthermore, in step S3.3, based on formula (7c), the transmission power p of the i-th user in the k-th user group is calculated by solving the following optimization problem: k,i :

[0073]

[0074] stp k,i ≥0, (11b)

[0075] log2(1+SINR k,i )≥R min (11c)

[0076] Furthermore, in step S3.4, according to the obtained precoding vector f of the kth user group, k , the three-dimensional position q of the kth UAV k , the smart reflection surface phase shift matrix Θ and the transmission power p of the i-th user in the k-th user group k,i , calculate the demodulation order u of the i-th user in the k-user group k (i) and the demodulation order u of the tth user in the k-user group k (t), as follows:

[0077] When the channel conditions of the i-th user and the t-th user in the k-th user group are different, the demodulation order is calculated using the following formula:

[0078]

[0079] When the channel conditions of the i-th user and the t-th user in the k-th user group are the same, the demodulation order is calculated using the following formula:

[0080]

[0081] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0082] This invention provides a resource allocation method for a RIS-assisted multi-UAV communication system. Compared to existing UAV communication systems, this method enhances the signal coverage of UAVs by leveraging intelligent reflective surfaces. Furthermore, using NOMA technology, multiple users can share the same spectrum resources while ensuring a minimum user transmission rate, enabling massive user access and improving spectrum efficiency. By jointly optimizing UAV position deployment, the intelligent reflective surface phase shift matrix, beamforming vectors, transmission power, and demodulation order, the system significantly reduces energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1Schematic diagram of the relationship between the minimum transmission rate threshold and the total power consumption of the system according to an embodiment of the present invention;

[0084] Figure 2 This is a schematic diagram showing the relationship between the number of drone antennas and the total system power consumption according to an embodiment of the present invention;

[0085] Figure 3 Schematic diagram of the relationship between the number of reflective units and the total power consumption of the system according to an embodiment of the present invention;

[0086] Figure 4 Schematic diagram of the relationship between the path loss coefficient and the total system power consumption according to an embodiment of the present invention;

[0087] Figure 5 This is a flowchart of the steps of a resource allocation method for a RIS-assisted multi-UAV communication system of the present invention. DETAILED DESCRIPTION

[0088] The specific implementation of the present invention is further described in detail below with reference to the embodiments and drawings, but the implementation of the present invention is not limited thereto.

[0089] Example:

[0090] A resource allocation method for RIS-assisted multi-UAV communication system, such as Figure 5 As shown, the following steps are included:

[0091] S1. Introducing intelligent reflective surfaces into a multi-UAV assisted non-orthogonal multiple access system to build an intelligent reflective surface assisted multi-UAV communication system;

[0092] The intelligent reflective surface assisted multi-UAV communication system is specifically as follows:

[0093] K quadrotor drones provide wireless communication services for K user groups, each of which includes multiple users; the kth drone serves the kth user group. In this embodiment, K=2;

[0094] In this embodiment, all drones carry N t = 32 antennas, and each user has 1 antenna;

[0095] Since the direct path between the drone and the user is blocked by obstacles, a smart reflective surface device with N = 100 reflective units is deployed on the surface of the building to reflect the signal transmitted from the drone to the user;

[0096] The channel g from the kth UAV to the smart reflective surface device k as follows:

[0097]

[0098] Where ρ0 is the channel power gain when the reference distance is 1m, q k is the three-dimensional position of the kth UAV, k = 1~K, w r =(0,250,20)m is the position of the smart reflective surface; and is the array response of the kth UAV ( and The Chinese meanings of cannot be the same. Please use different Chinese meanings to distinguish them and explain the Chinese meanings separately). and Obtained through channel estimation; where, represents the array response generated by the antenna unit; Represents the array response generated by the RIS unit.

[0099] The channel from the kth drone to the i-th user in the k-th user group as follows:

[0100]

[0101] in, is the position of the i-th user in the k-th user group; α ug is the path loss between the UAV and the user; κ ug is the Ricean factor of the channel between the drone and the user; and are the direct path component and the non-line-of-sight path component of the i-th user in the k-th user group served by the k-th UAV, respectively, k = 1~K;

[0102] The channel from the smart reflector device to the i-th user in the k-th user group is as follows:

[0103]

[0104] Among them, α rg is the path loss between the smart reflector and the user; k rg is the Ricean factor of the channel between the smart reflector and the user; and are the direct path component and the non-direct path component of the i-th user in the k-th user group, respectively;

[0105] Assuming that all drones use non-orthogonal multiple access technology to serve the user group, the signal-to-interference-and-noise ratio (SINR) of the i-th user in the k-th user group is k,i It is expressed as follows:

[0106]

[0107] Among them, f kis the precoding vector of the kth user group; p k,i is the transmission power of the i-th user in the k-th user group; Θ is the phase shift matrix of the smart reflection surface; σ 2 is additive white noise; the achievable rate R of the i-th user in the k-th user group k,i It is expressed as follows:

[0108] R k,i =log2(1+SINR k,i ). (5)

[0109] S2. Establish a mathematical model for minimizing the total energy consumption of a multi-UAV communication system assisted by an intelligent reflective surface;

[0110] Establish a mathematical model for minimizing the total energy consumption of a multi-UAV communication system assisted by an intelligent reflector, including determining the mathematical expressions of optimization variables, objective functions, and constraints;

[0111] The total energy consumption of the intelligent reflector-assisted multi-UAV communication system is expressed as:

[0112] P sum =P+P RIS +P UAV ; (6)

[0113] in, is the total transmission power of all UAVs, M k is the number of users in the kth user group, P RIS =1W is the power consumption of the smart reflector, P UAV =10W is the power consumed by the drone;

[0114] The optimization variables of the mathematical model for minimizing total energy consumption include:

[0115] 1) The three-dimensional position q of the kth UAV k ;

[0116] 2) Transmission power p of the i-th user in the k-th user group k,i ;

[0117] 3) smart reflection surface phase shift matrix Θ;

[0118] 4) Precoding vector f of the kth user group k ;

[0119] 5) User demodulation order u;

[0120] 6) The constraints of the mathematical model for minimizing total energy consumption include:

[0121] a) Transmission power constraint: p k,i ≥0;

[0122] b) Minimum transmission rate constraint: log2(1+SINR k,i )≥R min ; R min is the transmission rate threshold;

[0123] c) Minimum safe distance constraints between drones: Δ min is the minimum safe distance between drones;

[0124] d) Reflection unit phase coefficient constraint: θ n ∈[0,2π);θ n is the phase coefficient of the nth reflective unit in the intelligent reflective surface device, n = 1 to N;

[0125] e) Constraints on the demodulation order of all users: is a feasible set of demodulation orders, which can be obtained by brute force algorithm;

[0126] f) Constraints on the demodulation order of a single user: u k (i)>u k (t), u k (i) and u k (t) are the demodulation order of the i-th user in the k-th user group and the demodulation order of the t-th user in the k-th user group, respectively, and {u k (i),u k (t)∈u};

[0127] g) Constraints on the precoding vector: ||f k || 2 ≤1;

[0128] The mathematical model based on minimizing total energy consumption is as follows:

[0129]

[0130] stp k,i ≥0, (7b)

[0131] log2(1+SINR k,i )≥R min , (7c)

[0132]

[0133] θ n ∈[0,2π), (7e)

[0134]

[0135]

[0136]

[0137] ||f k || 2 ≤1. (7h)

[0138] S3. Jointly optimize the UAV position deployment and resource allocation, including the following steps:

[0139] S3.1. According to the maximum ratio transmission technique, the optimal precoding vector f for the kth user group is obtained. k ; Using the convex approximation algorithm, the three-dimensional position q of the kth UAV is obtained k , where k = 1, 2, ..., K;

[0140] The precoding vector f of the kth user group k The calculation formula is:

[0141]

[0142] in,(.) H Represents conjugate transpose; According to the constraint function formula (7c) and the precoding vector f k The calculation formula (8) is used to solve the optimization problem (9) to obtain the three-dimensional position q of the k-th UAV. k :

[0143]

[0144] stlog2(1+SINR k,i )≥R min (9b)

[0145]

[0146] q j represents the three-dimensional position of the j-th UAV, j = 1~K and j≠k.

[0147] S3.2. Using algebraic transformation methods, slack variables are introduced to transform the total energy consumption minimization problem into the differential form of two equivalent convex functions; then, a closed-loop expression for the smart reflector phase shift matrix Θ is derived using a Gaussian randomization process.

[0148] Based on formula (8), the precoding vector f of the kth user group is obtained k And the three-dimensional position q of the kth UAV obtained by formula (9a)-formula (9c) k , and according to the phase coefficient θ of the nth reflection unit n, calculate the smart reflection surface phase shift matrix Θ, the formula is as follows:

[0149]

[0150] Among them, e is the natural base; is a complex vector; is a complex vector The Nth element of ; U is a unitary matrix; ∑ is a diagonal matrix; S is a complex vector with mean 0 and variance 1; is the Nth element of the smart reflection surface phase shift matrix Θ, i.e., the phase coefficient; j is an imaginary number.

[0151] S3.3. According to the convex optimization tool, combined with the demodulation order u, the transmission power p of the i-th user in the k-th user group is obtained k,i where k = 1, 2, ..., K, i∈M k ;

[0152] Based on formula (7c) and solving the following optimization problem, calculate the transmission power p of the i-th user in the k-th user group k,i :

[0153]

[0154] stp k,i ≥0, (11b)

[0155] log2(1+SINR k,i )≥R min (11c)

[0156] S3.4, based on the principles of superposition coding technology and continuous interference elimination technology, combined with the three-dimensional position q of the kth UAV k , the transmission power p of the i-th user in the k-th user group k,i , the precoding vector f of the kth user group k Impact on the total energy consumption of the system, and obtain the demodulation order u of all users k (i), and then the demodulation order u is obtained;

[0157] According to the obtained precoding vector f of the kth user group k , the three-dimensional position q of the kth UAV k , the smart reflection surface phase shift matrix Θ and the transmission power p of the i-th user in the k-th user group k,i , calculate the demodulation order u of the i-th user in the k-user group k (i) and the demodulation order u of the tth user in the k-user group k (t), as follows:

[0158] When the channel conditions of the i-th user and the t-th user in the k-th user group are different, the demodulation order is calculated using the following formula:

[0159]

[0160] When the channel conditions of the i-th user and the t-th user in the k-th user group are the same, the demodulation order is calculated using the following formula:

[0161]

[0162] In this embodiment, a simulation effect diagram of a resource allocation method for a RIS-assisted multi-UAV communication system is shown in FIG. Figure 1 shown.

[0163] Figure 1 The other parameters are: minimum transmission rate threshold ξ = 0.5, 1, ..., 2.5. This embodiment shows the total power consumption of the proposed resource allocation scheme under different minimum SINRs and compares it with the "order reversal algorithm" and the "alternating optimization scheme". The number of user groups is K = 2, and the minimum SINR value varies from 0.5 to 2.5. Figure 1 As shown, the total power consumption of all schemes decreases as the value of the minimum SINR increases. This is because as the value of the minimum SINR increases, a larger transmission power needs to be allocated to the combined channel to meet the user's QoS requirements. In addition, the performance of the present invention in terms of power consumption is better than the "sequential reversal algorithm" and the "alternating optimization scheme". This is because the present invention takes into account NOMA technology and uses the same resource block to serve multiple users. This can obtain a higher SE and thus improve system performance. In addition, the present invention uses multiple UAVs to establish communication connections with edge users, thereby meeting the minimum rate required by edge users, thereby reducing transmission power. From Figure 1 It can be seen that when the minimum transmission rate threshold ξ is larger, the total system power consumption is larger.

[0164] Example 2:

[0165] In this embodiment, a simulation effect diagram of a resource allocation method for a RIS-assisted multi-UAV communication system is shown in FIG. Figure 2 shown.

[0166] Figure 2 Other parameters are: minimum transmission rate threshold ξ = 1.5, number of reflection units N = 100, number of antennas N t=4,8,…,32. This embodiment studies the total power consumption of the present invention under different numbers of days. To show the performance gain, this embodiment compares the power minimization scheme without considering RIS and the power minimization scheme of the RIS-assisted UAV-OFDMA system. The minimum SINR value ξ is set to 1.5 and the number of reflection units N is set to 100. Figure 2 As shown in Figure 2, as the number of antennas increases, the total power consumption of all resource allocation schemes decreases. In fact, configuring more antennas on the UAV can achieve higher diversity gain to balance the additional power consumed by the RF link, thereby consuming less transmit power. In addition, since the proposed scheme uses RIS to effectively improve the received signal power strength, it consumes less power compared to the "no RIS" scheme. Figure 2 It can be seen that when the number of antennas N t The more it is, the greater the total energy consumption of the system.

[0167] Example 3:

[0168] In this embodiment, a simulation effect diagram of a resource allocation method for a RIS-assisted multi-UAV communication system is shown in FIG. Figure 3-4 shown. Figure 3-4 Other parameters are: minimum SINR and number of antennas are ξ = 1.5 and N respectively t =32.

[0169] exist Figure 3 In this embodiment, the total power consumption of the present invention under different numbers of RIS reflection units is studied. Figure 3 As shown in FIG, the total power consumption of all resource allocation schemes decreases as the number of RIS reflector units increases. This is because the RIS optimization scheme proposed in the present invention can enhance the passive beamforming gain by controlling the phase shift coefficients of a large number of RIS reflector units, thereby reducing system power consumption. In addition, in this embodiment, Figure 3 It can be seen that placing the RIS close to the UAV can significantly improve performance, which indicates that selecting a suitable RIS location can enhance the passive beamforming gain. Figure 4 The total power consumption of all resource allocation schemes under different path loss coefficients is compared.

[0170] from Figure 4 As can be seen from the above, the total power consumption of all resource allocation schemes increases with the increase of path loss coefficient. In particular, when the path loss coefficient is α ug =α rg =2, all schemes have good system performance. However, when α ug =α rgWhen α > 2, the total power consumption increases significantly, especially for the “Withour RIS” solution. This is because as the path loss coefficient increases, the signal strength of the UAV-user link and the RIS-user link decreases. In this case, in order to meet the user’s minimum QoS requirements, a higher transmit power needs to be allocated. In addition, although α ug , α rg Increasing the value of will reduce system performance, but the performance of the present invention in terms of power consumption is still better than the "RIS-OMA" and "Without-RIS" solutions.

Claims

1. A resource allocation method for a RIS-assisted multi-UAV communication system, characterized in that: The following steps are involved: S1. Introduce intelligent reflective surface into multi-UAV assisted non-orthogonal multiple access system to build intelligent reflective surface assisted multi-UAV communication system; S2. Establish a mathematical model for minimizing the total energy consumption of a multi-UAV communication system assisted by an intelligent reflective surface; S3, jointly optimize the UAV position deployment and resource allocation; In step S1, the intelligent reflective surface assisted multi-UAV communication system is specifically as follows: K quadrotor drones provide wireless communication services for K user groups, each of which includes multiple users; the kth drone serves the kth user group; All drones carry N t antennas, and each user has 1 antenna; Since the direct path between the drone and the user is blocked by obstacles, a smart reflective surface device equipped with N reflective units is deployed on the surface of the building to reflect the signal transmitted from the drone to the user; The channel g from the kth UAV to the smart reflective surface device k as follows: Where ρ0 is the channel power gain when the reference distance is 1m, q k is the 3D position of the kth UAV, k = 1~K, w r is the location of the smart reflective surface; represents the array response of the kth UAV generated by the antenna unit; represents the array response of the kth UAV generated by the RIS unit, and Obtained through channel estimation; Channel from the kth drone to the i-th user in the k-th user group as follows: in, is the position of the i-th user in the k-th user group; α ug is the path loss between the UAV and the user; κ ug is the Rice factor of the channel between the drone and the user; and are the direct path component and the non-direct line-of-sight path component of the i-th user in the k-th user group served by the k-th UAV, k = 1~K; The channel from the smart reflector device to the i-th user in the k-th user group is as follows: Among them, α rg is the path loss between the smart reflector and the user; κ rg is the Ricean factor of the channel between the smart reflector and the user; and They are the direct path component and the non-direct path component of the i-th user in the k-th user group, respectively; Assuming that all drones use non-orthogonal multiple access technology to serve user groups, the SINR of the i-th user in the k-th user group is k,i It is expressed as follows: Among them, f k is the precoding vector of the kth user group; p k,i is the transmission power of the i-th user in the k-th user group; Θ is the smart reflection surface phase shift matrix; σ 2 is additive white noise; the achievable rate R of the i-th user in the k-th user group k,i It is expressed as follows: R k,i =log2(1+SINR k,i ); (5) In step S2, a mathematical model for minimizing the total energy consumption of the intelligent reflective surface-assisted multi-UAV communication system is established, including determining the mathematical expressions of optimization variables, objective functions, and constraint conditions; The total energy consumption of the intelligent reflective surface assisted multi-UAV communication system is expressed as: R sum =P+P RIS +P UAV ; (6) in, is the total transmission power of all UAVs, M k is the number of users in the kth user group, P RIS is the power consumed by the smart reflector, P UAV is the power consumed by the drone; The optimization variables of the mathematical model for minimizing total energy consumption include: 1) The three-dimensional position q of the kth UAV k ; 2) The transmission power p of the i-th user in the k-th user group k,i ; 3) Smart reflection surface phase shift matrix Θ; 4) Precoding vector f of the kth user group k ; 5) The demodulation order u of the user; The constraints of the mathematical model for minimizing total energy consumption include: a) Transmission power constraint: p k，i ≥0; b) Minimum transmission rate constraint: log2(1+SINR k,i )≥R min ; R min is the transmission rate threshold; c) Minimum safe distance constraints between drones: Δ min is the minimum safe distance between drones; d) Reflection unit phase coefficient constraint: θ n ∈[0,2π);θ n is the phase coefficient of the nth reflection unit in the intelligent reflection surface device, n = 1 to N; e) Constraints on the demodulation order of all users: is a feasible set of demodulation orders, obtained by brute force algorithm; f) Constraints on demodulation order of a single user: u k (i)>u k (t), u k (i) and u k (t) are the demodulation order of the i-th user in the k-th user group and the demodulation order of the t-th user in the k-th user group, respectively, and {u k (i),u k (t)∈u}; g) Constraints on precoding vectors: ∥f k ∥ 2 ≤1; The mathematical model based on minimizing total energy consumption is as follows: s.t.p k,i ≥0,(7b) log2(1+SINR k,i )≥R min ,(7c) i n ∈[0,2π),(7e) ∥f k ∥ 2 ≤1;(7h) Step S3 includes the following steps: S3.

1. According to the maximum ratio transmission technique, the optimal precoding vector f of the kth user group is obtained. k ; Using the convex approximation algorithm, the three-dimensional position q of the kth UAV is obtained k , where k = 1, 2, ..., K; S3.2, using the algebraic transformation method, introducing slack variables to transform the total energy consumption minimization problem into the difference form of two equivalent convex functions; and then using the Gaussian randomization process to obtain the closed-loop expression of the smart reflector phase shift matrix Θ; S3.3, according to the convex optimization tool, combined with the user demodulation order u, the transmission power p of the i-th user in the k-th user group is obtained k,i , where k = 1, 2, ..., K, i∈M k ; S3.4, based on the principle of superposition coding technology and continuous interference elimination technology, combined with the three-dimensional position q of the kth UAV k , the transmission power p of the i-th user in the k-th user group k,i , the precoding vector f of the kth user group k The impact on the total energy consumption of the system, and the demodulation order u of all users is obtained k (i), and then the total demodulation order u is obtained; In step S3.1, the precoding vector f of the kth user group k The calculation formula is: in,(.) H represents the conjugate transpose; according to the constraint function formula (7c) and the precoding vector f k The calculation formula (8) is used to solve the optimization problem (9) to obtain the three-dimensional position q of the kth UAV. k : s.t.log2(1+SINR k,i )≥R min (9b) q j represents the three-dimensional position of the j-th UAV, j = 1~K and j≠k; In step S3.2, the precoding vector f of the kth user group is obtained based on formula (8): k And the three-dimensional position q of the kth UAV obtained by formula (9a)-formula (9c) k , and according to the phase coefficient θ of the nth reflection unit n , calculate the smart reflection surface phase shift matrix Θ, the formula is as follows: Among them, e is the natural base; is a complex vector; is a complex vector The Nth element of ; U is a unitary matrix; ∑ is a diagonal matrix; S is a complex vector with a mean of 0 and a variance of 1; is the Nth element of the smart reflection surface phase shift matrix Θ, i.e., the phase coefficient; j is an imaginary number; In step S3.3, the transmission power p of the i-th user in the k-th user group is calculated based on formula (7c) and solving the following optimization problem: k,i : s.t.p k,i ≥0,(11b) log2(1+SINR k,i )≥R min ;(11c) In step S3.4, according to the obtained precoding vector f of the kth user group k , the 3D position q of the kth UAV k , the smart reflection surface phase shift matrix Θ and the transmission power p of the i-th user in the k-th user group k,i , calculate the demodulation order u of the i-th user in the k-user group k (i) and the demodulation order u of the tth user in the k-user group k (t), as follows: When the channel conditions of the i-th user and the t-th user in the k-th user group are different, the demodulation order is calculated by the following formula: When the channel conditions of the i-th user and the t-th user in the k-th user group are the same, the demodulation order is calculated by the following formula:

Images