UAV resource allocation method based on RIS model in multi-layer UAV swarm

Abstract

The present invention discloses a method for allocating drone resources based on a RIS model in a multi-layer drone cluster, and the implementation steps are as follows: 1. Install RIS on each drone of each layer of the multi-layer drone cluster network; 2. Obtain the channel characteristics of the composite channel; 3. Generate a probability density distribution; 4. Generate a power optimization model; 5. Solve the power optimization model, and the ground transmitter allocates power to each drone. The present invention solves the problem of small user capacity in the traditional multi-layer drone cluster resource allocation method by installing RIS on each drone of each layer. The present invention generates a probability distribution of instantaneous channel power gain of a multi-layer drone cluster based on the RIS model, and substitutes it into the power optimization model, so that the channel model and power allocation scheme of the multi-layer drone cluster can adapt to harsh environments.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication technology, and further relates to a method for allocating drone resources based on a RIS model in a multi-layer drone cluster in the field of emergency communication technology. The present invention can be applied to any multi-layer drone cluster formed by any number of drones with different attributes, to achieve the scheduling of power resources, to meet the power resource requirements of mission objectives in harsh environments, and to achieve real-time effective allocation of power resources carried by multi-layer drone clusters. Background Art

[0002] In a typical emergency wireless communication network, in order to timely and efficiently search and rescue trapped people after a rainstorm disaster, drones are usually placed as temporary base stations or temporary relays that rescue vehicles cannot reach. Communication transmission based on multi-layer unmanned aerial vehicle (UAV) clusters can be used to quickly establish a temporary communication network. Multi-layer UAV clusters are composed of various functionally heterogeneous drones. Different drones can carry different types of airborne equipment, which makes drone clusters have the advantages of low mission execution cost, high flexibility and maneuverability in harsh conditions such as rainstorms. The existing multi-UAV resource allocation method mainly determines the drone that serves a specific user based on the relative position between drones and users, and then allocates resources to different drones and users. When facing the resource allocation problem of multi-layer UAV clusters in harsh environments, due to the complex and changeable channel conditions, the existing UAV resource allocation method cannot meet the user's demand for communication capacity and has a small adaptability range. Reconfigurable Intelligent Surface (RIS) was first used in traditional ground wireless communications. Related research has verified that its application scope can be extended to drone communications and can achieve system capacity improvement without adding additional hardware facilities.

[0003] Nanjing University of Posts and Telecommunications has disclosed a method for allocating drone resources in a drone cluster in its patent document "Power allocation and trajectory design method for multi-drone assisted mobile user communication" (application number: 202110153592.0 Application date: 2021.02.04 Application publication number: CN 112969185 A). The implementation steps of this method are: first, construct a downlink transmission system for multi-drone assisted mobile user wireless communication, and cluster users using the K means clustering method according to the user's mobile trajectory; second, at the same time, users in the same cluster are provided with communication services by a drone, and the optimal drone communication position and transmission power in each cluster are calculated with the goal of maximizing the minimum average throughput of the user; third, power allocation is performed according to the calculated optimal transmission power corresponding to each drone. The disadvantage of this method is that in the power allocation method, a user can only receive the power sent by one drone at any time, and it does not consider that multiple drones allocate power to one user at the same time, so the user capacity of this method is small. In addition, when dealing with the power allocation problem of multi-layer UAV clusters in harsh environments, this method does not consider factors such as channel path loss, rain attenuation, shadow fading, multipath fading, etc. Therefore, this method cannot be applied to the power allocation problem of multi-layer UAV clusters in harsh environments.

[0004] Z.Yao, W.Cheng, W.Zhang, and H.Zhang published a resource allocation method based on the RIS model in a drone cluster in their paper "Resource Allocation for 5G-UAV Based Emergency Wireless Communications" (IEEE Journal on Selected Areas in Communications, pp.1-1, 2021). The specific steps of this method are: first, using Gaussian hypergeometric functions to establish a user channel model and power gain expression based on the Fisher-Snedecor F channel; second, solving the optimal power allocation method under the model; finally, jointly considering resources such as bandwidth and power, solving the optimal bandwidth-power allocation method under the model. However, the method still has the disadvantage that the Fisher-Snedecor F channel used is in the form of an uncorrected hypergeometric function and is modeled using Matlab. Therefore, the corresponding power allocation scheme can neither conform to the power allocation scheme under the Nakagami-m channel model (the Nakagami-m channel is a special case of the channel with path loss and shadow fading below zero), nor can it adapt to the corresponding Python-based deep reinforcement learning and other methods, which greatly limits the scope of application of the multi-layer UAV channel model based on the RIS model. In addition, when dealing with the power allocation problem of multi-layer UAV clusters in harsh environments, this method does not consider factors such as channel rain attenuation, so this method cannot be applied to the power allocation problem of multi-layer UAV clusters in harsh environments. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and propose a drone resource allocation method based on the RIS model in a multi-layer drone cluster, so as to solve the problem that the user capacity in the multi-layer drone cluster resource allocation method is small, the application scope of the existing resource allocation method is small, and it cannot be applied to harsh environments.

[0006] The specific idea of ​​implementing the present invention is: the present invention adopts the reconfigurable intelligent surface RIS technology, installs RIS on each drone of each layer, and utilizes the passive reflection characteristics of RIS to passively reflect the communication signal transmitted by the drone, so as to increase the channel capacity of the ground receiving terminal users without additionally increasing the transmission power of the drone, thereby solving the problem of small user capacity in the multi-layer drone cluster resource allocation method. The present invention adopts Fox's H function and the fitted Fisher-Snedecor F channel model to solve the multi-layer drone cluster channel characteristics obtained by the communication between the drone and the terminal using the pilot signal, and generates the probability distribution of the instantaneous channel power gain of the multi-layer drone cluster based on the RIS model, so that it can fit the channel model of the multi-layer drone cluster in the harsh environment, and solves the problem that the multi-layer drone channel model based on the RIS model has a small application range. The present invention substitutes the probability distribution of the instantaneous channel power gain of the multi-layer drone cluster into the power optimization model to obtain the power allocation scheme that maximizes the user capacity, solving the problem that the existing power allocation method cannot be used in harsh environments.

[0007] The steps of the present invention include the following:

[0008] Step 1: Install RIS on each drone in each layer of the multi-layer drone cluster network;

[0009] Step 2: Obtain the channel characteristics of the composite channel:

[0010] (2a) The ground transmitter sends a pilot signal to each UAV relay in the relay cluster. Each UAV relay transmits the pilot signal back to the ground transmitter. The ground transmitter calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by the ground transmitter to the relay cluster. The relay cluster refers to a multi-layer UAV cluster consisting of multiple UAV relays carrying RIS.

[0011] (2b) Each UAV relay sends a pilot signal to the ground receiving end, and the ground receiving end transmits the pilot signal back to each UAV relay in the relay cluster. Each UAV relay calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by each UAV relay to the ground receiving end, and fits the shadow fading factor and multipath fading factor of the communication signal sent by each UAV relay to the ground receiving end in the Fisher-Snedecor F channel model;

[0012] (2c) Each UAV relay sends the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor from the UAV relay end to the ground receiving end to the ground transmitting end. The ground transmitting end uses Fox's H function to obtain the channel characteristics of the overall channel from the ground transmitting end to the relay cluster and then to the ground receiving terminal;

[0013] Step 3, generate the probability density distribution f(h) as follows:

[0014]

[0015] Wherein, h represents the instantaneous power gain of the signal received by the ground receiving terminal, P represents the total number of layers of the multi-layer UAV cluster, Π represents the multiplication operation, p represents the serial number of the layer number of the multi-layer UAV cluster, K represents the total number of UAVs in the p-th layer, k represents the serial number corresponding to the k-th UAV relay in the p-th layer, N represents the total number of RIS on the k-th UAV relay in the p-th layer, n represents the serial number corresponding to the n-th RIS on the k-th UAV relay in the p-th layer, Γ represents the gamma function evaluation operation, and m p,k,n represents the multipath fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end, ms p,k,n represents the shadow fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end;

[0016] Step 4: Generate a power optimization model as follows:

[0017]

[0018]

[0019] Where P represents the model of the ground transmitting terminal transmission power, represents the power P(h) of the instantaneous signal transmitted by the ground transmitting terminal that maximizes the value of the expression in the brackets, B represents the bandwidth of the signal transmitted by the ground transmitting terminal, h represents the instantaneous power gain of the signal received by the ground receiving terminal, δ represents the root mean square value of the composite channel noise power, st represents the restricted condition of the model of the ground transmitting terminal's transmission power, Represents the average power value of the signal transmitted by the ground transmitting terminal, P hm Indicates the peak value of the power of the signal transmitted by the ground transmitting terminal;

[0020] Step 5: Solve the power optimization model. The ground transmitter allocates power P(h) to each drone relay as follows:

[0021]

[0022] Among them, h0 represents the instantaneous channel gain threshold from the ground transmitter to the relay and then to the ground receiving terminal, and λ represents the ground transmitter according to Defined power allocation parameters.

[0023] Compared with the prior art, the present invention has the following advantages:

[0024] First, since the present invention installs RIS on each drone of each layer of the multi-layer drone cluster network, it overcomes the defect of small user capacity in the prior art, so that the present invention has the characteristic of large user capacity served by the multi-layer drone cluster, and has the advantage of increasing the channel capacity of ground receiving terminal users without additionally increasing the drone transmission power.

[0025] Second, since the present invention adopts Fox's H function and the fitted Fisher-Snedecor F channel model to obtain the channel characteristics of the composite channel and generate the probability distribution of the instantaneous channel power gain of the multi-layer UAV cluster based on the RIS model, the problem that the existing channel model cannot fit the multi-layer UAV cluster channel in harsh environments is solved, so that the present invention can fit the characteristics of the channel model in harsh environments, and has the advantages of strong channel model fitting and applicability.

[0026] Third, since the present invention substitutes the probability distribution of the instantaneous channel power gain of the multi-layer UAV cluster that is applicable to harsh environments into the power optimization model to generate a power allocation plan that maximizes user capacity, it solves the problem that the existing UAV power allocation method cannot be used in harsh environments. The present invention has the characteristic of being applicable to harsh environments and has the advantage of a wide range of applicability of the UAV power allocation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of the present invention.

[0028] Figure 2 It is a simulation diagram of simulation experiment 1 of the present invention;

[0029] Figure 3 It is a simulation diagram of simulation experiment 2 of the present invention;

[0030] Figure 4 It is a simulation diagram of simulation experiment 3 of the present invention;

[0031] Figure 5 It is a simulation diagram of simulation experiment 4 of the present invention;

[0032] Figure 6 This is a simulation diagram of simulation experiment 5 of the present invention. DETAILED DESCRIPTION

[0033] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0034] Reference Figure 1 , the implementation steps of the present invention are further described in detail.

[0035] Step 1: Install RIS on each drone in each layer of the multi-layer drone cluster network.

[0036] A multi-layer drone cluster means that when the ground launch terminal is working, each drone in the drone cluster acts as a drone air relay, and when the ground launch terminal is not working, each drone in the drone cluster acts as a drone air base station.

[0037] Step 2: Obtain channel characteristics of the composite channel.

[0038] When each drone acts as an aerial relay, the channel characteristics of the composite channel are generated according to the following steps:

[0039] In the first step, the ground transmitter sends a pilot signal to each UAV relay in the relay cluster. Each UAV relay transmits the pilot signal back to the ground transmitter. The ground transmitter calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by the ground transmitter to the relay cluster.

[0040] The instantaneous path loss rain attenuation value experienced by the communication signal sent by the ground transmitter V to the kth UAV relay D in the pth layer is expressed as represents the distance from the ground transmitting terminal to the kth UAV relay in the pth layer, α represents the path loss factor of the composite channel, A represents the path loss from the ground transmitting terminal to the kth UAV relay on the pth layer, rVD,p,k It represents the rainfall loss from the ground transmitting terminal to the kth UAV relay in the pth layer. The relay cluster refers to a multi-layer UAV cluster consisting of multiple UAV relays carrying RIS.

[0041] In the second step, each UAV relay sends a preamble signal to the ground receiving end, and the ground receiving end transmits the preamble signal back to each UAV relay in the relay cluster. Each UAV relay calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by each UAV relay to the ground receiving end, and fits the shadow fading factor and multipath fading factor of the communication signal sent by each UAV relay to the ground receiving end in the Fisher-Snedecor F channel model.

[0042] The instantaneous path loss and rain attenuation value experienced by the communication signal sent by the kth UAV relay on the pth layer to the ground receiving end is expressed as represents the distance from the kth UAV relay in the pth layer to the ground receiving terminal, A represents the path loss from the kth UAV in the pth layer to the ground receiving terminal. rDU,p,k represents the rainfall loss from the kth UAV in the pth layer to the ground receiving terminal, m p,k,n represents the multipath fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end, ms p,k,nIt represents the shadow fading factor from the nth RIS on the kth UAV relay on the pth layer to the receiving end.

[0043] In the third step, each UAV relay sends the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor from the UAV relay end to the ground receiving end to the ground transmitting end. The ground transmitting end uses Fox's H function to obtain the channel characteristics H of the overall channel from the ground transmitting end to the relay cluster and then to the ground receiving terminal. PDF .

[0044] H PDF The expression is as follows:

[0045]

[0046] in,

[0047] represents the path loss and rain attenuation experienced by the k-th UAV relay in the p-th layer of the composite channel, represents the average power gain from the nth RIS on the kth UAV relay at the pth layer to the ground receiving terminal,

[0048] When each drone is used as an aerial base station, the channel characteristics of the composite channel are generated according to the following steps:

[0049] In the first step, each UAV base station sends a pilot signal to the ground receiving end, and the ground receiving end transmits the pilot signal back to each UAV base station in the UAV cluster. Each UAV base station calculates the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor experienced by the communication signal sent by each UAV base station to the ground receiving end;

[0050] The instantaneous path loss and rain attenuation value experienced by the communication signal sent by the kth UAV relay on the pth layer to the ground receiving end is expressed as represents the distance from the kth UAV relay in the pth layer to the ground receiving terminal, α represents the path loss factor of the composite channel, A represents the path loss from the kth UAV in the pth layer to the ground receiving terminal. rDU,p,k represents the rainfall loss from the kth UAV in the pth layer to the ground receiving terminal, m p,k,n represents the multipath fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end, ms p,k,n It represents the shadow fading factor from the nth RIS on the kth UAV relay on the pth layer to the receiving end.

[0051] In the second step, each UAV base station sends the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor from the UAV base station to the ground receiving terminal to all other UAV base stations. Each UAV base station in the UAV cluster considers the direct communication signals transmitted by all UAV base stations to the ground receiving terminal, and obtains the channel characteristics H of the overall channel from all UAV base stations to the ground receiving terminal. PDF .

[0052] H PDF The expression is as follows:

[0053]

[0054] in,

[0055] represents the path loss and rain attenuation experienced by the k-th UAV relay in the p-th layer of the composite channel, represents the average power gain from the nth RIS on the kth UAV relay at the pth layer to the ground receiving terminal,

[0056]

[0057] Step 3, generate probability density distribution f(h):

[0058]

[0059] Among them, h represents the instantaneous power gain of the signal received by the ground receiving terminal, P represents the total number of layers of the multi-layer UAV cluster, Π represents the multiplication operation, p represents the serial number of the layer number of the multi-layer UAV cluster, K represents the total number of UAVs in the p-th layer, k represents the serial number corresponding to the k-th UAV relay in the p-th layer, N represents the total number of reconfigurable intelligent surfaces RIS on the k-th UAV relay in the p-th layer, n represents the serial number corresponding to the n-th RIS on the k-th UAV relay in the p-th layer, and Γ represents the gamma function evaluation operation.

[0060] Step 4: Convert the optimization problem of the transmission power of the ground base station in the multi-layer UAV cluster when communicating with the ground user with the help of the communication UAV relay into a convex optimization problem, and generate the power optimization model as follows:

[0061]

[0062]

[0063] Where P represents the model of the ground transmitting terminal transmission power, represents the power P(h) of the instantaneous signal transmitted by the ground transmitting terminal that maximizes the value of the expression in the brackets, B represents the bandwidth of the signal transmitted by the ground transmitting terminal, h represents the instantaneous power gain of the signal received by the ground receiving terminal, δ represents the root mean square value of the composite channel noise power, st represents the restricted condition of the model of the ground transmitting terminal's transmission power, Represents the average power value of the signal transmitted by the ground transmitting terminal, P hm Indicates the peak power of the signal transmitted by the ground transmitting terminal.

[0064] Step 5: Solve the power optimization model. The ground transmitter allocates power P(h) to each drone relay as follows:

[0065]

[0066] Among them, h0 represents the instantaneous channel gain threshold from the ground transmitter to the relay and then to the ground receiving terminal, and λ represents the ground transmitter according to Defined power allocation parameters.

[0067] The effect of the present invention is further described below in conjunction with simulation experiments:

[0068] 1. Simulation experiment conditions:

[0069] The hardware platform of the simulation experiment of the present invention is: the processor is AMD Ryzen 7 3800X, the graphics card is NVIDIA GeForce GTX 1050Ti, the main frequency is 4.3GHz, and the memory is 32GB.

[0070] The software platform for the simulation experiment of the present invention is: Windows 10 operating system, PyCharm2021.1.1 and python 3.6.

[0071] 2. Simulation content and results analysis:

[0072] The simulation experiments of the present invention include 5.

[0073] The simulation experiment 1 of the present invention adopts the present invention and an existing technology to obtain the channel characteristics of the composite channel using the input drone parameters and wireless channel parameters, and obtains the probability density distribution diagram of the instantaneous channel gain and the result of the optimal transmission power.

[0074] The parameters of the drone in the simulation experiment 1 of the present invention include the number of drone layers P=2, the number of drones per layer K=1, and the number of reconfigurable smart surface units on each drone N=2. The wireless channel parameters include the signal frequency f=30GHz of the ITU-R rainfall fading model, D rain=5km, R = 100mm / h, the polarization tilt angle of the drone relative to the horizontal plane τ = 0, the path elevation angle θ = 45°, and the distance between each layer of drones and the transmitter and receiver d = L DU =L VD =400m, α=0.3, The shadow fading coefficient ms=5, and the multipath fading coefficient m=1~10.

[0075] A prior art reconfigurable reflective surface RIS used in simulation experiment 1 is:

[0076] Cui Tiejun et al. proposed the concepts of digitally coded metamaterials and programmable metamaterials in “Coding metamaterials, digital metamaterials and programmable metamaterials, Light-Sci. Appl., vol. 3, pp. 1-9, Oct. 2014”, referred to as reconfigurable reflective surfaces RIS.

[0077] Combine the following Figure 2 The simulation diagram of the present invention is further described.

[0078] Figure 2 (a) Using PyCharm2021.1.1 and Python 3.6 software platforms and the method of the present invention, a probability density distribution diagram of the instantaneous channel gain when the signal sent by each drone relay to the ground receiving end in the simulation experiment 1 of the present invention experiences a shadow fading coefficient of ms=5 and a multipath fading coefficient of m=1~10 is drawn. Figure 2 The x-axis coordinate in (a) is the multipath fading coefficient m, the y-axis coordinate is the instantaneous power gain of the channel, and the z-axis coordinate is the probability density distribution of the instantaneous channel gain.

[0079] Depend on Figure 2 (a) It can be seen that the average signal-to-noise ratio of each layer of the fixed channel increases with the decrease of the multipath fading coefficient m. It can be seen that the instantaneous signal-to-noise ratio / channel gain probability density function curve obtained by the simulation is shifted to the left as a whole, indicating that the central value of the instantaneous signal-to-noise ratio of the channel decreases. Therefore, it is shown that the invention can be used to obtain a channel model that meets the changing multipath fading of the channel in a harsh environment.

[0080] Figure 2 (b) Using PyCharm2021.1.1 and python 3.6 software platforms and the method of the present invention, the optimal power distribution diagram is drawn when the signal sent by each drone relay to the ground receiving end in the simulation experiment 1 of the present invention experiences a shadow fading coefficient of ms=5 and a multipath fading coefficient of m=1~10. Figure 2The x-axis coordinate in (b) is the instantaneous power gain of the channel, the y-axis coordinate is the multipath fading coefficient m, and the z-axis coordinate is the optimal allocated power.

[0081] Depend on Figure 2 (b) It can be seen that as the multipath fading coefficient m continues to decrease, the multipath fading continues to increase, and the power required to be allocated to the emergency communication channel also increases accordingly. It can be seen that the present invention can be consistent with the optimal power allocation scheme of the Nakagami-m channel as a special case of the channel. Therefore, it is shown that the use of the present invention can meet the power allocation situation of the channel fading in a harsh environment as multipath fading.

[0082] Simulation experiment 2 of the present invention adopts the present invention and an existing technology to obtain the channel characteristics of the composite channel using the input drone parameters and wireless channel parameters, and obtains the probability density distribution diagram of the instantaneous channel gain and the result of the optimal transmission power.

[0083] The parameters of the drone in the simulation experiment 2 of the present invention include the number of drone layers P = 2, the number of drones per layer K = 1, and the number of reconfigurable smart surface units on each drone N = 2. The wireless channel parameters include the signal frequency f = 30 GHz of the ITU-R rainfall fading model, D rain =5km, R = 100mm / h, the polarization tilt angle of the drone relative to the horizontal plane τ = 0, the path elevation angle θ = 45°, and the distance between each layer of drones and the transmitter and receiver d = L DU =L VD =400m, α=0.3, The multipath fading coefficient m=5, and the shadow fading coefficient ms=1.1~10.

[0084] A prior art reconfigurable intelligent surface RIS used in simulation experiment 2 is:

[0085] Cui Tiejun et al. proposed the concepts of digitally coded metamaterials and programmable metamaterials in “Coding metamaterials, digital metamaterials and programmable metamaterials, Light-Sci. Appl., vol. 3, pp. 1-9, Oct. 2014”, referred to as reconfigurable intelligent surfaces (RIS).

[0086] Combine the following Figure 3 The simulation diagram of the present invention is further described.

[0087] Figure 3(a) Using PyCharm2021.1.1 and Python 3.6 software platforms and the method of the present invention, a probability density distribution diagram of the instantaneous channel gain when the signal sent by each drone relay to the ground receiving end in the simulation experiment 2 of the present invention experiences a multipath fading coefficient of m=5 and a shadow fading coefficient of ms=1.1~10 is drawn. Figure 3 The x-axis coordinate in (a) is the shadow fading coefficient ms, the y-axis coordinate is the instantaneous power gain of the channel, and the z-axis coordinate is the probability density distribution of the instantaneous channel gain.

[0088] Depend on Figure 3 (a) It can be seen that the average signal-to-noise ratio of each layer of the fixed channel increases with the decrease of the shadow fading coefficient ms, and the instantaneous signal-to-noise ratio / channel gain probability density function curve obtained by simulation shifts to the left as a whole, indicating that the central value of the instantaneous signal-to-noise ratio of the channel decreases. Therefore, it is shown that the invention can obtain a channel model that conforms to the changing shadow fading of the channel in a harsh environment.

[0089] Figure 3 (b) Using PyCharm2021.1.1 and python 3.6 software platforms and the method of the present invention, the optimal power distribution diagram is drawn when the signal sent by each drone relay to the ground receiving end in the simulation experiment 2 of the present invention experiences a shadow fading coefficient of ms=5 and a multipath fading coefficient of m=1~10. Figure 3 The x-axis coordinate in (b) is the instantaneous power gain of the channel, the y-axis coordinate is the multipath fading coefficient m, and the z-axis coordinate is the optimal allocated power.

[0090] Depend on Figure 3 (b) It can be seen that as ms=1.1-10 decreases, shadow fading increases, and the power required to be allocated to the emergency communication channel also increases, which is consistent with the shadow fading situation. Therefore, it is shown that the invention can meet the power allocation situation of the channel fading in the harsh environment as shadow fading.

[0091] The simulation experiment 3 of the present invention adopts the present invention and an existing technology to obtain the channel characteristics of the composite channel using the input drone parameters and wireless channel parameters to obtain the probability density distribution result of the instantaneous channel gain.

[0092] The parameters of the drone in the simulation experiment 3 of the present invention include the number of drone layers P=2, the number of drones per layer K=1, and the number of intelligent reflection units on each drone N=2. The wireless channel parameters include the signal frequency f=30GHz of the ITU-R rainfall fading model, D rain = 5km, different rain speeds R = 95 ~ 110mm / h, the polarization tilt angle of the drone relative to the horizontal plane τ = 0, the path elevation angle θ = 45°, and the distance between the transmitter and the receiver of each layer of drones d = LDU =L VD =400m, α=0.3, The multipath fading coefficient m=5, and the shadow fading coefficient ms=5.

[0093] A prior art reconfigurable reflective surface RIS used in simulation experiment 3 is:

[0094] Cui Tiejun et al. proposed the concepts of digitally coded metamaterials and programmable metamaterials in “Coding metamaterials, digital metamaterials and programmable metamaterials, Light-Sci. Appl., vol. 3, pp. 1-9, Oct. 2014”, referred to as reconfigurable reflective surfaces RIS.

[0095] Combine the following Figure 4 The simulation diagram of the present invention is further described.

[0096] Figure 4 Using PyCharm2021.1.1 and python 3.6 software platforms and utilizing the method of the present invention, a probability density distribution diagram of the instantaneous channel gain when the signal sent by the ground transmitting terminal to each drone relay and then to the ground receiving end in the simulation experiment 3 of the present invention experiences different rainfall attenuations that vary with the increase of rain speed is drawn. Figure 4 The x-axis coordinate in is the instantaneous power gain of the channel, and the y-axis coordinate is the probability density distribution of the instantaneous channel gain. Figure 4 The straight line marked with small dots in the middle represents the case where the rainfall speed is 95 mm / h, the straight line marked with large dots represents the case where the rainfall speed is 100 mm / h, the straight line marked with a lower triangle represents the case where the rainfall speed is 105 mm / h, and the straight line marked with an upper triangle represents the case where the rainfall speed is 110 mm / h.

[0097] Depend on Figure 4 It can be seen that as the rain speed increases, the corresponding link rain attenuation also increases. Taking rain attenuation as an important factor affecting the average signal-to-noise ratio of each layer (the trend of path loss is the same as rain attenuation), the instantaneous signal-to-noise ratio of each layer of the fixed channel fluctuates up and down, and the instantaneous signal-to-noise ratio / channel gain probability density function curve obtained by simulation shifts to the left as a whole, indicating that the central value of the instantaneous signal-to-noise ratio of the channel decreases, and the concentration of the instantaneous signal-to-noise ratio increases. The greater the rain speed of the rainstorm, the smaller the power gain of the channel, and the worse the communication conditions, which shows that the present invention can be applied to harsh environments such as rainstorms.

[0098] Simulation experiment 4 of the present invention adopts the present invention and an existing technology to obtain the channel characteristics of the composite channel using the input drone parameters and wireless channel parameters, and obtains the probability density distribution diagram of the instantaneous channel gain and the result of the optimal transmission power.

[0099] The parameters of the drone in the simulation experiment 4 of the present invention include the number of drone layers P=2, the number of drones per layer K=1, and the number of intelligent reflection units on each drone N=2. The wireless channel parameters include the signal frequency f=30GHz of the ITU-R rainfall fading model, D rain =5km, R = 100mm / h, the polarization tilt angle of the drone relative to the horizontal plane τ = 0, the path elevation angle θ = 45°, and the distance between each layer of drones and the transmitter and receiver d = L DU =L VD =380~410m,α=0.3, The multipath fading coefficient m=5, and the shadow fading coefficient ms=5.

[0100] A prior art reconfigurable reflective surface RIS used in simulation experiment 4 is:

[0101] Cui Tiejun et al. proposed the concepts of digitally coded metamaterials and programmable metamaterials in “Coding metamaterials, digital metamaterials and programmable metamaterials, Light-Sci. Appl., vol. 3, pp. 1-9, Oct. 2014”, referred to as reconfigurable reflective surfaces RIS.

[0102] Combine the following Figure 5 The simulation diagram of the present invention is further described.

[0103] Figure 5 (a) Using PyCharm2021.1.1 and Python 3.6 software platforms and the method of the present invention, a probability density distribution diagram of the instantaneous channel gain when the signal sent by the ground transmitting terminal to each drone relay and then to the ground receiving end in the simulation experiment 4 of the present invention passes through different layers of drones at different lengths is drawn. Figure 5The x-axis coordinate in (a) is the instantaneous power gain of the channel, and the y-axis coordinate is the probability density distribution of the instantaneous channel gain. The straight line marked with small dots represents the case where the first layer distance is 410m and the second layer distance is 410m (each layer has 2 reflection units), the straight line marked with large dots represents the case where the first layer distance is 410m and the second layer distance is 400m, the straight line marked with lower triangles represents the case where the first layer distance is 400m and the second layer distance is 400m, the straight line marked with upper triangles represents the case where the first layer distance is 410m and the second layer distance is 390m, the straight line marked with left triangles for short spacing represents the case where the first layer distance is 410m and the second layer distance is 380m, and the straight line marked with left triangles for long spacing represents the case where the first layer distance is 400m and the second layer distance is 390m.

[0104] Depend on Figure 5 (a) It can be seen that when the average SNR of one layer is fixed (such as setting the path length to 400m) and the other layer increases (such as setting the path length from 410m to 400m), the instantaneous signal-to-noise ratio of each layer of the fixed channel fluctuates up and down, and the simulated instantaneous signal-to-noise ratio / channel gain probability density function curve moves to the right, indicating that the central value of the instantaneous signal-to-noise ratio of the channel increases, while the probability of the central value decreases. And the large difference between the average SNRs of the two layers causes the instantaneous power gain to be more concentrated toward the center. Therefore, it is shown that the present invention can meet the channel model of a multi-layer drone cluster.

[0105] Figure 5 (b) Using PyCharm2021.1.1 and python 3.6 software platforms and the method of the present invention, a transmission power diagram of the signal sent by the ground transmitting terminal to each drone relay and then to the ground receiving end in the simulation experiment 4 of the present invention is drawn when the length of the signal passing through different layers of drones is different. Figure 5 The x-axis coordinate in (b) is the instantaneous power gain of the channel, and the y-axis coordinate is the optimal transmission power. Among them, the straight line marked with a cross represents the situation where the first layer distance is 410m and the second layer distance is 410m (each layer has 2 reflection units). The straight line marked with a five-pointed star represents the situation where the first layer distance is 410m and the second layer distance is 400m. The straight line marked with meters represents the situation where the first layer distance is 400m and the second layer distance is 400m. The straight line marked with an upper triangle represents the situation where the first layer distance is 410m and the second layer distance is 390m. The straight line marked with a lower triangle represents the situation where the first layer distance is 410m and the second layer distance is 380m. The straight line marked with a hexagon represents the situation where the first layer distance is 400m and the second layer distance is 390m.

[0106] Depend on Figure 5(b) It can be seen that when the average signal-to-noise ratio of the layers of the drone cluster is different, the fading becomes more severe as the link length increases, and more power is required to meet basic needs. It can be seen that when the sum of the average power gains of multiple layers is equal or approximately equal under different conditions, their instantaneous power gain probability distribution and power allocation scheme are close. However, if the communication conditions of a layer are poor (low average signal-to-noise ratio), the power required for this layer will increase, and the difference in average SNR between the two layers requires more power to be allocated. Therefore, it is shown that the present invention can meet the situation of power resource allocation under a multi-layer drone cluster.

[0107] Simulation experiment 5 of the present invention adopts the present invention and a prior art to obtain the channel characteristics of the composite channel using the input drone parameters and wireless channel parameters to obtain the probability density distribution result of the instantaneous channel gain.

[0108] The parameters of the drone in the simulation experiment 5 of the present invention include the number of drone layers P=1, the number of drones per layer K=1, and the number of intelligent reflection units is set to N=1-3. The wireless channel parameters include the signal frequency f=30GHz of the ITU-R rainfall fading model, D rain = 5km, different rain speed R = 100mm / h, the polarization tilt angle of the drone relative to the horizontal plane τ = 0, the path elevation angle θ = 45°, and the distance between the transmitter and the receiver of each layer of drones d = L DU =L VD =400m, α=0.3, The multipath fading coefficient m=5, and the shadow fading coefficient ms=5.

[0109] A prior art reconfigurable reflective surface RIS used in simulation experiment 5 is:

[0110] Cui Tiejun et al. proposed the concepts of digitally coded metamaterials and programmable metamaterials in “Coding metamaterials, digital metamaterials and programmable metamaterials, Light-Sci. Appl., vol. 3, pp. 1-9, Oct. 2014”, referred to as reconfigurable reflective surfaces RIS.

[0111] Combine the following Figure 6 The simulation diagram of the present invention is further described.

[0112] Figure 6Using PyCharm2021.1.1 and python 3.6 software platforms and utilizing the method of the present invention, a probability density distribution diagram of the instantaneous channel gain of the signal sent by the ground transmitting terminal to each drone relay and then to the ground receiving end in the simulation experiment 5 of the present invention is drawn when it is forwarded by different smart transmitting surfaces. Figure 6 The x-axis coordinate is the instantaneous power gain of the channel, and the y-axis coordinate is the probability density distribution of the instantaneous channel gain. The straight line marked with dots represents the case where the number of intelligent transmitting units is 1, the straight line marked with upper triangles represents the case where the number of intelligent transmitting units is 2, and the straight line marked with lower triangles represents the case where the number of intelligent transmitting units is 3.

[0113] Depend on Figure 4 It can be seen that as the number of intelligent reflective units increases, the instantaneous signal-to-noise ratio / channel gain probability density function curve obtained by simulation shifts to the right as a whole, indicating that the central value of the instantaneous signal-to-noise ratio of the channel increases. The larger the central value of the power gain, the better the channel condition, and the larger the corresponding channel capacity. Therefore, it is shown that the use of the present invention can increase the user capacity without additionally increasing the transmission power of the drone.

[0114] The above simulation experiments show that: the present invention adopts the reconfigurable intelligent surface RIS technology, installs RIS on each drone of each layer, and utilizes the passive reflection characteristics of RIS to passively reflect the communication signals transmitted by the drone, which can increase the channel capacity of the ground receiving terminal users without additionally increasing the transmission power of the drone, thereby solving the problem of small user capacity in the multi-layer drone cluster resource allocation method. The present invention adopts Fox's H function and the fitted Fisher-Snedecor F channel model to solve the multi-layer drone cluster channel characteristics obtained by the communication between the drone and the terminal using the pilot signal, and generates the probability distribution of the instantaneous channel power gain of the multi-layer drone cluster based on the RIS model, so that it can fit the channel model of the multi-layer drone cluster in the harsh environment, and solves the problem that the multi-layer drone channel model based on the RIS model has a small application range. The present invention substitutes the probability distribution of the instantaneous channel power gain of the multi-layer drone cluster into the power optimization model to obtain a power allocation scheme that maximizes the user capacity, which solves the problem that the existing power allocation method cannot be used in harsh environments, and is a very practical multi-layer drone cluster resource allocation method.

Claims

1. A method for allocating drone resources in a multi-layer drone cluster based on the RIS model, characterized in that: A reconfigurable intelligent surface RIS is installed on each drone of each layer of the multi-layer drone cluster network to generate a probability density function of the instantaneous channel power gain of the overall channel; the steps of the allocation method include the following: Step 1: Install RIS on each drone in each layer of the multi-layer drone cluster network; Step 2: Obtain the channel characteristics of the composite channel: (2a) The ground transmitter sends a pilot signal to each UAV relay in the relay cluster. Each UAV relay transmits the pilot signal back to the ground transmitter. The ground transmitter calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by the ground transmitter to the relay cluster. The relay cluster refers to a multi-layer UAV cluster consisting of multiple UAV relays carrying RIS. (2b) Each UAV relay sends a pilot signal to the ground receiving end, and the ground receiving end transmits the pilot signal back to each UAV relay in the relay cluster. Each UAV relay calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by each UAV relay to the ground receiving end, and fits the shadow fading factor and multipath fading factor of the communication signal sent by each UAV relay to the ground receiving end in the Fisher-Snedecor F channel model; (2c) Each UAV relay sends the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor from the UAV relay end to the ground receiving end to the ground transmitting end. The ground transmitting end uses Fox's H function to obtain the channel characteristics H of the overall channel from the ground transmitting end to the relay cluster and then to the ground receiving terminal. PDF ; When each drone acts as an aerial relay, the channel characteristic H of the composite channel is PDF It is generated by following the steps below: In the first step, the ground transmitter sends a pilot signal to each UAV relay in the relay cluster. Each UAV relay transmits the pilot signal back to the ground transmitter. The ground transmitter calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by the ground transmitter to the relay cluster. The instantaneous path loss rain attenuation value experienced by the communication signal sent by the ground transmitter V to the kth UAV relay D in the pth layer is expressed as represents the distance from the ground transmitting terminal to the kth UAV relay in the pth layer, α represents the path loss factor of the composite channel, A represents the path loss from the ground transmitting terminal to the kth UAV relay on the pth layer, rVD ,p,k represents the rainfall loss from the ground transmitting terminal to the kth UAV relay in the pth layer. The relay cluster refers to a multi-layer UAV cluster consisting of multiple UAV relays carrying RIS; In the second step, each UAV relay sends a pilot signal to the ground receiving end, and the ground receiving end transmits the pilot signal back to each UAV relay in the relay cluster. Each UAV relay calculates the instantaneous path loss and rain attenuation value experienced by the communication signal sent by each UAV relay to the ground receiving end, and fits the shadow fading factor and multipath fading factor of the communication signal sent by each UAV relay to the ground receiving end in the Fisher-Snedecor F channel model. The instantaneous path loss and rain attenuation value experienced by the communication signal sent by the kth UAV relay on the pth layer to the ground receiving end is expressed as represents the distance from the kth UAV relay in the pth layer to the ground receiving terminal, A represents the path loss from the kth UAV in the pth layer to the ground receiving terminal. rDU ,p,k represents the rainfall loss from the kth UAV in the pth layer to the ground receiving terminal, m p,k,n represents the multipath fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end, ms p,k,n represents the shadow fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end; In the third step, each UAV relay sends the instantaneous path loss rain attenuation value, shadow fading factor and multipath fading factor from the UAV relay end to the ground receiving end to the ground transmitting end. The ground transmitting end uses Fox's H function to obtain the channel characteristics H of the overall channel from the ground transmitting end to the relay cluster and then to the ground receiving terminal. PDF ; H PDF The expression is as follows: in, represents the path loss and rain attenuation experienced by the k-th UAV relay in the p-th layer of the composite channel, represents the average power gain from the nth RIS on the kth UAV relay at the pth layer to the ground receiving terminal, Step 3, generate the probability density distribution f(h) as follows: Wherein, h represents the instantaneous power gain of the signal received by the ground receiving terminal, P represents the total number of layers of the multi-layer UAV cluster, Π represents the multiplication operation, p represents the serial number of the layer number of the multi-layer UAV cluster, K represents the total number of UAVs in the p-th layer, k represents the serial number corresponding to the k-th UAV relay in the p-th layer, N represents the total number of RIS on the k-th UAV relay in the p-th layer, n represents the serial number corresponding to the n-th RIS on the k-th UAV relay in the p-th layer, Γ represents the gamma function evaluation operation, and m p,k,n represents the multipath fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end, ms p,k,n represents the shadow fading factor from the nth RIS on the kth UAV relay at the pth layer to the receiving end; Step 4: Generate a power optimization model as follows: Where P represents the model of the ground transmitting terminal transmission power, represents the power P(h) of the instantaneous signal transmitted by the ground transmitting terminal that maximizes the value of the expression in the brackets, B represents the bandwidth of the signal transmitted by the ground transmitting terminal, h represents the instantaneous power gain of the signal received by the ground receiving terminal, δ represents the root mean square value of the composite channel noise power, st represents the restricted condition of the model of the ground transmitting terminal's transmission power, Represents the average power value of the signal transmitted by the ground transmitting terminal, P hm Indicates the peak value of the power of the signal transmitted by the ground transmitting terminal; Step 5: Solve the power optimization model. The ground transmitter allocates power P(h) to each drone relay as follows: Among them, h0 represents the instantaneous channel gain threshold from the ground transmitter to the relay and then to the ground receiving terminal, and λ represents the ground transmitter according to Defined power allocation parameters.

2. The method for allocating drone resources based on the RIS model in a multi-layer drone cluster according to claim 1 is characterized in that: The multi-layer UAV cluster described in step 1 means that when the ground transmitting terminal is working, each UAV in the UAV cluster acts as a UAV air relay, and when the ground transmitting terminal is not working, each UAV in the UAV cluster acts as a UAV air base station.

Images