A method for optimizing air-ground integrated wireless network coverage based on incomplete information

By building the drone communication network environment model and planar array antenna model, and using reinforcement learning algorithm to design the optimal path, the problem of maximizing the range of sight links and regional spectral efficiency in the drone network is solved, and the drone network is realized maximum coverage and maximum efficiency.

CN118368641BInactive Publication Date: 2024-11-19CHINESE PEOPLES ARMED POLICE FORCE NON-COMMISSIONED OFFICER SCHOOL
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410505053.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-25
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing research on drone wireless networks failed to consider optimal path planning while maximizing the line-of-sight link and maximizing regional spectral efficiency, resulting in the inability to achieve maximum coverage and maximum efficiency of drone networks.

Method used

By constructing the drone downlink communication network environment model, planar array antenna model and regional spectral efficiency model, and using reinforcement learning algorithm to design the optimal path of the drone network under constraints, we ensure that the drone tracking process is carried out by using reinforcement learning algorithms. maximize regional spectral efficiency.

Benefits of technology

Achieve uniformly distributed equipment terminal services in three-dimensional urban areas, maximize the coverage area of ​​the visual link and non-visit link, improve the spectral efficiency and coverage of the drone network, and ensure the maximum signal reception by the equipment terminal. value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118368641B_ABST
    Figure CN118368641B_ABST
Patent Text Reader

Abstract

The present invention discloses an air-ground integrated wireless network coverage optimization method based on incomplete information, comprising the following steps: Step 1: Establishing a millimeter wave UAV downlink communication network environment model with the equipment terminal in the wireless network environment with the UAV as the center; Step 2: Building a planar array antenna model for the communication antenna equipment carried by the UAV based on the communication network environment model and combining the base station of the UAV in the wireless network environment; Step 3: Building a regional spectral efficiency model of the UAV according to the range of wireless network coverage of the planar array antenna model; Step 4: Using the reinforcement learning algorithm to calculate the maximum value of the regional spectral efficiency model during the UAV track process under constraints. The method of the present invention mainly determines the optimal method of wireless network coverage of the UAV through four steps, and can provide services for multiple outdoor equipment terminals evenly distributed, thereby obtaining higher spectral efficiency and a larger wireless communication network coverage area.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of unmanned aerial vehicle wireless network coverage, and in particular relates to an air-ground integrated wireless network coverage optimization method based on incomplete information. Background Art

[0002] In recent years, with the continuous development of technologies such as intelligent collaborative control and intelligent allocation of communication spectrum resources, the application of drones has gradually expanded from agriculture, firefighting, surveying and mapping, remote sensing, logistics and other fields. With their flexibility, high maneuverability and strong adaptability, drones can perform rescue, exploration and surveying in complex environments, thereby greatly improving productivity and efficiency and providing people with a safer working environment and life guarantee.

[0003] The continuous expansion of smart devices and data traffic in cities will lead to an increase in data demand. Due to the blocking effect of building obstacles, the traditional base station deployment and the line-of-sight link (LoS) information transmission of the device terminal (DT) will be severely limited. The flexibility and mobility of the drone base station (BS) can solve such problems. Compared with ground communication, it has less interference and higher transmission reliability, and can achieve high-speed communication transmission.

[0004] UAVs can be quickly deployed to areas where communication is required, are not restricted by the ground environment, and have strong flexibility and adaptability. In an emergency, a communication network can be quickly built to provide timely communication services; compared with traditional ground communication infrastructure, the manufacturing cost and maintenance cost of UAVs are lower, which can greatly reduce the communication cost; UAVs can fly in the air, have a higher communication coverage, can cover a wider area, and provide comprehensive communication services; multiple UAVs can form a UAV cluster network, and through distributed algorithms, the network can be self-repaired and expanded, and the robustness and scalability of the network can be improved. UAV communication has the advantages of high speed, rapid deployment, flexibility, low cost, wide coverage and strong scalability. However, existing research has not considered the maximization of regional spectral efficiency (ASE) and the maximum coverage of line-of-sight link (LoS) and optimal path planning together. Therefore, the purpose of the present invention is to use UAVs to achieve the maximum regional spectral efficiency ASE under the constraints of maximizing LoS coverage and optimal path. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention invents an air-ground integrated wireless network coverage optimization method based on incomplete information.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A method for optimizing air-ground integrated wireless network coverage based on incomplete information comprises the following steps:

[0008] Step 1: Establish a millimeter wave UAV downlink communication network environment model with the UAV as the center and the equipment terminals in the wireless network environment;

[0009] Step 2: Based on the communication network environment model and the base station of the UAV in the wireless network environment, a planar array antenna model is constructed for the communication antenna equipment carried by the UAV;

[0010] Step 3: Construct a regional spectral efficiency model of the UAV based on the wireless network coverage range of the planar array antenna model;

[0011] Step 4: Under constraints, use the reinforcement learning algorithm to design an optimal working path for the UAV network to maximize the regional spectral efficiency during the UAV trajectory.

[0012] The optimization method of the present invention mainly determines the optimal method for wireless network coverage of drones through four steps, specifically, first constructing a wireless communication network environment model, constructing a planar array antenna model in the wireless communication network environment model, and then constructing a drone regional spectral efficiency model, and finally obtaining the optimal working path of the drone network under constraints so that the regional spectral efficiency of the drone during the flight path is maximized.

[0013] Furthermore, the communication network environment model is a downlink signal transmission network of a UAV at a fixed height h, which is affected by line-of-sight links or non-line-of-sight links, where the number of line-of-sight links is recorded as L and the number of non-line-of-sight links is recorded as N;

[0014] Set the drone's base station to serve a size of The target rectangular area has Subgrid, As an outdoor subgrid set;

[0015] The base station of the drone is enabled to hover or land at an appropriate location, in the set Q = {1, 2, ..., q, ..., Q}, the hovering point q ∈ Q, to achieve the maximum coverage area of ​​potential device terminals, which is constrained by the flight time and hovering time slot of the drone through each sub-grid, and the constraints are:

[0016]

[0017] in is the side length of each sub-grid, is the area of ​​each subgrid; t k is the flight time of the UAV in the kth subgrid, t q For the hover time slot, is the number of bits requested, is the number of line-of-sight links and non-line-of-sight links to the device terminals to be served by the UAV from the Qth subgrid, is the achievable rate between the non-aviation vehicle from the Qth subgrid and the device terminal in the kth subgrid, v U is the uniform motion speed of the UAV, where o∈{L,N}.

[0018] Furthermore, the drone serves one device terminal at a time based on the proximity principle within the wireless network coverage area.

[0019] Furthermore, the planar array antenna model consists of each UAV antenna having N 2 The square uniform planar array of units is composed of , where the transmitting array response vector a(γ t ,ω t ) are the products of an array response vector along the x-axis and y-axis respectively:

[0020]

[0021] Among them, γ t is the azimuth of the UAV, γ t ∈(-π,π],ω t is the elevation angle of the drone, ω t ∈(-π,π]; is the product;

[0022] The a x (γ t ,ω t ) and a y (γ t ,ω t ) is calculated as:

[0023]

[0024] Among them, the distance between all antennas in the x-axis and y-axis directions is d, and the wavelength of the carrier is ι∈[x,y]; is the standardized transmit antenna position, n x and n y All are integers in [0, N-1].

[0025] Furthermore, when the beamforming vector of the planar array antenna model is , the UAV antenna gain expression is:

[0026]

[0027] Among them, the azimuth and elevation angles of the UAV are φ s and s ,φ s =γ t,ξ s =ω t .

[0028] Furthermore, the regional spectrum effect provided by the base station enabled by the drone is:

[0029]

[0030] in, σ 2 is the noise power with weak interference, o∈{L,N}; and The number of UAVs in the line-of-sight link and non-line-of-sight link from the qth hovering point, respectively. It is determined by the radius r of the drone’s wireless network coverage. L is a line-of-sight link, and N is a non-line-of-sight link;

[0031] Said is the received signal from the drone from hovering to the service equipment terminal, and the expression of the received signal is:

[0032]

[0033] in, is the distance between the hovering drone above the qth subgrid and the device terminal in the kth grid, α is the path loss index, β is the signal transmission frequency, and P U Transmit power for the drone.

[0034] Furthermore, the constraints of the reinforcement learning are:

[0035] C1:min|ΦP|

[0036] C2:

[0037] C3:

[0038] C4:

[0039] Among them, ΦP represents the set path sub-grid, C1 is the minimum path connecting the hovering position of Q, C2 is that all device terminals need to be served, C3 is the maximum time constraint, and C4 is the h and r related constraints.

[0040] The advantages of the present invention are: 1. The optimization method of the present invention can move in a specific three-dimensional urban area through the wireless communication network of the unmanned aerial vehicle, and provide services for multiple evenly distributed outdoor equipment terminals; the maximum spectral efficiency of the network coverage, that is, the maximum value of the signal received by the device terminal, is determined by maximizing the coverage area of ​​the line-of-sight link and non-line-of-sight link of the wireless communication network; according to the optimization method of the present invention, the unmanned aerial vehicle can detect the location of a higher line-of-sight link coverage area, and combine the maximization of regional spectral efficiency, the optimal path and the maximum wireless communication network coverage range to detect the optimal path for the unmanned aerial vehicle network operation, so that the regional spectral efficiency during the unmanned aerial vehicle track is maximized. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the main flow chart of the air-ground integrated wireless network coverage optimization method based on incomplete information;

[0042] Figure 2 It is a propagation scenario diagram of stereo line-of-sight link and non-line-of-sight link;

[0043] Figure 3 It is a schematic diagram of a uniform planar array antenna. DETAILED DESCRIPTION

[0044] A method for optimizing air-ground integrated wireless network coverage based on incomplete information comprises the following steps:

[0045] Step 1: Taking the UAV as the center and cooperating with the equipment terminals in the wireless network environment, a millimeter wave UAV downlink communication network environment model is established; specifically, the downlink signal transmission network of the UAV at a fixed height h is affected by the line-of-sight link or the non-line-of-sight link. The number of line-of-sight links is recorded as L, and the number of non-line-of-sight links is recorded as N;

[0046] Set the drone's base station (BS) to serve a The target rectangular area has Subgrid, As a set of outdoor subgrids; there is only one device terminal in each subgrid;

[0047] The base station of the drone is enabled to hover or land at an appropriate location, in the set Q = {1, 2, ..., q, ..., Q}, the hovering point q ∈ Q, to achieve the maximum coverage area of ​​potential device terminals, which is constrained by the flight time and hovering time slot of the drone through each sub-grid, and the constraints are:

[0048]

[0049] in is the side length of each sub-grid, is the area of ​​each subgrid; t k is the flight time of the UAV in the kth subgrid, t q For the hover time slot, is the number of bits requested, is the number of line-of-sight links and non-line-of-sight links to the device terminals to be served by the UAV from the Qth subgrid, is the achievable rate between the non-aviation vehicle from the Qth subgrid and the device terminal in the kth subgrid, v U is the uniform motion speed of the UAV (ignoring acceleration and deceleration), where o∈{L,N}.

[0050] The drone serves one device terminal at a time within the wireless network coverage area based on the proximity principle, that is, when the drone determines the optimal position, it will provide the ground device terminal with the wireless communication network service with the highest coverage based on the proximity principle according to the location of the ground device terminal or user needs.

[0051] Step 2: Based on the communication network environment model and the base station of the UAV in the wireless network environment, a planar array antenna model is constructed for the communication antenna equipment carried by the UAV; the planar array antenna model consists of each UAV antenna with N 2 The square uniform planar array of units is composed of , where the transmitting array response vector a(γ t ,ω t ) are the products of an array response vector along the x-axis and y-axis respectively:

[0052]

[0053] Among them, γ t is the azimuth of the UAV, γ t ∈(-π,π],ω t is the elevation angle of the drone, ω t ∈(-π,π]; is the product;

[0054] The a x (γ t ,ω t ) and a y (γ t ,ω t ) is calculated as:

[0055]

[0056] Among them, the distance between all antennas in the x-axis and y-axis directions is d, and the wavelength of the carrier is ι∈[x,y]; is the standardized transmit antenna position, n x and ny All are integers in [0, N-1].

[0057] When the beam forming vector of the planar array antenna model, the UAV antenna gain expression is:

[0058]

[0059] Among them, the azimuth and elevation angles of the UAV are φ s and s ,φ s =γ t ,ξ s =ω t .

[0060] Step 3: Construct a regional spectral efficiency model of the drone based on the wireless network coverage range of the planar array antenna model; the regional spectral effect provided by the base station enabled by the drone is:

[0061]

[0062] in, σ 2 is the noise power with weak interference, o∈{L,N}; and The number of UAVs in the line-of-sight link and non-line-of-sight link from the qth hovering point, respectively. It is determined by the radius r of the drone’s wireless network coverage.

[0063] Said is the received signal from the drone from hovering to the service equipment terminal, and the expression of the received signal is:

[0064]

[0065] in, is the distance between the hovering drone above the qth subgrid and the device terminal in the kth grid, α is the path loss index, β is the signal transmission frequency, and P U Transmit power for the drone.

[0066] Step 4: Under constraints, use the reinforcement learning algorithm to design an optimal working path for the UAV network to maximize the regional spectral efficiency during the UAV trajectory.

[0067] The constraints of the reinforcement learning are:

[0068] C1:min|ΦP|

[0069] C2:

[0070] C3:

[0071] C4:

[0072] Among them, ΦP represents the set path sub-grid, C1 is the minimum path connecting the hovering position of Q, C2 is that all device terminals need to be served, C3 is the maximum time constraint, C4 is the h and r related constraints, and P is the total degree of the grid passed by the drone.

[0073] The specific steps of the reinforcement learning method are:

[0074] Step 4-1: The drone first hovers at a predetermined position based on a predetermined process with a certain device terminal as the reference point and calculates the regional spectral efficiency of this position according to the constraint conditions; this step is set as the drone execution action a and the observation state s;

[0075] Step 4-2: Adjust the drone’s position, perform action a, transfer to the next state s’, and obtain reward g;

[0076] Step 4-3: Store (s, a, s', g) into the experience pool;

[0077] Step 4-4: Randomly sample data from the experience pool and update the wireless communication network;

[0078] Step 4-5: Compare the values ​​of the regional spectral efficiency of each drone observation state to determine whether s is a terminal state. If not, repeat steps 4-1 to 4-4. If it is a terminal state, proceed to the next step;

[0079] Step 4-6: If s is the terminal state, determine whether the number of iterations meets the requirement. If not, reset the agent state and repeat steps 4-1 to 4-6. If satisfied, proceed to the next step.

[0080] Step 4-7: Output the optimal working path of the wireless communication network.

Claims

1. A method for optimizing air-ground integrated wireless network coverage based on incomplete information, characterized in that: The steps include: Step 1: Establish a millimeter wave UAV downlink communication network environment model with the UAV as the center and the equipment terminals in the wireless network environment; The communication network environment model is a downlink signal transmission network of a UAV at a fixed height h, which is affected by line-of-sight links or non-line-of-sight links. The number of line-of-sight links is recorded as L, and the number of non-line-of-sight links is recorded as N. Set the drone's base station to serve a size of The target rectangular area has Subgrid, As an outdoor subgrid set; The base station of the drone is enabled to hover or land at an appropriate location, in the set Q = {1, 2, ..., q, ..., Q}, the hovering point q ∈ Q, to achieve the maximum coverage area of ​​potential device terminals, which is constrained by the flight time and hovering time slot of the drone through each sub-grid, and the constraints are: in is the side length of each sub-grid, is the area of ​​each subgrid; t k is the flight time of the UAV in the kth subgrid, t q is the hover slot, Q is the number of bits requested, is the number of line-of-sight links and non-line-of-sight links to the device terminals to be served by the UAV from the Qth subgrid, is the achievable rate between the non-aviation vehicle from the Qth subgrid and the device terminal in the kth subgrid, v U is the uniform motion speed of the UAV, where o∈{L,N}; The drone serves one device terminal at a time within the wireless network coverage area based on the principle of proximity; Step 2: Based on the communication network environment model and the base station of the UAV in the wireless network environment, a planar array antenna model is constructed for the communication antenna equipment carried by the UAV; The planar array antenna model consists of each UAV antenna with N 2 The square uniform planar array of units is composed of , where the transmitting array response vector a(γ t ,ω t ) are the products of an array response vector along the x-axis and y-axis respectively: Among them, γ t is the azimuth of the UAV, γ t ∈(-π,π],ω t is the elevation angle of the drone, ω t ∈(-π,π]; is the product; The a x (γ t ,ω t ) and a y (γ t ,ω t ) is calculated as: Among them, the distance between all antennas in the x-axis and y-axis directions is d, and the wavelength of the carrier is ι∈[x,y]; is the standardized transmit antenna position, n x and n y All are integers in [0, N-1]; When the beamforming vector of the planar array antenna model is , the UAV antenna gain expression is: Among them, the azimuth and elevation angles of the UAV are φ s and s ,φ s =γ t ,ξ s =ω t ; Step 3: Construct a regional spectral efficiency model of the UAV based on the wireless network coverage range of the planar array antenna model; The regional spectrum effect provided by the drone-enabled base station is: in, σ 2 is the noise power with weak interference, o∈{L,N}; and The number of UAVs in the line-of-sight link and non-line-of-sight link from the qth hovering point, respectively. It is determined by the radius r of the drone’s wireless network coverage. L is a line-of-sight link, and N is a non-line-of-sight link; Said is the received signal from the drone from hovering to the service equipment terminal, and the expression of the received signal is: in, is the distance between the hovering drone above the qth subgrid and the device terminal in the kth grid, α is the path loss index, β is the signal transmission frequency, and P U Transmit power for the drone; Step 4: Under constraints, use the reinforcement learning algorithm to design an optimal working path for the UAV network to maximize the regional spectral efficiency during the UAV trajectory.

2. The method for optimizing air-ground integrated wireless network coverage based on incomplete information according to claim 1, characterized in that: The constraints of the reinforcement learning are: C1:min|ΦP| Among them, ΦP represents the set path sub-grid, C1 is the minimum path connecting the hovering position of Q, C2 is that all device terminals need to be served, C3 is the maximum time constraint, and C4 is the h and r related constraints.

Citation Information

Patent Citations

  • Unmanned aerial vehicle heterogeneous network energy efficiency optimization method based on deep reinforcement learning

    CN114189891A

  • Robust beam forming design method under non-ideal channel in intelligent reflector-assisted communication

    CN114268350A

  • Network deployment method with maximum communication service coverage rate

    CN116033437A

  • Low-altitude network coverage optimization method based on digital twinning

    CN117320024A

  • System architecture and method for enhancing wireless networks with mini-satellites and pseudollites and adaptive antenna processing

    US10348394B1