A capacity analysis method for three-layer heterogeneous networks based on multi-hop drone-assisted communication

Abstract

A capacity analysis method for three-layer heterogeneous networks based on multi-hop UAV-assisted communication is proposed. A three-layer heterogeneous network capacity analysis model consisting of a terminal access layer, a UAV relay layer, and a backhaul link layer is established. In the model, the terminal is connected to an available base station through the relay function of the UAV. Firstly, the ground heterogeneous terminal is modeled as a Poisson point process using the random geometry method. The ergodic capacity closed expression of the terminal access layer using NOMA is analyzed using the properties of the Poisson point process. Then, by introducing a frequency hopping mechanism that can avoid interference between UAV clusters, a resource allocation scheme based on the relay UAV rate satisfaction function is proposed, and the relay link layer capacity of multi-hop transmission is analyzed. Finally, in the backhaul link, the communication between the backhaul UAV and the BS adopts a model based on independent non-orthogonal beam communication supporting SDMA, and the capacity closed expression of the backhaul link is derived.

Description

Technical Field

[0001] The invention relates to the field of communication network technology, and specifically to a three-layer heterogeneous network capacity analysis method based on multi-hop drone-assisted communication. Background Art

[0002] The future 6G will be a network that integrates "air, space, land and sea", "distributed self-organization" and "resources allocated on demand", and can provide support for perception, connection, computing and other needs in disaster and emergency scenarios. When a small disaster occurs in a local area, some infrastructure such as macro cellular networks will experience functional decline or even devastating blows and cannot meet the communication needs of the disaster area. Compared with large-scale natural disasters, small disasters in local areas are more common and difficult to predict, and the economic losses recovered by timely rescue are considerable. Therefore, in this case, the uploading of the victims' distress signals and survival conditions and environmental measurement information of IoT devices is crucial. Mobile information networks can meet the communication needs of disaster areas by timely building a reliable and high-performance emergency communication network.

[0003] Drones are now widely used as relay base stations in emergency scenarios due to their flexible mobility. For emergency communication scenarios with drone relays, existing research has designed optimization methods to find the optimal deployment location of drones to achieve optimal communication performance; designed an optimization model for uplink throughput to jointly optimize multiple factors to maximize network throughput; used D2D technology to analyze the delay performance under multi-hop information transmission; and used non-orthogonal multiple access technology to improve the spectrum efficiency of a single drone in emergency communications.

[0004] However, existing research has always allocated resources based on the node's benefit in improving network throughput when optimizing network throughput. In a multi-hop uplink scenario, communication resources are mostly concentrated in high-level nodes close to the destination, while low-level nodes have low benefits in improving network throughput due to relay transmission, and are allocated fewer communication resources. In addition, current research has not established a capacity analysis model for three-layer heterogeneous networks. Summary of the invention

[0005] The purpose of the present invention is to solve the problem of emergency communication capacity assessment in a scenario where base station functions are impaired, and to solve the problem of solving the closed expression of the ergodic capacity of the terminal access layer under non-orthogonal multiple access by means of random geometry methods, to solve the problem of multi-hop UAV relay layer capacity analysis by designing optimization problems, and to solve the problem of solving the closed expression of the capacity of the return link layer by analyzing the return link using independent beams.

[0006] A three-layer heterogeneous network capacity analysis method based on multi-hop UAV-assisted communication, the steps of the method are as follows:

[0007] Step 1: For a three-layer heterogeneous communication system including terminal access layer, drone relay layer, and backhaul link layer, the ergodic capacity of the terminal access layer is calculated based on the distribution density of heterogeneous terminals, the flight altitude of drones, access threshold and signal processing capability, Gaussian white noise spectral density, terminal access layer bandwidth, and the channel coefficient and fading factor of the Ruili fading channel.

[0008] Step 2: According to the frequency hopping sequence of the drone cluster, the position relationship and routing information of the drones, the channel coefficient and fading factor under the Ruili fading channel, initialize the following parameters: average persistence coefficient N p , the bandwidth of each frequency hopping subchannel Its proportion in the frequency hopping sequence τ k , the average achievable rate of each hop link

[0009] Step 3, based on the initialized parameters, the satisfaction function is used as the objective function, the amount of communication resources occupied by each information flow is used as the decision variable, and the restriction conditions of the working mechanism are used as constraints to form an optimization problem;

[0010] Step 4: Convert the optimization problem into a linear programming problem according to the constraints between decision variables. The optimal solution of this problem is the resource allocation when the satisfaction is the highest, and the capacity of the drone relay layer is obtained.

[0011] Step 5: Calculate the interference gain based on the positional relationship between multiple backhaul drones and the base station and the beam pattern, and then obtain a closed-form expression; substitute the interference gain, the backhaul link layer bandwidth, and the channel coefficient and fading factor under the Ruili fading channel to obtain the capacity of the backhaul link:

[0012] Step 6, compare the obtained three-layer network capacities, and the network capacity of the smallest layer is the network capacity of the entire system.

[0013] Furthermore, in the three-layer heterogeneous communication system, the ground terminal transmits the signal back to the available base station through multiple relay drones, realizing emergency communication in disaster scenarios; represents the capacity of the ground terminal access layer, that is, the sum of the ergodic capacities of all relay UAVs communicating with multiple ground terminals using non-orthogonal multiple access, and R represents the set of relay UAVs; represents the capacity of the drone relay layer, which is the sum of the multi-hop throughput rates of all relay drones and the return drones in their cluster; It represents the capacity of the backhaul link layer, which is the sum of the backhaul link capacities of all backhaul UAVs communicating with the base station.

[0014] Furthermore, in step 1, the distribution of ground heterogeneous terminals is modeled as a Poisson point process using the random geometry method. The properties of the Poisson point process are used to derive the coverage probability P of the terminal with the kth distance ranking from the drone under non-orthogonal multiple access communication. k (τ) and ergodic capacity R k The mathematical expectation of ; the specific closed expression is as follows:

[0015]

[0016]

[0017] The network capacity of the terminal access layer is expressed as

[0018]

[0019] Furthermore, in step 2, the drone relay layer includes relay drones for communicating between ground terminals and aerial drones, and return drones for communicating between aerial drones and ground base stations; a return drone acts as a cluster head to form a multi-hop connected drone cluster with multiple relay drones, and frequency hopping sequences and CSMA / CA are used within the cluster to avoid inter-cluster and intra-cluster interference.

[0020] Furthermore, in step 2, a variable with three parameters is set in combination with the time-division resource utilization mode of the system: Indicates drone U i The information flow per unit time in link L j Upper occupied channel F k Normalized ratio of communication resources; compare it with the corresponding link and the achievable rate under the Rayleigh fading channel Combined, the transmission rate of the information flow on the hop link is obtained.

[0021] Furthermore, in step 3, combined with the constraints generated by the intra-cluster hierarchical networking, the inter-cluster frequency hopping mechanism, and the single-transmit-single-receive full-duplex working mechanism of the drone, an optimization problem about the intra-cluster throughput is formed. The upper bound of the intra-cluster throughput is the capacity. The optimization problem is as follows:

[0022]

[0023]

[0024]

[0025]

[0026]

[0027] Furthermore, in step 3, a function representing the rate satisfaction of the relay drone is set as the objective function to increase the weight of the relay drone with a high hop count when allocating resources, so that the drone relay layer resources are not concentrated in the relay drone close to the return drone, and the resources are evenly allocated; the satisfaction function is expressed as follows:

[0028]

[0029] Where a>0.

[0030] Furthermore, in step 4, the optimization problem formed in step 3 is converted into a linear programming problem, which has an optimal solution, denoted as This is the network capacity of the drone relay layer.

[0031] Furthermore, in step 5, a multi-access array antenna is deployed at the base station to achieve simultaneous communication with multiple backhaul drones that transmit independent beams; the transmit and receive beam patterns are derived as The interference gain between return UAVs at the time:

[0032]

[0033] This will get you a return drone The closed-form expression of the capacity under the Rayleigh fading channel is:

[0034]

[0035] The network capacity of the backhaul link layer is expressed as:

[0036]

[0037] Furthermore, in step 6, the capacity expression of the three-layer heterogeneous network is given:

[0038]

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] (1) The frequency hopping mechanism is applied to avoid interference; (2) The capacity analysis method of the present invention considers the influencing factors more comprehensively, and the analysis result is more in line with the network characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of a three-layer heterogeneous network structure in an embodiment of the present invention.

[0042] Figure 2 It is a timing resource diagram of two drone clusters under a specific frequency hopping sequence in an embodiment of the present invention.

[0043] Figure 3 It is a schematic diagram of the return link layer beam in an embodiment of the present invention.

[0044] Figure 4 It is a schematic table of parameter settings and physical meanings in the embodiments of the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0046] The present invention Figure 1 The three-layer heterogeneous communication system under emergency communication shown in the figure is taken as an example. In this scenario, some base stations are damaged due to accidents and cannot provide communication services. The terminals in the affected area first communicate with the relay drones through non-orthogonal multiple access. After decoding, the relay drone forwards the signal through multiple relay drones and transmits the signal to the cluster head backhaul drone. The backhaul drone uses an array antenna to generate an independent beam to backhaul the signal to the base station. The problem to be solved is the communication capacity provided by the communication system, which is decomposed into three sub-problems: solving the capacity of the terminal access layer, the drone relay layer, and the backhaul link layer respectively, where the minimum value is the system capacity. The terminal access layer contains all ground terminals. The drone relay layer contains multiple clusters of drones, and the drone cluster consists of a backhaul drone and multiple relay drones. In the backhaul link layer, the backhaul drone backhauls the signal to a base station with sound functions.

[0047] Based on the model diagram, the present invention proposes an analysis method for the capacity of a three-layer heterogeneous network based on multi-hop drone-assisted communication, and constructs a three-layer heterogeneous communication system for scenarios where local disasters cause base stations to be unable to provide services to terminals. In the terminal access layer, the terminal modeled as a Poisson point process distribution communicates with the relay drone through non-orthogonal multiple access, and the closed expression of the ergodic capacity of the terminal access layer is analyzed. In the drone relay layer, a resource allocation scheme is proposed to achieve balanced resource allocation of the drone relay layer. According to the constraints generated by interference avoidance and other working mechanisms, an optimization problem is designed to solve the upper bound of the throughput of the drone relay layer, and the solution obtained is the capacity of the drone relay layer. In the return link layer, based on a model of independent non-orthogonal beam communication supporting space division multiple access, a closed expression for the capacity of the return link is derived. The variables involved in this method are as follows: Figure 4 As shown, the method comprises the following steps:

[0048] Step 1: Derive the closed-form expression of the ergodic capacity of the terminal access layer. According to the parameters required by the expression: the distribution density of heterogeneous terminals, the flight altitude of the drone, the access threshold and signal processing capability, the Gaussian white noise spectrum density, the bandwidth of the terminal access layer, the channel coefficient and fading factor of the Ruili channel, the ergodic capacity of the terminal access layer is calculated.

[0049] First, according to the distribution density of ground terminals, the probability density function of the distance between the ground terminal with the kth position from the drone and the drone is derived:

[0050]

[0051] Then, by analyzing the access link signal interference and noise ratio, the access link signal interference and noise ratio of the terminal with the kth position under perfect decoding non-orthogonal multiple access is obtained:

[0052]

[0053] Among them, SINR k The Laplace transform of the interference part in the denominator can be written as:

[0054]

[0055] Therefore, using the exponential distribution characteristics of the Rayli fading channel coefficient, the coverage probability and ergodic capacity of the terminal with the rank k are:

[0056]

[0057]

[0058] Combined with the signal processing capabilities of drones ps Reunion with Relay Drone The number of drones in the terminal access layer can be obtained:

[0059]

[0060] Step 2: According to the frequency hopping sequence of the drone cluster, the location relationship and routing information of the drones, the channel coefficient and the fading factor, initialize the parameters: the average persistence coefficient N p , the bandwidth of each frequency hopping subchannel Its proportion in the frequency hopping sequence τ k , the average achievable rate of each hop link

[0061] Interference avoidance mechanism: In order to avoid interference between adjacent drone clusters, a frequency hopping mechanism is used in the drone relay layer. Assume that frequency band 2 is divided into N r subchannels, then the bandwidth of the kth subchannel is defined as There are multiple drone relay clusters in the drone relay layer. A backhaul drone serves as the cluster head and forms a relay cluster with multiple relay drones. The relay drones are connected to the backhaul drones through multi-hop links. A mutually orthogonal frequency hopping sequence is planned for each adjacent cluster so that multiple adjacent clusters use orthogonal sub-channels to communicate at the same time, avoiding interference between adjacent drone clusters. In order to avoid interference within the drone cluster, a carrier sense and avoidance mechanism is adopted, and the backoff delay in carrier sense avoidance is ignored. Therefore, under the two interference avoidance mechanisms, which subchannel is used in a certain time slot within the cluster is determined by the above-mentioned orthogonal frequency hopping sequence. Figure 2 As shown, the sequence of cluster A and cluster B consists of 5 subchannels, where the drones transmit on orthogonal subchannels at the same time, avoiding mutual interference. In cluster A, only one drone (for example, U1) starts transmitting on subchannel 2, while all other drones in cluster A are waiting in the queue for transmission due to CSMA / CA. After one time slot, cluster A will switch to the next subchannel for a new transmission, and another drone (for example, U2) can transmit on this subchannel, thus avoiding intra-cluster interference. If U1 has not completed the transmission, it can continue to transmit for a while. Since subchannels 2 and 4 are orthogonal, there will still be no intra-cluster interference at this time. The ratio of transmission time to time slot is defined as the average duration coefficient N p . N p The higher the value, the stronger the concurrent transmission capability of the network and the higher the network capacity. p Unlimited increase will lead to inter-cluster and intra-cluster interference. This depends on the design of the frequency hopping sequence. The average achievable rate of the link under the Rayleigh fading channel can be expressed as With τ k Determined by the frequency hopping sequence.

[0062] Step 3: Based on the initialization parameters, the satisfaction function is used as the objective function, the amount of communication resources occupied by each information flow is used as the decision variable, and the restrictions of various working mechanisms are used as constraints to form an optimization problem.

[0063] Combined with the system's time-sharing resource utilization, design a variable with three parameters Indicates drone U i The information flow per unit time in link L j Upper occupied channel F k The normalized ratio of communication resources. Compare it with the achievable rate on the corresponding link and channel. Combined with, the transmission rate of the information flow on the hop link can be obtained

[0064] Combining the constraints of intra-cluster hierarchical networking, inter-cluster frequency hopping mechanism, and the single-transmit-single-receive full-duplex working mechanism of drones, an optimization problem about intra-cluster throughput is formed, and the upper bound of intra-cluster throughput is capacity. The optimization problem is as follows:

[0065]

[0066]

[0067]

[0068]

[0069]

[0070] In the formula, (1.1) is the satisfaction function, and the ratio of the throughput rate of the drone in the relay link to the rate required by the user is taken as the independent variable of the function. The higher the ratio, the more satisfied the user is. We hope that each user has a balanced satisfaction level, so the characteristics of the satisfaction function should be monotonically increasing and the first-order derivative function decreases sharply, that is, Information flow constraints for sharing a link. Each drone not only needs to transmit its own information flow upward, but also relay the information flow of the drones in the next layer. There are multiple information flows on each link. The time slot length of the resource block occupied by multiple information flows sharing a transmitter per unit time is less than the unit time. (1.3) is the resource block constraint within a single channel. For the frequency hopping mechanism, the channel resources of the drones in the cluster are determined by the frequency hopping sequence. The channel F occupied by each information flow k The sum of x does not exceed channel F k The ratio of the frequency hopping sequence. (1.4) is the rate constraint. When analyzing the sum of the throughput rates of a multi-hop network, the information flow rate on each hop must be equal to ensure the maximum sum of the rates. (1.5) constrains the range of the variable x. x is the normalized ratio, so the range is [0,1].

[0071] Step 4: According to the constraints between decision variables, the optimization problem is converted into a linear programming problem, which has an optimal solution. The solution is the resource allocation when the satisfaction is the highest. This is the capacity of the drone relay layer. The problem is solved using the MATLAB toolkit CVX.

[0072] Step 5: Based on the positional relationship between multiple backhaul drones and the base station and the beam pattern, the interference gain is calculated to derive a closed-form expression. The capacity of the backhaul link can be calculated by substituting the interference gain, the bandwidth of the backhaul link layer, the channel coefficient and the fading factor into the expression.

[0073] Depend on Figure 3 Backhaul link layer beam diagram, the transmit and receive beam patterns are derived as The interference gain between return UAVs at the time:

[0074]

[0075] The interference gain is related to the beam overlap range. Based on this, the closed-form expression of the capacity when a base station with multiple access array antennas communicates with multiple return UAVs transmitting independent beams at the same time under the Ruili fading channel is derived:

[0076]

[0077] Step 6: Compare the network capacities of the three layers obtained. The network capacity of the smallest layer is the network capacity of the entire system. The network capacity of the three-layer network is expressed as:

[0078]

[0079] The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiment. Any equivalent modifications or changes made by ordinary technicians in this field based on the contents disclosed by the present invention should be included in the protection scope recorded in the claims.

Claims

1. A three-layer heterogeneous network capacity analysis method based on multi-hop UAV-assisted communication, characterized by: The method comprises the following steps: Step 1: For a three-layer heterogeneous communication system including terminal access layer, drone relay layer, and backhaul link layer, the ergodic capacity of the terminal access layer is calculated based on the distribution density of heterogeneous terminals, the flight altitude of drones, access threshold and signal processing capability, Gaussian white noise spectral density, terminal access layer bandwidth, and the channel coefficient and fading factor of the Ruili fading channel. Step 2: According to the frequency hopping sequence of the drone cluster, the position relationship and routing information of the drones, the channel coefficient and fading factor under the Ruili fading channel, initialize the following parameters: average persistence coefficient N p , the bandwidth of each frequency hopping subchannel Its proportion in the frequency hopping sequence τ k , the average achievable rate of each hop link In step 2, the drone relay layer includes relay drones for communicating between ground terminals and aerial drones, and return drones for communicating between aerial drones and ground base stations. A return drone acts as a cluster head and forms a multi-hop connected drone cluster with multiple relay drones. Frequency hopping sequences and CSMA / CA are used within the cluster to avoid inter-cluster and intra-cluster interference. Step 3, based on the initialized parameters, the satisfaction function is used as the objective function, the amount of communication resources occupied by each information flow is used as the decision variable, and the restriction conditions of the working mechanism are used as constraints to form an optimization problem; Step 4: Convert the optimization problem into a linear programming problem according to the constraints between decision variables. The optimal solution of this problem is the resource allocation when the satisfaction is the highest, and the capacity of the drone relay layer is obtained. Step 5: Calculate the interference gain according to the position relationship between multiple backhaul UAVs and the base station and the beam pattern, and then obtain a closed expression; Substituting the interference gain, the backhaul link layer bandwidth, and the channel coefficient and fading factor under the Rayleigh fading channel, the capacity of the backhaul link is obtained. Step 6: Compare the network capacities of the three layers obtained, and take the smallest network capacity among the three as the network capacity of the entire system.

2. According to claim 1, a three-layer heterogeneous network capacity analysis method based on multi-hop drone-assisted communication is characterized in that: In the three-layer heterogeneous communication system, the ground terminal transmits the signal back to the available base station through multiple relay drones, realizing emergency communication in disaster scenarios; represents the capacity of the ground terminal access layer, that is, the sum of the ergodic capacities of all relay UAVs communicating with multiple ground terminals using non-orthogonal multiple access, and R represents the set of relay UAVs; represents the capacity of the drone relay layer, which is the sum of the multi-hop throughput rates of all relay drones and the return drones in their cluster; It represents the capacity of the backhaul link layer, which is the sum of the backhaul link capacities of all backhaul UAVs communicating with the base station.

3. According to claim 2, a three-layer heterogeneous network capacity analysis method based on multi-hop drone-assisted communication is characterized in that: In step 1, the distribution of ground heterogeneous terminals is modeled as a Poisson point process using the random geometry method. The properties of the Poisson point process are used to derive the coverage probability P of the terminal with the kth distance ranking from the drone under non-orthogonal multiple access communication. k (τ) and ergodic capacity R k The mathematical expectation of ; the specific closed expression is as follows: Where r 2 is the integral differential element, corresponding to dr; B1 represents the bandwidth of the device access layer. The three layers of the entire model divide all available bandwidth into three parts, namely B1, B2, and B3; The network capacity of the terminal access layer is expressed as follows: In the formula, Representative Set The number of elements in, N ps Represents the number of ground terminal signals processed simultaneously by the drone.

4. According to claim 1, a three-layer heterogeneous network capacity analysis method based on multi-hop drone-assisted communication is characterized in that: In step 2, a variable with three parameters is set according to the system's time-sharing resource utilization method: Indicates drone U i The information flow per unit time in link L j Upper occupied channel F k Normalized ratio of communication resources; compare it with the corresponding link and the achievable rate under the Rayleigh fading channel Combined, the transmission rate of the information flow on the hop link is obtained.

5. According to claim 1, a three-layer heterogeneous network capacity analysis method based on multi-hop drone-assisted communication is characterized in that: In step 3, the constraints generated by the intra-cluster hierarchical networking, the inter-cluster frequency hopping mechanism, and the single-transmit-single-receive full-duplex working mechanism of the drone are combined to form an optimization problem about the intra-cluster throughput. The upper bound of the intra-cluster throughput is the capacity. The optimization problem is as follows: At the drone relay layer, the frequency band B2 is further divided into multiple sub-channels for frequency modulation. represents the set of all subchannels; L j 'With L j Correspondingly, it refers to any two links on the routing path; Represents a collection of links, is the set of all first-hop return drone-relay drone links, Represents a set of links with constraints on the sending and receiving mechanisms; 6. The method for analyzing the capacity of a three-layer heterogeneous network based on multi-hop UAV-assisted communication according to claim 5 is characterized in that: In step 3, a function representing the rate satisfaction of the relay drone is set as the objective function to increase the weight of the relay drone with a high hop count when allocating resources, so that the drone relay layer resources are not concentrated in the relay drone close to the return drone, and the resources are evenly allocated; the satisfaction function is expressed as follows: Where a>0, a represents the parameter of the adjusted satisfaction function. The larger a is, the more balanced the resource allocation result is.

7. The method for analyzing the capacity of a three-layer heterogeneous network based on multi-hop UAV-assisted communication according to claim 1 is characterized in that: In step 4, the optimization problem formed in step 3 is converted into a linear programming problem, which has an optimal solution, denoted as is the network capacity of the drone relay layer, where represents the optimal solution to the linear programming problem.

8. The method for analyzing the capacity of a three-layer heterogeneous network based on multi-hop UAV-assisted communication according to claim 1 is characterized in that: In step 5, a multi-access array antenna is deployed at the base station to achieve simultaneous communication with multiple return UAVs that transmit independent beams; the transmit and receive beam patterns are derived as The interference gain between return UAVs at the time: in, Refers to the antenna gain of the main lobe of the transmit beam, Refers to the antenna gain of the transmit beam sidelobe, Refers to the antenna gain of the main lobe of the receive beam, Refers to the antenna gain of the receiving beam sidelobe; where M is defined as the base station of M; U j b with U i b represents any backhaul drone communicating with base station M; represents the angle between the lines connecting the two arbitrary return drones and the base station; thus, the return drone is obtained The closed-form expression of the capacity under the Rayleigh fading channel is: In the formula, Represents UAV i b The distance to the base station M; therefore, the backhaul link layer network capacity expression is:

9. The method for analyzing the capacity of a three-layer heterogeneous network based on multi-hop UAV-assisted communication according to claim 1 is characterized in that: In step 6, the capacity expression of the three-layer heterogeneous network is given:

Images