CN109753082B - Multi-unmanned aerial vehicle network cooperative communication method - Google Patents

Multi-unmanned aerial vehicle network cooperative communication method Download PDF

Info

Publication number
CN109753082B
CN109753082B CN201811635024.9A CN201811635024A CN109753082B CN 109753082 B CN109753082 B CN 109753082B CN 201811635024 A CN201811635024 A CN 201811635024A CN 109753082 B CN109753082 B CN 109753082B
Authority
CN
China
Prior art keywords
unmanned aerial
aerial vehicle
link
network
drone
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811635024.9A
Other languages
Chinese (zh)
Other versions
CN109753082A (en
Inventor
许文俊
王译锌
张平
张治�
林家儒
李绍胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Original Assignee
Beijing University of Posts and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Priority to CN201811635024.9A priority Critical patent/CN109753082B/en
Publication of CN109753082A publication Critical patent/CN109753082A/en
Application granted granted Critical
Publication of CN109753082B publication Critical patent/CN109753082B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a network cooperative communication method for multiple unmanned aerial vehicles, which comprises the steps that an unmanned aerial vehicle is led and a slave unmanned aerial vehicle flies to a set track according to a preset flight track and then starts to monitor, and fire disaster data are collected; the slave unmanned aerial vehicle collects the acquired data to the master unmanned aerial vehicle in an ad-hoc mode according to the distributed link flow, the power of the link and the information transmission rate; and leading the unmanned aerial vehicle to send the collected data to a ground receiving station. According to the communication requirement in fire monitoring, in an unmanned aerial vehicle network with dynamically changed network topology, the communication capacity of the network is improved by jointly considering data acquisition rate control of the unmanned aerial vehicle, transmission rate arrangement of a channel and resource allocation. The scheme for controlling the joint information source and channel rate in the unmanned aerial vehicle fire monitoring network effectively improves the communication performance of the network, thereby providing technical support in the communication field for the network to be applied to actual fire monitoring, ensuring the communication capability of the network and having great significance for fire rescue.

Description

Multi-unmanned aerial vehicle network cooperative communication method
Technical Field
The invention relates to the field of rate control in the field of wireless communication, in particular to a network cooperative communication method for multiple unmanned aerial vehicles.
Background
In an urban fire monitoring network formed by unmanned aerial vehicles, the acquisition rate and the communication rate of data are controlled according to network conditions, so that congestion is reduced, and more effective auxiliary fire rescue can be realized. When all unmanned aerial vehicles simultaneously collect and send data without limitation, communication resources can be wasted, meanwhile, serious congestion of a network is caused, and the final monitoring result of a fire disaster and a rescue decision are influenced.
At present, research on unmanned aerial vehicle communication focuses on research on communication system performance, for example, joint trajectory optimization and power control of communication from a single unmanned aerial vehicle to a ground receiving station are researched in Energy-efficiency communication with project optimization, so as to optimize Energy efficiency of the system; in Delay-constrained throughput optimization in UAV-enabled OFDM systems, joint trajectory optimization and power control of a Delay-constrained system are studied to optimize the throughput of the system. The existing research focuses on optimizing the communication performance of a point-to-point system, and performance improvement brought by network formation of unmanned aerial vehicles is not considered. In addition, the existing scheme is not considered for the scene of urban fire monitoring. The research of utilizing unmanned aerial vehicle to carry out fire control mainly focuses on the design of fire rescue unmanned aerial vehicle, if have the chinese patent of publication number 207311836U, designed each module of utilizing in unmanned aerial vehicle of high-rise fire rescue, for example airborne control system, navigation module, information acquisition modulation module, make a video recording camera module, luminous guide signal lamp etc.. This patent is only from the design angle of unmanned aerial vehicle itself, does not consider from the angle of communication how to design the scene that just can be better help city condition of a fire control.
Based on the analysis of the prior art, the unmanned aerial vehicle communication network and the application of the unmanned aerial vehicle communication network in the aspect of urban fire monitoring have the following defects:
(1) in the existing scheme, the communication performance optimization of a point-to-point system is researched, the assistance of other unmanned aerial vehicles is not considered, and in a fire monitoring scene, due to the shielding of urban buildings, the quality of a communication link is poor, so that a point-to-point network cannot be competent for a fire monitoring task.
(2) Existing research is developed under a pure communication scene, and for a task-driven network, optimization needs to be performed in combination with task execution requirements, rather than purely considering an abstract communication scene.
(3) The existing scheme optimizes the gravity center at the track of the unmanned aerial vehicle, and the intention is to improve the communication performance of the system through track optimization. However, in the scene of fire monitoring, the unmanned aerial vehicle has a fixed monitoring target, and therefore the track of the unmanned aerial vehicle is also fixed, and the communication performance cannot be improved through track optimization.
Disclosure of Invention
In view of this, the present invention provides a method for cooperative communication between multiple unmanned aerial vehicles, so as to improve communication performance of a network.
Based on the above purpose, the invention provides a multi-unmanned aerial vehicle network cooperative communication method, which is suitable for a multi-unmanned aerial vehicle networking system, wherein the multi-unmanned aerial vehicle networking system is formed by connecting a leading unmanned aerial vehicle and N following unmanned aerial vehicles, and comprises the following steps:
the unmanned aerial vehicle receives the set flight trajectory, forms unmanned aerial vehicle formation and forms an unmanned aerial vehicle communication network consisting of the slave unmanned aerial vehicle, the lead unmanned aerial vehicle and the ground receiving station, wherein the topology of the unmanned aerial vehicle communication network is used
Figure BDA0001929848140000021
It is shown that,
Figure BDA0001929848140000022
the vertices of the topology are represented and,
Figure BDA0001929848140000023
representing links, sets of vertices
Figure BDA0001929848140000024
Figure BDA0001929848140000025
Show from drone, D shows lead drone, G shows groundA surface receiving station, a link set being denoted by
Figure BDA0001929848140000026
Wherein
Figure BDA0001929848140000027
Is a link between the slave drones,
Figure BDA0001929848140000028
is a link from drone to drone,/DGThe link from the unmanned aerial vehicle to the ground receiving station is led;
in time period k, is
Figure BDA0001929848140000029
Each link in the network is allocated with different flow and is recorded as a link
Figure BDA00019298481400000210
The distributed flow rate is fl,kThe slave drones cycle in the same center, leading the drone to hover in a fixed position, the amount of data S that any two slave drones and the channel between the drone and the ground receiving station can carry over a period of timel,kExpressed according to shannon's theorem as:
Figure BDA0001929848140000031
wherein WlIs the bandwidth of the link, pl,kIs the power allocated to the link for link l for period k, hl,kIs the channel fading coefficient, σ, of the time period k2Is the power of additive white gaussian noise,
Figure BDA0001929848140000032
is the set of links between slave drones during time period k,
Figure BDA0001929848140000033
represents a set of time slots, BkAnd EkRepresents the start and end times of period k;
the formation of drones keeps the same mode in flight and runs with a fixed trajectory, leading the maximum data volume S that the link between drone and slave drone can carryl，kExpressed according to shannon's theorem as:
Figure BDA0001929848140000034
wherein h isl,tIs the channel fading coefficient at time t,
Figure BDA0001929848140000035
is the set of links from drone to drone within epoch k;
the multi-unmanned aerial vehicle communication network distributes link flow f, link power p and information transmission rate lambda to each link, so that the flow arranged by any one link is ensured not to exceed the maximum capacity which can be borne by the link in unit time;
according to the link flow of established distribution, the data of gathering are sent to leading unmanned aerial vehicle through unmanned aerial vehicle communication network from unmanned aerial vehicle, lead unmanned aerial vehicle and send the data that assemble for ground receiving station.
As an alternative embodiment, the maximum data volume S that can be carried by the link between the drone D and the ground receiving station G of said formula (3) isl,kThe affine function after introducing the auxiliary variable α is expressed in the form of:
Figure BDA0001929848140000036
wherein the content of the first and second substances,
Figure BDA0001929848140000041
for the set of drones communicating with the ground receiving station G during the period k, λi,kData transfer rate, p, for unmanned plane i in period ki,kTransmit power for drone i, αi,kLet K be the auxiliary variable for epoch corresponding to drone i, K be the total number of instants, λ, f,
Figure BDA0001929848140000042
In the form of a vector of transmission rates, traffic schedules and transmit powers,
Figure BDA0001929848140000043
is a feasible set.
As an alternative embodiment, the equation (12) is solved by introducing lagrangian multiplier β and KKT conditions, and the expression of lagrangian multiplier β introduced by equation (12) is:
Figure BDA0001929848140000044
and according to the KKT condition:
Figure BDA0001929848140000045
in the formula (DEG)*Representing the optimal solution of the corresponding variable solved, e being the base of the natural logarithm, βi,kIs the Lagrangian multiplier for drone i in epoch k, βi,kLagrange multiplier for drone i in time k;
wherein a two-layer loop algorithm is employed to solve equation (13): link traffic f, link power p, and information transmission rate λ.
As an alternative embodiment, the lagrangian function expression of equation (13) is:
Figure BDA0001929848140000046
specifically, the method can be decomposed into an equation (18) and an equation (19), where the equation (18) is a rate control problem and the equation (19) is a power control problem:
Figure BDA0001929848140000047
Figure BDA0001929848140000048
wherein formula (7) is: af ═ SgIn a multi-unmanned networkAssociation matrix
Figure BDA0001929848140000049
And data rate vector
Figure BDA00019298481400000410
The flow conservation expression of (1); the formula (8) is:
Figure BDA00019298481400000411
pi,maxis the peak transmit power of drone i.
As an optional embodiment, the two-layer loop algorithm specifically includes the following steps:
in step 301, parameters are initialized, λ, f, p in feasible domain are selected, initial value r-m-0 is given,
definition α ═ α0，β＝β0，v＝v0Wherein r and m represent the number of external iterations and the number of internal iterations, respectively;
step 302, updating the external iteration times r;
step 303, starting an internal loop and updating an internal iteration coefficient m;
step 304, solving the rate control problem of equation (18);
step 305, solving the power control problem of equation (19);
step 306, updating the dual variable v according to the sub-gradient algorithmm+1A value of (d);
step 307, when | vm+1-νmIf | is less than | the inner loop stops, the next step executes outer loop step 308, the condition | vm+1-νmIf | < is not satisfied, return to step 303, where is a predefined threshold;
step 308, updating the parameter variables α according to the equations (21) and (22)r+1And βr+1：
Figure BDA0001929848140000051
Figure BDA0001929848140000052
Wherein
Figure BDA0001929848140000053
And
Figure BDA0001929848140000054
is the optimal solution obtained by the m-th inner loop, ξαξ step size for updating of auxiliary variable αβThe update step size for the auxiliary variable β;
step 309, when | αr+1-αr< sigma and | βr+1-βrIf the | < sigma is satisfied at the same time, the outer loop is stopped, the step 310 is executed next, otherwise, the step 302 is returned to;
and step 310, outputting the link flow f, the link power p and the information transmission rate lambda which are allocated to each link.
As an alternative embodiment, the step 306 includes the following specific steps: the dual variable ν is calculated as:
Figure BDA0001929848140000061
wherein [. ]]+＝max{·,0}，νIs the update step size of the dual variable v.
As an alternative embodiment, the unmanned aerial vehicle communication network consisting of the slave unmanned aerial vehicles, the lead unmanned aerial vehicle and the ground receiving station adopts ad-hoc networking.
A multi-unmanned aerial vehicle network cooperative communication method is applied to urban fire rescue and is executed according to the following steps:
leading the unmanned aerial vehicle and the slave unmanned aerial vehicle to fly to a set track according to a preset flight track and then start monitoring, and acquiring fire disaster data;
confirming the distributed link flow f, the link power p and the information transmission rate lambda according to the multi-unmanned-aerial-vehicle network cooperative communication method, and summarizing the acquired data to the leading unmanned aerial vehicle through the multi-unmanned-vehicle network; and leading the unmanned aerial vehicle to send the collected data to a ground receiving station.
From the above, the multi-unmanned-aerial-vehicle network cooperative communication method provided by the invention firstly proposes a design for monitoring urban fire by using a network formed by unmanned aerial vehicles, and compared with the design of the existing single fire rescue unmanned aerial vehicle, the multi-unmanned-vehicle network cooperative communication method provided by the invention improves the fire monitoring capability and provides assistance for urban fire rescue in a manner of cooperative networking of the plurality of unmanned aerial vehicles.
Secondly, the invention provides a communication mode that the unmanned aerial vehicle fire monitoring network adopts a wireless self-organizing networking aiming at the condition that the urban building shelters to cause poor communication link quality so as to influence the performance of a point-to-point network, provides combined information source rate and channel rate control aiming at a task-driven unmanned aerial vehicle fire monitoring network, and provides a scheme for solving through a two-layer cyclic algorithm, so that the communication performance of the network is ensured, and the overall monitoring performance of the network is improved through dynamic rate allocation.
Drawings
FIG. 1 is a schematic diagram of an unmanned aerial vehicle network working scene oriented to urban fire monitoring according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating a cooperative communication method of multiple unmanned aerial vehicles in a scene of urban fire monitoring by the unmanned aerial vehicles according to the embodiment of the invention;
FIG. 3 is a flowchart of an algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.
By means of rapid popularization of unmanned aerial vehicles, the invention proposes that unmanned aerial vehicles form a network and are applied to urban fire monitoring. At city conflagration frequent today, when the conflagration that takes place, the sensor in the building probably has suffered destruction of certain degree, utilizes the multiple sensor that unmanned aerial vehicle was equipped with, can effectively know the condition of conflagration, provides the guidance for the rescue of conflagration. Aiming at the communication challenge introduced by the specific scene, the designed unmanned aerial vehicle network adopts an ad-hoc mode for networking, and carries out optimization design aiming at the scheme, and jointly considers the data acquisition rate control of the unmanned aerial vehicle, the rate control of a channel and the resource allocation, thereby improving the performance of the unmanned aerial vehicle network.
As shown in figure 1, the mobile unmanned aerial vehicle fire monitoring network consists of a leading unmanned aerial vehicle D with long-distance communication capability and a group of slave unmanned aerial vehicles with only short-distance communication capability
Figure BDA0001929848140000071
And (4) forming. According to the actual situation, a multi-unmanned aerial vehicle communication network is established, and the following steps are executed to monitor the fire disaster, as shown in fig. 2:
step 201, deploying positions of the unmanned aerial vehicles according to fire disasters of the site, and monitoring by flying different unmanned aerial vehicles to the arranged heights as required to form unmanned aerial vehicle formations.
According to the need for fire monitoring, all slave drones fly in the same mode: they spiral around the building at a constant speed in a horizontal plane, but observe different floors at different heights. Meanwhile, the unmanned aerial vehicle is led to hover at the fixed position of the top of the building, and the fire is observed from the top.
Step 202, comprehensively considering the link data volume between any two slave unmanned aerial vehicles, the link data volume between all the slave unmanned aerial vehicles and the leading unmanned aerial vehicle, and the link data volume from the leading unmanned aerial vehicle to the ground receiving station, the multi-unmanned aerial vehicle communication network allocates link flow f, link power p and information transmission rate lambda to each link, so that the flow arranged by any one link is ensured not to exceed the maximum capacity which can be borne by the link in unit time, and the sum of the data flow transmitted by all the links is not more than the total data volume of the links between the leading unmanned aerial vehicle and the ground receiving station; each slave unmanned aerial vehicle and the link thereof transmit data according to the allocated rate; all drones can collect data through multiple sensors to detect fire behavior, building structural damage and survivor conditions.
And step 203, the slave unmanned aerial vehicle collects data to the slave unmanned aerial vehicle in an ad-hoc mode through the network according to the current network topology.
And step 204, transmitting the acquired data back to a ground receiving station in real time through a network so as to assist in decision of fire rescue.
Because the shielding of the building causes serious attenuation of communication signals, the invention adopts a line-of-sight channel model, namely, the communication cannot be carried out when a direct path does not exist between any two unmanned aerial vehicles. In this case, some slave drones may not be able to communicate with the leading drone and therefore cannot send the collected data to the leading drone, so the network is organised in an ad-hoc manner, the leading drone, in addition to sending the data it generates itself, also undertakes the task of assisting other slave drones, so that fires of different heights can be fed back to the control centre.
The network operates in the following mode: first, data relating to a fire is collected from the course of the unmanned aerial vehicle moving on its trajectory. And then, collecting the data collected from the unmanned aerial vehicle to the leading unmanned aerial vehicle through an ad-hoc network, and finally returning the collected data to the ground receiving station by the leading unmanned aerial vehicle to assist in fire rescue decision making.
Topology of unmanned aerial vehicle fire monitoring network
Figure BDA0001929848140000081
In the description that follows,
Figure BDA0001929848140000082
the vertices of the topology are represented and,
Figure BDA0001929848140000083
representing links in which the vertices aggregate to
Figure BDA0001929848140000084
Consists of a slave unmanned aerial vehicle, a lead unmanned aerial vehicle and a ground receiving station, wherein,
Figure BDA0001929848140000085
representing slave drone, D leading drone and G ground receiving station. Set of links as
Figure BDA0001929848140000086
Wherein
Figure BDA0001929848140000087
Is a link between the slave drones,
Figure BDA0001929848140000088
is a link from drone to drone,/DGIs the link leading the unmanned aerial vehicle to the ground receiving station. Due to the movement of the drone,
Figure BDA0001929848140000089
is time-varying.
Since the slave drones in this scenario run in a fixed trajectory, the topology of the network cycles with a period T. Further, the period T may be divided into K periods, and the topology of the network is constant in each period. Set of divided time slots as
Figure BDA0001929848140000091
With BkAnd EkIndicating the start and end times, t, of the period kk＝Ek-BkRepresenting the duration of time period k. In time period k, the set of links is represented as
Figure BDA0001929848140000092
In particular, during time period k, the set of links between slave drones may be represented as
Figure BDA0001929848140000093
In which there is a directed edge<i,j>Representing a link with slave unmanned aerial vehicle i as a transmitting end and j as a receiving end, di,jTo the distance between slave drones i and j, dmaxIs the maximum communication distance from the drone. In a similar manner to that described above,
Figure BDA0001929848140000094
represents the set of links from drone to drone, where di,DFor leading unmanned aerial vehicle from unmanned aerial vehicle iD, distance between D. Each link is assigned channels that are orthogonal to each other to avoid interference.
In epoch k, the correlation matrix is used
Figure BDA0001929848140000095
Representing relationships of nodes and edges
Figure BDA0001929848140000096
Wherein a isi,l,kIs a matrix
Figure BDA0001929848140000097
I row and l column, |, represents the number of collection elements. According to graph theory, two vertices are connected if vertex j is reached from fixed point i through a set of directed edges. The set of drones communicating with the ground receiving station during time k is denoted as
Figure BDA0001929848140000098
Collection
Figure BDA0001929848140000099
Can the slave drone transmit data to the ground receiving station. At time k, is unmanned aerial vehicle
Figure BDA00019298481400000915
Assigned data transmission rate of lambdai,kThus, the amount of data that drone i can transmit is
Figure BDA00019298481400000911
In time period k, is
Figure BDA00019298481400000912
Each link in the network is allocated with different flow and is recorded as a link
Figure BDA00019298481400000913
The distributed flow rate is fl,k. Since all drones circulate around the same center around the building, the channel between any two slave drones remains unchanged for a period of time. In addition, the channel leading the drone to the ground receiving station is also constant. For the above-mentioned channel, the amount of data S that can be carried in a period of timel,kCan be expressed according to shannon's theorem as:
Figure BDA00019298481400000914
wherein WlIs the bandwidth of the link, pl,kIs the power allocated to the link for link l for period k, hl,kIs the channel fading coefficient, σ, of the time period k2Is the power of additive white gaussian noise,
Figure BDA0001929848140000101
is the set of links between slave drones during time period k.
It is also possible for the drone to hover outside the center of the circular motion of the drone to better observe the fire. Thus, the channel fading coefficients between the leading drone and the trailing drone are dynamically changing. Similarly, according to shannon's theorem, the amount of data that can be carried from the link between the drone and the lead drone can be calculated as
Figure BDA0001929848140000102
Wherein h isl,tIs the channel fading coefficient at time t,
Figure BDA0001929848140000103
is the set of links from drone to drone within epoch k.
To ensure that any link can complete data transmission at any time period, the traffic allocated to each link cannot exceed its maximum load, i.e. the traffic allocated to each link cannot exceed its maximum load
f≤S (4)
Where f and S are vector representations of the corresponding link scheduling traffic and the total amount of data that the link can carry.
The law of conservation of flow implies that the amount of flow into any one node plus the data that generates the data itself should be equal to the amount of flow out. For each drone, it may be denoted as
Figure BDA0001929848140000104
Wherein
Figure BDA0001929848140000105
Is the outgoing link for drone i,
Figure BDA0001929848140000106
is the inbound link for drone i. All the generated data are transmitted to the ground receiving station, and the flow conservation law of the ground receiving station can be expressed as
Figure BDA0001929848140000107
The expression for the law of conservation of flow described above is redundant because
Figure BDA0001929848140000108
That is, the matrices are linearly dependent. Will matrix
Figure BDA0001929848140000109
The last row of (6) is removed to obtain a simplified correlation matrix
Figure BDA0001929848140000111
And data rate vector
Figure BDA0001929848140000112
Thus, a compact expression of the law of conservation of flow throughout the network can be written as
Af＝Sg(7)
Unmanned plane
Figure BDA0001929848140000113
May be expressed as
Figure BDA0001929848140000114
For any drone, its peak transmit power is limited, which can be expressed as
Figure BDA0001929848140000115
Wherein p isi,maxIs the peak transmit power of drone i.
Thus, the energy efficiency maximization problem of the unmanned aerial vehicle fire monitoring network can be expressed as
Figure BDA0001929848140000116
And uses the utility function log (-) to ensure fairness. Where lambda is the assigned data transmission rate vector,
Figure BDA00019298481400001113
a transmit power vector allocated for the link.
Since the integral in equation (3) is difficult to handle, we replace solving the integral directly with its lower bound, i.e.
Figure BDA0001929848140000118
Whereinh l,k＝minhl,t,t∈[Bk,Ek]. Thus, the vector form of the S update can be expressed asSThus, the original problem (9) can be converted into
Figure BDA0001929848140000119
It is clear that the constraints in equation (11) relate to optimizing the set of variables
Figure BDA00019298481400001111
Form aThe feasible set of the convex is recorded as
Figure BDA00019298481400001112
However, because the goal of the optimization problem contains fractional terms, it is still tricky to convert the goal in equation (11) to form and restate the problem with the introduction of the auxiliary variable α
Figure BDA0001929848140000121
αi,kFor the purpose of making the problem easy to handle, we further convert equation (12) to parametric subtraction form and introduce another lagrange multiplier β.
Suppose that
Figure BDA0001929848140000122
Is a solution of formula (12) and is present as β*So that for the parameter variable α - α*And β ═ β*，
Figure BDA0001929848140000123
A KKT condition satisfying the following formula:
Figure BDA0001929848140000124
where e is the base of the natural logarithm, βi,kIs the lagrange multiplier, λ, of drone i in epoch ki,kFor transmission rate of drone i, pi,kTransmitting power for the unmanned aerial vehicle i; and, according to the KKT condition, have
Figure BDA0001929848140000125
Wherein (·)*And representing the optimal solution of the corresponding variable obtained by solving.
Therefore, the solution of equation (12) can be obtained by solving equation (13) while ensuring equation (14).
The whole algorithm can be divided into two parts. The inner loop applies dual decomposition to solve the convex optimization problem until a convergence criterion is reached, which means that the inner dual and original problem are reached to reach the optimal solution. The outer loop updates the parameter variables α and β using the output of the inner loop until the convergence criterion of the outer loop is reached. Otherwise, the internal optimization is restarted using the updated α and β to continue the iteration. Let r and m denote the number of external iterations and the number of internal iterations, respectively, as shown in FIG. 3:
step 301, initializing parameters, selecting λ, f, p in feasible domain, and defining r-m-0, α - α0，β＝β0，v＝v0。
And step 302 and step 303, starting the internal loop algorithm, and updating the external iteration number r and the internal iteration number m according to the loop number.
Assume that the last updated parameter variable for the outer loop is αrAnd βr. Lagrangian function of equation (14) with the introduction of the Lagrangian multiplier
Figure BDA0001929848140000131
Is composed of
Figure BDA0001929848140000132
1) Original problems: original problem
Figure BDA0001929848140000133
Can be decomposed into two independent sub-problems about the optimization variables
By unfolding formula (15), can be obtained
Figure BDA0001929848140000134
Observing the above formula, it can be decomposed into two independent sub-problems, i.e.
Figure BDA0001929848140000135
Step 304, solving the source rate and link rate control subproblem, namely equation (18), and solving the power control subproblem, namely equation (19):
Figure BDA0001929848140000136
equation (18) is a linear programming problem that can be solved using a number of tools, such as the matlab tool.
Step 305, solving the power control subproblem, namely equation (19):
Figure BDA0001929848140000141
equation (19) is a convex optimization problem that is easily solved by convex optimization methods such as the interior point method.
2) The dual problem is as follows:
step 306, dual variable vm+1The solution may be based on a sub-gradient algorithm. The solution obtained by the m-th inner loop is recorded as
Figure BDA0001929848140000142
Therefore, according to the sub-gradient descent algorithm, the dual variable of the m +1 th inner loop is updated according to the following equation
Figure BDA0001929848140000143
Wherein [. ]]+＝max{·,0}，νIs the update step size of the dual variable v.
By solving the two sub-problems and iteratively calculating equation (20), the optimal solution can be found when the convergence criterion is reached, step 307. When v is greater thanm+1-νmIf, | < then, the inner loop should stop, where it is a predefined threshold.
A outer loop algorithm
Step 308, update αr+1And βr+1The outer loop optimization problem is in the form of parametric subtraction, with an objective function as shown in equation (13) where α is the auxiliary variable and β is the lagrangian multiplier introduced by the constraint in equation (12) the updating of the parametric variables using the sub-gradient method is as follows:
Figure BDA0001929848140000144
Figure BDA0001929848140000145
wherein
Figure BDA0001929848140000146
And
Figure BDA0001929848140000147
is the optimal solution obtained by the m-th inner loop, ξαξ step size for updating of auxiliary variable αβIs the update step size of the auxiliary variable β.
Step 309, similarly, when | αr+1-αr< sigma and | βr+1-βrIf | is less than σ, the outer loop is stopped.
And step 310, outputting the link flow f, the link power p and the information transmission rate lambda value after the last iteration when the outer loop is finished.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the idea of the invention, also features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.
The embodiments of the invention are intended to embrace all such alternatives, modifications and variances that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements and the like that may be made without departing from the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (6)

1. A multi-unmanned aerial vehicle network cooperative communication method is suitable for a multi-unmanned aerial vehicle networking system, and is characterized in that the multi-unmanned aerial vehicle networking system is formed by connecting a leading unmanned aerial vehicle and N following unmanned aerial vehicles, and comprises the following steps:
the unmanned aerial vehicle receives the set flight trajectory, forms unmanned aerial vehicle formation and forms an unmanned aerial vehicle communication network consisting of the slave unmanned aerial vehicle, the lead unmanned aerial vehicle and a ground receiving station, wherein the topology of the unmanned aerial vehicle communication network is represented by G (V, L), V represents a topological structure vertex, L represents a link, a vertex set V (GUDUN), N represents the slave unmanned aerial vehicle, D represents the lead unmanned aerial vehicle, G represents the ground receiving station, and a link set L (L) (L) is represented by { GUDUN }NNU LNDU lDGIn which L isNNIs a link between slave drones, LNDIs a link from drone to drone,/DGThe link from the unmanned aerial vehicle to the ground receiving station is led;
at time period k, is LkEach link in the network is allocated different traffic, denoted as link L ∈ LkThe distributed flow rate is fl,kThe slave drones cycle in the same center, leading the drone to hover in a fixed position, the amount of data S that any two slave drones and the channel between the drone and the ground receiving station can carry over a period of timel，kExpressed according to shannon's theorem as:
Figure FDA0002387724510000011
wherein WlIs the bandwidth of the link, pl,kIs the power allocated to the link for link l for period k, hl,kIs the channel fading coefficient, σ, of the time period k2Is the power of additive white Gaussian noise, LNN,kIs a link set between slave drones in a period K, and K represents a time slot set;
the unmanned aerial vehicle formation keeps the same mode to move in a fixed track during flightMaximum data volume S that link between leading unmanned aerial vehicle and following unmanned aerial vehicle can bearl，kExpressed according to shannon's theorem as:
Figure FDA0002387724510000012
wherein h isl,tIs the channel fading coefficient, L, at time tND,kIs the set of links from drone to drone in epoch k, BkAnd EkRepresents the start and end times of period k;
formula (3) leads the biggest data bulk S that the link between unmanned aerial vehicle D and the ground receiving station G can bearl，kThe affine function after introducing the auxiliary variable α is expressed in the form of:
Figure FDA0002387724510000021
wherein N iskFor the set of drones communicating with the ground receiving station G during the period k, λi，kData transfer rate, p, for unmanned plane i in period ki，kTransmit power for drone i, αi，kFor K periods, corresponding to the auxiliary variable of drone i, K being the total number of moments, λ, f, pLIn the form of vectors of transmission rate, traffic arrangement and transmission power, X is a feasible set;
the formula (12) is solved by introducing Lagrangian multiplier beta and KKT conditions, and the expression of the Lagrangian multiplier beta introduced by the formula (12) is as follows:
Figure FDA0002387724510000022
and according to the KKT condition:
Figure FDA0002387724510000023
in the formula (DEG)*Representing the optimal solution of the corresponding variable obtained by the solution, e being a natural pairBase of number, βi,kIs the Lagrangian multiplier for drone i in epoch k, βi,kThe lagrangian multiplier for drone i in epoch k,
wherein a two-layer loop algorithm is employed to solve equation (13): obtaining an optimized unmanned aerial vehicle communication network by using the link flow f, the power p of the link and the information transmission rate lambda;
the slave unmanned aerial vehicle sends the collected data to the leading unmanned aerial vehicle through the optimized unmanned aerial vehicle communication network, and the leading unmanned aerial vehicle sends the collected data to the ground receiving station.
2. The cooperative multi-drone network communication method according to claim 1, wherein the lagrangian function expression of the equation (13) is: l (v)m；·)＝L(νm；λ,f)+L(νm；pL) Specifically, the method can be decomposed into equation (18) and equation (19), where equation (18) is a rate control problem and equation (19) is a power control problem:
Figure FDA0002387724510000031
Figure FDA0002387724510000032
wherein formula (7) is: af ═ SgFor the incidence matrix in the multi-unmanned aerial vehicle network
Figure FDA0002387724510000033
And data rate vector
Figure FDA0002387724510000034
The flow conservation expression of (1); the formula (8) is: p is a radical ofi,k≤pi,max,
Figure FDA0002387724510000035
k∈K，pi,maxIs the peak transmit power of drone i.
3. The cooperative communication method for multiple unmanned aerial vehicle networks according to claim 1, wherein the double-layer loop algorithm specifically comprises the following steps:
in step 301, parameters are initialized, λ, f, and p in feasible domain are selected, initial value r ═ m ═ 0 is given, and definition α ═ α0，β＝β0，v＝v0Wherein r and m represent the number of external iterations and the number of internal iterations, respectively;
step 302, updating the external iteration times r;
step 303, starting an internal loop and updating an internal iteration coefficient m;
step 304, solving the rate control problem of equation (18);
step 305, solving the power control problem of equation (19);
step 306, updating the dual variable v according to the sub-gradient algorithmm+1A value of (d);
step 307, when | vm+1-νmIf | is less than | the inner loop stops, the next step executes outer loop step 308, the condition | vm+1-νmIf | < is not satisfied, return to step 303, where is a predefined threshold;
step 308, updating the parameter variables α according to the equations (21) and (22)r+1And βr+1：
Figure FDA0002387724510000036
Figure FDA0002387724510000041
Wherein
Figure FDA0002387724510000042
And
Figure FDA0002387724510000043
is the optimal solution obtained by the m-th inner loop, ξαξ step size for updating of auxiliary variable αβThe update step size for the auxiliary variable β;
step 309, when | αr+1-αr< sigma and | βr+1-βrIf the | < sigma is satisfied at the same time, the outer loop is stopped, the step 310 is executed next, otherwise, the step 302 is returned to;
and step 310, outputting the link flow f, the link power p and the information transmission rate lambda which are allocated to each link.
4. The cooperative communication method for multiple unmanned aerial vehicle networks as claimed in claim 3, wherein the step 306 comprises the following steps: the dual variable ν is calculated as:
Figure FDA0002387724510000044
wherein [. ]]+＝max{·,0}，νIs the update step size of the dual variable v.
5. The cooperative communication method for multi-drone networks according to claim 1, wherein the drone communication network composed of slave drones, lead drones and ground receiving stations is ad-hoc networked.
6. The multi-unmanned-aerial-vehicle network cooperative communication method according to any one of claims 1 to 5, wherein the multi-unmanned-aerial-vehicle network cooperative communication method is applied to urban fire rescue and is executed according to the following steps:
leading the unmanned aerial vehicle and the slave unmanned aerial vehicle to fly to a set track according to a preset flight track and then start monitoring, and acquiring fire disaster data;
confirming the distributed link flow f, the link power p and the information transmission rate lambda according to the multi-unmanned-aerial-vehicle network cooperative communication method, and summarizing the acquired data to the leading unmanned aerial vehicle through the multi-unmanned-vehicle network; and leading the unmanned aerial vehicle to send the collected data to a ground receiving station.
CN201811635024.9A 2018-12-29 2018-12-29 Multi-unmanned aerial vehicle network cooperative communication method Active CN109753082B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811635024.9A CN109753082B (en) 2018-12-29 2018-12-29 Multi-unmanned aerial vehicle network cooperative communication method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811635024.9A CN109753082B (en) 2018-12-29 2018-12-29 Multi-unmanned aerial vehicle network cooperative communication method

Publications (2)

Publication Number Publication Date
CN109753082A CN109753082A (en) 2019-05-14
CN109753082B true CN109753082B (en) 2020-08-14

Family

ID=66404310

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811635024.9A Active CN109753082B (en) 2018-12-29 2018-12-29 Multi-unmanned aerial vehicle network cooperative communication method

Country Status (1)

Country Link
CN (1) CN109753082B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110392357B (en) * 2019-06-21 2022-03-11 广东工业大学 Resource allocation control method of unmanned aerial vehicle-assisted Internet of things communication device
CN110312265B (en) * 2019-06-24 2021-01-01 清华大学 Power distribution method and system for unmanned aerial vehicle formation communication coverage
CN110611527B (en) * 2019-09-17 2021-01-15 北京邮电大学 Array antenna, beam forming method and sensing and communication integrated system
CN111142562B (en) * 2019-12-27 2021-04-30 北京理工大学 Formation transformation control method under hybrid condition constraint based on stress matrix
CN111343609B (en) * 2020-02-28 2022-09-02 西南电子技术研究所（中国电子科技集团公司第十研究所） Unmanned aerial vehicle measurement and control chain power and rate combined control method
CN112034891B (en) * 2020-09-21 2022-03-29 北京邮电大学 Method and device for controlling mobility of self-organizing network in flight
CN114268903B (en) * 2021-12-28 2022-09-30 北京航空航天大学 Geographic information assisted unmanned aerial vehicle relay position deployment and power distribution method
CN117351706A (en) * 2023-10-07 2024-01-05 广东艾百智能科技有限公司 Highway number monitoring and data closed-loop analysis system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105163380A (en) * 2015-09-23 2015-12-16 浙江理工大学 Distributed cross-layer optimization method for MIMO (Multiple Input Multiple Output) wireless multi-hop network
WO2016069598A1 (en) * 2014-10-27 2016-05-06 Sikorsky Aircraft Corporation Communication link accessibility aware navigation
CN108243480A (en) * 2018-01-11 2018-07-03 南京华讯方舟通信设备有限公司 A kind of unmanned plane ad hoc network chance routing algorithm based on the prediction of link available bandwidth and forward travel distance
CN108834173A (en) * 2018-06-04 2018-11-16 杭州电子科技大学 A kind of centralized optimizing distribution method of wireless multi-hop network
CN109067490A (en) * 2018-09-29 2018-12-21 郑州航空工业管理学院 Cellular Networks join lower multiple no-manned plane and cooperate with mobile edge calculations method for distributing system resource

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9853715B2 (en) * 2014-02-17 2017-12-26 Ubiqomm Llc Broadband access system via drone/UAV platforms

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016069598A1 (en) * 2014-10-27 2016-05-06 Sikorsky Aircraft Corporation Communication link accessibility aware navigation
CN105163380A (en) * 2015-09-23 2015-12-16 浙江理工大学 Distributed cross-layer optimization method for MIMO (Multiple Input Multiple Output) wireless multi-hop network
CN108243480A (en) * 2018-01-11 2018-07-03 南京华讯方舟通信设备有限公司 A kind of unmanned plane ad hoc network chance routing algorithm based on the prediction of link available bandwidth and forward travel distance
CN108834173A (en) * 2018-06-04 2018-11-16 杭州电子科技大学 A kind of centralized optimizing distribution method of wireless multi-hop network
CN109067490A (en) * 2018-09-29 2018-12-21 郑州航空工业管理学院 Cellular Networks join lower multiple no-manned plane and cooperate with mobile edge calculations method for distributing system resource

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Network Building and Communication Strategy;Fu Xiaowei 等;《2016 14th International Conference on Control, Automation, Robotics and Vision 》;IEEE;20170202;第1-5页 *
多天线协作传输系统的容量分析与功率分配;张静美等;《北京邮电大学学报》;北京邮电大学;20051031;第28卷(第5期);第11-15页 *
数据链指挥下无人机群超机动对空对地攻击研究;邵凝宁;《中国优秀博硕士学位论文全文数据库 (硕士) 工程科技Ⅱ辑》;中国学术期刊（光盘版）电子杂志社;20061015(第10期);第1-107页 *
无人机战场通信网络及其关键技术研究;王宇;《中国优秀博硕士学位论文全文数据库 (硕士) 信息科技辑》;中国学术期刊（光盘版）电子杂志社;20060715(第7期);第1-61页 *
高动态航空自组网MAC协议研究与设计;苏宝霞;《中国优秀硕士学位论文全文数据库 信息科技辑》;中国学术期刊（光盘版）电子杂志社;20170315(第3期);第1-59页 *

Also Published As

Publication number Publication date
CN109753082A (en) 2019-05-14

Similar Documents

Publication Publication Date Title
CN109753082B (en) Multi-unmanned aerial vehicle network cooperative communication method
CN110381444B (en) Unmanned aerial vehicle track optimization and resource allocation method
Zhang et al. Energy-efficient trajectory optimization for UAV-assisted IoT networks
CN108924788B (en) Energy consumption balancing method in wireless ultraviolet light cooperation unmanned aerial vehicle formation network
Tan et al. UAV-aided edge/fog computing in smart IoT community for social augmented reality
CN110730031B (en) Unmanned aerial vehicle track and resource allocation joint optimization method for multi-carrier communication
CN108768497B (en) Unmanned aerial vehicle auxiliary wireless sensor network and node scheduling and flight parameter design method thereof
CN110730495B (en) Unmanned aerial vehicle data distribution optimization method under energy constraint
CN110380773A (en) A kind of track optimizing and resource allocation methods of unmanned plane multi-hop relay communication system
Pinto et al. A framework for analyzing fog-cloud computing cooperation applied to information processing of UAVs
CN112911537B (en) Method for minimizing task time of multi-unmanned aerial vehicle information acquisition system
CN110147040B (en) Flight trajectory and power distribution joint optimization method for energy-carrying transmission of unmanned aerial vehicle
CN114048689B (en) Multi-unmanned aerial vehicle aerial charging and task scheduling method based on deep reinforcement learning
CN110913402A (en) High-coverage-efficiency unmanned aerial vehicle ad hoc network clustering method for jointly optimizing communication and formation
CN112564767B (en) Continuous coverage method based on self-organizing optimization cooperation in unmanned aerial vehicle network
Valentino et al. Opportunistic computational offloading system for clusters of drones
CN113259946A (en) Ground-to-air full coverage power control and protocol design method based on centralized array antenna
Giyenko et al. Intelligent unmanned aerial vehicle platform for smart cities
Yeduri et al. Energy and throughput management in delay-constrained small-world UAV-IoT network
Wang et al. Computation efficiency maximization for UAV-assisted relaying and MEC networks in urban environment
CN113776531A (en) Multi-unmanned-aerial-vehicle autonomous navigation and task allocation algorithm of wireless self-powered communication network
Zhang et al. Joint computation offloading and trajectory design for aerial computing
CN114945182B (en) Multi-unmanned aerial vehicle relay optimization deployment method in urban environment
CN112020001A (en) Time slot resource allocation method for multi-station multi-machine system of unmanned aerial vehicle
Kumar et al. Battery and Energy Management in UAV‐Based Networks

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant