CN116781144A - Method, device and storage medium for carrying edge server by unmanned aerial vehicle - Google Patents
Method, device and storage medium for carrying edge server by unmanned aerial vehicle Download PDFInfo
- Publication number
- CN116781144A CN116781144A CN202310845782.8A CN202310845782A CN116781144A CN 116781144 A CN116781144 A CN 116781144A CN 202310845782 A CN202310845782 A CN 202310845782A CN 116781144 A CN116781144 A CN 116781144A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- load
- grid
- current
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003860 storage Methods 0.000 title claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims abstract description 53
- 238000004891 communication Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 230000002441 reversible effect Effects 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 7
- 230000005484 gravity Effects 0.000 claims description 4
- 238000009434 installation Methods 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 23
- 230000008859 change Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000020595 eating behavior Effects 0.000 description 3
- 238000013439 planning Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000009440 infrastructure construction Methods 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The application aims to provide a method, a device and a storage medium for carrying an edge server on an unmanned aerial vehicle, which comprise the following steps: selecting a map of a region to be served, and rasterizing the map to obtain a grid corresponding to the map; the load of each grid in any time period is obtained, and the grid loads in the time period are sequenced to obtain a grid load sequence; and determining the deployment position and coverage range of the unmanned aerial vehicle for providing service in the time period according to the rasterization load sequence. According to the method, the unmanned aerial vehicle base stations are utilized to provide calculation power, the number and the positions of deployed unmanned aerial vehicle base stations can be flexibly adjusted in a high-speed maneuver mode according to the real-time requirements of the users of the Internet of vehicles, and the resource utilization rate is improved; and the ground pressure is not required to be considered, the installation and configuration are simple and convenient, and the application prospect is wide.
Description
Technical Field
The application relates to the field of vehicle networking edge computing infrastructure construction planning, in particular to a method, a device and a storage medium for carrying an edge server on an unmanned aerial vehicle.
Background
In recent years, edge computing has rapidly developed, which sinks computing to the network edge, and is close to end users, and is a key driver for the development of internet of vehicles. Edge computation is characterized by high bandwidth, ultra low latency, and real-time access to network information. And the real-time communication and coordination command of vehicles in the Internet of vehicles are not separated from a high-speed and low-delay network. The characteristics of the edge calculation exactly meet the development requirement of the Internet of vehicles, and have important significance for promoting the development and application of the Internet of vehicles.
In edge computing technology, edge servers play an important role. For the end user, the edge server plays a role of a service provider, provides more various calculation processing modes, and brings the calculation resources close to the end user, so that the delay of the service request is reduced. For cloud servers, edge servers serve as a first layer of data processing, helping to reduce data transmission over the core network.
The deployment of the edge server is a key problem, and the current deployment method is mainly divided into two forms of static deployment and dynamic deployment. Static deployment, i.e., deploying edge servers in fixed locations. However, the running of the vehicle exhibits a strong randomness under the influence of various factors. Thus, the speed of change of the vehicle flow may be rapid during certain periods of time, which may lead to large fluctuations in the regional load. For example, servers somewhere in certain time periods may overload the data volume due to sudden increases in traffic flow. To support such transient peaks, simply increasing server capacity would greatly increase infrastructure construction costs. Accordingly, during off-peak hours, another server node may be idle due to less traffic, which wastes valuable computing resources.
Therefore, static deployment may not meet the requirements, and dynamic deployment of edge servers becomes a more flexible and efficient deployment approach. The dynamic deployment can allocate server resources according to real-time traffic flow information, so that resource waste and unbalanced regional load are avoided.
Currently, most research selects a Unmanned Aerial Vehicle (UAV) as a mobile carrier for a server to provide edge computing services. Unmanned aerial vehicles are widely used in many different scenes at present, and can carry various devices to complete various special tasks. Unmanned aerial vehicle technology has brought the great progress for the construction in fields such as disaster monitoring, communication assistance, military reconnaissance. Unmanned aerial vehicles have two basic modes of low-altitude flight and hover. In hover situations, the drone may act as a stable over-the-air edge computing platform. As a mobile node, the unmanned aerial vehicle can change the position according to the dynamic change of the load, so that the resource allocation is optimized, and the defect that the position of a fixed site cannot be changed and the calculation force is inconvenient to adjust is overcome.
Currently, drones may be applied to provide a reliable low-latency communication link for users due to their hover stability and line-of-sight transmission characteristics. Therefore, the deployment mode of the edge computing server can only be fixedly deployed on the base station in the past, and the edge computing server can be deployed on various mobile carriers, so that more flexible services are provided. In terms of mobile deployment, current research is mainly focused on mobile vehicles such as Unmanned Aerial Vehicles (UAVs) and traveling vehicles; the prior art mainly focuses on the computational offloading between mobile servers and vehicles and the selection of mobile servers; and on the premise of ensuring the service quality, the track and the position of the unmanned aerial vehicle are adjusted through dynamic planning and a decision algorithm so as to adapt to the requirements of mobile users.
However, most of the solutions do not consider the dynamic adjustment of the position of the unmanned aerial vehicle base station according to the space-time variation of the load, so that the resource utilization rate of the conventional solution in which the unmanned aerial vehicle is mounted with the edge calculation server is not high.
In view of this, a method, apparatus and storage medium for enhancing the loading of an edge server using an unmanned aerial vehicle as a carrier have been desired.
Disclosure of Invention
The application aims to provide a method, a device and a storage medium for carrying an edge server on an unmanned aerial vehicle, which are used for at least solving one technical problem in the prior art.
The technical scheme of the application is as follows:
a method for an unmanned aerial vehicle to carry an edge server, comprising:
selecting a map of a region to be served, and rasterizing the map to obtain a grid corresponding to the map;
the load of each grid in any time period is obtained, and the grid loads in the time period are sequenced to obtain a grid load sequence;
and determining the deployment position and coverage range of the unmanned aerial vehicle for providing service in the time period according to the rasterization load sequence.
The determining the deployment position and coverage range of the unmanned aerial vehicle for providing service in the time period according to the rasterization load sequence comprises the following steps:
the number of unmanned aerial vehicles is not limited, and according to the grids and the loads corresponding to the grids, the hovering position and the coverage range of each unmanned aerial vehicle are sequentially selected according to the direction from the outer side to the inner side of the map until all loads are distributed;
and arranging the unmanned aerial vehicle according to the load quantity in a reverse order according to the rasterization load sequence, selecting the first M positions with the largest load as final hovering positions, and dispatching the unmanned aerial vehicle to execute corresponding tasks.
According to the grids and the loads corresponding to the grids, selecting a hovering position and a coverage range of each unmanned aerial vehicle in turn according to a direction from the outer side to the inner side of the map, wherein the method comprises the following steps:
let unmanned aerial vehicle initial calculation power be C curr =c, coverage is
From the edge of the map, selecting any edge position g as a starting point, adding a grid g into the coverage area S=S-U { g } of the unmanned aerial vehicle, and distributing the load of the grid to the current unmanned aerial vehicle w g =0 and updating the current calculation of the unmanned aerial vehicle to the original calculation minus the intra-grid load C curr ＝C curr -w g ；
Extending from the starting point g to the outer layer, and sequentially adding grids surrounding the starting point g into the coverage area of the unmanned aerial vehicle;
and taking the comparison result of the distance from the grid h to any grid in the coverage area and the communication radius of the unmanned aerial vehicle as a strategy distributed by the current unmanned aerial vehicle or taking the comparison result of the current calculation power and the load of the grid as the strategy distributed by the current unmanned aerial vehicle to obtain the position and the coverage area of any unmanned aerial vehicle.
And if the distance from the grid h to any grid in the coverage range exceeds twice the communication radius of the unmanned aerial vehicle, ending the current unmanned aerial vehicle distribution.
If the calculated force is greater than the load C added to the grid curr ≥w h Adding the grid into the current range S=S U { h }, and distributing the grid load to the current unmanned plane w h =0, update the current calculation force to the current calculation force minus the grid load C curr ＝C curr -w h ；
If the current calculation power is smaller than the load C added to the grid curr ＜w h Adding a grid into the current range S=S U { h }, distributing a grid load part to the current unmanned aerial vehicle so as to ensure that the calculation power of the unmanned aerial vehicle is completely used up, updating the grid load to be the grid load minus the current calculation power w of the unmanned aerial vehicle h ＝w h -C curr Updating current calculation power of unmanned aerial vehicle to C curr And (4) ending the current unmanned aerial vehicle allocation.
Ending the current unmanned aerial vehicle allocation, comprising:
the load is taken as a mass, the center of gravity of the coverage area S is selected as a candidate position, and all grids with load 0 on the map are deleted.
The unmanned aerial vehicle is arranged according to the rasterization load sequence and the load quantity is reversed, and the first M positions with the largest load are selected as the final hovering positions, and the method comprises the following steps:
traversing and comparing the load amounts of all grids, and selecting the area with the largest load amount as a target area of the first unmanned aerial vehicle;
and arranging the loading amounts of the remaining grids in a reverse order, and dispatching the remaining unmanned aerial vehicle to the area with the largest loading amount in the remaining grids until all grids are distributed, so as to complete all grid coverage, or dispatching all unmanned aerial vehicles.
The unmanned aerial vehicle is arranged according to the rasterization load sequence and the load quantity is reversed, and the first M positions with the largest load are selected as the final hovering positions, and the method further comprises the following steps:
if the load of the candidate positions of the unmanned aerial vehicles is the same, selecting an area which is closer to the base station as a target position of the unmanned aerial vehicle according to the distance between the hovering position of the unmanned aerial vehicle with the same load and the base station.
An electronic device, comprising:
a storage medium for storing a computer program,
and a processing unit for performing data exchange with the storage medium, wherein the processing unit executes the computer program when performing the unmanned aerial vehicle-mounted edge server, so as to perform the steps of the unmanned aerial vehicle-mounted edge server method.
A computer-readable storage medium:
the computer readable storage medium has a computer program stored therein;
the computer program, when run, performs the steps of the method of the unmanned aerial vehicle to mount an edge server as described above.
The beneficial effects of the application at least comprise:
the method of the application is to grid the map into a plurality of small grids. And then, acquiring the load in each grid from the historical or real-time data, namely, calculating the total amount of task unloading calculation of the Internet of vehicles users in the grid. According to the load data, a heuristic algorithm is used for scheduling the unmanned aerial vehicle base station to move to a specified position to cover the ground load as much as possible so as to achieve the aim of providing low-delay calculation unloading service for the ground vehicle; according to the method, the unmanned aerial vehicle base stations are utilized to provide calculation power, the number and the positions of deployed unmanned aerial vehicle base stations can be flexibly adjusted in a high-speed maneuver mode according to the real-time requirements of the users of the Internet of vehicles, and the resource utilization rate is improved; and the ground pressure is not required to be considered, the installation and configuration are simple and convenient, and the application prospect is wide.
Drawings
FIG. 1 is a schematic illustration of rasterization of a map in a method of the present application;
fig. 2 is a schematic view of an initial position of the unmanned aerial vehicle;
FIG. 3 is a schematic view of a grid g joining unmanned aerial vehicle coverage;
FIG. 4 is a schematic view of sequential addition of grids near grid g to the unmanned coverage area;
fig. 5 is an expanded schematic view of a drone for coverage according to an example;
fig. 6 is an expanded schematic view of the unmanned aerial vehicle performing coverage according to the remaining examples;
fig. 7 is an expanded schematic diagram of the unmanned aerial vehicle continuing to perform coverage according to the remaining examples;
fig. 8 is a schematic diagram of repeated coverage extension of the unmanned aerial vehicle under the condition of existence of calculation power;
FIG. 9 is a schematic view of coverage of an unmanned aerial vehicle when the unmanned aerial vehicle is under-calculated;
FIG. 10 is a schematic diagram of unique movement and load distribution of the current drone due to insufficient computing power;
FIG. 11 is a schematic diagram of unique movement and load distribution of the current drone due to the over-distance end;
FIG. 12 is a schematic diagram of acquiring a maximum load area;
FIG. 13 is a schematic illustration of a dispatch drone according to load;
fig. 14 is a schematic view of dispatching a drone by number with the same load;
FIG. 15 is a graph showing server utilization at 0.2C for four algorithms according to an embodiment of the present application;
FIG. 16 is a graph showing server utilization at 0.6C for four algorithms according to an embodiment of the present application;
FIG. 17 is a graph showing server utilization at C for four algorithms according to an embodiment of the present application;
FIG. 18 shows the average server utilization for four algorithms in an embodiment of the application.
Detailed Description
The application is further described below with reference to the accompanying drawings.
Term interpretation:
edge server: an edge server is a server deployed at the edge of a network that is capable of providing computing, networking, and storage functions. The edge servers enable low latency, high bandwidth and mass access services by storing computation and data close to end users and field applications. It is able to receive a calculation request from an end user, complete the corresponding calculation task, and then return the result to the end user, a process known as calculation offloading. Edge servers are typically deployed at the edges of the internet of vehicles, such as base stations, road side units, various types of mobile units, such as vehicles, drones, etc., and are capable of handling computing requests generated by surrounding end nodes. In addition, the edge server can also communicate with a remote cloud computing and Internet of vehicles control center, upload refined data for summarization and receive configuration instructions of the control center. The appearance of the edge server brings better possibility of computing resource management and application performance optimization for technical applications such as real-time application, large-scale data processing, the Internet of things and the like.
Unmanned aerial vehicle basic station: the unmanned aerial vehicle base station is a mobile communication device, and communication coverage can be realized in the air through the unmanned aerial vehicle. It generally includes a transmitter, a receiver, an antenna, a power supply, etc., and can provide radio signal coverage, support data transmission, call, location tracking, etc. The unmanned aerial vehicle base station can provide mobile communication service for users in areas without ground infrastructure support, such as disaster areas, seas, deserts and the like. In addition, the unmanned aerial vehicle base station can be used for rapidly deploying a temporary communication network, such as providing temporary communication support in the scenes of large-scale activities, emergency rescue and the like. The unmanned aerial vehicle base station can also be integrated with other equipment, such as unmanned aerial vehicle carrying edge servers and the like, so that the application range and the functions of the unmanned aerial vehicle base station are further improved.
Unmanned aerial vehicle carries on edge server: on the basis of the unmanned aerial vehicle base station, the computing capacity is increased, for example, a small server is carried, namely, the unmanned aerial vehicle carrying edge server is formed. The unmanned aerial vehicle carries with the edge server and can move and hover in the internet of vehicles, provide the calculation uninstallation service for ground vehicles.
Specific example I:
once deployed, the fixed site edge computing server in the prior art cannot be changed in position and coverage, dynamic adjustment and expansion become extremely difficult due to limitation of urban land feature, deployment cost and the like, load of space-time dynamic change cannot be effectively tracked and adapted, and the resource utilization rate of the scheme of the existing unmanned aerial vehicle carrying the edge computing server is not high, and most schemes do not consider to dynamically adjust the position of an unmanned aerial vehicle base station according to the space-time change of the load.
Aiming at the defects of the existing scheme, the application aims at:
the unmanned aerial vehicle base station is utilized to provide calculation power, and the number and the positions of deployed unmanned aerial vehicle base stations can be flexibly adjusted according to the real-time requirements of the Internet of vehicles users; the present application provides one embodiment herein.
Specific example I:
the scheme is a deployment scheme of the base station unit of the unmanned aerial vehicle, and comprises the following steps:
step one, rasterizing and refining data;
as shown in fig. 1, a map of a certain area is first selected, the map is divided into a plurality of small squares according to specifications, each small square is called a grid, the actual size is 1km by 1km, and this step is called rasterization of the map.
Next, each grid load for a certain period of time is acquired. A scheme based on history and real-time detection may be used. The history-based scheme uses contemporaneous historical data to replace real-time loads, and performs space-time division on the loads: the vehicle position information is processed according to the already divided grids of the map, so that the time is discretized. Dividing the traffic flow position information into a plurality of time periods according to a certain time interval, and inducing the position information of the corresponding time period and place into the corresponding map grid position. The load of any grid in any period can be obtained by using the number of vehicles as a load measurement index. And finally, arranging the loads of the grids of the map, and simultaneously removing grid positions with zero loads, such as forests and water areas, so as to finally obtain a plurality of grid load sequences in different time periods.
Step two, a mobile station deploys a strategy;
carrying out unmanned aerial vehicle position planning according to the load condition of a given period; based on the rasterized load sequence, the embodiment provides a heuristic algorithm to determine the deployment position and coverage of the unmanned aerial vehicle in a given period. And determining the hovering position of each unmanned aerial vehicle in a certain given period and the coverage range of each unmanned aerial vehicle by taking the computing capacity C, the communication radius R and the number M of the unmanned aerial vehicles of the single unmanned aerial vehicle as constraint conditions. The main idea of the algorithm is to simulate the eating behavior of a person. When a person eats a cake, each mouth starts from the edge of the remaining cake and bites down substantially the same quality of cake. The 'cake eating behavior' selects the optimal solution in each step, thereby finally achieving the purpose of global suboptimal solution. At the same time, small fragments can be avoided in this way. Each drone is responsible for covering a range of fixed sizes "bite down" on the map until all loads on the map are completely covered. This has the advantage that the computational power and communication range of the drone can be maximally utilized while minimizing the number of drones used, thereby saving costs and improving efficiency.
The algorithm proposed in this embodiment is divided into two steps altogether;
the first step, without limiting the number of unmanned aerial vehicles, a batch of candidate positions are determined: simulating cake eating behaviors, sequentially selecting a hovering position and a coverage area of each unmanned aerial vehicle, belonging to grids and corresponding loads of the unmanned aerial vehicle until all loads are distributed;
secondly, arranging unmanned aerial vehicles at all candidate positions in a reverse order according to the load quantity, selecting the first M positions with the largest load as final hovering positions, and dispatching the unmanned aerial vehicles to execute corresponding tasks. If the number of unmanned aerial vehicles required is smaller than M, the number of unmanned aerial vehicles actually required is dispatched.
The detailed steps are as follows:
step 1: determining a candidate list;
1.1, simulating an allocation process of the unmanned plane u, and determining a candidate position. The initial calculation force of the unmanned aerial vehicle is C curr =c, coverage is
1.1.1, starting from the edge of the map, selecting an edge position g, namely the upper left corner of the current map, as a starting point, adding a grid g into the coverage area S=S { g } of the unmanned aerial vehicle, and distributing the load of the grid to the current unmanned aerial vehicle w entirely g =0 and updating the current calculation of the unmanned aerial vehicle to the original calculation minus the intra-grid load C curr ＝C curr -w g 。
1.1.2, then starting from the starting point and expanding towards the outer layer. Attempting to sequentially add surrounding grids surrounding the starting point to the unmanned coverage: the grids surrounding the current range are sequentially added clockwise from the right neighbor grid of the current range.
If the distance from the grid h to any grid in the coverage area exceeds twice the communication radius of the unmanned aerial vehicle, jumping to end the current unmanned aerial vehicle allocation.
If the calculated force is greater than the load C added to the grid curr ≥w h Adding the grid into the current range S=S U { h }, and distributing the grid load to the current unmanned plane w h =0, update the current calculation force to the current calculation force minus the grid load C curr ＝C curr -w h 。
If the current calculation power is smaller than the load C added to the grid curr ＜w h Adding a grid into the current range S=S U { h }, distributing a grid load part to the current unmanned aerial vehicle so as to ensure that the calculation power of the unmanned aerial vehicle is completely used up, updating the grid load to be the grid load minus the current calculation power w of the unmanned aerial vehicle h ＝w h -C curr Updating current calculation power of unmanned aerial vehicle to C curr =0. Jump to "end current drone allocation".
Ending the current unmanned aerial vehicle allocation: considering the load as a mass, the center of gravity of the coverage area S is selected as a candidate location, and all grids with load 0 on the map are deleted.
And 1.2, jumping to 1.1, and repeating the simulation allocation process of the next unmanned aerial vehicle until all loads on the map are allocated.
Step 2: determining an actual dispatch scheme;
and (3) for the positions and coverage areas of the unmanned aerial vehicles in the step (1), sorting according to the descending order of loads distributed to the unmanned aerial vehicles, selecting the first M positions with the largest loads as final hovering positions, and dispatching the unmanned aerial vehicles to execute tasks. If the situation that the loads of the candidate positions of the unmanned aerial vehicles are the same occurs, calculating the distance between the hovering positions of the unmanned aerial vehicles with the same loads and the base station, and preferentially selecting the area which is closer to the base station as the target position of the unmanned aerial vehicle;
in this embodiment, as shown in fig. 2-11, the following specific examples are used to describe the actual allocation scheme in detail:
as shown in fig. 2, a process of allocation of a drone u is simulated to determine a candidate location. Selecting an edge position g from the edges of the map, such as the upper left corner of the current map, as a starting point; adding a grid g into the unmanned aerial vehicle coverage area S=S U { g }, and distributing the load of the grid to the current unmanned aerial vehicle, namely, w g =0; and updating the current calculation power of the unmanned aerial vehicle to the original calculation power minus the load C in the grid curr ＝C curr -w g Then there is an initial calculation force C of unmanned plane A curr =c=60; initial coverage of unmanned aerial vehicle a:
and then extends from the starting point to the outer layer. Attempting to sequentially add surrounding grids surrounding the starting point to the unmanned coverage: sequentially adding grids surrounding the current range clockwise from the neighbor grids on the right side of the current range; as in fig. 3, grid g joins the drone coverage, updates: coverage area s=s { g } of unmanned aerial vehicle a; load of grid g: w (w) g =0; remaining computing power of unmanned aerial vehicle a: c (C) curr ＝C curr -w g =59; as shown in FIG. 4, tasteThe method comprises the steps of (1) trying to sequentially add grids surrounding a current range into the coverage area of an unmanned plane A clockwise from a right neighbor grid of the current range;
attempting to join the grid h, and if the distance from the grid h to any grid in the coverage area exceeds twice the communication radius of the unmanned aerial vehicle, ending the current unmanned aerial vehicle distribution; if the calculated force is greater than the load C added to the grid h curr ≥w h Then, adding the grid h into the current range S=S U { h }, and distributing the grid load to the current unmanned plane w h =0, update the current calculation force to the current calculation force minus the grid load C curr ＝C curr -w h The method comprises the steps of carrying out a first treatment on the surface of the As in fig. 5, the remaining computing power of unmanned aerial vehicle a: c (C) curr =59; load of grid h: w (w) h =5; if C curr ≥w h The grid h is added into the coverage area of the unmanned aerial vehicle; updating the coverage area S=S { h }; load w of grid h h =0; remaining power C of unmanned aerial vehicle A curr ＝C curr -w h =54; as in fig. 6, the remaining computing power C of the unmanned aerial vehicle a curr =54; load w of grid i i =9; if C curr ≥w i Adding a grid i into the coverage of the unmanned aerial vehicle, and updating the coverage S=S { i }; load w of grid i i =0; remaining power C of unmanned aerial vehicle A curr ＝C curr -w i =45; as shown in fig. 7, the remaining computing power C of the unmanned aerial vehicle a curr =54; load w of grid i j =6; if C curr ≥w j Then the grid j is added into the coverage area of the unmanned aerial vehicle; updating the coverage area S=S { j }; load w of grid j j =0; remaining power C of unmanned aerial vehicle A curr ＝C curr -W j =39; the steps of fig. 4-7 are repeated as described above with respect to fig. 8, and steps as described with respect to fig. 9 are performed: remaining power C of unmanned aerial vehicle A curr =1; load w of grid z z =6; if C curr ＜w i The grid Z is added into the coverage area of the unmanned aerial vehicle; updating the coverage area S=S { z }; grid z load:
w z ＝w z -C curr =5; remaining power C of unmanned aerial vehicle A curr ＝0；
The condition determination shown in fig. 10 is performed, and the determination of the condition for the end is made, including: if the current calculation power is smaller than the load C added to the grid curr ＜w h Adding a grid into the current range S=S { h }, enabling a grid load part to be distributed to the current unmanned aerial vehicle, and enabling the computational power of the unmanned aerial vehicle to be completely used up; remaining computing power C of unmanned aerial vehicle a curr =0, ending the current unmanned plane position movement and load distribution; or alternatively, the process may be performed,
as shown in fig. 11, the distance L from the newly added grid to any grid in the coverage area exceeds twice the communication radius R of the drone, i.e. the grid load is updated to the grid load minus the current computing power w of the drone h ＝w h -C curr Updating current calculation power of unmanned aerial vehicle to C curr And (4) ending the current unmanned plane position movement and load distribution.
The method for judging the distance L between the newly added grid and any grid in the coverage area can be as follows: considering the load as a mass, the center of gravity of the coverage area S is selected as a candidate location, and all grids with load 0 on the map are deleted.
Based on the specific embodiment I, when the unmanned aerial vehicle is specifically used, firstly, after traversing, the grid areas are sequenced according to the size of the load, and the unmanned aerial vehicle is dispatched from the largest load bearing grid area, and then the unmanned aerial vehicle is dispatched in sequence. In fig. 12, the numbers of the areas are shown in circles, and the bolded numbers are the loads to be borne by the grid areas. Through traversal comparison, the area with the number of 1 is the largest in load, so that the unmanned aerial vehicle is dispatched to the area with the number of 1; and then, carrying out reverse order arrangement on the load capacity of the residual area, and dispatching the unmanned aerial vehicle to the largest load area. And (5) cycling the steps until all areas are covered or the unmanned aerial vehicle is dispatched, and ending the algorithm.
Then, as shown in fig. 13, the unmanned aerial vehicle is dispatched in the order of load magnitude, and when the situation that the loads of the plurality of areas are the same is faced, the second step is performed, otherwise, the step is skipped. And calculating the distance between the centers of the areas with the same load and the base station, comparing and judging, and dispatching the unmanned aerial vehicle to the area with the small distance preferentially. The area of No. 13 and No. 15 bear the same load as the area of No. 24, and the area of No. 24 is closer to the base station after calculation, so that the unmanned aerial vehicle is dispatched to the area of No. 24.
Finally, the unmanned aerial vehicle is dispatched according to the load size sequence, when the conditions that the loads of a plurality of areas are the same and the distances between the centers of the areas and the base station are the same are faced, a third step is carried out, and otherwise, the step is skipped. The plurality of areas with the same loads and the same distance are compared and ordered, and the areas with the small numbers are dispatched firstly according to the principle from small to large. As shown in fig. 14, the area with the number 15 and the area with the number 18 have the same load, and the unmanned aerial vehicle is dispatched to the area with the number 15 first after comparison.
And (3) verification:
in this embodiment, there are provided server utilization rates of four algorithms under different total calculation capacities, the four algorithms including: the cake eating algorithm Pie-eating, heaviest-first algorithm, fixed Homogeneous algorithm and Fixed Proportional algorithm are described in the application. As shown in fig. 15-17, the utilization of the unmanned aerial vehicle cake-eating algorithm described in this embodiment is always stable and close to 1 at 0.2C, i.e., where the budget for purchasing the calculated capacity is limited. This means that the drone is fully utilized, better utilizing the funds. In contrast, the performance of the other three algorithms fluctuates significantly and is well below 1 for many periods of time; the reference to C is herein made to the experiment where the fixed site Fixed Homogeneous is to meet the minimum computational capacity of the load requirements of all grids.
Because all lattices are satisfied at 0.5C under the unmanned aerial vehicle cake-eating algorithm in the embodiment, unmanned aerial vehicles are not added any more under the condition that C is more than or equal to 0.5C, and the server utilization rate is kept at 80%. The performance of the unmanned plane heaviest priority algorithm is very similar to the fixed ratio method, while the fixed uniformity algorithm does not surprisingly exhibit the worst utilization. The average server utilization of the algorithm over 336 time periods at different computing capacities is summarized in fig. 18, again verifying the superiority of the unmanned aerial vehicle cake-eating algorithm described in this example.
Specific example II:
an electronic device, comprising: a storage medium and a processing unit; the storage medium is used for executing the computer program through the processing unit when the CNN accelerator architecture is optimized, and the steps of the method, the device and the storage medium for carrying out the edge server on the unmanned aerial vehicle according to the specific embodiment I are performed.
The application also provides an embodiment:
a computer-readable storage medium having a computer program stored therein; the computer program, when executed, performs the steps of the method, apparatus and storage medium for loading an edge server on a drone as described in embodiment I.
In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present application, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The foregoing disclosure is merely illustrative of some embodiments of the application, and the application is not limited thereto, as modifications may be made by those skilled in the art without departing from the scope of the application. The above-mentioned inventive sequence numbers are merely for description and do not represent advantages or disadvantages of the implementation scenario.
Claims (10)
1. A method for an unmanned aerial vehicle to mount an edge server, comprising:
selecting a map of a region to be served, and rasterizing the map to obtain a grid corresponding to the map;
the load of each grid in any time period is obtained, and the grid loads in the time period are sequenced to obtain a grid load sequence;
and determining the deployment position and coverage range of the unmanned aerial vehicle for providing service in the time period according to the rasterization load sequence.
2. The method for carrying an edge server on a unmanned aerial vehicle according to claim 1, wherein the determining the deployment location and coverage of the unmanned aerial vehicle providing the service in the period of time according to the rasterized load sequence comprises:
the number of unmanned aerial vehicles is not limited, and according to the grids and the loads corresponding to the grids, the hovering position and the coverage range of each unmanned aerial vehicle are sequentially selected according to the direction from the outer side to the inner side of the map until all loads are distributed;
according to the rasterization load sequence, arranging the unmanned aerial vehicle in a load reverse order, selecting the first M positions with the largest load as final hovering positions, and dispatching the unmanned aerial vehicle to execute corresponding tasks; m is a user set point.
3. The method for loading an edge server on an unmanned aerial vehicle according to claim 2, wherein the sequentially selecting a hover position and a coverage area of each unmanned aerial vehicle according to the grid and a load corresponding to each grid from the outside to the inside of the map comprises:
let unmanned aerial vehicle initial calculation power be C curr =c, coverage is
From the edge of the map, selecting any edge position g as a starting point, and adding the grid g into the unmanned plane coverage area S=S { g }, the loads of the grids are all distributed to the current unmanned plane w g =0 and updating the current calculation of the unmanned aerial vehicle to the original calculation minus the intra-grid load C curr ＝C curr -w g ；
Extending from the starting point g to the outer layer, and sequentially adding grids surrounding the starting point g into the coverage area of the unmanned aerial vehicle;
and (3) setting the grid h which is tried to be added, and comparing the distance from the grid h to any grid in the coverage range with the communication radius of the unmanned aerial vehicle, or using the comparison result of the current calculation power and the load of the grid as the strategy of current unmanned aerial vehicle allocation to obtain the position and the coverage range of any unmanned aerial vehicle.
4. A method for an unmanned aerial vehicle to mount an edge server according to claim 3, wherein:
and if the distance from the grid h to any grid in the coverage range exceeds twice the communication radius of the unmanned aerial vehicle, ending the current unmanned aerial vehicle distribution.
5. A method for an unmanned aerial vehicle to mount an edge server according to claim 3, wherein:
if the calculated force is greater than the load C added to the grid curr ≥w h Adding the grid into the current range S=S U { h }, and distributing the grid load to the current unmanned plane w h =0, update the current calculation force to the current calculation force minus the grid load C curr ＝C curr -w h ；
If the current calculation power is smaller than the load C added to the grid curr ＜w h Adding a grid into the current range S=S U { h }, distributing a grid load part to the current unmanned aerial vehicle so as to ensure that the calculation power of the unmanned aerial vehicle is completely used up, updating the grid load to be the grid load minus the current calculation power w of the unmanned aerial vehicle h ＝w h -C curr Updating current calculation power of unmanned aerial vehicle to C curr And (4) ending the current unmanned aerial vehicle allocation.
6. A method of edge server piggybacking of drones according to claim 4 or 5, wherein said ending the current drone allocation comprises:
the load is taken as a mass, the center of gravity of the coverage area S is selected as a candidate position, and all grids with load 0 on the map are deleted.
7. The method for loading an edge server on an unmanned aerial vehicle according to claim 2, wherein the arranging the unmanned aerial vehicle in a reverse order according to the rasterized load sequence, selecting the first M positions with the largest load as the final hover positions, comprises:
traversing and comparing the load amounts of all grids, and selecting the area with the largest load amount as a target area of the first unmanned aerial vehicle;
and arranging the loading amounts of the remaining grids in a reverse order, and dispatching the remaining unmanned aerial vehicle to the area with the largest loading amount in the remaining grids until all grids are distributed, so as to complete all grid coverage, or dispatching all unmanned aerial vehicles.
8. The method for loading an edge server on an unmanned aerial vehicle according to claim 2, wherein the unmanned aerial vehicle is arranged according to the gridding load sequence in a reverse order of load capacity, and the selecting the first M positions with the largest load as the final hovering positions further comprises:
if the load of the candidate positions of the unmanned aerial vehicles is the same, selecting an area which is closer to the base station as a target position of the unmanned aerial vehicle according to the distance between the hovering position of the unmanned aerial vehicle with the same load and the base station.
9. An electronic device, comprising:
a storage medium for storing a computer program,
a processing unit, performing data exchange with the storage medium, for executing the computer program by the processing unit when performing the unmanned aerial vehicle-mounted edge server, and performing the steps of the unmanned aerial vehicle-mounted edge server method according to any one of claims 1 to 8.
10. A computer-readable storage medium, characterized by:
the computer readable storage medium has a computer program stored therein;
the computer program, when run, performs the steps of the method of the unmanned aerial vehicle-mounted edge server of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310845782.8A CN116781144A (en) | 2023-07-11 | 2023-07-11 | Method, device and storage medium for carrying edge server by unmanned aerial vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310845782.8A CN116781144A (en) | 2023-07-11 | 2023-07-11 | Method, device and storage medium for carrying edge server by unmanned aerial vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116781144A true CN116781144A (en) | 2023-09-19 |
Family
ID=87992903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310845782.8A Pending CN116781144A (en) | 2023-07-11 | 2023-07-11 | Method, device and storage medium for carrying edge server by unmanned aerial vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116781144A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952285A (en) * | 2024-03-27 | 2024-04-30 | 广东工业大学 | Dynamic scheduling method for unmanned aerial vehicle mobile charging station |
-
2023
- 2023-07-11 CN CN202310845782.8A patent/CN116781144A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952285A (en) * | 2024-03-27 | 2024-04-30 | 广东工业大学 | Dynamic scheduling method for unmanned aerial vehicle mobile charging station |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108243245B (en) | Wireless access network based on mixed fog calculation and resource allocation method thereof | |
| CN112381265B (en) | Unmanned aerial vehicle-based charging and task unloading system and task time consumption optimization method thereof | |
| CN110602633B (en) | Explosive flow-oriented mobile edge computing unmanned aerial vehicle cluster auxiliary communication method | |
| CN112104502B (en) | Time-sensitive multitask edge computing and cache cooperation unloading strategy method | |
| CN109213712B (en) | Service providing method and device for machine type communication system and electronic equipment | |
| CN112153145A (en) | Method and device for unloading calculation tasks facing Internet of vehicles in 5G edge environment | |
| CN115361689B (en) | Collaborative deployment method for fixed site and unmanned aerial vehicle-mounted edge server | |
| CN115297171B (en) | Edge computing and unloading method and system for hierarchical decision of cellular Internet of vehicles | |
| CN116781144A (en) | Method, device and storage medium for carrying edge server by unmanned aerial vehicle | |
| CN112512013A (en) | Learning pruning-based vehicle networking mobile edge computing task unloading method and system | |
| CN117255368B (en) | Edge dynamic integration method for vehicle-mounted edge server and cooperative fixed edge server | |
| CN116451934B (en) | Multi-unmanned aerial vehicle edge calculation path optimization and dependent task scheduling optimization method and system | |
| CN116547648A (en) | Method and apparatus for supporting application mobility in a multiple access edge computing platform architecture | |
| Tang et al. | Machine learning for space–air–ground integrated network assisted vehicular network: A novel network architecture for vehicles | |
| CN114363803A (en) | Energy-saving multi-task allocation method and system for mobile edge computing network | |
| Zhang et al. | Mobile edge computing for UAVs | |
| CN112579290B (en) | Computing task migration method of ground terminal equipment based on unmanned aerial vehicle | |
| CN114153515A (en) | Expressway internet of vehicles task unloading algorithm based on 5G millimeter wave communication | |
| CN115955711A (en) | Air-ground 6G network resource allocation method oriented to optimal energy efficiency | |
| Midya et al. | Pso based optimized resource allocation in three tier cloud architecture for vanet | |
| Tian et al. | Deep Reinforcement Learning‐Based Dynamic Offloading Management in UAV‐Assisted MEC System | |
| CN114980127A (en) | Calculation unloading method based on federal reinforcement learning in fog wireless access network | |
| CN112243239A (en) | Unmanned aerial vehicle deployment method based on overpass and related device | |
| CN116546597A (en) | Low-energy-consumption scheduling method for heterogeneous unmanned aerial vehicle auxiliary mobile edge calculation | |
| CN116600342A (en) | Multi-unmanned aerial vehicle edge network task unloading optimization method based on user mobile perception |
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 |