CN117412262A - Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception - Google Patents
Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception Download PDFInfo
- Publication number
- CN117412262A CN117412262A CN202311473938.0A CN202311473938A CN117412262A CN 117412262 A CN117412262 A CN 117412262A CN 202311473938 A CN202311473938 A CN 202311473938A CN 117412262 A CN117412262 A CN 117412262A
- Authority
- CN
- China
- Prior art keywords
- urllc
- resource allocation
- optimization
- constraint
- unmanned aerial
- 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 46
- 230000006855 networking Effects 0.000 title claims abstract description 7
- 230000008447 perception Effects 0.000 title claims abstract description 5
- 238000005457 optimization Methods 0.000 claims abstract description 29
- 238000013468 resource allocation Methods 0.000 claims abstract description 26
- 230000005540 biological transmission Effects 0.000 claims abstract description 8
- 239000003795 chemical substances by application Substances 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 6
- 238000004891 communication Methods 0.000 claims abstract description 6
- 230000002787 reinforcement Effects 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000007774 longterm Effects 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 9
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 230000014509 gene expression Effects 0.000 claims description 2
- 238000004088 simulation Methods 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
- H04W72/044—Wireless resource allocation based on the type of the allocated resource
- H04W72/0453—Resources in frequency domain, e.g. a carrier in FDMA
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/51—Allocation or scheduling criteria for wireless resources based on terminal or device properties
- H04W72/512—Allocation or scheduling criteria for wireless resources based on terminal or device properties for low-latency requirements, e.g. URLLC
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a decoupling method for supporting the cooperative calculation problem of the air-ground vehicle networking of URLLC perception. The method focuses on a network of Unmanned Aerial Vehicles (UAVs), roadside units base stations (RSUs), cooperative Vehicles (CVs), and User Vehicles (UV), in an effort to optimize unmanned aerial vehicle trajectories, offloading decisions, and computing resources. And calculating UV and server transmission rates through the position information and the OFDMA channel model, and formulating constraint conditions by combining the UV task buffer queue and the URLLC communication requirements. And converting URLLC constraint into a virtual queue stability problem by utilizing the Lyapunov optimization theory, and simultaneously decoupling the total problem. The decoupled problems include offloading decisions, optimization of the unmanned trajectory and transmission delay of channel resource allocation, and optimization of server computing resource allocation. The first sub-problem is solved by multi-agent deep reinforcement learning, and the other problem is solved by adopting a greedy algorithm. The invention converts the problem of long-term task unloading into the problem of short-term optimization, and effectively aims at optimizing challenges in dynamic space.
Description
Technical Field
The invention relates to the field of air-ground vehicle collaborative computing networks, in particular to a decoupling method for air-ground vehicle networking collaborative computing problems perceived by URLLC
Background
With the rapid development of 5G and internet of vehicles (IoV) technology, automobiles have evolved from pure vehicles to human mobile living space. This transition motivates personalized services such as autopilot and online entertainment. With the advent of new applications, the task of vehicle handling has become more and more complex and low latency requirements for autopilot and the like have generally been stringent, which has made it inadequate to rely solely on the onboard processing capabilities of the vehicle.
Cooperative computing network (AVC) utilizing air-to-ground vehicles 2 N) is expected to solve the above-mentioned problems. The network consists of roadside units (RSUs) which are arranged beside roads and have more calculation resources, vehicles which are provided with idle resources around and provide Vehicle Cooperative Calculation (VCC), unmanned aerial vehicles which are flexible to deploy and controllable in cost and can greatly expand the network coverage range.
Existing research efforts on task offloading in space-floor vehicle collaborative computing networks have less consideration for the URLLC communication requirements arising from today's rapid 5G development, as well as lack of consideration for overall system delay. The invention provides a method for decoupling the space-to-ground Internet of vehicles collaborative computing problem perceived by URLLC, which can jointly optimize the track of an unmanned aerial vehicle, unload decisions and resource allocation and achieve the minimum total delay of a system.
Disclosure of Invention
The invention discloses a decoupling method for supporting the cooperative calculation problem of the air-ground vehicle networking of URLLC perception. The invention mainly aims at unmanned aerial vehicle track, unloading decision and computing resource optimization in an air-ground vehicle cooperative computing network consisting of UAVs, RSUs, CVs and UV, and the method comprises the following steps: step one, giving out transmission rate expressions between UV and each server according to UV, CV, UAV and RSU position information and an orthogonal frequency division multiple access channel model, deducing three constraint condition formulas of URLLC by utilizing a UV task buffer queue and combining with the communication requirements of the URLLC, and comprehensively describing task unloading, unmanned plane track planning and resource allocation problems of the system with minimum total delay from end to end; step two, utilizing a Lyapunov optimization theory to convert URLLC constraint into a virtual queue stability problem, and further converting the total problem of long-term and short-term optimized coupling into a series of short-term determinable minimum drift and punishment problems; decoupling the total problem into an unloading decision according to an optimization variable, and optimizing two sub-problems of transmission delay optimization and server-side computing resource allocation by combining unmanned plane tracks and channel resource allocation; and step four, solving the decoupled problem 1 by using a multi-agent deep reinforcement learning-based method due to information uncertainty, and solving the problem 2 by using a greedy algorithm.
The invention provides an air-ground vehicle cooperative computing network (AVC) 2 N) contains K CVs, N RSUs, M UAVs and I UV. Suppose that the drone can communicate with UV using OFDMA. Meanwhile, the receiver is assumed to compensate the Doppler effect caused by the mobility of the unmanned aerial vehicle. Thus, the unloading rate of the UAV may be expressed as
Wherein B is 0 (t) refers to the total channel bandwidth of the present time slot; upsilon (v) i (t) represents the proportion of channel resources allocated to UV i;UV emission power; g 0 UV refers to the power gain of UAV; d, d i,j (t) represents the three-dimensional distance between UV i and UAV j; />Is the noise power. The unloading rates of CV and RSU are
The update formula of the UV local buffer queue is as follows
Q(t+1)＝max{Q(t)+τA(t)-W(t),0}
Wherein A (t) is the arrival rate of the task; w (t) is the off-load task amount. According to the requirement of URLLC communication, setting the excessive backlog quantity of the queue as
Wherein,maximum queue delay that can be tolerated by UV. According to the probability of the occurrence of the overstock, three indexes of the mean and the variance of the overstock are given out
Where ε represents the probability of a tolerable extreme backlog to occur; sigma (sigma) max And xi max To constrain the mean and variance of the overstock. Integrating end-to-end upload, computation, feedback delay and energy consumption, giving the sum of system costs for all UV task offloading in the system as
Wherein the method comprises the steps ofThe method comprises the steps of uploading UV i in t time slots, calculating, feeding back and switching delay required by an unloading server;
task offloading with minimized total system delay, unmanned aerial vehicle trajectory planning and resource allocation problems can be described as follows:
C 3 ：β i，j (t)∈(0，1)j∈[1，K+M+N]
C 5 ：υ i (t)∈(0，1)i∈[1，I]
C 7 -C 9 : unmanned aerial vehicle flight restraint
C 10 -C 12 : URLLC constraint
The optimization targets are task unloading decisions, unmanned plane track planning and frequency spectrum and computing resource allocation; c (C) 1 And C 2 Refers to unloading decision constraints, C 3 And C 4 To calculate resource allocation constraints, C 5 And C 6 Constraint for channel resource allocation, C 7 To C 9 For unmanned aerial vehicle flight constraint, C 10 To C 12 Is a URLLC constraint.
The problem solving can be divided into the following steps:
1) To handle URLLC constraint C 10 To C 12 Three virtual queues are established as follows: and->
2) Subsequent definition ofThe lyapunov function and the drift plus penalty function are constructed in order as follows:
further deriving delta V L[Θ(t)]And converts the original problem into the following single slot optimization problem:
s.t.C 1 ～C 9
3) The total problem is decoupled into two sub-problems according to the optimization variables. The sub-problem 1 is an unloading decision, and the transmission delay optimization of the combination of the unmanned plane track and the channel resource allocation can be described as:
s.t.C 1 ,C 2 ,C 5 ～C 9
problem 2 is server-side computing resource allocation optimization, described as:
s.t.C 3 ,C 4
4) The sub-problem 1 has information uncertainty due to the flight of the unmanned aerial vehicle. The traditional method is difficult to solve, so that the method based on multi-agent deep reinforcement learning can be adopted for solving. And setting UAVs, UV and communication operators as agents, respectively learning flight positions, unloading decisions and spectrum allocation, and finally obtaining the optimal solution of the total rewards maximization.
5) The sub-problem 2 is that each UV served by the edge server allocates computing resources to maximize the objective function, where it can be solved by a greedy algorithm, i.e. the available resources on the edge side are preferentially allocated to the terminals that can maximize the objective function until all computing resources are allocated.
6) Solving the two sub-problems sequentially according to the method to obtain a multidimensional optimization strategy of the system in the time slot, updating the backlog of the task of the next time slot according to the queue, and further solving the optimization problem of the next time slot.
The technical method of the invention has the following advantages:
firstly, task unloading in an air-ground vehicle cooperative computing network and unmanned aerial vehicle flight track and resource allocation are comprehensively considered, so that the total delay of the system under the constraint of URLLC is minimized. Secondly, the problem of unloading long-term queue tasks with URLLC constraint is effectively converted into the problem of optimizing in short-term time slots by utilizing Lyapunov optimization, and the problem of optimizing in dynamic space is solved by utilizing an efficient method after decoupling. Finally, the method can make a strategy of multidimensional optimization variable in each time slot and reasonably control the total delay of the system.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical methods in the prior art, the drawings of the embodiments will be briefly described below. It should be clear that the figures described below represent only some embodiments of the invention, from which other figures can be obtained for a person skilled in the art without having to perform inventive work.
Fig. 1 is a plot of system average delay versus time slot.
Fig. 2 is a plot of local queuing delay versus time slot.
FIG. 3 is a graph showing the variation of the overstock amount with respect to the time slot.
Detailed Description
The invention provides a method for decoupling a space-to-ground Internet of vehicles collaborative computing problem perceived by URLLC, and the embodiment is described in detail below with reference to the accompanying drawings.
The specific implementation scene of the invention is a Beijing real road simulation scene. Consider that on a bi-directional four lane road, there are 6 UV and 3 CV's with traveling speeds randomly distributed within [50,70] km/h. And simultaneously comprises 2 UAVs initially positioned 40m above the two ends of the road and one RSU deployed on the road side.
The simulation time dimension of the invention is 200, the length of each time slot is 0.1s, and the method of the invention is executed in each time slot and obtains the optimized result. The simulation parameters were selected as follows: UV task arrival per time slot of [2,2.5]Or [0.5,1.5 ]]MB; the transmitting power of the UV uploading and the feedback of the server is 20dBm; the power gain of the channel is-50 dB; the noise power is-100 dB; the total available channel bandwidth of each time slot is 50MHz; the computational resources of CV, UAV, RSU are distributed in [30,70 respectively]GHz，[130,170]GHz，[230,270]GHz; the ratio of the feedback data to the uploading data is 0.2; the maximum tolerated queue delay is 0.1s; the switching cost is 0.25s; 1000 CPU cycles are required to calculate one bit of data; v value of 10 3 。
The specific implementation steps are as follows:
1) The offloading rates within the network are calculated from the distribution of UV, CV, UAV and RSU in the scene and the required implementation parameters are generated.
2) And for any time slot, according to the objective function and the constraint condition of the sub-problem 1 after decoupling, obtaining the optimal flight position of the UAV, and the unloading decision and the channel allocation strategy of the UV by a multi-agent deep reinforcement learning method.
3) Sub-problem 2 is established and the proportion of computing resources allocated by each server to the UV offloaded to it is solved by a greedy algorithm.
4) And updating the UV queue backlog according to the optimized result of the time slot.
5) The above-described optimization method is performed at each time slot until the optimization time is over.
Fig. 1 shows the change of the total delay of the system relative to the time slots, and the curve slightly fluctuates in a steady trend, because the distance between the positions of the edge servers relative to the vehicle reaching the task amount per time slot varies to different degrees, so that the unloading delay fluctuates. From the comparison between the methods, the system delay of the method is obviously lower than that of the other two methods, because the method provided by the invention jointly optimizes the unloading decision, optimizes the flight path of the unmanned aerial vehicle and the frequency spectrum, and calculates the resource allocation, thereby realizing the optimal delay performance. Numerical results show that when the time slot is 50, the delay of the method is 27.33% and 55.20% of the unmanned aerial vehicle flight optimization method and the unmanned aerial vehicle resource allocation method respectively.
Fig. 2 and fig. 3 show the local average queuing delay and the local excess backlog amount varying with the time slot, respectively, and it can be seen that after the optimization by the method of the present invention, the queues can be kept at a lower backlog and delay level, no continuously increased queue backlog occurs, and the URLLC can be effectively ensured. Due to the reasonable comprehensive strategy, the queuing delay and the overstock amount of the method are significantly lower than those of the other two methods. Numerical results show that compared with the method without a resource allocation strategy, the method is reduced by 82.41% and 88.58% on average in queuing delay and excessive backlog amount respectively. Compared with the unmanned plane-free flight optimization method, the method is reduced by 92.68% and 97.80% on average.
The foregoing merely represents the preferred embodiments of the disclosure and an explanation of the technical principles used. It is clear to a person skilled in the art that the scope of the invention disclosed herein is not limited to the specific combinations of features described above, but comprises other technical methods, provided that these methods do not deviate from the inventive concepts presented herein. In other words, the different technical methods are constructed by arbitrarily combining the above technical features or features having similar functions, and are not necessarily limited to the specific combination manner and are also encompassed by the present invention. Such as the features described above, have similar functionality as disclosed in (but not limited to) the present disclosure.
Claims (4)
1. The invention discloses a decoupling method for supporting a URLLC perception air-ground vehicle networking collaborative computing problem, which mainly aims at unmanned aerial vehicle tracks in an air-ground vehicle collaborative computing network consisting of UAVs, RSUs, CVs and UV, unloading decisions and computing resource optimization. The method comprises the following steps:
step 1, a transmission rate expression of a UV task unloading channel is given according to UV, CV, UAV, RSU position information and OFDMA channel model, three constraint condition formulas of the URLLC are deduced by utilizing a UV task buffer queue and combining with the communication requirements of the URLLC, and task unloading with minimum system delay is described comprehensively through end-to-end delay, and unmanned plane track planning and resource allocation are achieved;
step 2, converting URLLC constraint into a virtual queue stability problem by utilizing a virtual queue, then constructing a Lyapunov function and a drift plus penalty function, and further converting the total problem of long-term and short-term optimized coupling into a series of short-term determinable drift plus penalty minimization problems;
step 3, decoupling the resolvable problem in the converted time slot into 1) unloading decision according to the optimization variable, and optimizing the transmission delay of combining the unmanned plane track and the channel resource allocation 2) calculating the resource allocation optimization at the server side;
and 4, solving the decoupled problem 1 by using a multi-agent deep reinforcement learning-based method due to information uncertainty, and finally obtaining an optimal solution of the original problem by using a greedy algorithm to solve the problem 2.
2. The decoupling method for the air-ground vehicle networking collaborative computing problem according to claim 1, wherein the task offloading, unmanned aerial vehicle trajectory planning and resource allocation problems with the minimum system delay in step 1 can be described as follows:
C 3 ：β i，j (t)∈(0，1)j∈[1，K+M+N]
C 5 ：υ i (t)∈(0，1)i∈[1，I]
C 7 -C 9 : unmanned aerial vehicle flight restraint
C 10 -C 12 : URLLC constraint
Wherein i is E [1, I]Represents the ith, j E [1, K+M+N ] of the I UV]Refer to the jth of k+m+n servers, where j=1, 2, …, K corresponds to K CVs, j=k+1, k+2, …, k+m represents M UAVs, j=k+m+1, k+m+2, …, k+m+n refers to N RSUs, t represents a slot number; t (T) i (t) is the total delay of the t time slot terminal i, which consists of end-to-end uploading, calculation, feedback and switching delay;refers to the unloading decision of UV in one time slot; /> Refers to the position coordinates of the unmanned aerial vehicle; /> Computing resource allocation policy, beta i，j (t) represents the proportion of computing resources allocated by server j to UV i; />Indicating channel resource allocation strategy, v i (t) represents the proportion of channel resources to which UV i is allocated; c (C) 1 And C 2 Refers to offloading decision constraints, b i，j (t) =1 means that t slots UV i select to offload tasks to server j, and each slot UV selects at most one server; c (C) 3 And C 4 For computing resource allocation constraint, the server gives a ratio of computing resources allocated to one UV to total computing resources between (0, 1), and the sum of the ratios of resources allocated to all UV is less than or equal to 1; c (C) 5 And C 6 For the constraint of channel resource allocation, the ratio of the channel resource allocated by one UV to the total channel bandwidth is (0, 1), and the sum of the ratios of the resources allocated to all the UV is less than or equal to 1; c (C) 7 To C 9 For unmanned aerial vehicle flight constraints, the flight boundary of the UAV and the maximum flight distance per slot are limited, while the minimum distance to prevent unmanned aerial vehicle collisions is considered. C (C) 10 To C 12 The constraint of the excess backlog probability on the UV side and the constraint of the mathematical mean and variance of the excess backlog are respectively represented for URLLC constraint.
3. The system delay minimization problem according to claim 2, the solving step of which is:
first, based on Lyapunov optimization theory, the original problem is transformed into the following short-term optimization problem of minimizing drift plus penalty function:
s.t.C 1 ～C 9
wherein Q is i (t) represents the task volume of UV local buffer queue backlogAnd->The virtual queue amounts respectively refer to queue backlog, excessive backlog mean value and excessive backlog variance; v refers to a parameter that balances queue stability and total system delay; secondly, decoupling the problem into 1) unloading decision by distinguishing optimization variables, and optimizing transmission delay of combining unmanned plane track and channel resource allocation 2) calculating resource allocation optimization at a server side; finally, solving two sub-problems, wherein the problem 1 is solved by a multi-agent deep reinforcement learning-based method, and the problem 2 is solved by a greedy algorithm.
4. According to the method, the problem of unloading the long-term queue task with the URLLC constraint can be effectively converted into the problem of optimizing in a short-term time slot, and the problem of optimizing in a dynamic space can be effectively solved by using an efficient method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311473938.0A CN117412262A (en) | 2023-11-07 | 2023-11-07 | Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311473938.0A CN117412262A (en) | 2023-11-07 | 2023-11-07 | Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117412262A true CN117412262A (en) | 2024-01-16 |
Family
ID=89488842
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311473938.0A Pending CN117412262A (en) | 2023-11-07 | 2023-11-07 | Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117412262A (en) |
-
2023
- 2023-11-07 CN CN202311473938.0A patent/CN117412262A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Overcoming endurance issue: UAV-enabled communications with proactive caching | |
| CN111132077B (en) | Multi-access edge computing task unloading method based on D2D in Internet of vehicles environment | |
| CN109286913B (en) | Energy consumption optimization method of unmanned aerial vehicle mobile edge computing system based on cellular network connection | |
| CN111935303B (en) | Task unloading method based on intention perception in air-ground integrated Internet of vehicles | |
| CN114143355B (en) | Low-delay safety cloud side end cooperation method for power internet of things | |
| CN111757361B (en) | Task unloading method based on unmanned aerial vehicle assistance in fog network | |
| CN110730031A (en) | Unmanned aerial vehicle track and resource allocation joint optimization method for multi-carrier communication | |
| CN111629443B (en) | Optimization method and system for dynamic spectrum slicing frame in super 5G Internet of vehicles | |
| CN112929849B (en) | Reliable vehicle-mounted edge calculation unloading method based on reinforcement learning | |
| CN114650567B (en) | Unmanned aerial vehicle auxiliary V2I network task unloading method | |
| CN114630397B (en) | Unmanned aerial vehicle access selection method based on time slot division | |
| CN115629873A (en) | System and method for controlling unloading of vehicle-road cloud cooperative tasks and stability of task queue | |
| CN113627013A (en) | System throughput maximization method based on unmanned aerial vehicle binary unloading edge calculation | |
| CN112996121A (en) | U2U distributed dynamic resource allocation method for intra-cluster communication | |
| CN114979135B (en) | Unmanned aerial vehicle flight height and task allocation control method based on mobile edge calculation | |
| CN116882270A (en) | Multi-unmanned aerial vehicle wireless charging and edge computing combined optimization method and system based on deep reinforcement learning | |
| CN117412262A (en) | Air-ground vehicle networking collaborative computing problem decoupling method based on URLLC perception | |
| CN116528250A (en) | Unmanned aerial vehicle auxiliary MEC resource optimization method based on NOMA | |
| CN115967430A (en) | Cost-optimal air-ground network task unloading method based on deep reinforcement learning | |
| CN114866979A (en) | Unmanned aerial vehicle-assisted edge calculation time delay minimization method | |
| CN115134370A (en) | Multi-unmanned-aerial-vehicle-assisted mobile edge calculation unloading method | |
| CN115765826A (en) | Unmanned aerial vehicle network topology reconstruction method for on-demand service | |
| CN114513814A (en) | Edge network computing resource dynamic optimization method based on unmanned aerial vehicle auxiliary node | |
| CN117241300B (en) | Unmanned aerial vehicle-assisted general sense calculation network fusion method | |
| CN115665798A (en) | Problem decoupling method for air ground 6G network long-term task unloading optimization |
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 |