CN-115905900-B - Unmanned aerial vehicle assisted thermal sensing MEC network task scheduling and resource allocation method

Abstract

The application discloses a unmanned aerial vehicle assisted thermal aware MEC network task scheduling and resource allocation method, which minimizes the cruising time of an unmanned aerial vehicle by jointly optimizing user admittance policies, task scheduling, unmanned aerial vehicle hovering tracks, flight time and calculation and communication resource allocation under the condition of considering CPU thermal constraints. The optimization problem is further decomposed into three sub-problems, and iterative processing is performed sequentially. Simulation results show that compared with other algorithms, the method provided by the application has a better performance improvement effect.

Inventors

• ZHAO MINGXIONG
• HAO YUYU
• BAO LINGYAN
• Ren Daizheng
• Ge Xueqing
• SHI PENG

Assignees

• 云南大学

Dates

Publication Date

20260512

Application Date

20221104

Claims (5)

1. 1. An unmanned aerial vehicle assisted thermal aware MEC network task scheduling and resource allocation method is characterized by comprising the following steps: Step S1, pair Given hover trajectory Q, computing and communication resource allocation Time of flight Obtaining an optimal user access strategy by adopting a K-means clustering algorithm Combining greedy algorithm and B & B algorithm to obtain optimal task scheduling strategy Wherein a greedy algorithm is used to derive forwarding decisions Computing and caching decisions using B & B algorithms Obtaining In the process of obtaining Post-update And according to the new result Designing a scheduling strategy for the task which is not forwarded, so that the task is immediately calculated or temporarily cached; Step S2, pair Based on optimal user admission policy Optimal task scheduling strategy Carry-in Calculating optimal time of flight using CVX solver ; Step S3, pair Converting non-convex constraints to convex constraints using successive convex approximations, at a given global lower limit Optimizing temporary hover trajectory at ith iteration in inner loop of (c) Until the tolerance precision is converged to obtain the best At the same time update again For subsequent optimization; Step S4, pair Using given conditions The next-order Taylor expansion will Converting into a convex problem, and then obtaining optimal calculation and communication resource allocation strategies based on CVX iteration according to auxiliary variables ; Step S5, sequential iterative optimization 、 、 、 Updating related variables to obtain optimal cruising conditions of the MEC network based on the unmanned plane; Wherein 1) User admittance policy and task scheduling policy; 2) Time of flight arrangement; 3) Hover locus optimization; 4) Calculating and communicating resource allocation policies; Step S4 comprises the steps of: Based on the obtained Under the condition of meeting the CPU temperature, calculation and buffer constraints, adopting a CVX solver to solve the following formula: (22a) (22b) (22c) Wherein, the The size of the task is calculated and the task is calculated, (22d) (22e) (22f) (7m) (7n) (7o) (7p) (7q) (7r) (7s) (7t) (20) (21) Optimal communication and computing resource allocation policies are obtained to support data transmission for time-sensitive computing tasks.
2. 2. The method according to claim 1, wherein step S1 comprises the steps of: step S11, obtaining an optimal user access strategy of each hover point by adopting a K-means clustering algorithm ; Step S12 from the offloaded task Selecting a task to be processed, and transmitting the task to be processed to a base station for processing by adopting a greedy algorithm; step S13, obtaining the optimal scheduling decision of calculation or buffer memory by adopting a B & B algorithm 。
3. 3. The method according to claim 2, wherein the step S11 is to divide N observations/users into K clusters/groups by using a K-means clustering method, wherein K is less than or equal to N, each observation belongs to a cluster center with the nearest average value, and the distance between different cluster centers is the farthest; Given a series of user locations And a set of user groups K-means clustering minimizes the sum of squares within the clusters, with the goal of finding: (9) Wherein, the Is a cluster Mean value of the user positions; A two-stage heuristic is used to find the clustering solution to initialize the position of the K-center point: step S111, the nth user is allocated to the nearest center at the t-th iteration based on: Wherein the nth user belongs to the kth cluster; Step S112, updating the cluster with the average value of the user coordinates in the group according to the following formula Is located at the center point of (c): Wherein, the Is the cluster at the t-th iteration The number of users in (a); and step S113, repeating the steps S111-S112 until the clustering is stable.
4. 4. The method according to claim 1, characterized in that for time-sensitive calculation tasks, step S2 is obtained by calculation by a ground based station integration server; The step S1 comprises the steps of making a forwarding decision through a greedy algorithm, then sending the recently offloaded task to a base station, and reducing hover time according to delay requirements; step S114, according to the optimal user admission strategy in step S1 Aggregation for clustered users of jth hover point The representation is made of a combination of a first and a second color, And The size and index of the users in the cluster, respectively; Step S115: And Representing forwarding and local computation sets of the user, respectively, for all From the slave Is the first to select Individual user, and will be The tasks unloaded by the users are forwarded to the base station, and whether the operation meets constraints (7 b) - (7 r) or not is judged: (7b) (7c) (7d) (7e) (7f); (7g) (7h) (7i) (7j) (7k) (7l) (7m) (7n) (7o) (7p) (7q) (7r) if the judgment result is yes, the method comprises the following steps: Then let the ; Step S116, repeating the steps S114-115 until no task can be transferred, and reducing the optimal flight time obtained in the step S2 Deriving the j-th point ; And step S117, repeating the steps S114-117 for all the hovering points to make a forwarding decision.
5. 5. The method according to claim 4, comprising the steps of: step S6, setting initial temperature , And Is the number of internal and external circulation iteration at the current temperature, Initializing hover trajectories for cooling factors from clustered users Each user can access the unmanned aerial vehicle at any hovering point; Step S7 based on the current trajectories satisfying constraints (7 i), (7 k), (7 l), (7 q), (7 r) and (12 b) Generating a hover trajectory in an ith iteration Calculation of , Is given by The cruising time of the unmanned aerial vehicle; (7i) (7k) (7l) (7q) Wherein, the (7r) ; Wherein q j+1 is the coordinate of the UAV at the j+1th hover point, q j is the coordinate of the UAV at the j-th hover point, v j is the flying speed of the UAV between the j-th hover point and the j+1th hover point, v max is the maximum flying speed of the UAV; step S8 if By using Update initial hover trajectory if Probability can be used Will be Replaced by Specifically, a random value Is generated by standard normal distribution when Time setting Otherwise discard 。

Description

Unmanned aerial vehicle assisted thermal sensing MEC network task scheduling and resource allocation method Technical Field The application relates to the technical field of communication, in particular to a heat-sensing MEC network task scheduling and resource allocation method assisted by an unmanned aerial vehicle. Background Because of the uniqueness of Unmanned Aerial Vehicles (UAVs) in terms of mobility and cost, unmanned aerial vehicles are used as airborne MEC (mobile edge computing) servers in the prior art, effectively expanding their service coverage in areas of resource shortage. The unmanned aerial vehicle network can provide various services for the internet of things equipment by adjusting the track of the unmanned aerial vehicle, such as calculation and unloading data acquisition and content caching. However, the computing power, size and weight of the on-board MEC server are relatively less, smaller, lighter, as compared to the MEC server integrated on the base station, subject to hardware costs, user experience design and deployment environment. In order to provide better computing services and to ensure greater maneuverability for the drone, the computing power, size and weight of the onboard MEC server need to be taken into account comprehensively. For example, the Dajiang manifield 2 platform uses Intel Kuui processor i7-8550U with a CPU dominant frequency of 1.8GHz. Meanwhile, DJI Manifold has a length (width) and a thickness of only 11 cm and 2.6 cm, respectively, and its weight is less than 200 g. On the other hand, with the development of semiconductor technology used in chips and the exponential increase of their performance, more and more embedded real-time systems are expected to be implemented on these power density computing platforms, which further brings new challenges to heat dissipation and temperature control of chips. The existing literature on the unmanned aerial vehicle supporting the MEC network mainly focuses on energy consumption of the unmanned aerial vehicle in calculation, hovering or flight, and does not disclose how to solve the problem of excessive CPU temperature in the previous situation. Although the hardware limitation (i.e., the CPU temperature) of the on-board MEC server can be effectively alleviated or solved by DVFS scheduling, and the computing resource allocation efficiency is improved, the bottleneck of computing resources and energy supply still prevents popularization and application of the MEC network supported by the unmanned aerial vehicle. To further improve the efficiency of utilization of communication and computing resources in the MEC network supported by the drone, the drone accesses all hover points in a "hover-flight-hover" mode. While this mode may help minimize total endurance and total energy consumption, or maximize the minimum average throughput for all users, the computing resources of the drone are wasted during flight due to the omission of the computing resources of the drone. Unmanned Aerial Vehicle (UAV) supported Mobile Edge Computing (MEC) can serve areas of resource shortage, such as rural and disaster relief areas. However, the calculation capability and the energy supply of the unmanned aerial vehicle installed by the MEC server are limited, and when the unmanned aerial vehicle is applied to an application with large calculation amount, the serious heat dissipation caused by the limitation of hardware (such as limitation of size and weight) can cause the excessive high temperature of a CPU. In the existing method, the influence of the factor of the CPU temperature on the calculation result of the hover-flight-hover mode and the hover time is not considered, and the minimum hover time of the unmanned aerial vehicle cannot be realized under the condition of considering the factor of the CPU temperature. Disclosure of Invention The application provides an unmanned aerial vehicle assisted thermal sensing MEC network task scheduling and resource allocation method, which is used for solving the technical problems that in the prior art, the computing capacity and energy supply of a mobile edge computing server supported by an unmanned aerial vehicle are limited, when the unmanned aerial vehicle supported mobile edge computing server is applied to an application with large computing capacity for a long time, the CPU temperature is too high due to hardware limitation, and the cruising time of the unmanned aerial vehicle cannot be shortest in the existing hovering-flying-hovering mode. The application provides a unmanned aerial vehicle assisted thermal sensing MEC network task scheduling and resource allocation method, which comprises the following steps: Step S1, pair Given hover trajectory Q, computing and communication resource allocationTime of flightObtaining an optimal user access strategy by adopting a K-means clustering algorithmCombining greedy and B & B algorithm to obtain optimal task scheduling strategyWherein a greedy