CN-122028114-A - Three-layer MEC network resource allocation method and system combining user-small cell-macro cell

CN122028114ACN 122028114 ACN122028114 ACN 122028114ACN-122028114-A

Abstract

The invention discloses a three-layer MEC network resource allocation method and system of a combined user-small base station-macro base station, wherein the method comprises the following specific steps of firstly, acquiring basic configuration information of a network model; the method comprises the steps of establishing an optimization model P1 by taking the time delay of a minimized network model as a target, taking task allocation proportion of user tasks on a user local area, a small base station and a macro base station, a user-small base station matching result and calculation resources of the small base station as optimization variables, setting constraint conditions, decomposing the optimization model in the step II, dividing the optimization model into a matching sub-problem and a resource allocation sub-problem according to different constraint conditions, solving the matching sub-problem in the step III, and solving the resource sub-allocation problem in the step III. The joint optimization scheme effectively solves the problems of task matching and resource allocation under a three-layer MEC network architecture, and improves the time delay performance of the system.

Inventors

DU CHENJIE
Hong Yidian
WANG XUESONG
ZHOU JUNJIE
FENG WEI
XIA XIAOWEI
HE HONG

Assignees

浙江万里学院

Dates

Publication Date: 20260512
Application Date: 20260318

Claims (8)

1. The three-layer MEC network resource allocation method combining the user, the small base station and the macro base station is characterized by comprising the following specific steps: step one, acquiring basic configuration information of a network model; Secondly, taking the time delay of the minimized network model as a target, taking the task allocation proportion of user tasks on the user local area, the small base station and the macro base station, the matching result of the user and the small base station and the computing resource of the small base station as optimization variables, establishing an optimization model P1, and setting constraint conditions; decomposing the optimization model in the second step, and dividing the optimization model into a matching sub-problem and a resource allocation sub-problem according to different constraint conditions; step four, solving the matching sub-problem in the step three; And step five, solving the resource sub-allocation problem in the step three.
2. The method for three-layer MEC network resource allocation of combined user-small cell-macro cell as claimed in claim 1, wherein in the first step, the basic configuration information includes location information of the user, small cell and macro cell, task size, task transmission rate, small cell computing resource required to be processed by the user, and N is used to obtain the base station information Representing the user, using M The small base station is represented and the macro base station is represented by the sequence number 0.
3. The method for allocating three-layer MEC network resources of joint user-small cell-macro cell as claimed in claim 2, wherein in step two, the optimization model P1 and constraint conditions are as follows: (1) Wherein, the Representing the time delay required for user n to complete the task, The result of the matching is indicated, Represents the task division ratio among the local base station, the small base station and the macro base station, Representing the computing resources allocated to user n by small base station m, Representing the total computing resources of the small base station m; Constraint C1 indicates that the time delay required by each user to process the task cannot exceed a set value, constraint C2 indicates that each user is connected with at most one small base station, constraint C3 indicates that the task division ratio and the necessary sum are 1, constraint C4 indicates that the computing resources allocated to the user by the small base station cannot be negative, and constraint C5 indicates that the total computing resources allocated by the small base station cannot exceed the computing resources owned by the small base station.
4. The method for three-layer MEC network resource allocation of a joint user-small base station-macro base station as claimed in claim 3, wherein in step three, the optimization model is decomposed, constraint conditions C1 and C2 are divided into matching sub-problems, and constraint conditions C1, C3-C5 are divided into resource allocation sub-problems.
5. The method for allocating resources of the three-layer MEC network combining the user, the small base station and the macro base station according to any one of claims 1 to 4 is characterized in that in the fourth step, a wireless channel model between the user, the small base station and the macro base station comprehensively considers large-scale path loss and small-scale fading, wherein the small-scale fading adopts a Rayleigh fading model, a matching sub-problem is solved by adopting a three-way matching algorithm, and a result is substituted into the resource allocation sub-problem.
6. The method for three-layer MEC network resource allocation of a joint user-small cell-macro cell as claimed in claim 5 wherein step four is as follows: firstly, each user generates a preference list which contains the cost generated when the user unloads the task to each small base station, ascending order is carried out according to the cost, and each user initially selects the small base station with the minimum cost and sends out a matching request; Secondly, the small base station calculates the priority of the small base station according to the cost data in the user preference list; And finally, performing counter selection on the small base stations, starting from the small base station with the highest priority, if the matching request received by the small base station is smaller than the self load, receiving all the small base stations, and continuing to select by the next small base station, otherwise, the small base station only receives the users with low cost until the self load reaches the upper limit, and the other users forward to the next small base station which does not make selection according to the self preference list, wherein after each small base station makes selection, obtaining an initial matching result, and substituting the initial matching result into the resource allocation sub-problem.
7. The method for allocating three-layer MEC network resources of combined user-small base station-macro base station as claimed in claim 6, wherein in the fifth step, the macro base station covers the whole network scene, the user, the small base station and the macro base station jointly process tasks, and the tasks are processed in parallel, so that the time delay is calculated as an NP difficult problem, and the particle swarm algorithm is adopted for solving.
8. A three-tier MEC network resource allocation system combining a user-small base station-macro base station for performing the method according to any of claims 1-7, comprising in particular the following modules: the information acquisition module is used for acquiring basic configuration information of the network model; The optimization model building module is used for building an optimization model P1 by taking the time delay of the minimized network model as a target, taking the task allocation proportion of user tasks on the user local area, the small base station and the macro base station, the matching result of the user and the small base station and the computing resource of the small base station as optimization variables, and setting constraint conditions; The problem division module is used for decomposing the optimization model and dividing the optimization model into a matching sub-problem and a resource allocation sub-problem according to different constraint conditions; And the problem solving module is used for solving the matching sub-problem and the resource sub-allocation problem.

Description

Three-layer MEC network resource allocation method and system combining user-small cell-macro cell Technical Field The invention belongs to the technical field of mobile edge computing network computing and unloading, and particularly relates to a matching and resource allocation joint optimization method and system under a Mobile Edge Computing (MEC) scene of multiple users and multiple base stations. Background With the wide application of the internet of things, mobile Edge Computing (MEC) is receiving more and more attention from industry as a novel computing architecture close to a data source. The method relies on computing and storage resources deployed at the edge side, and provides efficient and feasible technical support for time delay sensitive scenes. The intelligent network-connected automobile and the automatic driving system can assist in realizing intelligent traffic regulation and control, urban security monitoring and environmental condition monitoring, support low-delay collaborative perception and dynamic path planning in intelligent network-connected automobiles and automatic driving, and can also assist in various applications such as remote monitoring, real-time physiological data analysis, early disease risk discovery and the like in the aspect of medical health. Mobile devices, while equipped with powerful processors, are limited by their own hardware resources, resulting in an inability to independently sustain a large number of latency-intensive computing tasks, and for this reason, mobile edge computing relieves the mobile device of the task burden by offloading some of the tasks to the network edge. However, the performance of mobile edge computing systems is limited by multiple factors such as power allocation, channel conditions, etc., especially in the three-tier model, the complexity of the offloading decisions increases significantly, which presents a continuing challenge for algorithm design. Traditionally, genetic algorithms are often used to solve such problems, but the computational overhead itself is large and the quality of the solution is somewhat random. Based on the method, the invention discloses a method for jointly optimizing task unloading and resource allocation of a multi-user three-layer cellular architecture MEC network, and after an optimization model is established, the problem of task matching is solved by utilizing three-way matching of a user-small base station and a macro base station, and the problem of resource allocation of a finished task is solved by utilizing a particle swarm algorithm. Disclosure of Invention Aiming at the problems in the prior art, the invention discloses a three-layer MEC network resource allocation method and a system for a combined user-small base station-macro base station. Under the condition that the positions of user equipment, a small base station and a macro base station are known, the invention aims to minimize the time delay of a system model, establishes an optimization problem, decomposes the optimization problem into two sub-problems, utilizes a three-way matching algorithm to solve the task matching sub-problem of a decomposable task in the user local area, the small base station and the macro base station, utilizes a particle swarm optimization algorithm to solve the task segmentation proportion and the resource allocation sub-problem required by completing the task, In order to achieve the purpose of the invention, the invention adopts the following technical scheme: a three-layer MEC network resource allocation method combining a user, a small cell and a macro cell comprises the following specific steps: The initialization stage, obtaining basic configuration information of a network model through information interaction; Secondly, taking the time delay of the minimized network model as a target, taking the task allocation proportion of user tasks on the user local area, the small base station and the macro base station, the matching result of the user and the small base station and the computing resource of the small base station as optimization variables, establishing an optimization model P1, and setting constraint conditions; decomposing the optimization model in the second step, and dividing the optimization model into a matching sub-problem and a resource allocation sub-problem according to different constraint conditions; step four, solving the matching sub-problem in the step three; And step five, solving the resource sub-allocation problem in the step three. Preferably, in the first step, the required basic configuration information includes the position information of the user, the small cell and the macro cell, the task size required to be processed by the user, the task transmission rate, the small cell computing resource, and the task of the user adopts a partial unloading mode. Subsequently use NRepresenting the user, using MThe small base station is represented and the macro base station is rep