[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN109862592B - Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation - Google Patents

Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation Download PDF

Info

Publication number
CN109862592B
CN109862592B CN201811484129.9A CN201811484129A CN109862592B CN 109862592 B CN109862592 B CN 109862592B CN 201811484129 A CN201811484129 A CN 201811484129A CN 109862592 B CN109862592 B CN 109862592B
Authority
CN
China
Prior art keywords
base station
mobile
task
mobile edge
calculation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811484129.9A
Other languages
Chinese (zh)
Other versions
CN109862592A (en
Inventor
范文浩
苑文
唐碧华
吴帆
刘元安
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Original Assignee
Beijing University of Posts and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Priority to CN201811484129.9A priority Critical patent/CN109862592B/en
Publication of CN109862592A publication Critical patent/CN109862592A/en
Application granted granted Critical
Publication of CN109862592B publication Critical patent/CN109862592B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a resource management and scheduling method in a mobile edge computing environment based on multi-base station cooperation, which comprises the following steps: the mobile edge computing intelligent base station utilizes the five units of receiving, controlling, buffering, computing and transmitting to allocate and schedule resources. When the mobile terminal has a new calculation task, uploading a migration request to the intelligent base station to which the mobile terminal belongs; if the task is determined to be executed in the intelligent base station by the management algorithm and the task data is cached, directly executing the task; if the task data is not cached, the data request of the task is sent to the cloud end; if the algorithm determines to execute the task in the adjacent base station, the intelligent base station sends a calculation migration request to the adjacent base station, and the judgment about the cache is the same as before; and if the algorithm determines to execute the task at the cloud end, the intelligent base station sends the migration request to the cloud end. Therefore, the method of the invention can simultaneously optimize the aspects of transmission and calculation time delay, buffer allocation, system gain and the like.

Description

Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation
Technical Field
The invention relates to the technical field of mobile edge computing, in particular to a resource management method in a mobile edge computing environment based on multi-base-station cooperation, which is used for optimizing system computing time and transmission time overhead under the combined action of a computing migration technology and a data caching technology.
Background
In recent years, with the rapid development of mobile internet and internet of things, the number and variety of mobile information applications, and the breadth and depth of application fields have increased explosively. The functions of a mobile terminal bearing mobile information applications are richer than ever before, and a mobile terminal required by a consumer includes various applications such as augmented reality, virtual reality, live webcasting and the like. The features of mobile terminals have gradually evolved from simple communication tools to a powerful individual integrating communication, entertainment, and office functions.
These typical applications with high computational complexity and high latency sensitivity not only load the mobile cloud in the computing and storage resources, but also cause system network congestion and service quality degradation. Under the promotion of various basic supporting information technologies (mobile intelligent terminals, fifth-generation mobile communication, cloud computing and the like), the end-pipe-cloud information communication industry is deeply transformed to the moving direction, and the contradiction between the continuously increasing service quality requirement of mobile application and the mobile cloud load environment with limited resources is increasingly remarkable. Therefore, Mobile Edge Computing (MEC) is introduced, and is a method for breaking through the Computation and storage resource limitation of a Mobile cloud and reducing the load of the Mobile cloud. Task calculation and data storage are sunk to the mobile edge node through the mobile edge calculation intelligent base station, network load and the requirement on network return bandwidth can be effectively reduced, and service response delay is reduced. When the service entity is located in the intelligent base station, the complicated network nodes in the wired domain can be ignored, and the data interaction between the intelligent base station and the terminal can be completed only through uplink and downlink wireless transmission. Therefore, the intelligent base station can realize quick and sensitive end-base station-cloud interaction, and can greatly improve the user experience quality of the time delay sensitive service.
In the process of implementing the invention, the inventor finds that the following problems exist in the prior art: in the prior art, the calculation time overhead and the transmission time overhead are main problems when a mobile cloud is combined with a mobile edge computing intelligent base station to perform calculation migration and data caching. The research on the problem at home and abroad has the problems of incomplete modeling objects, incomplete model architecture, poor model effect and the like, and in the existing research based on the mobile communication network scene, most of research works only consider the problems of calculation migration or data caching. In joint resource management, too, only the computation migration is combined with the radio resource optimization. However, in most mobile applications, the computing and storage requirements of the terminal are tightly coupled with each other, that is, both the computing process and the data access work are involved in one task, and are related to each other, so that the problem of joint optimization of computing and storage resources needs to be further explored in the research of resource management technology. In the existing research work, most documents are based on the mobile edge computing resource management problem of a single base station, including the single-user problem and the multi-user problem in the single-base station environment. In the architecture design of a future mobile communication system, a cooperation function between base stations is covered, the base stations in the network can be virtualized into a resource whole through close cooperation, and each terminal user in the network is flexibly served in a distributed computing and distributed storage mode.
Disclosure of Invention
Technical problem to be solved
The technical problem to be solved by the invention is as follows: the embodiment of the invention provides a method for scheduling computing resources and data resources of mobile edge computing. The intelligent base station for computing the mobile edge is utilized to realize computing migration and data caching so as to share computing load pressure and data load pressure of the mobile cloud, a new scheme is used for reasonably migrating computing tasks, and task data are cached from the mobile cloud to the intelligent base station supporting the mobile edge computing. Meanwhile, a corresponding intelligent algorithm is designed, the global optimization problems of transmission delay and calculation resource occupation are solved, the calculation migration and data caching probabilities are determined, the overall operation speed of the system is improved, and the optimization of user experience is realized in various scenes.
(II) technical scheme
In order to solve the technical problem, the invention covers the whole system under the mobile cloud environment, and comprises a mobile cloud end, a mobile edge computing intelligent base station, a mobile terminal covered by the mobile edge computing intelligent base station and an adjacent mobile edge computing intelligent base station within a certain distance from the mobile edge computing intelligent base station. According to the invention, the computing and storage pressure of the mobile cloud is relieved by the mobile edge computing intelligent base station. Since each mobile terminal has poor computing power, we neglect the possibility of being a migration destination of the computing task, thereby ensuring the service quality. All tasks of the mobile terminal need to be migrated to a mobile cloud or any adjacent base station supporting mobile edge computing through a mobile edge computing intelligent base station to which the mobile terminal belongs
In a first aspect, the present invention provides a method for deploying a mobile edge computing intelligent base station, where a mobile edge computing intelligent base station is defined to be composed of five units:
a receiving unit: the mobile terminal is used for receiving task processing requests from mobile terminals covered by the mobile edge computing intelligent base station, receiving computing migration execution results from surrounding mobile edge computing intelligent base stations, receiving computing migration execution results and data request execution results from a mobile cloud end, and receiving computing migration requests from surrounding mobile edge computing intelligent base stations;
a control unit: the method comprises the steps that a control unit utilizes a cooperative resource management scheduling algorithm to calculate offline to obtain a migration cache probabilistic model of a certain type of calculation tasks for a certain type of terminals, when a receiving unit receives a task processing request of a mobile terminal, the control unit generates a random number H belonging to [0,1] based on uniform distribution, then H is compared with a migration cache probability vector H, and the calculation execution place of the calculation tasks is determined; when the receiving unit receives a calculation migration request sent by a surrounding mobile edge calculation intelligent base station, the control unit sends a calculation task to the calculation unit for processing; when the receiving unit receives a task calculation request, if the control unit finds that the cache unit does not have data required by a corresponding task at the moment, the control unit requests the corresponding data to the cloud end through the sending unit, after the receiving unit receives the data returned by the cloud end, the control unit generates a random number c [0,1] based on uniform distribution, compares c with a migration cache probability vector H calculated offline, and determines whether the received data needs to be cached in the cache unit of the intelligent base station;
a cache unit: the system comprises a cache module, a data processing module and a data processing module, wherein the cache module is used for caching data required by corresponding tasks based on a cooperative resource management algorithm so as to reduce data access delay to a cloud end;
a calculation unit: the system comprises a mobile terminal, a cloud computing load calculation module and a cloud computing load calculation module, wherein the mobile terminal is covered by a part of mobile edge computing intelligent base stations;
a transmission unit: the mobile terminal is used for sending the calculation result to the mobile terminal, sending the migration request to the mobile cloud or a surrounding mobile edge calculation intelligent base station and sending the data request of the corresponding task to the cloud.
It is worth noting that the mobile edge computing smart base station has limited computing and storage resources and cannot provide computing and caching services for all tasks in the cloud. Therefore, if too many tasks on the mobile terminal are migrated to the mobile edge computing smart base station, an overload phenomenon occurs. In the existing algorithms, the load of the intelligent base station for mobile edge computing is attempted to be relieved, and the computing migration task of the mobile terminal is rejected, delayed or queued, but the algorithms can cause the service quality of the system to be reduced. However, the resource management model provided by the invention can balance the relationship among all the constraint conditions, and minimize the transmission delay and the calculation delay of the system.
In a second aspect, the present invention provides a resource scheduling method, including the steps of:
the method comprises the following steps: when the mobile terminal has a new calculation task, the mobile terminal transmits an uploading migration request to a mobile edge calculation intelligent base station to which the mobile terminal belongs;
step two: if the task is determined to be executed in the intelligent base station for mobile edge computing by the cooperative resource management algorithm, and the intelligent base station for mobile edge computing caches data of the task, the task is directly executed in the intelligent base station for mobile edge computing, and then a migration response is returned to the mobile terminal;
step three: and if the task is determined to be executed in the mobile edge computing intelligent base station and no cache data exists in the mobile edge computing intelligent base station, the mobile edge computing intelligent base station sends a data request of the task to the mobile cloud terminal, and then the cloud terminal responds the return data to the mobile edge computing intelligent base station. The next process of subsequent execution and delivery is the same as step two;
step four: if the algorithm decides to execute the task in the adjacent mobile edge computing intelligent base station, the mobile edge computing intelligent base station sends a migration request to the adjacent mobile edge computing intelligent base station. If the adjacent mobile edge computing intelligent base station has cached computing data before, the adjacent mobile edge computing intelligent base station executes the task and returns a migration response to the mobile edge computing intelligent base station, and then the adjacent mobile edge computing intelligent base station returns the migration response to the mobile terminal;
step five: if the adjacent mobile edge computing intelligent base station decides to execute the task and does not cache computing data of the task in advance, the adjacent mobile edge computing intelligent base station requests data from the cloud end, and then the cloud end returns a data response;
step six: and if the task is determined to be executed by the cloud terminal by the algorithm, the mobile edge computing intelligent base station sends a migration request to the cloud terminal, then the cloud terminal executes the task and returns a migration response to the mobile edge computing intelligent base station, and then the migration response is returned to the mobile terminal.
In the present invention, each sub-task is randomly and independently generated for each mobile terminal, subject to a poisson distribution, eiIndicating the request rate of the mobile terminal, i.e. the mobile terminal generates e per secondiThe request ratio of each task, mobile terminal and each task (Rj) is pij(pij∈[0,1]) And the ratio is different for different tasks. Note that ∑ Sj∈Rpij1 because of pijIn effect representing the proportion of Rj in the task generated by the mobile terminal.
We assume that the request message length and the response message length are fixed and approximately equal in the migration signal transmission or data cache transmission between the mobile terminal, the mobile edge computing smart base station, the adjacent mobile edge computing smart base station, and the mobile cloud. Each task Rj is described by an ordered vector < Cj, Vj, Wj, Zj, Qj >, characterized in that: 1) cj, the amount of computation required to complete Rj; 2) the size of the migration request of Vj, Rj (including the necessary description and parameters of Rj); 3) the size of the migration response of Wj, Rj (including the execution result of Rj); 4) zj, size of data cache request; 5) the size of the data cache response for Qj, Rj (including the data needed to perform Rj).
When Rj migrates to the moving edge computing smart base station, it can be counted at the moving edgeAnd (4) executing the intelligent base station calculation, or further migrating to the adjacent mobile edge intelligent base station (the migration probability of all the adjacent mobile edge intelligent base stations is equal). For task Rj, we define αijL (i belongs to M, j belongs to R) is used for calculating the migration probability of the intelligent base station from the mobile terminal to the mobile edge, and beta is definedijkL (i belongs to M, j belongs to R, K belongs to K) is used for calculating the migration probability of the intelligent base station from the mobile terminal to the adjacent mobile edge, so that the migration probability from the mobile terminal to the mobile cloud end is
Figure GDA0002667384200000061
In addition, we define the cache probability of the intelligent base station calculated from the mobile cloud end to the mobile edge asjL (j belongs to R), calculating the cache probability of the intelligent base station from the mobile cloud end to the adjacent mobile edge to be pijk|(j∈R,k∈K)。
In the present invention, all tasks executing on the mobile cloud share its computing resources. By defining the service rate of the mobile cloud as eta, if Rj executed on the mobile cloud is selected, the time consumed for completing Rj is
Figure GDA0002667384200000062
Figure GDA0002667384200000063
Where the denominator is the stable processing speed (the calculated amount of processing per second).
Figure GDA0002667384200000067
Is the amount of computation per second performed by the mobile cloud task, which is determined to be migrated to the mobile cloud. It can be seen that as the number of tasks performed at the mobile cloud end increases, the processing speed of the mobile cloud end is decreasing, which means that the task arrival rate cannot exceed the service rate of the mobile cloud end.
All tasks migrated from the mobile cloud to the mobile edge computing smart base station share the computing resources of the mobile edge computing smart base station. By movingThe service rate of the edge computing intelligent base station is defined as
Figure GDA0002667384200000064
If Rj is selected to be performed on the moving edge computing smart base station, the time consumed to complete Rj is
Figure GDA0002667384200000065
Figure GDA0002667384200000066
Where the denominator is the speed of the stabilization process (the amount of computation processed per second) of the moving edge computing smart base station.
Figure GDA0002667384200000074
Is the total amount of computation of tasks per second performed by the mobile edge computing smart base station that is determined to be migrated to the mobile edge computing smart base station. It can be seen that as the number of tasks executed by the mobile edge computing smart base station increases, the processing speed of the mobile edge computing smart base station gradually decreases. It should be noted that hard constraints mean that the task arrival rate cannot exceed the service rate of the mobile edge computing smart base station.
All tasks migrated from the mobile cloud to the adjacent mobile edge computing smart base station share the computing resources of the adjacent mobile edge computing smart base station. Calculating the service rate of the intelligent base station by the adjacent mobile edge as
Figure GDA0002667384200000075
If Rj is selected to
Figure GDA0002667384200000076
Execution, then the time consumed to complete Rj is
Figure GDA0002667384200000071
Figure GDA0002667384200000072
Where the denominator is the speed of stabilization processing (the amount of computation processed per second) of the neighboring mobile edge computing smart base station.
Figure GDA0002667384200000077
Is the total amount of computation of tasks per second performed by the neighboring mobile edge computing smart base station that is determined to migrate to the neighboring mobile edge computing smart base station. It can be seen that as the tasks executed by the adjacent moving edge computing smart base stations increase, the processing speed of the adjacent moving edge computing smart base stations gradually decreases. It should be noted that hard constraints mean that the task arrival rate cannot exceed the service rate of the neighboring mobile edge computing smart base station.
All the mobile edge computing intelligent base stations, including the mobile edge computing intelligent base station and the adjacent mobile edge computing intelligent base stations, share the wireless resources under the coverage of the mobile cloud. The present invention ignores the effects of inter-and intra-site interference caused by computational migration because of their very small size. Two communication modes (migration signal transmission and data cache signal transmission) are arranged between the mobile cloud and the mobile edge computing intelligent base station.
The invention defines the signal transmission rate from the mobile edge computing intelligent base station to the mobile cloud end as
Figure GDA0002667384200000073
If Rj is judged by the algorithm to need to compute migration, the time consumed for sending the migration request signal from the mobile edge computing smart base station to the mobile cloud can be defined as the time consumed for computing the migration request signal
Figure GDA0002667384200000081
Figure GDA0002667384200000082
Signal transmission from mobile cloud to mobile edge computing smart base stationAt a rate of
Figure GDA0002667384200000083
The time consumed for receiving the migration response signal from the mobile cloud to the mobile edge computing smart base station can be defined as the time consumed for receiving the migration response signal
Figure GDA0002667384200000084
Figure GDA0002667384200000085
For data caching signal transmission, the mobile edge computing intelligent base station utilizes a cooperative resource management algorithm to decide a resource caching strategy optimized for the task Rj, and the time consumed for sending a data caching request signal from the mobile edge computing intelligent base station to the mobile cloud can be defined as
Figure GDA0002667384200000086
Figure GDA0002667384200000087
By analogy, the time consumed for receiving the data cache response signal from the mobile cloud end to the mobile edge computing intelligent base station can be obtained as
Figure GDA0002667384200000088
Figure GDA0002667384200000089
The adjacent mobile edge computing intelligent base station can also decide whether to cache the computing data requested by the task Rj or not by a cooperative resource management algorithm. The invention defines the transmission rate of data cache signals from an intelligent base station to a mobile cloud end calculated from adjacent mobile edges as
Figure GDA00026673842000000810
The time consumed for sending the Rj request for calculating the data buffer signal from the intelligent base station to the mobile cloud end from the adjacent mobile edge is defined as the time
Figure GDA0002667384200000091
Figure GDA0002667384200000092
The data buffer signal transmission rate of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as
Figure GDA0002667384200000093
The time consumed for receiving the Rj request and computing the data buffer signal response of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as the time consumed for responding to the Rj request
Figure GDA0002667384200000094
Figure GDA0002667384200000095
For data security, we assume that data signals cannot be transmitted between mobile edge computing smart base stations, including mobile edge computing smart base stations and neighboring mobile edge computing smart base stations. Therefore, there is only one type of communication between the moving edge computing smart base station and the neighboring moving edge computing smart base station, i.e., computing mobility signaling. We define the calculated migration signaling rate from a moving edge calculating intelligent base station to an adjacent moving edge calculating intelligent base station as
Figure GDA0002667384200000096
Accordingly, the method can be used for solving the problems that,
Figure GDA0002667384200000097
representing computation migration from neighboring moving edge computing smart base stations to moving edge computing smart base stationsShifting the signal transmission rate.
The time consumed for transmitting the calculation migration request signal of the task Rj from the moving edge calculation intelligent base station to the adjacent moving edge calculation intelligent base station is defined as
Figure GDA0002667384200000098
Figure GDA0002667384200000099
Similarly, the time consumed for transmitting the calculation migration response signal of the task Rj from the adjacent mobile edge calculation intelligent base station to the mobile edge calculation intelligent base station is defined as the time consumed for calculating the migration response signal
Figure GDA0002667384200000101
Figure GDA0002667384200000102
The mobile edge calculates that all mobile terminals under the coverage of the intelligent base station share the wireless resource of the mobile terminals. In this context, we neglect the computing power of each mobile terminal, so there is only one communication between the mobile terminal and the mobile edge computing smart base station, i.e. computing migration signal data transmission.
The invention defines the uplink data transmission rate from the mobile terminal to the mobile edge computing intelligent base station as
Figure GDA0002667384200000103
The time consumed for sending the computation migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station can be defined as the time consumed for computing the migration request of the task Rj
Figure GDA0002667384200000104
Figure GDA0002667384200000105
Defining the downlink data transmission rate from the mobile edge to the mobile terminal as
Figure GDA0002667384200000106
The time consumed for transmitting the calculation migration response of the task Rj from the mobile edge calculation smart base station to the mobile terminal can be defined as the time consumed for transmitting the calculation migration response of the task Rj to the mobile terminal
Figure GDA0002667384200000107
Figure GDA0002667384200000108
The total time spent in the present invention to complete Rj includes: (1) time spent executing in the mobile cloud if Rj is selected for execution in the mobile cloud; (2) calculating the time consumed by migration, and calculating the intelligent base station if Rj is selected to be migrated to the mobile edge; (3) and calculating the time consumed by migration, and if Rj is selected, further migrating to an adjacent mobile edge calculation intelligent base station.
In (1), the time consumption is generated by transmitting an Rj migration request from the mobile terminal to the mobile edge computing smart base station, transmitting an Rj migration request from the mobile edge computing smart base station to the mobile cloud, performing Rj at the mobile cloud, transmitting an Rj migration response from the mobile cloud to the mobile edge computing smart base station, and transmitting the migration response. Rj of the intelligent base station is calculated from the mobile edge to the mobile terminal, i.e.,
Figure GDA0002667384200000111
Figure GDA0002667384200000112
in (2), the time consumption needs to be divided into two cases: 1) the mobile edge computing intelligent base station has data cache required by a task Rj; 2) the mobile edge computing intelligent base station does not have data cache required by the task Rj, and data can be acquired only by sending a data cache request to the mobile cloud. These two cases are discussed separately below:
1) the total time consumption includes calculating the transmission delay of the migration request for the task Rj from the mobile terminal to the mobile edge computing intelligent base station, calculating the calculation delay of the task Rj at the mobile edge computing intelligent base station and calculating the migration response for the task Rj from the mobile edge computing the transmission delay of the intelligent base station to the mobile terminal, that is,
Figure GDA0002667384200000113
2) the total time consumption includes the transmission delay of the computing migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station, the transmission delay of the data cache request of the task Rj from the mobile edge computing smart base station to the mobile cloud, the transmission delay of the data cache response of the task Rj from the mobile cloud to the mobile edge computing smart base station, the computing delay of Rj performed at the mobile edge computing smart base station, and the transmission delay of the computing migration response of the task Rj from the mobile edge computing smart base station to the mobile terminal, that is,
Figure GDA0002667384200000114
Figure GDA0002667384200000115
in (3), the time consumption needs to be divided into two cases: 1) the adjacent mobile edge computing intelligent base station has data cache required by a task Rj; 2) the adjacent mobile edge computing intelligent base station does not have data cache required by the task Rj, and needs to send a data cache request to the mobile cloud to acquire data. These two cases are discussed separately below:
1) the total time consumption comprises the steps of transmitting the calculation migration request of the task Rj from the mobile terminal to the mobile edge calculation intelligent base station for transmission delay, transmitting the calculation migration request of the task Rj from the mobile edge calculation intelligent base station to the adjacent mobile edge calculation intelligent base station for transmission delay, calculating the calculation delay of the task Rj executed by the adjacent mobile edge calculation intelligent base station, and calculating the task RjThe migration response calculates the transmission delay from the intelligent base station to the mobile edge from the adjacent mobile edge, and calculates the transmission delay from the intelligent base station to the mobile terminal from the mobile edge from the calculated migration response of the task Rj, that is,
Figure GDA0002667384200000121
Figure GDA0002667384200000122
2) the total time consumption includes transmitting the compute migration request of task Rj from the mobile terminal to the mobile edge compute smart base station, transmitting the compute migration request of task Rj from the mobile edge compute smart base station to the adjacent mobile edge compute smart base station, transmitting the data cache request of task Rj from the adjacent mobile edge compute smart base station to the mobile cloud, transmitting the data cache response of task Rj from the mobile cloud to the adjacent mobile edge compute smart base station, computing the compute delay of task Rj at the adjacent mobile edge compute smart base station, transmitting the compute migration response of task Rj from the adjacent mobile edge compute smart base station to the mobile edge compute smart base station, and transmitting the compute migration response of task Rj from the mobile edge compute smart base station to the mobile terminal, that is to say that the first and second electrodes,
Figure GDA0002667384200000123
Figure GDA0002667384200000124
in summary, the overall time consumption for completing task Rj is as follows:
Figure GDA0002667384200000131
therefore, the moving edge calculates the overall time consumption of all mobile terminals covered by the smart base station as t:
Figure GDA0002667384200000132
(III) advantageous effects
The invention provides a resource allocation method and a base station service deployment scheme based on a scene of multiple mobile terminals and multiple intelligent base stations. The deployment scheme of the invention comprises a mobile terminal, an intelligent base station to which the mobile terminal belongs, adjacent base stations of the intelligent base stations and a cloud server, and comprehensively considers the whole mobile communication system, divides mobile edge computing application according to tasks, caches data required by high-frequency tasks in a plurality of intelligent base stations according to probability, firstly judges the computing location of the task after the intelligent base station to which the mobile terminal belongs receives a task request sent by the terminal, then determines whether the computing location has data corresponding to the task, if the required data is cached in the computing location, directly executes the corresponding task, otherwise, only needs to acquire the uncached computing data from the cloud server, so that the flexibility of the intelligent base stations is improved, and further the service processing efficiency is improved. The model will simultaneously optimize the contents of transmission delay, calculation delay, buffer allocation, system gain, etc. The optimized migration probability and the data caching probability are calculated as the result after modeling, so that the model does not need to be called for calculation for many times in mobile communication, and the overall efficiency of the system is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a structural diagram of an intelligent base station deployment scheme in mobile edge computing according to the present invention.
Fig. 2 is an overall architecture diagram of a multi-intelligent base station and a multi-mobile terminal in the mobile edge calculation provided by the present invention.
Fig. 3 is a flow chart of cooperative resource management scheduling of multiple intelligent base stations and multiple mobile terminals in mobile edge computing according to the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
As shown in fig. 1, a mobile edge computing smart base station deployment scenario structure diagram according to the present invention. The mobile edge computing intelligent base station comprises five units: the reception unit 101: the mobile terminal is used for receiving a service request from a mobile terminal covered by the mobile edge computing intelligent base station, receiving a computing migration execution result from the surrounding mobile edge computing intelligent base stations, and receiving a computing migration execution result and a data request execution result from the mobile cloud; the control unit 102: if the control unit receives the calculation task request, the control unit is used for determining the execution place of the calculation task; if the control unit receives a data request required by a task, the control unit is used for determining whether the data needs to be cached and the cache location of the data; the calculation unit 103: the system comprises a mobile terminal, a computing unit, a cloud computing unit and a cloud computing unit, wherein the computing unit is used for computing a part of computing tasks requested by the mobile terminal covered by the intelligent base station at a mobile edge so as to reduce the pressure of cloud computing load; the cache unit 104: the system comprises a cache module, a data processing module and a data processing module, wherein the cache module is used for caching data required by corresponding tasks based on a cooperative resource management algorithm so as to reduce data access delay to a cloud end; the transmission unit 105: the mobile terminal is used for sending the calculation result to the mobile terminal, sending the migration request to the mobile cloud or a surrounding mobile edge calculation intelligent base station and sending the data request of the corresponding task to the cloud.
As shown in fig. 2, an overall architecture diagram of a multi-intelligent-base-station multi-mobile-terminal in a mobile edge calculation according to the present invention is shown. The invention relates to an overall system in a mobile CLOUD environment, which comprises a Mobile Terminal (MT)201, a mobile edge computing base station (MEC-BS)202, an adjacent mobile edge computing base station (MEC-NBS)203 within a certain distance from the MEC-BS and a mobile CLOUD end (CLOUD) 204.
The steps of computing migration and data caching under the mobile edge computing environment established by the invention are as follows: 1) when the MT has a new computing task, it will upload a migration request (OL-REQ) to its MEC-BS; 2) if the task is determined by the cooperative resource management algorithm to be executed in the MEC-BS and the MEC-BS buffers data of the task, the task is executed directly in the MEC-BS and then a migration response (OL-RSP) is returned to the MT; 3) if it is determined that the task is performed in the MEC-BS and the data is not buffered in the MEC-BS, the MEC-BS transmits a data request (DA-REQ) of the task to the cloud, and the cloud returns a data response (DA-RSP) to the MEC-BS. The next process of subsequent execution and delivery is the same as 2); 4) if the algorithm decides to perform a task in the MEC-NBS, the MEC-BS will send a migration request to the MEC-NBS. If the MEC-NBS has cached the calculation data before, it will execute the task and return the migration response to the MEC-BS, and then the MEC-BS returns the migration response to the MT; 5) if the MEC-NBS decides to execute a task and it does not pre-cache the task's computational data, it will request data from the cloud and the cloud will return a data response. 6) If the task is determined by the algorithm to be migrated to the cloud for execution, the MEC-BS sends a migration request to the cloud, and then the cloud executes the task and returns a migration response to the MEC-BS, and then returns the migration response to the MT.
Fig. 3 is a flow chart of cooperative resource management scheduling of multiple intelligent base stations and multiple mobile terminals in mobile edge computing according to the present invention. The invention provides a cooperative resource management scheduling of multiple intelligent base stations and multiple mobile terminals, which comprises the following steps:
step 301: the mobile terminal generates a calculation task and sends a task processing request to the edge intelligent base station;
by way of example, each task Rj is described by an ordered vector < Cj, Vj, Wj, Zj, Qj >, characterized in that: 1) cj, the amount of computation required to complete Rj; 2) the size of the migration request of Vj, Rj (including the necessary description and parameters of Rj); 3) the size of the migration response of Wj, Rj (including the execution result of Rj); 4) zj, size of data cache request; 5) the size of the data cache response for Qj, Rj (including the data needed to perform Rj).
Step 302: the method comprises the steps that a migration cache probability vector H under the condition that the total task processing time t is minimum is obtained offline by utilizing a genetic algorithm, when a certain mobile terminal has a new calculation task, aiming at calculation migration, firstly, a mobile edge calculation intelligent base station generates a random number H epsilon [0,1] based on uniform distribution, then, H is compared with the migration cache probability vector H, and if H falls into a representative interval representing the mobile edge calculation intelligent base station, the calculation task is migrated to the belonging edge base station for calculation; if the representative interval of the intelligent base station which represents the moving edge calculation adjacent is included, the calculation task is transferred to the adjacent base station of the edge base station to be calculated; otherwise, the computing task is migrated to the cloud for computing;
step 303: if the calculation task is judged to be executed by the intelligent calculation base station at the belonged mobile edge by the cooperative resource management scheduling algorithm, whether the belonged base station has cache data of the corresponding task or not needs to be judged;
as an example, the present invention defines α for task RjijCalculating the migration probability of the intelligent base station as the migration probability from the mobile terminal to the mobile edge | (i ∈ M, j ∈ R)
Step 304: if the calculation task is judged to be executed by the adjacent base station of the intelligent base station at the belonged moving edge by the cooperative resource management scheduling algorithm, whether the adjacent base station has cache data of the corresponding task or not needs to be judged;
as an example, the present invention defines β for task RjijkL (i belongs to M, j belongs to R, K belongs to K) is used for calculating the migration probability of the intelligent base station from the mobile terminal to the adjacent mobile edge, and correspondingly, the migration probability from the mobile terminal to the mobile cloud is
Figure GDA0002667384200000161
Step 305: when a certain mobile terminal has a new calculation task, aiming at data caching, firstly, generating a random number c which belongs to [0,1] based on uniform distribution, then comparing c with a migration caching probability vector H calculated off-line, and if c falls into a representative interval representing a mobile edge calculation intelligent base station, caching the data of the calculation task to the belonging edge base station; if the data of the calculation task falls into the representative interval representing the moving edge calculation adjacent intelligent base station, the data of the calculation task is cached to the adjacent base station of the edge base station; otherwise, the data of the computing task will not be cached;
as an embodiment, the invention defines the cache probability of the intelligent base station calculated from the mobile cloud to the mobile edge for the task Rj asjL (j belongs to R), calculating the cache probability of the intelligent base station from the mobile cloud end to the adjacent mobile edge to be pijk|(j∈R,k∈K)。
Step 306: and after the running place has data of the corresponding task, executing the calculation task, and returning an execution result to the mobile terminal through the mobile edge calculation intelligent base station to which the execution result belongs.
As an embodiment, in the present invention, all tasks executing on the mobile cloud share its computing resources. By defining the service rate of the mobile cloud as eta, if Rj executed on the mobile cloud is selected, the time consumed for completing Rj is
Figure GDA0002667384200000171
Figure GDA0002667384200000172
All tasks migrated from the mobile cloud to the mobile edge computing smart base station share the computing resources of the mobile edge computing smart base station. Calculating the service rate of the intelligent base station by defining the moving edge as
Figure GDA0002667384200000173
If Rj is selected to be performed on the moving edge computing smart base station, the time consumed to complete Rj is
Figure GDA0002667384200000174
Figure GDA0002667384200000175
From moving toAll tasks migrated from the mobile cloud to the adjacent mobile edge computing intelligent base station share the computing resources of the adjacent mobile edge computing intelligent base station. Calculating the service rate of the intelligent base station by the adjacent mobile edge as
Figure GDA0002667384200000176
If Rj is selected to
Figure GDA0002667384200000177
Execution, then the time consumed to complete Rj is
Figure GDA0002667384200000178
Figure GDA0002667384200000179
All the mobile edge computing intelligent base stations, including the mobile edge computing intelligent base station and the adjacent mobile edge computing intelligent base stations, share the wireless resources under the coverage of the mobile cloud. The present invention ignores the effects of inter-and intra-site interference caused by computational migration because of their very small size. Two communication modes (migration signal transmission and data cache signal transmission) are arranged between the mobile cloud and the mobile edge computing intelligent base station.
The invention defines the signal transmission rate from the mobile edge computing intelligent base station to the mobile cloud end as
Figure GDA0002667384200000181
If Rj is judged by the algorithm to need to compute migration, the time consumed for sending the migration request signal from the mobile edge computing smart base station to the mobile cloud can be defined as the time consumed for computing the migration request signal
Figure GDA0002667384200000182
Figure GDA0002667384200000183
From moving toThe signal transmission rate from the mobile cloud end to the mobile edge computing intelligent base station is
Figure GDA0002667384200000184
The time consumed for receiving the migration response signal from the mobile cloud to the mobile edge computing smart base station can be defined as the time consumed for receiving the migration response signal
Figure GDA0002667384200000185
Figure GDA0002667384200000186
For data caching signal transmission, the mobile edge computing intelligent base station utilizes a cooperative resource management algorithm to decide a resource caching strategy optimized for the task Rj, and the time consumed for sending a data caching request signal from the mobile edge computing intelligent base station to the mobile cloud can be defined as
Figure GDA0002667384200000187
Figure GDA0002667384200000188
By analogy, the time consumed for receiving the data cache response signal from the mobile cloud end to the mobile edge computing intelligent base station can be obtained as
Figure GDA0002667384200000189
Figure GDA00026673842000001810
The adjacent mobile edge computing intelligent base station can also decide whether to cache the computing data requested by the task Rj or not by a cooperative resource management algorithm. The invention defines the transmission rate of data cache signals from an intelligent base station to a mobile cloud end calculated from adjacent mobile edges as
Figure GDA0002667384200000191
The time consumed for sending the Rj request for calculating the data buffer signal from the intelligent base station to the mobile cloud end from the adjacent mobile edge is defined as the time
Figure GDA0002667384200000192
Figure GDA0002667384200000193
The data buffer signal transmission rate of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as
Figure GDA0002667384200000194
The time consumed for receiving the Rj request and computing the data buffer signal response of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as the time consumed for responding to the Rj request
Figure GDA0002667384200000195
Figure GDA0002667384200000196
For data security, we assume that data signals cannot be transmitted between mobile edge computing smart base stations, including mobile edge computing smart base stations and neighboring mobile edge computing smart base stations. Therefore, there is only one type of communication between the moving edge computing smart base station and the neighboring moving edge computing smart base station, i.e., computing mobility signaling. We define the calculated migration signaling rate from a moving edge calculating intelligent base station to an adjacent moving edge calculating intelligent base station as
Figure GDA0002667384200000197
Accordingly, the method can be used for solving the problems that,
Figure GDA0002667384200000198
representing computation from adjacent moving edgesAnd calculating the transfer signal transmission rate from the intelligent base station to the mobile edge.
The time consumed for transmitting the calculation migration request signal of the task Rj from the moving edge calculation intelligent base station to the adjacent moving edge calculation intelligent base station is defined as
Figure GDA0002667384200000199
Figure GDA00026673842000001910
Similarly, the time consumed for transmitting the calculation migration response signal of the task Rj from the adjacent mobile edge calculation intelligent base station to the mobile edge calculation intelligent base station is defined as the time consumed for calculating the migration response signal
Figure GDA00026673842000001911
Figure GDA0002667384200000201
The mobile edge calculates that all mobile terminals under the coverage of the intelligent base station share the wireless resource of the mobile terminals. In this context, we neglect the computing power of each mobile terminal, so there is only one communication between the mobile terminal and the mobile edge computing smart base station, i.e. computing migration signal data transmission.
The invention defines the uplink data transmission rate from the mobile terminal to the mobile edge computing intelligent base station as
Figure GDA0002667384200000202
The time consumed for sending the computation migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station can be defined as the time consumed for computing the migration request of the task Rj
Figure GDA0002667384200000203
Figure GDA0002667384200000204
Defining the downlink data transmission rate from the mobile edge to the mobile terminal as
Figure GDA0002667384200000205
The time consumed for transmitting the calculation migration response of the task Rj from the mobile edge calculation smart base station to the mobile terminal can be defined as the time consumed for transmitting the calculation migration response of the task Rj to the mobile terminal
Figure GDA0002667384200000206
Figure GDA0002667384200000207
The total time spent in the present invention to complete Rj includes: (1) time spent executing in the mobile cloud if Rj is selected for execution in the mobile cloud; (2) calculating the time consumed by migration, and calculating the intelligent base station if Rj is selected to be migrated to the mobile edge; (3) and calculating the time consumed by migration, and if Rj is selected, further migrating to an adjacent mobile edge calculation intelligent base station.
In (1), the time consumption is generated by transmitting an Rj migration request from the mobile terminal to the mobile edge computing smart base station, transmitting an Rj migration request from the mobile edge computing smart base station to the mobile cloud, performing Rj at the mobile cloud, transmitting an Rj migration response from the mobile cloud to the mobile edge computing smart base station, and transmitting the migration response. Rj of the intelligent base station is calculated from the mobile edge to the mobile terminal, i.e.,
Figure GDA0002667384200000208
Figure GDA0002667384200000211
in (2), the time consumption needs to be divided into two cases: 1) the mobile edge computing intelligent base station has data cache required by a task Rj; 2) the mobile edge computing intelligent base station does not have data cache required by the task Rj, and data can be acquired only by sending a data cache request to the mobile cloud. These two cases are discussed separately below:
1) the total time consumption includes calculating the transmission delay of the migration request for the task Rj from the mobile terminal to the mobile edge computing intelligent base station, calculating the calculation delay of the task Rj at the mobile edge computing intelligent base station and calculating the migration response for the task Rj from the mobile edge computing the transmission delay of the intelligent base station to the mobile terminal, that is,
Figure GDA0002667384200000212
2) the total time consumption includes the transmission delay of the computing migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station, the transmission delay of the data cache request of the task Rj from the mobile edge computing smart base station to the mobile cloud, the transmission delay of the data cache response of the task Rj from the mobile cloud to the mobile edge computing smart base station, the computing delay of Rj performed at the mobile edge computing smart base station, and the transmission delay of the computing migration response of the task Rj from the mobile edge computing smart base station to the mobile terminal, that is,
Figure GDA0002667384200000213
Figure GDA0002667384200000214
in (3), the time consumption needs to be divided into two cases: 1) the adjacent mobile edge computing intelligent base station has data cache required by a task Rj; 2) the adjacent mobile edge computing intelligent base station does not have data cache required by the task Rj, and needs to send a data cache request to the mobile cloud to acquire data. These two cases are discussed separately below:
1) the total time consumption comprises the steps of transmitting the calculation migration request of the task Rj from the mobile terminal to the mobile edge calculation intelligent base station for transmission delay, transmitting the calculation migration request of the task Rj from the mobile edge calculation intelligent base station to the adjacent mobile edge calculation intelligent base station for transmission delay, and calculating the intelligent base station at the adjacent mobile edgeThe station performs the calculation of the time delay of the task Rj, calculates the transfer response of the task Rj from the adjacent mobile edge to the transmission time delay of the intelligent base station to the mobile edge, and calculates the transfer response of the task Rj from the mobile edge to the transmission time delay of the intelligent base station to the mobile terminal, that is,
Figure GDA0002667384200000221
Figure GDA0002667384200000222
2) the total time consumption includes transmitting the compute migration request of task Rj from the mobile terminal to the mobile edge compute smart base station, transmitting the compute migration request of task Rj from the mobile edge compute smart base station to the adjacent mobile edge compute smart base station, transmitting the data cache request of task Rj from the adjacent mobile edge compute smart base station to the mobile cloud, transmitting the data cache response of task Rj from the mobile cloud to the adjacent mobile edge compute smart base station, computing the compute delay of task Rj at the adjacent mobile edge compute smart base station, transmitting the compute migration response of task Rj from the adjacent mobile edge compute smart base station to the mobile edge compute smart base station, and transmitting the compute migration response of task Rj from the mobile edge compute smart base station to the mobile terminal, that is to say that the first and second electrodes,
Figure GDA0002667384200000223
Figure GDA0002667384200000224
in summary, the overall time consumption for completing task Rj is as follows:
Figure GDA0002667384200000225
Figure GDA0002667384200000231
therefore, the moving edge calculates the overall time consumption of all mobile terminals covered by the smart base station as t:
Figure GDA0002667384200000232
in summary, the present invention provides a resource allocation method and a base station service deployment scheme based on a scenario of multiple mobile terminals and multiple intelligent base stations, which includes a mobile terminal, an intelligent base station to which the mobile terminal belongs, a base station adjacent to the intelligent base station and a cloud server, comprehensively considers the whole mobile communication system, divides the mobile edge computing application according to tasks and caches data required by high-frequency tasks in multiple intelligent base stations according to probabilities, after the intelligent base station to which the mobile terminal belongs receives a task request sent by the terminal, firstly, the calculation place of the task is judged, then, whether the calculation place has data corresponding to the task or not is determined, if the required data is cached in the calculation place, and directly executing the corresponding task, otherwise, only obtaining the uncached computing data from the cloud server, so that the flexibility of the intelligent base station is improved, and further the service processing efficiency is improved. The model simultaneously optimizes the contents of transmission delay, calculation delay, cache allocation, system gain and the like. The optimized migration probability and the data caching probability are calculated as the result after modeling, so that the model does not need to be called for calculation for many times in mobile communication, and the overall efficiency of the system is improved.
The above embodiments are only for illustrating the invention and are not to be construed as limiting the invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention, therefore, all equivalent technical solutions also belong to the scope of the invention, and the scope of the invention is defined by the claims.

Claims (8)

1. A resource management and scheduling method in a mobile edge computing environment based on multi-base station cooperation comprises the following steps:
step 301: the mobile terminal generates a calculation task and sends a task processing request to the edge intelligent base station;
step 302: a probabilistic model of a certain type of calculation tasks of a certain type of terminals is obtained through offline calculation of a cooperative resource management scheduling algorithm, when a mobile edge calculation intelligent base station receives a task processing request, the intelligent base station generates a random number H epsilon [0,1] based on uniform distribution, then H is compared with a migration cache probability vector H, and the calculation execution place of the calculation tasks is determined;
step 303: if the calculation task is judged to be executed by the intelligent calculation base station at the belonged mobile edge by the cooperative resource management scheduling algorithm, whether the belonged base station has cache data of the corresponding task or not needs to be judged;
step 304: if the calculation task is judged to be executed by the adjacent base station of the intelligent base station at the belonged moving edge by the cooperative resource management scheduling algorithm, whether the adjacent base station has cache data of the corresponding task or not needs to be judged;
step 305: if the operation place judged by the cooperative resource management scheduling algorithm does not have cache data of a corresponding task, requesting the cloud end, returning the corresponding data to a requesting party by the cloud end, generating a random number c [0,1] by the mobile edge computing intelligent base station based on uniform distribution, comparing c with a migration cache probability vector H calculated off-line, and determining whether the data needs to be cached in the intelligent base station;
step 306: and after the running place has data of the corresponding task, executing the calculation task, and returning an execution result to the mobile terminal through the mobile edge calculation intelligent base station to which the execution result belongs.
2. The method of claim 1, wherein for compute migration and data caching of an application running on a mobile terminal, the computation time includes a time delay for running a computation task, a time delay for transmitting a compute migration signal and a data caching signal; the calculation time delay mainly exists in a mobile edge calculation intelligent base station, a mobile edge calculation intelligent base station adjacent base station and a cloud end; in the invention, on the mobile cloudAll tasks executed share its computing resources; representing service rate of mobile cloud end by defining eta, epsiloniIndicating the request rate of the mobile terminal, i.e. the mobile terminal generates e per secondiIndividual task, pij(pij∈[0,1]) Representing the request ratio of the mobile terminal to each task (Rj), cjIndicates the amount of computation, α, required to complete RjijL (i belongs to M, j belongs to R) represents the migration probability, beta, of the intelligent base station calculated from the mobile terminal to the mobile edgeijkL (i belongs to M, j belongs to R, K belongs to K) represents the migration probability of the intelligent base station calculated from the mobile terminal to the adjacent mobile edge, if the Rj executed on the mobile cloud end is selected, the time consumed for completing the Rj is the time
Figure FDA0002692548380000021
Figure FDA0002692548380000022
All tasks transferred from the mobile cloud to the mobile edge computing intelligent base station share the computing resources of the mobile edge computing intelligent base station, and the service rate of the mobile edge computing intelligent base station is defined as
Figure FDA0002692548380000023
If Rj is selected to be performed on the moving edge computing smart base station, the time consumed to complete Rj is
Figure FDA0002692548380000024
Figure FDA0002692548380000025
All tasks migrated from the mobile cloud to the adjacent mobile edge computing smart base station share the computing resources of the adjacent mobile edge computing smart base station by defining the service rate of the adjacent mobile edge computing smart base stationIs composed of
Figure FDA0002692548380000026
If Rj is selected to
Figure FDA0002692548380000027
Execution, then the time consumed to complete Rj is
Figure FDA0002692548380000028
Figure FDA0002692548380000029
The transmission delay mainly exists between the mobile terminal and the mobile edge computing intelligent base station, between the mobile edge computing intelligent base station and the adjacent base stations thereof, and between each mobile edge computing intelligent base station and the cloud, and the signal transmission rate from the mobile edge computing intelligent base station to the mobile cloud is defined as
Figure FDA00026925483800000210
The migration request for Rj has a size vjIf Rj is determined by the algorithm to require computation migration, the time consumed for sending the migration request signal from the mobile edge computing smart base station to the mobile cloud may be defined as the time consumed for computing the migration request signal from the mobile edge computing smart base station to the mobile cloud
Figure FDA00026925483800000211
Figure FDA00026925483800000212
Calculating the signal transmission rate of the intelligent base station from the mobile cloud end to the mobile edge into
Figure FDA00026925483800000213
The magnitude of the migration response of Rj is wjReceiving the intelligent base station from the mobile cloud to the mobile edge computingThe time consumed for the migration response signal of (a) can be defined as
Figure FDA0002692548380000031
Figure FDA0002692548380000032
For data caching signal transmission, the mobile edge computing intelligent base station determines a resource caching strategy optimized for a task Rj by using a cooperative resource management algorithm, wherein the size of a data caching request of the Rj is zjThe time consumed for sending the data cache request signal from the mobile edge computing smart base station to the mobile cloud can be defined as the time consumed for sending the data cache request signal from the mobile edge computing smart base station to the mobile cloud
Figure FDA0002692548380000033
Figure FDA0002692548380000034
By analogy, the time consumed for receiving the data cache response signal from the mobile cloud end to the mobile edge computing intelligent base station can be obtained as
Figure FDA0002692548380000035
Wherein q isjSize of data cache response for Rj:
Figure FDA0002692548380000036
the adjacent mobile edge computing intelligent base station can determine whether to cache the computing data requested by the task Rj or not by a cooperative resource management algorithm, and the transmission rate of data cache signals from the adjacent mobile edge computing intelligent base station to the mobile cloud end is defined as
Figure FDA0002692548380000037
The time consumed for sending the Rj request for calculating the data buffer signal from the intelligent base station to the mobile cloud end from the adjacent mobile edge is defined as the time
Figure FDA0002692548380000038
Figure FDA0002692548380000039
The data buffer signal transmission rate of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as
Figure FDA00026925483800000310
The time consumed for receiving the Rj request and computing the data buffer signal response of the intelligent base station from the mobile cloud end to the adjacent mobile edge is defined as the time consumed for responding to the Rj request
Figure FDA00026925483800000311
Figure FDA0002692548380000041
For data security, we assume that data signals cannot be transmitted between the mobile edge computing intelligent base stations, including the mobile edge computing intelligent base station and the adjacent mobile edge computing intelligent base stations, and therefore, only one communication type, namely, computing migration signal transmission, is between the mobile edge computing intelligent base station and the adjacent mobile edge computing intelligent base stations, and we define the computing migration signal transmission rate from the mobile edge computing intelligent base station to the adjacent mobile edge computing intelligent base stations as
Figure FDA0002692548380000042
Accordingly, the method can be used for solving the problems that,
Figure FDA0002692548380000043
representing a calculated migration signal transmission rate from the neighboring moving edge calculating intelligent base station to the moving edge calculating intelligent base station;
the time consumed for transmitting the calculation migration request signal of the task Rj from the moving edge calculation intelligent base station to the adjacent moving edge calculation intelligent base station is defined as
Figure FDA0002692548380000044
Figure FDA0002692548380000045
Similarly, the time consumed for transmitting the calculation migration response signal of the task Rj from the adjacent mobile edge calculation intelligent base station to the mobile edge calculation intelligent base station is defined as the time consumed for calculating the migration response signal
Figure FDA0002692548380000046
Figure FDA0002692548380000047
All mobile terminals under the coverage of the mobile edge computing intelligent base station share the wireless resources of the mobile terminals, and in the text, the computing capacity of each mobile terminal is ignored, so that only one type of communication exists between the mobile terminals and the mobile edge computing intelligent base station, namely, the data transmission of a computing migration signal;
the invention defines the uplink data transmission rate from the mobile terminal to the mobile edge computing intelligent base station as
Figure FDA0002692548380000048
The time consumed for sending the computation migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station can be defined as the time consumed for computing the migration request of the task Rj
Figure FDA0002692548380000049
Figure FDA00026925483800000410
Defining the downlink data transmission rate from the mobile edge to the mobile terminal as
Figure FDA0002692548380000051
The time consumed for transmitting the calculation migration response of the task Rj from the mobile edge calculation smart base station to the mobile terminal can be defined as the time consumed for transmitting the calculation migration response of the task Rj to the mobile terminal
Figure FDA0002692548380000052
Figure FDA0002692548380000053
3. The method of claim 2, wherein the total time consumption for completing task Rj comprises: (1) time spent executing in the mobile cloud if Rj is selected for execution in the mobile cloud; (2) calculating the time consumed by migration, and calculating the intelligent base station if Rj is selected to be migrated to the mobile edge; (3) calculating the time consumed by migration, and if Rj is selected, further migrating to an adjacent mobile edge to calculate an intelligent base station;
in (1), generating time consumption by transmitting an Rj migration request from a mobile terminal to a mobile edge computing smart base station, transmitting the Rj migration request from the mobile edge computing smart base station to a mobile cloud, performing Rj at the mobile cloud, transmitting an Rj migration response from the mobile cloud to the mobile edge computing smart base station, and transmitting the migration response, Rj of the smart base station is computed from the mobile edge to the mobile terminal, that is,
Figure FDA0002692548380000054
Figure FDA0002692548380000055
in (2), the time consumption needs to be divided into two cases: 1) the mobile edge computing intelligent base station has data cache required by a task Rj; 2) the mobile edge computing intelligent base station does not have a data cache required by the task Rj, and needs to send a data cache request to the mobile cloud to acquire data, and the two situations are respectively discussed as follows:
1) the total time consumption includes calculating the transmission delay of the migration request for the task Rj from the mobile terminal to the mobile edge computing intelligent base station, calculating the calculation delay of the task Rj at the mobile edge computing intelligent base station and calculating the migration response for the task Rj from the mobile edge computing the transmission delay of the intelligent base station to the mobile terminal, that is,
Figure FDA0002692548380000056
wherein,ji (j belongs to R) represents the cache probability of the intelligent base station calculated from the mobile cloud end to the mobile edge;
2) the total time consumption includes the transmission delay of the computing migration request of the task Rj from the mobile terminal to the mobile edge computing smart base station, the transmission delay of the data cache request of the task Rj from the mobile edge computing smart base station to the mobile cloud, the transmission delay of the data cache response of the task Rj from the mobile cloud to the mobile edge computing smart base station, the computing delay of Rj performed at the mobile edge computing smart base station, and the transmission delay of the computing migration response of the task Rj from the mobile edge computing smart base station to the mobile terminal, that is,
Figure FDA0002692548380000061
Figure FDA0002692548380000062
in (3), the time consumption needs to be divided into two cases: 1) the adjacent mobile edge computing intelligent base station has data cache required by a task Rj; 2) the adjacent mobile edge computing intelligent base station does not have a data cache required by the task Rj, and needs to send a data cache request to the mobile cloud to acquire data, and the two cases are respectively discussed as follows:
1) the total time consumption includes calculating a transmission delay of the migration calculation request for the task Rj from the mobile terminal to the mobile edge computing smart base station, calculating a transmission delay of the migration calculation request for the task Rj from the mobile edge computing smart base station to the neighboring mobile edge computing smart base station, calculating a calculation delay of the smart base station for executing the task Rj at the neighboring mobile edge, calculating a transmission delay of the migration calculation response for the task Rj from the neighboring mobile edge computing smart base station to the mobile edge computing smart base station, and calculating a transmission delay of the migration calculation response for the task Rj from the mobile edge computing smart base station to the mobile terminal, that is,
Figure FDA0002692548380000063
Figure FDA0002692548380000064
wherein, pijkL (j belongs to R, K belongs to K) represents the cache probability of the intelligent base station calculated from the mobile cloud end to the adjacent mobile edge;
2) the total time consumption includes transmitting the compute migration request of task Rj from the mobile terminal to the mobile edge compute smart base station, transmitting the compute migration request of task Rj from the mobile edge compute smart base station to the adjacent mobile edge compute smart base station, transmitting the data cache request of task Rj from the adjacent mobile edge compute smart base station to the mobile cloud, transmitting the data cache response of task Rj from the mobile cloud to the adjacent mobile edge compute smart base station, computing the compute delay of task Rj at the adjacent mobile edge compute smart base station, transmitting the compute migration response of task Rj from the adjacent mobile edge compute smart base station to the mobile edge compute smart base station, and transmitting the compute migration response of task Rj from the mobile edge compute smart base station to the mobile terminal, that is to say that the first and second electrodes,
Figure FDA0002692548380000071
Figure FDA0002692548380000072
4. method according to claim 3, characterized in that the overall time consumption for completing a task Rj is as follows:
Figure FDA0002692548380000073
meanwhile, the total task processing time of the system is equal to the sum of all task time delays applied by all mobile terminals in the coverage area of the base station, and can be expressed as:
Figure FDA0002692548380000074
5. the method of claim 4, wherein a migration cache probability vector H under the condition of minimum overall task processing time t can be obtained offline by using a genetic algorithm, when a certain mobile terminal has a new calculation task, for calculation migration, first, the intelligent base station for calculating the moving edge generates a random number H e [0,1] based on uniform distribution, then, H is compared with the migration cache probability vector H, and if H falls into a representative interval representing the intelligent base station for calculating the moving edge, the calculation task is migrated to the edge base station to be calculated; if the representative interval of the intelligent base station which represents the moving edge calculation adjacent is included, the calculation task is transferred to the adjacent base station of the edge base station to be calculated; otherwise, the computing task is migrated to the cloud for computing;
when a certain mobile terminal has a new calculation task, aiming at data caching, firstly, generating a random number c which belongs to [0,1] based on uniform distribution, then comparing c with a migration caching probability vector H calculated off-line, and if c falls into a representative interval representing a mobile edge calculation intelligent base station, caching the data of the calculation task to the belonging edge base station; if the data of the calculation task falls into the representative interval representing the moving edge calculation adjacent intelligent base station, the data of the calculation task is cached to the adjacent base station of the edge base station; otherwise, the data for the computing task will not be cached.
6. A novel base station apparatus with control, computation and storage capabilities located on a mobile edge compute node, comprising:
a receiving unit: the mobile terminal is used for receiving task processing requests from mobile terminals covered by the mobile edge computing intelligent base station, receiving computing migration execution results from surrounding mobile edge computing intelligent base stations, receiving computing migration execution results and data request execution results from a mobile cloud end, and receiving computing migration requests from surrounding mobile edge computing intelligent base stations;
a control unit: the method comprises the steps that a control unit utilizes a cooperative resource management scheduling algorithm to calculate offline to obtain a migration cache probabilistic model of a certain type of calculation tasks for a certain type of terminals, when a receiving unit receives a task processing request of a mobile terminal, the control unit generates a random number H belonging to [0,1] based on uniform distribution, then H is compared with a migration cache probability vector H, and the calculation execution place of the calculation tasks is determined; when the receiving unit receives a calculation migration request sent by a surrounding mobile edge calculation intelligent base station, the control unit sends a calculation task to the calculation unit for processing; when the receiving unit receives a task calculation request, if the control unit finds that the cache unit does not have data required by a corresponding task at the moment, the control unit requests the corresponding data to the cloud end through the sending unit, after the receiving unit receives the data returned by the cloud end, the control unit generates a random number c [0,1] based on uniform distribution, compares c with a migration cache probability vector H calculated offline, and determines whether the received data needs to be cached in the cache unit of the intelligent base station;
a cache unit: the system comprises a cache module, a data processing module and a data processing module, wherein the cache module is used for caching data required by corresponding tasks based on a cooperative resource management algorithm so as to reduce data access delay to a cloud end;
a calculation unit: the system comprises a mobile terminal, a cloud computing load calculation module and a cloud computing load calculation module, wherein the mobile terminal is covered by a part of mobile edge computing intelligent base stations;
a transmission unit: the mobile terminal is used for sending the calculation result to the mobile terminal, sending the migration request to the mobile cloud or a surrounding mobile edge calculation intelligent base station and sending the data request of the corresponding task to the cloud.
7. The method according to any one of claims 1 to 5, wherein probabilistic calculation migration and data caching operations are performed on a certain type of calculation tasks generated by a certain type of mobile terminals according to a migration caching probability vector obtained by a cooperative resource management algorithm in the intelligent base station for mobile edge calculation, so as to achieve the purpose of minimizing the overall task processing time.
8. The novel base station device as claimed in claim 6, wherein probabilistic calculation migration and data caching operations are performed on a certain type of calculation tasks generated by a certain type of mobile terminals according to a migration cache probability vector obtained by a cooperative resource management algorithm in the intelligent base station based on the mobile edge calculation, so as to achieve the purpose of minimizing the overall task processing time.
CN201811484129.9A 2018-12-06 2018-12-06 Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation Active CN109862592B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811484129.9A CN109862592B (en) 2018-12-06 2018-12-06 Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811484129.9A CN109862592B (en) 2018-12-06 2018-12-06 Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation

Publications (2)

Publication Number Publication Date
CN109862592A CN109862592A (en) 2019-06-07
CN109862592B true CN109862592B (en) 2020-11-10

Family

ID=66890593

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811484129.9A Active CN109862592B (en) 2018-12-06 2018-12-06 Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation

Country Status (1)

Country Link
CN (1) CN109862592B (en)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110134507B (en) * 2019-07-11 2019-10-22 电子科技大学 A kind of cooperative computing method under edge calculations system
CN110248206B (en) * 2019-07-29 2020-08-28 北京邮电大学 Resource allocation method and device for edge network system and electronic equipment
CN112422598A (en) * 2019-08-22 2021-02-26 中兴通讯股份有限公司 Resource scheduling method, intelligent front-end equipment, intelligent gateway and distributed system
CN110809043B (en) * 2019-11-01 2023-08-11 腾讯云计算（北京）有限责任公司 Data processing method, device, equipment and storage medium based on Internet of things
CN112882826B (en) * 2019-11-30 2023-03-03 华为技术有限公司 Resource cooperative scheduling method and device
CN111274031B (en) * 2020-01-16 2023-07-25 国家电网有限公司信息通信分公司 Method and device for dynamic migration authentication of end-cloud cooperative edge service
CN111432438B (en) * 2020-03-26 2021-10-15 中国科学院计算技术研究所 Base station processing task real-time migration method
CN111506408B (en) * 2020-04-15 2023-06-16 辽宁大学 Edge computing task scheduling method based on associated data set
CN112003921B (en) * 2020-08-18 2022-07-22 东北大学 Method for actively caching and replacing hot data in edge computing environment
CN112148492B (en) * 2020-09-28 2023-07-28 南京大学 Service deployment and resource allocation method considering multi-user mobility
CN116491159A (en) * 2020-11-23 2023-07-25 中兴通讯股份有限公司 Method and apparatus for joint service of user equipment through radio access network node
US11853810B2 (en) 2021-01-07 2023-12-26 International Business Machines Corporation Edge time sharing across clusters via dynamic task migration based on task priority and subtask result sharing
CN112511570B (en) * 2021-02-07 2021-05-18 浙江地芯引力科技有限公司 Internet of things data integrity checking system and method based on special chip
CN114285854B (en) * 2022-03-03 2022-07-05 成都工业学院 Edge computing system and method with storage optimization and security transmission capability
CN115460147A (en) * 2022-07-29 2022-12-09 新华三技术有限公司 Data processing method and device
CN118118526B (en) * 2024-04-23 2024-07-23 浙江大学 Cloud edge cooperative data acquisition and control method for new energy power station

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107018539A (en) * 2016-01-27 2017-08-04 中兴通讯股份有限公司 The ambulant processing method and processing device of application
CN107734558A (en) * 2017-10-26 2018-02-23 北京邮电大学 A kind of control of mobile edge calculations and resource regulating method based on multiserver
CN107846704A (en) * 2017-10-26 2018-03-27 北京邮电大学 A kind of resource allocation and base station service arrangement method based on mobile edge calculations
CN108076488A (en) * 2016-11-14 2018-05-25 华为技术有限公司 For the methods, devices and systems of cell switching
CN108337691A (en) * 2018-01-19 2018-07-27 南京邮电大学 A kind of services migrating method and system of mobile edge user oriented movement
CN108809695A (en) * 2018-04-28 2018-11-13 国网浙江省电力有限公司电力科学研究院 A kind of distribution uplink unloading strategy towards mobile edge calculations

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10887239B2 (en) * 2015-09-08 2021-01-05 Parallel Wireless, Inc. RAN for multimedia delivery
US10004019B2 (en) * 2015-09-08 2018-06-19 Parallel Wireless, Inc. RAN for multimedia delivery

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107018539A (en) * 2016-01-27 2017-08-04 中兴通讯股份有限公司 The ambulant processing method and processing device of application
CN108076488A (en) * 2016-11-14 2018-05-25 华为技术有限公司 For the methods, devices and systems of cell switching
CN107734558A (en) * 2017-10-26 2018-02-23 北京邮电大学 A kind of control of mobile edge calculations and resource regulating method based on multiserver
CN107846704A (en) * 2017-10-26 2018-03-27 北京邮电大学 A kind of resource allocation and base station service arrangement method based on mobile edge calculations
CN108337691A (en) * 2018-01-19 2018-07-27 南京邮电大学 A kind of services migrating method and system of mobile edge user oriented movement
CN108809695A (en) * 2018-04-28 2018-11-13 国网浙江省电力有限公司电力科学研究院 A kind of distribution uplink unloading strategy towards mobile edge calculations

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《移动边缘计算中移动终端与智能基站资源协同优化技术研究》;刘梦颉;《北京邮电大学硕士学位论文》;20180120;全文 *

Also Published As

Publication number Publication date
CN109862592A (en) 2019-06-07

Similar Documents

Publication Publication Date Title
CN109862592B (en) Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation
CN110198278B (en) Lyapunov optimization method for vehicle networking cloud and edge joint task scheduling
Alvi et al. BEST-MAC: Bitmap-assisted efficient and scalable TDMA-based WSN MAC protocol for smart cities
Zhang et al. Toward transcoding as a service: energy-efficient offloading policy for green mobile cloud
CN108509276A (en) Video task dynamic migration method in edge computing environment
CN107734558A (en) A kind of control of mobile edge calculations and resource regulating method based on multiserver
CN107846704A (en) A kind of resource allocation and base station service arrangement method based on mobile edge calculations
CN111475274B (en) Cloud collaborative multi-task scheduling method and device
CN114637608B (en) Calculation task allocation and updating method, terminal and network equipment
KR20200017589A (en) Cloud server for offloading task of mobile node and therefor method in wireless communication system
CN114363962B (en) Collaborative edge server deployment and resource scheduling method, storage medium and system
Zhang et al. DMRA: A decentralized resource allocation scheme for multi-SP mobile edge computing
Qu et al. Robust offloading scheduling for mobile edge computing
CN112333774B (en) Edge network control system and control method
CN111263401A (en) Multi-user cooperative computing unloading method based on mobile edge computing
Balasubramanian et al. Exploring computing at the edge: A multi-interface system architecture enabled mobile device cloud
Chen et al. DDPG-based computation offloading and service caching in mobile edge computing
Balasubramanian et al. Edge-boost: Enhancing multimedia delivery with mobile edge caching in 5G-D2D networks
KR102298698B1 (en) Method and apparatus for service caching in edge computing network
Zhu et al. Speed-aware and customized task offloading and resource allocation in mobile edge computing
Somesula et al. Cooperative service placement and request routing in mobile edge networks for latency-sensitive applications
Fan et al. Terminalbooster: Collaborative computation offloading and data caching via smart basestations
Yang et al. Optimal task scheduling in communication-constrained mobile edge computing systems for wireless virtual reality
Fan et al. Exploiting joint computation offloading and data caching to enhance mobile terminal performance
CN117376984B (en) Crown block scheduling data transmission method and readable storage medium

Legal Events

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