CN116647880B - Base station cooperation edge computing and unloading method and device for differentiated power service - Google Patents

Base station cooperation edge computing and unloading method and device for differentiated power service Download PDF

Info

Publication number
CN116647880B
CN116647880B CN202310922331.XA CN202310922331A CN116647880B CN 116647880 B CN116647880 B CN 116647880B CN 202310922331 A CN202310922331 A CN 202310922331A CN 116647880 B CN116647880 B CN 116647880B
Authority
CN
China
Prior art keywords
unloading
task
equipment
base station
offloaded
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
CN202310922331.XA
Other languages
Chinese (zh)
Other versions
CN116647880A (en
Inventor
金燊
邢宁哲
申昉
纪雨彤
赵阳
陈雅琳
张佳乐
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.)
State Grid Corp of China SGCC
State Grid Jibei Electric Power Co Ltd
Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
State Grid Jibei Electric Power Co Ltd
Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd
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 State Grid Corp of China SGCC, State Grid Jibei Electric Power Co Ltd, Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202310922331.XA priority Critical patent/CN116647880B/en
Publication of CN116647880A publication Critical patent/CN116647880A/en
Application granted granted Critical
Publication of CN116647880B publication Critical patent/CN116647880B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/08—Load balancing or load distribution
    • H04W28/09—Management thereof
    • H04W28/0925—Management thereof using policies
    • H04W28/095—Management thereof using policies based on usage history, e.g. usage history of devices
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/0268—Traffic management, e.g. flow control or congestion control using specific QoS parameters for wireless networks, e.g. QoS class identifier [QCI] or guaranteed bit rate [GBR]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Abstract

The invention provides a base station cooperation edge computing and unloading method and device for differentiated power service, which are applied to the technical field of power system automation. The method is applied to first power equipment, the first power equipment is connected with associated equipment, the associated equipment comprises an access base station and a cooperative base station, and the method comprises the following steps: acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading tasks to be unloaded according to an optimal task unloading mode; the base station cooperation edge calculation unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes. According to the method, the optimal task unloading mode determined by the base station cooperative edge calculation unloading model is calculated so as to effectively unload the task to be unloaded.

Description

Base station cooperation edge computing and unloading method and device for differentiated power service
Technical Field
The invention relates to the technical field of power system automation, in particular to a base station cooperation edge computing and unloading method and device for differentiated power service.
Background
With the continuous expansion of the power distribution network, the number of power equipment is also rapidly increasing, and various time delay sensitive and computation intensive power services are continuously emerging.
In an electric power communication network, due to limited computing capacity of base stations, and in addition to uneven distribution of electric power equipment and diversity of electric power traffic types, overload of base stations may occur in an electric power traffic intensive area or in a specific period of time.
The existing unloading method of the corresponding tasks of the power service often adopts a queuing or retransmission mode so as to relieve the pressure of the base station under high load. However, since the quality of service (Quality of Service, qoS) of the power equipment is degraded by queuing and retransmission, how to reasonably utilize the computing resources of the power communication network is a problem to be solved.
Disclosure of Invention
The invention provides a base station cooperation edge computing and unloading method and device for differentiated power service, which are used for constructing a base station cooperation edge computing and unloading model according to equipment energy consumption information and equipment time delay information respectively corresponding to historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, and the computing resources of a power communication network can be reasonably utilized on the premise of guaranteeing the QoS of the time delay sensitive service, so that an optimal task unloading mode corresponding to first power equipment can be determined from different task unloading modes, and the tasks to be unloaded in the first power equipment can be effectively unloaded.
The invention provides a base station cooperation edge computing and unloading method for differentiated power service, which is applied to first power equipment, wherein the first power equipment is connected with associated equipment, the associated equipment comprises an access base station of the first power equipment and a cooperation base station connected with the access base station, and the method comprises the following steps:
acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded;
inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model;
unloading the task to be unloaded according to the optimal task unloading mode;
the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
According to the base station cooperation edge computing and unloading method for the differentiated power service, which is provided by the invention, the base station cooperation edge computing and unloading model is obtained based on the following steps: acquiring a historical task to be offloaded and a state space corresponding to the second power equipment, and an offloading decision action space corresponding to the second power equipment when offloading the historical task to be offloaded; based on the historical to-be-offloaded task, when the to-be-offloaded task is offloaded according to the different task offloading modes, respectively corresponding equipment energy consumption information and equipment time delay information, and taking the accumulated minimum time delay information of the second power equipment as an optimization target, constructing a reward function corresponding to the second power equipment; and constructing a base station cooperation edge calculation unloading model according to the state space, the unloading decision action space and the rewarding function.
According to the base station cooperation edge computing and unloading method for the differentiated power service provided by the invention, when the historical task to be unloaded is unloaded according to the different task unloading modes, respectively corresponding equipment energy consumption information and equipment time delay information, and taking the accumulated minimum time delay information of the second power equipment as an optimization target, constructing a reward function corresponding to the second power equipment, wherein the method comprises the following steps: acquiring corresponding equipment energy consumption information and equipment time delay information respectively when the historical task to be offloaded is offloaded according to the different task offloading modes; constructing accumulated minimum delay information of the second power equipment by taking the energy consumption information of the plurality of equipment and the delay information of the plurality of equipment as constraint conditions; and constructing a reward function corresponding to the second power equipment according to the accumulated minimum time delay information.
According to the base station cooperation edge calculation unloading method for the differentiated power service provided by the invention, the base station cooperation edge calculation unloading model is constructed according to the state space, the unloading decision action space and the rewarding function, and the method comprises the following steps: determining an initial Q value based on the state space, the offload decision action space, and the reward function; s1, determining a target Q value according to the initial Q value, a state space at the next moment and an unloading decision action space at the next moment; determining the target Q value as a new initial Q value, and repeatedly executing the step S1 until the difference between the target Q value and a preset Q value is minimum; and determining an unloading model corresponding to the target Q value with the minimum difference as the base station cooperation edge calculation unloading model.
According to the base station cooperation edge computing and unloading method for the differentiated power service provided by the invention, when the historical task to be unloaded is unloaded according to different task unloading modes, corresponding equipment energy consumption information and equipment time delay information respectively are obtained, and the method comprises the following steps: under the condition that the task unloading mode is the mode of unloading in the associated equipment, acquiring the data quantity of the historical task to be unloaded and the CPU period required for completing the historical task to be unloaded; determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to the unloading mode in the associated equipment according to the data volume and the CPU period; under the condition that the task unloading mode is the local processing mode, acquiring a CPU period required for completing the historical task to be unloaded; and determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to the local processing mode according to the period of the central processing unit.
According to the base station cooperation edge computing and unloading method for differentiated power service provided by the invention, the state space corresponding to the second power equipment is acquired, and the method comprises the following steps: acquiring the current residual capacity of the second power equipment and the current calculation capacity corresponding to the associated equipment connected with the second power equipment; and determining a state space corresponding to the second power equipment according to the current residual electric quantity and the current calculation capacity.
The invention also provides a base station cooperation edge computing and unloading device facing differentiated power service, which is applied to first power equipment, wherein the first power equipment is connected with associated equipment, the associated equipment comprises an access base station of the first power equipment and a cooperation base station connected with the access base station, and the device comprises:
the acquisition module is used for acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded;
the processing module is used for inputting the current state information into the base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
The invention also provides power equipment, which comprises a memory, an unloader and a computer program stored in the memory and capable of running on the unloader, wherein the unloader realizes the base station cooperation edge computing unloading method facing the differentiated power service when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by an unloader, implements a base station cooperative edge computing unloading method for differentiated power services as described in any one of the above.
The invention also provides a computer program product comprising a computer program which when executed by an unloader implements a base station cooperation edge computing unloading method for differentiated power services as described in any one of the above.
The invention provides a base station cooperation edge computing unloading method and device for differentiated power service, which are applied to first power equipment, wherein the first power equipment is connected with associated equipment, the associated equipment comprises an access base station of the first power equipment and a cooperation base station connected with the access base station, and the method is used for acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge calculation unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, and the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment. According to the method, according to the equipment energy consumption information and the equipment time delay information which correspond to the historical to-be-offloaded tasks respectively when the to-be-offloaded tasks are offloaded according to different task offloading modes, the built base station cooperation edge calculation offloading model can reasonably utilize the calculation resources of the power communication network on the premise of guaranteeing the time delay sensitive service QoS, and the optimal task offloading mode corresponding to the first power equipment is determined from the different task offloading modes so as to effectively offload the to-be-offloaded tasks in the first power equipment.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a scenario of a base station cooperative edge computing and unloading method for differentiated power service provided by the invention;
fig. 2 is a schematic flow chart of a base station cooperative edge computing and unloading method for differentiated power service provided by the invention;
FIG. 3 is a schematic flow chart of a Q-learning offloading decision algorithm provided by the present invention;
fig. 4 is a schematic structural diagram of a base station cooperative edge computing and unloading device for differentiated power service provided by the invention;
fig. 5 is a schematic structural diagram of the power device provided by the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Fig. 1 is a schematic view of a scenario of a base station cooperation edge computing and unloading method for differentiated power service. In fig. 1, the power equipment is connected to an access base station, which is connected to a cooperative base station, the number of which is 4.
The power device may also be referred to as a User Equipment (UE), and refers to a device that processes a task corresponding to a power service, where the task may also be referred to as a task to be offloaded/a computing task.
Optionally, the power device may include at least one of: distribution automation (Power distribution automation) equipment, power generation capacity prediction (Power generation forecast) equipment, power load monitoring (Power load monitoring) equipment and the like.
Because of the diversity of the power equipment, the power business corresponding to each power equipment also has certain variability.
The access base station and the cooperative base station in fig. 1 may be collectively referred to as an association device of the above-mentioned power device. Each base station deploys a server.
Alternatively, the server may be a Multi-access edge computing (Multi-access Edge Computing, MEC) server, which has a sink cache function and a computing function, and is configured to perform computing tasks that need to be offloaded in the power device.
The connection mode between the power equipment and the access base station and the connection mode between the access base station and the cooperative base station are both wireless communication technologies.
Alternatively, the wireless communication technology may include, but is not limited to, one of the following: fourth generation communication technology (the 4 Generation mobile communication technology,4G) and fifth generation communication technology (the 5 Generation mobile communication technology,5G), and the like.
It should be noted that, the execution subject related to the embodiment of the present invention may be a base station cooperation edge computing and unloading device for differentiated power service, and may also be a power device.
The embodiment of the present invention is further described below by taking a first power device as an example, where the first power device is connected to an associated device, and the associated device may include an access base station of the first power device and a cooperative base station connected to the access base station.
As shown in fig. 2, a flow chart of a base station cooperative edge computing and unloading method for differentiated power service provided by the invention may include:
201. and acquiring current state information corresponding to the first power equipment when unloading the task to be unloaded.
Alternatively, the current state information s may include: the remaining power e of the first power device and the current calculation capacity c of the corresponding associated device of the first power device.
Under the scene facing the differentiated power service, when the first power equipment unloads the task to be unloaded, the current state information corresponding to the first power equipment can be acquired first so as to accurately determine the optimal task unloading mode corresponding to the task to be unloaded.
202. And inputting the current state information into the base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model.
The base station cooperation edge calculation unloading model is obtained by training equipment energy consumption information E and equipment time delay information T which correspond to historical tasks to be unloaded respectively when the tasks are unloaded according to different task unloading modes.
The equipment energy consumption information E refers to energy consumption generated when the target equipment corresponding to the task unloading mode unloads the task to be unloaded.
The device time delay information T refers to total delay time generated when the target device corresponding to the task unloading mode unloads the task to be unloaded.
The task unloading manner may include: local processing means and means for offloading in the associated device. The manner of offloading in the associated device may include: the manner of offloading in the access base station and the manner of offloading in the cooperative base station.
After the first power equipment acquires the current state information corresponding to the first power equipment, the current state information can be input into a pre-built base station cooperation edge calculation unloading model, so that an optimal task unloading mode output by the base station cooperation edge calculation unloading model can be obtained, and the tasks to be unloaded in the first power equipment can be effectively unloaded based on the optimal task unloading mode.
In some embodiments, the base station cooperative edge computation offload model may be derived based on the steps of: the method comprises the steps that a first power device acquires a historical task to be offloaded and a state space corresponding to a second power device, and an offloading decision action space corresponding to the second power device when offloading the historical task to be offloaded; when the first power equipment is used for unloading according to different task unloading modes based on historical tasks to be unloaded, respectively corresponding equipment energy consumption information and equipment time delay information, and constructing a reward function corresponding to the second power equipment by taking accumulated minimum time delay information of the second power equipment as an optimization target; the first power equipment constructs a base station cooperation edge calculation unloading model according to the state space, the unloading decision action space and the rewarding function.
Wherein the first electrical device belongs to the second electrical device. It will be appreciated that the first power device is a device that applies a base station cooperative edge computing offload model and the second power device is a device that trains the base station cooperative edge computing offload model.
First, the first power device may construct a system state space corresponding to the plurality of second power devices based on the power communication network. Wherein the power communication network may include:second power device +.>Individual base stations, this->The power equipment set corresponding to the second power equipment is available +.>Indicating that this->Available to individual base stationsAnd (3) representing.
The system state space corresponding to the plurality of second power devices is availableA representation;representation->Residual capacity corresponding to the second power equipment, < >>Representing the current residual capacity corresponding to the ith second power equipment in the plurality of second power equipment; />Representing the current computing capacity of a plurality of second power devices corresponding to the associated devices, the number of associated devices being +.>Personal (S)>Representing the current computing capacity of the mth base station in all associated devices.
Second, the communication period is discretized in time and the time slotIn the method, each second power device only has one historical task to be offloaded to be processed, and each second power device is randomly distributed in the coverage area of the base station. Each second power device has a computation-intensive power business, i.e., a history of tasks to be offloaded that are available to be completed Indicating (I)>Representing the data volume of historical tasks to be offloaded; />The CPU (Central Processing Unit, CPU) period required by the completion of the historical task to be offloaded is represented, and the CPU period required by different historical task to be offloaded is large in difference; />Representing the maximum tolerated delay of the historical task to be offloaded.
The first power device then sets the historical offloaded task to binary offloaded, which means that each second power device should execute the historical offloaded task by local processing or offloading in an associated device, and cannot split the historical offloaded task. Wherein the first power device can adopt binary variableRepresenting historical task waiting to be offloaded->Whether or not to be processed locally and at +.>Under the condition of (1), the history task to be offloaded is described to be processed in a local processing mode; the first power device may employ a binary variable +.>Representing historical task waiting to be offloaded->Whether or not by the access base station->Processed and at->In the case of (a) the history task to be offloaded is described by +.>Unloading in a medium unloading mode; the first power device may employ binary variablesRepresenting historical task waiting to be offloaded- >Whether or not by the cooperative base station->Processed and at->In the case of (a) specifying historical tasks to be offloaded by +.>Unloading is carried out in a way of unloading. In this way, since each history task to be offloaded corresponds to only one task offloading mode, the task offloading mode corresponding to the ith second power device is available +.>And (3) representing.
Then, the first power equipment acquires respective corresponding task unloading modes of a plurality of second power equipment at the current moment to acquire a task decision strategy, and further can determine unloading decision action spaces corresponding to the plurality of second power equipment, wherein the unloading decision action spaces are availableA representation; />Showing an unloading decision action space corresponding to an ith second power device of the plurality of second power devices,/and>，/>and representing the set of all task unloading modes at the current moment.
Finally, for the ith second power device in the plurality of second power devices, the first power device may acquire a historical task to be offloaded and a state space corresponding to the second power deviceAnd the corresponding unloading decision action space when the second power equipment unloading history is to unload tasks +.>The method comprises the steps of carrying out a first treatment on the surface of the The first power equipment constructs a reward function corresponding to the second power equipment by taking accumulated minimum delay information of the second power equipment as an optimization target based on the equipment energy consumption information and the equipment delay information respectively corresponding to the historical to-be-offloaded tasks when the historical to-be-offloaded tasks are offloaded according to different task offloading modes >The method comprises the steps of carrying out a first treatment on the surface of the The first power device is dependent on the state space +.>Unloading decision action space->And bonus function->And constructing a base station cooperation edge calculation unloading model so as to be used for determining an optimal task unloading mode of the historical task to be unloaded by the first power equipment.
In some embodiments, when the power device acquires the device energy consumption information and the device time delay information corresponding to the historical task to be offloaded when the task to be offloaded is offloaded according to different task offloading modes, the method may include: under the condition that the task unloading mode is the mode of unloading in the associated equipment, the first power equipment acquires the data volume of the historical task to be unloaded and the CPU period required by completing the historical task to be unloaded; determining corresponding equipment energy consumption information and equipment time delay information when a historical task to be offloaded is offloaded according to the method of offloading in the associated equipment according to the data volume and the period of the central processing unit; under the condition that the task unloading mode is a local processing mode, the first power equipment acquires a CPU period required by completing the historical task to be unloaded; and determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to a local processing mode according to the period of the central processing unit.
For different task unloading modes, the process of the device energy consumption information corresponding to the second power device acquired by the first power device is different, and the process of the device time delay information corresponding to the second power device acquired by the first power device is also different.
Specifically, in the case that the task unloading mode is a mode of unloading in the associated device, the first power device may determine device energy consumption information and device time delay information corresponding to the historical task to be unloaded according to a data amount of the historical task to be unloaded and a period of the central processor required for completing the historical task to be unloaded; and under the condition that the task unloading mode is a local processing mode, the power equipment can determine the equipment energy consumption information and the equipment time delay information corresponding to the historical task to be unloaded only according to the period of the central processing unit.
Optionally, when the task unloading mode of the first power device is a local processing mode, determining, according to a period of the central processing unit, corresponding device energy consumption information and device delay information when the historical task to be unloaded is unloaded according to the local processing mode, may include: the first power equipment determines corresponding equipment energy consumption information of a historical task to be offloaded when the task to be offloaded is offloaded according to a local processing mode according to a first energy consumption formula; and determining corresponding equipment delay information of the historical task to be offloaded when the task to be offloaded is offloaded according to the local processing mode according to the first delay formula.
Wherein, the first energy consumption formula is:；
the first delay formula is:；
the method comprises the steps of representing equipment energy consumption information corresponding to historical tasks to be offloaded when the tasks to be offloaded are offloaded according to a local processing mode;representing the effective switched capacitance depending on the chip architecture; />Representing the CPU period required for completing the historical task to be offloaded;representing a computing power of the second power device; />And the device delay information corresponding to the historical task to be offloaded when the task to be offloaded is offloaded according to the local processing mode is represented.
Optionally, when the task unloading mode of the first power device is a mode of unloading in the associated device and the associated device is an access base station, determining, according to the data amount and the period of the central processor, corresponding device energy consumption information when the historical task to be unloaded is unloaded in the mode of unloading in the associated device, may include: and the first power equipment determines corresponding equipment energy consumption information of the historical task to be offloaded when the task to be offloaded is offloaded in the access equipment according to the second energy consumption formula.
Wherein, the second energy consumption formula is:；
the method comprises the steps of representing equipment energy consumption information corresponding to historical tasks to be offloaded when the tasks to be offloaded are offloaded in an access equipment; / >Representing the transmitting power of the second power equipment when transmitting data to the access base station; />Representing historical tasks to be offloaded in an access deviceThe corresponding transmission delay information is carried out in an unloading mode; />Representing the data volume of historical tasks to be offloaded; />Representing the transmission rate/communication rate at which data is transmitted between the second power device and the access base station.
Optionally, when the task unloading mode of the first power device is a mode of unloading in the associated device and the associated device is an access base station, determining, according to the data amount and the period of the central processor, corresponding device delay information when the historical task to be unloaded is unloaded in the mode of unloading in the associated device, may include: the first power equipment determines transmission delay information corresponding to a historical task to be offloaded when the task to be offloaded is offloaded in an access equipment according to a transmission delay formula; according to an unloading delay formula, corresponding unloading delay information of a historical task to be unloaded when unloading is carried out in an access device in an unloading mode is determined; and the first power equipment determines corresponding equipment delay information of the historical task to be offloaded when the task to be offloaded is offloaded in the access equipment according to the second delay formula.
Wherein, the transmission delay formula is:；
the unloading delay formula is:；
the second delay formula is:；
representing unloading delay information; />Representing the computing power of the access base station; />And the device delay information corresponding to the historical task to be offloaded when the task to be offloaded is offloaded in the access device is represented.
It should be noted that, if the second power device selects to calculate the historical task to be offloaded by offloading to the access base station, first, the second power device uploads the relevant data to the access base station through the wireless communication technology, and the MEC server in the access base station may allocate a part of the calculation resources to execute the historical task to be offloaded, and then, the MEC server returns the calculation result. Because the data volume of the calculation result is often far smaller than the data volume of the historical task to be offloaded, the downlink rate of the data is generally very high, and therefore the time delay information transmitted back to the second power equipment by the calculation result can be ignored.
In addition, since the computing power of each access base station is limited, it is possible, therefore,indicating that the calculation task load of the second power equipment when the historical task to be offloaded is offloaded to the access base station for processing cannot exceed the maximum calculation capacity of the access base station, < + > >Representing the maximum computational capacity.
Optionally, when the task unloading mode of the first power device is a mode of unloading in the associated device and the associated device is a cooperative base station, determining, according to the data amount and the period of the central processor, corresponding device energy consumption information when the historical task to be unloaded is unloaded in the mode of unloading in the associated device, may include: and the first power equipment determines corresponding equipment energy consumption information of the historical task to be offloaded when the task to be offloaded is offloaded in the cooperative equipment according to the third energy consumption formula.
Wherein, the third energy consumption formula is:；
and the device energy consumption information corresponding to the historical task to be offloaded when the task to be offloaded is offloaded in the cooperative device is represented.
Optionally, when the task unloading mode of the first power device is a mode of unloading in the associated device and the associated device is a cooperative base station, determining, according to the data amount and the period of the central processor, corresponding device delay information when the historical task to be unloaded is unloaded in the mode of unloading in the associated device, may include: and the first power equipment determines corresponding equipment energy consumption information of the historical task to be offloaded when the task to be offloaded is offloaded in a mode of being offloaded in the cooperative base station according to a third time delay formula.
The third time delay formula is as follows:；/>。
representing transmission delay information between base stations; />Indicating the transmission rate at which data is transmitted between base stations.
Assuming that the power communication system corresponding to the power communication network adopts orthogonal frequency division multiple access (Orthogonal Frequency Division Multiple Access, OFDMA), each base station equally distributes subcarriers to the second power devices associated with the base station, at this time, no inter-device interference exists between the second power devices, and data transmission is performed between the base stations through optical fiber networking, and the first power device can accurately determine according to the following conditionsTransmission delay information between base stations.
In addition, in the base station cooperative unloading mode, when the second power equipment transmits the historical task to be unloaded to the cooperative base station for processing, the equipment time delay information consists of three parts, namely the transmission time delay of the second power equipment and the access base station, the transmission time delay between the base stations and the processing time delay, and at the moment, the first power equipment can accurately determine the corresponding equipment energy consumption information of the historical task to be unloaded when unloading is performed in the cooperative base station according to the third time delay formula.
In some embodiments, when the first power device performs unloading according to different task unloading modes based on the historical task to be unloaded, the device energy consumption information and the device delay information respectively corresponding to the task to be unloaded, and the accumulated minimum delay information of the second power device is used as an optimization target, and constructing a reward function corresponding to the second power device may include: the method comprises the steps that first power equipment acquires corresponding equipment energy consumption information and equipment time delay information respectively when historical tasks to be offloaded are offloaded according to different task offloading modes; the first power equipment takes the energy consumption information of a plurality of equipment and the time delay information of a plurality of equipment as constraint conditions to construct the accumulated minimum time delay information of the second power equipment; and the first power equipment constructs a reward function corresponding to the second power equipment according to the accumulated minimum time delay information.
In the process of constructing a reward function corresponding to the second power equipment, the first power equipment can firstly acquire equipment energy consumption information and equipment time delay information respectively corresponding to historical tasks to be offloaded when the historical tasks to be offloaded are offloaded according to different task offloading modes, namely, how many equipment energy consumption information and how many equipment time delay information exist in the task offloading modes; and then, the power equipment firstly builds the accumulated minimum time delay information of the second power equipment according to all the equipment energy consumption information and the equipment time delay information, and further builds a reward function corresponding to the second power equipment so as to prepare a base station cooperation edge calculation unloading model with higher subsequent training accuracy.
Optionally, the first power device uses the energy consumption information of the multiple devices and the time delay information of the multiple devices as constraint conditions, and constructing the accumulated minimum time delay information of the second power device may include: and the first power equipment determines the accumulated minimum delay information of the second power equipment according to the optimization formula.
Wherein, the optimization formula is:；
constraint conditions of the optimization formulaThe method comprises the following steps:
first constraint:；
second constraint:；
third constraint:；
fourth constraint: The method comprises the steps of carrying out a first treatment on the surface of the The method comprises the steps of,
fifth constraint:。
it should be noted that the first constraint condition is an offloading decision constraint, which indicates that the second power device is unique to an offloading decision of the historical task to be offloaded.
The second constraint indicates that the offloading policy is binary offloading.
The third constraint represents a maximum computational capacity constraint of the associated device.
The fourth constraint condition indicates that the computational task load of the second power device when the to-be-offloaded task offload value is processed by the access base station cannot exceed the maximum computational capacity (computational load) of the access base station.
The fifth constraint indicates that the total latency information of the power traffic cannot exceed the maximum tolerable latency of the power traffic,indicating the maximum tolerable delay.
Optionally, the first power device constructs a reward function corresponding to the second power device according to the accumulated minimum delay information, which may include: and the first power equipment determines a reward function corresponding to the second power equipment according to the reward formula.
Wherein, the rewarding formula is that；
Representing a bonus function.
As can be seen from the above reward formula, the optimization goal is to accumulate the minimum delay information, but the reinforcement learning is more focused on maximizing the reward, so that the reward function is inversely related to the accumulated minimum delay information.
It should be noted that, when the generated optimal task offloading mode does not satisfy the constraint conditionWhen it is, generate +.>And enabling the neural network in the base station cooperative edge computing unloading model to regenerate an optimal task unloading mode until the optimal task unloading mode corresponding to the historical task to be unloaded is determined.
In some embodiments, the first power device constructing the base station cooperative edge computing offload model from the state space, the offload decision action space, and the reward function may include: the first power equipment determines an initial Q value (Q-Table parameter) according to the state space, the unloading decision action space and the rewarding function; the first power equipment S1 determines a target Q value according to an initial Q value, a state space at the next moment and an unloading decision action space at the next moment; the first power equipment determines a target Q value as a new initial Q value, and repeatedly executes the step S1 until the difference between the target Q value and a preset Q value is minimum; and the first power equipment determines an unloading model corresponding to the target Q value with the smallest difference as a base station cooperation edge calculation unloading model.
Wherein the preset Q value is availableAnd (3) representing.
The first power equipment can firstly determine an initial Q value (Q-Table) according to the state space, the unloading decision action space and the rewarding function of the second power equipment at the current moment; then, in step S1, the first power device determines a Q value corresponding to the next moment, which is referred to as a target Q value, according to the initial Q value, a state space of the next moment, and an unloading decision action space of the next moment; then, as the second power equipment can repeatedly interact with the environment, the first power equipment can acquire rewarding functions corresponding to different task unloading decisions; at this time, the first power device determines the target Q value as a new initial Q value, and repeatedly executes the above step S1 until the difference between the finally obtained target Q value and the preset Q value is minimum; and finally, determining an unloading model corresponding to the target Q value with the minimum difference value as a base station cooperation edge calculation unloading model by the first power equipment, wherein the base station cooperation edge calculation unloading model is accurate.
Optionally, the determining, by the first power device, the target Q value according to the initial Q value, the state space at the next moment, and the unloading decision action space at the next moment may include: and the first power equipment determines a target Q value according to a target formula.
Wherein, the target formula is:
；
representing a target Q value; />A state space representing the current time; />Representing learning rate for reflecting the influence of current learning knowledge on previous learning knowledge,/>A bonus function representing a next time;representing discount factors->Predictive Q value indicating the next moment, +.>A state space representing the next time; />Representing the unloading decision action space at the next moment.
The second power equipment can repeatedly interact with the environment, so that the first power equipment can obtain rewards under different actions (namely rewarding functions under different task unloading decisions), and then the Q-Table parameters are adjusted through the Bellman equation and a value iteration method to find an optimal task unloading mode.
Based on this, the first power device may input the current state of the second power device into the base station cooperative edge computing offload model, take action and interact with the environment according to the Q-value function (i.e., the above-mentioned target formula), and approximate with a value iteration estimate To determine the state space->What task offloading mode is to be adopted is optimal. In each round, the above target formula may be adopted to perform the Q value update, so that the first power device updates the Q table through the target formula to learn past experience to generate a better task offloading mode.
However, since the above process (i.e., Q-learning offload decision algorithm) may converge to a local search area, the search performance may be poor, and to overcome this, the first power device may employ a method ofA greedy strategy improves the search performance of the Q-learning offload decision algorithm. This->Greedy strategies build on the concept of exploration, which is to randomly choose an action to find rewards under different actions as much as possible to increase the knowledge of the environment. Thus, each second power device has the possibility to choose a random behaviour from a discrete even distribution, availableAnd (3) representing.
Exemplary, as shown in fig. 3, a flow chart of the Q-learning unloading decision algorithm provided by the present invention is shown. The second power equipment can repeatedly interact with the environment, so that the first power equipment can acquire rewarding functions under different task unloading decisions, and further adjust Q-Table parameters through a Bellman equation and a value iteration method to find an optimal task unloading mode. Based on the method, a Markov decision process modeling is carried out on the task unloading mode according to the Q-learning unloading decision algorithm, Q-Table parameters are updated through a Belman equation and a value iteration method to achieve a better task unloading mode, and the access quantity of the power service can be effectively improved on the premise of guaranteeing the QoS of the power service.
In some embodiments, the obtaining, by the power device, a state space corresponding to the second power device may include: the method comprises the steps that a first power device obtains the current residual capacity of a second power device and the current calculation capacity corresponding to associated equipment connected with the second power device; the first power equipment determines a state space corresponding to the second power equipment according to the current residual electric quantity and the current calculation capacity.
Wherein the current residual power is availableIndicating that the current computing capacity is available +.>And (3) representing. That is, the state space corresponding to the second power device is available +.>And (3) representing.
203. And unloading the task to be unloaded according to the optimal task unloading mode.
After the first power equipment obtains the optimal task unloading mode corresponding to the task to be unloaded, the task to be unloaded can be effectively unloaded according to the optimal task unloading mode.
In the embodiment of the invention, the current state information corresponding to the first power equipment when unloading the task to be unloaded is acquired; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; and unloading the task to be unloaded according to the optimal task unloading mode. According to the method, according to the equipment energy consumption information and the equipment time delay information which correspond to the historical to-be-offloaded tasks respectively when the to-be-offloaded tasks are offloaded according to different task offloading modes, the built base station cooperation edge calculation offloading model can reasonably utilize the calculation resources of the power communication network on the premise of guaranteeing the time delay sensitive service QoS, and the optimal task offloading mode corresponding to the first power equipment is determined from the different task offloading modes so as to effectively offload the to-be-offloaded tasks in the first power equipment.
The base station cooperative edge computing and unloading device for the differentiated power service, which is provided by the invention, is described below, and the base station cooperative edge computing and unloading device for the differentiated power service, which is described below, and the base station cooperative edge computing and unloading method for the differentiated power service, which is described above, can be referred to correspondingly.
As shown in fig. 4, a schematic structural diagram of a base station cooperative edge computing and unloading device for differentiated power service provided by the present invention is applied to a first power device, where the first power device is connected with an associated device, and the associated device includes an access base station of the first power device and a cooperative base station connected with the access base station, where the device may include:
the obtaining module 401 is configured to obtain current state information corresponding to the first power device when unloading a task to be unloaded;
the processing module 402 is configured to input the current state information into a base station cooperative edge computing and offloading model, so as to obtain an optimal task offloading mode output by the base station cooperative edge computing and offloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
Optionally, the obtaining module 401 is further configured to obtain a historical task to be offloaded and a state space corresponding to the second power device, and an offloading decision action space corresponding to the second power device when offloading the historical task to be offloaded;
the processing module 402 is further configured to construct a reward function corresponding to the second power device by taking the accumulated minimum delay information of the second power device as an optimization target based on the device energy consumption information and the device delay information respectively corresponding to the historical task to be offloaded when the task to be offloaded is offloaded according to the different task offloading modes; and constructing a base station cooperation edge calculation unloading model according to the state space, the unloading decision action space and the rewarding function.
Optionally, the processing module 402 is specifically configured to obtain device energy consumption information and device time delay information corresponding to the historical task to be offloaded when the historical task to be offloaded is offloaded according to the different task offloading modes; constructing accumulated minimum delay information of the second power equipment by taking the energy consumption information of the plurality of equipment and the delay information of the plurality of equipment as constraint conditions; and constructing a reward function corresponding to the second power equipment according to the accumulated minimum time delay information.
Optionally, the processing module 402 is specifically configured to determine an initial Q value according to the state space, the unloading decision action space, and the reward function; s1, determining a target Q value according to the initial Q value, a state space at the next moment and an unloading decision action space at the next moment; determining the target Q value as a new initial Q value, and repeatedly executing the step S1 until the difference between the target Q value and a preset Q value is minimum; and determining an unloading model corresponding to the target Q value with the minimum difference as the base station cooperation edge calculation unloading model.
Optionally, the processing module 402 is specifically configured to obtain, when the task offloading mode is the mode of offloading in the associated device, a data amount of the historical task to be offloaded and a cpu cycle required for completing the historical task to be offloaded; determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to the unloading mode in the associated equipment according to the data volume and the CPU period; under the condition that the task unloading mode is the local processing mode, acquiring a CPU period required for completing the historical task to be unloaded; and determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to the local processing mode according to the period of the central processing unit.
Optionally, the obtaining module 401 is specifically configured to obtain a current remaining power of the second power device and a current computing capacity corresponding to an associated device connected to the second power device; and determining a state space corresponding to the second power equipment according to the current residual electric quantity and the current calculation capacity.
As shown in fig. 5, the electrical device provided by the present invention may include: an unloader (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530, and a communication bus 540, wherein the unloader 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The unloader 510 may invoke logic instructions in the memory 530 to perform a base station cooperative edge computing offload method for differentiated power services, the method being applied to a first power device connected to an associated device including an access base station of the first power device and a cooperative base station connected to the access base station, the method comprising: acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
Further, the logic instructions in the memory 530 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by an unloader, being capable of executing a base station cooperative edge computation offload method for differentiated power services provided by the methods described above, the method being applied to a first power device, the first power device being connected to an associated device, the associated device including an access base station of the first power device and a cooperative base station connected to the access base station, the method comprising: acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
In still another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by an unloader, implements a base station cooperative edge computation offload method for differentiated power services provided by the above methods, the method being applied to a first power device connected to an associated device, the associated device including an access base station of the first power device and a cooperative base station connected to the access base station, the method comprising: acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded; inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training corresponding equipment energy consumption information and equipment time delay information respectively based on historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. The base station cooperation edge computing and unloading method for the differentiated power service is characterized by being applied to first power equipment, wherein the first power equipment is connected with associated equipment, the associated equipment comprises an access base station of the first power equipment and a cooperation base station connected with the access base station, and the method comprises the following steps:
acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded;
inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model;
Unloading the task to be unloaded according to the optimal task unloading mode;
the base station cooperation edge computing and unloading model is obtained by training equipment energy consumption information and equipment time delay information respectively corresponding to historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment;
the base station cooperation edge calculation unloading model is obtained based on the following steps:
acquiring a historical task to be offloaded and a state space corresponding to a second power device, and an offloading decision action space corresponding to the second power device when offloading the historical task to be offloaded;
based on the historical to-be-offloaded task, when the to-be-offloaded task is offloaded according to the different task offloading modes, respectively corresponding equipment energy consumption information and equipment time delay information, and taking the accumulated minimum time delay information of the second power equipment as an optimization target to construct a reward function corresponding to the second power equipment;
and constructing the base station cooperation edge calculation unloading model according to the state space, the unloading decision action space and the rewarding function.
2. The method according to claim 1, wherein the constructing, based on the historical task to be offloaded and when the task to be offloaded is offloaded according to the different task offloading modes, the corresponding device energy consumption information and the device time delay information, with the accumulated minimum time delay information of the second power device as an optimization target, includes:
acquiring corresponding equipment energy consumption information and equipment time delay information respectively when the historical task to be offloaded is offloaded according to the different task offloading modes;
constructing accumulated minimum delay information of the second power equipment by taking the energy consumption information of the plurality of equipment and the delay information of the plurality of equipment as constraint conditions;
and constructing a reward function corresponding to the second power equipment according to the accumulated minimum time delay information.
3. The method according to claim 1 or 2, wherein said constructing the base station cooperation edge calculation offload model from the state space, the offload decision action space and the reward function comprises:
determining an initial Q value according to the state space, the offload decision action space, and the reward function;
S1, determining a target Q value according to the initial Q value, a state space at the next moment and an unloading decision action space at the next moment;
determining the target Q value as a new initial Q value, and repeatedly executing the step S1 until the difference between the target Q value and a preset Q value is minimum;
and determining an unloading model corresponding to the target Q value with the minimum difference as the base station cooperation edge calculation unloading model.
4. The method of claim 2, wherein the obtaining the device energy consumption information and the device latency information respectively corresponding to the historical task to be offloaded when the task to be offloaded is offloaded according to the different task offloading modes includes:
under the condition that the task unloading mode is the mode of unloading in the associated equipment, acquiring the data quantity of the historical task to be unloaded and the CPU period required by completing the historical task to be unloaded; determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be offloaded is offloaded according to the unloading mode in the associated equipment according to the data volume and the CPU period;
acquiring a CPU period required by completing the historical task to be offloaded under the condition that the task offloading mode is the local processing mode; and determining corresponding equipment energy consumption information and equipment time delay information when the historical task to be unloaded is unloaded according to the local processing mode according to the period of the central processing unit.
5. The method according to claim 1 or 2, wherein the acquiring the state space corresponding to the second power device comprises:
acquiring the current residual capacity of the second power equipment and the current calculation capacity corresponding to the associated equipment connected with the second power equipment;
and determining a state space corresponding to the second power equipment according to the current residual electric quantity and the current calculation capacity.
6. A base station cooperative edge computing and unloading device for differentiated power services, which is characterized by being applied to a first power device, wherein the first power device is connected with associated equipment, the associated equipment comprises an access base station of the first power device and a cooperative base station connected with the access base station, and the device comprises:
the acquisition module is used for acquiring current state information corresponding to the first power equipment when unloading a task to be unloaded;
the processing module is used for inputting the current state information into a base station cooperation edge calculation unloading model to obtain an optimal task unloading mode output by the base station cooperation edge calculation unloading model; unloading the task to be unloaded according to the optimal task unloading mode; the base station cooperation edge computing and unloading model is obtained by training equipment energy consumption information and equipment time delay information respectively corresponding to historical tasks to be unloaded when the tasks are unloaded according to different task unloading modes, wherein the task unloading modes comprise a local processing mode and a mode of unloading in the associated equipment;
The base station cooperation edge calculation unloading model is obtained based on the following steps:
acquiring a historical task to be offloaded and a state space corresponding to a second power device, and an offloading decision action space corresponding to the second power device when offloading the historical task to be offloaded;
based on the historical to-be-offloaded task, when the to-be-offloaded task is offloaded according to the different task offloading modes, respectively corresponding equipment energy consumption information and equipment time delay information, and taking the accumulated minimum time delay information of the second power equipment as an optimization target to construct a reward function corresponding to the second power equipment;
and constructing the base station cooperation edge calculation unloading model according to the state space, the unloading decision action space and the rewarding function.
7. A power device comprising a memory, an unloader and a computer program stored on the memory and executable on the unloader, wherein the unloader implements the base station cooperation edge computing unloading method for differentiated services according to any one of claims 1 to 5 when the program is executed by the unloader.
8. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by an unloader implements the base station cooperation edge computing offload method for differentiated services according to any one of claims 1 to 5.
CN202310922331.XA 2023-07-26 2023-07-26 Base station cooperation edge computing and unloading method and device for differentiated power service Active CN116647880B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310922331.XA CN116647880B (en) 2023-07-26 2023-07-26 Base station cooperation edge computing and unloading method and device for differentiated power service

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310922331.XA CN116647880B (en) 2023-07-26 2023-07-26 Base station cooperation edge computing and unloading method and device for differentiated power service

Publications (2)

Publication Number Publication Date
CN116647880A CN116647880A (en) 2023-08-25
CN116647880B true CN116647880B (en) 2023-10-13

Family

ID=87640409

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310922331.XA Active CN116647880B (en) 2023-07-26 2023-07-26 Base station cooperation edge computing and unloading method and device for differentiated power service

Country Status (1)

Country Link
CN (1) CN116647880B (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109951897A (en) * 2019-03-08 2019-06-28 东华大学 A kind of MEC discharging method under energy consumption and deferred constraint
CN112905315A (en) * 2021-01-29 2021-06-04 北京邮电大学 Task processing method, device and equipment in Mobile Edge Computing (MEC) network
CN113573342A (en) * 2021-09-27 2021-10-29 南京邮电大学 Energy-saving computing unloading method based on industrial Internet of things
CN113645637A (en) * 2021-07-12 2021-11-12 中山大学 Method and device for unloading tasks of ultra-dense network, computer equipment and storage medium
CN114217881A (en) * 2022-02-23 2022-03-22 北京航空航天大学杭州创新研究院 Task unloading method and related device
CN114340016A (en) * 2022-03-16 2022-04-12 北京邮电大学 Power grid edge calculation unloading distribution method and system
WO2022242468A1 (en) * 2021-05-18 2022-11-24 北京航空航天大学杭州创新研究院 Task offloading method and apparatus, scheduling optimization method and apparatus, electronic device, and storage medium
CN115408072A (en) * 2022-05-20 2022-11-29 北京航空航天大学杭州创新研究院 Rapid adaptation model construction method based on deep reinforcement learning and related device
CN116390160A (en) * 2023-01-17 2023-07-04 国网四川省电力公司信息通信公司 Cloud edge cooperation-based power task unloading method, device, equipment and medium
CN116489712A (en) * 2023-04-25 2023-07-25 北京交通大学 Mobile edge computing task unloading method based on deep reinforcement learning

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111245651B (en) * 2020-01-08 2022-03-29 上海交通大学 Task unloading method based on power control and resource allocation

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109951897A (en) * 2019-03-08 2019-06-28 东华大学 A kind of MEC discharging method under energy consumption and deferred constraint
CN112905315A (en) * 2021-01-29 2021-06-04 北京邮电大学 Task processing method, device and equipment in Mobile Edge Computing (MEC) network
WO2022242468A1 (en) * 2021-05-18 2022-11-24 北京航空航天大学杭州创新研究院 Task offloading method and apparatus, scheduling optimization method and apparatus, electronic device, and storage medium
CN113645637A (en) * 2021-07-12 2021-11-12 中山大学 Method and device for unloading tasks of ultra-dense network, computer equipment and storage medium
CN113573342A (en) * 2021-09-27 2021-10-29 南京邮电大学 Energy-saving computing unloading method based on industrial Internet of things
CN114217881A (en) * 2022-02-23 2022-03-22 北京航空航天大学杭州创新研究院 Task unloading method and related device
CN114340016A (en) * 2022-03-16 2022-04-12 北京邮电大学 Power grid edge calculation unloading distribution method and system
CN115408072A (en) * 2022-05-20 2022-11-29 北京航空航天大学杭州创新研究院 Rapid adaptation model construction method based on deep reinforcement learning and related device
CN116390160A (en) * 2023-01-17 2023-07-04 国网四川省电力公司信息通信公司 Cloud edge cooperation-based power task unloading method, device, equipment and medium
CN116489712A (en) * 2023-04-25 2023-07-25 北京交通大学 Mobile edge computing task unloading method based on deep reinforcement learning

Also Published As

Publication number Publication date
CN116647880A (en) 2023-08-25

Similar Documents

Publication Publication Date Title
CN110971706B (en) Approximate optimization and reinforcement learning-based task unloading method in MEC
CN111405568B (en) Computing unloading and resource allocation method and device based on Q learning
CN113225377B (en) Internet of things edge task unloading method and device
CN111405569A (en) Calculation unloading and resource allocation method and device based on deep reinforcement learning
CN111726826A (en) Online task unloading method in base station intensive edge computing network
CN113543176A (en) Unloading decision method of mobile edge computing system based on assistance of intelligent reflecting surface
CN112954736A (en) Policy-based computation offload of wireless energy-carrying internet-of-things equipment
CN113727362B (en) Unloading strategy method of wireless power supply system based on deep reinforcement learning
CN112511336B (en) Online service placement method in edge computing system
CN113573363B (en) MEC calculation unloading and resource allocation method based on deep reinforcement learning
CN115190033B (en) Cloud edge fusion network task unloading method based on reinforcement learning
CN113590279A (en) Task scheduling and resource allocation method for multi-core edge computing server
CN116260871A (en) Independent task unloading method based on local and edge collaborative caching
CN113645273A (en) Internet of vehicles task unloading method based on service priority
CN114090108B (en) Method and device for executing computing task, electronic equipment and storage medium
CN116233927A (en) Load-aware computing unloading energy-saving optimization method in mobile edge computing
CN115473896A (en) Electric power internet of things unloading strategy and resource configuration optimization method based on DQN algorithm
CN113747507B (en) 5G ultra-dense network-oriented computing resource management method and device
CN113821346B (en) Edge computing unloading and resource management method based on deep reinforcement learning
CN116541106B (en) Computing task unloading method, computing device and storage medium
CN116647880B (en) Base station cooperation edge computing and unloading method and device for differentiated power service
CN115756873B (en) Mobile edge computing and unloading method and platform based on federation reinforcement learning
CN114615705B (en) Single-user resource allocation strategy method based on 5G network
Wang et al. Partial task offloading strategy based on deep reinforcement learning
CN115480882A (en) Distributed edge cloud resource scheduling method and system

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