CN114363923A - Industrial Internet of things resource allocation method and system based on federal edge learning - Google Patents
Industrial Internet of things resource allocation method and system based on federal edge learning Download PDFInfo
- Publication number
- CN114363923A CN114363923A CN202111444856.4A CN202111444856A CN114363923A CN 114363923 A CN114363923 A CN 114363923A CN 202111444856 A CN202111444856 A CN 202111444856A CN 114363923 A CN114363923 A CN 114363923A
- Authority
- CN
- China
- Prior art keywords
- federal
- learning
- things
- industrial
- edge
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000013468 resource allocation Methods 0.000 title claims abstract description 16
- 238000005457 optimization Methods 0.000 claims abstract description 41
- 238000004891 communication Methods 0.000 claims abstract description 22
- 230000007774 longterm Effects 0.000 claims abstract description 21
- 238000005265 energy consumption Methods 0.000 claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000009826 distribution Methods 0.000 claims abstract description 8
- 238000006243 chemical reaction Methods 0.000 claims abstract description 6
- 238000003860 storage Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 8
- 230000004931 aggregating effect Effects 0.000 claims description 4
- 230000006855 networking Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013641 positive control Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides an industrial Internet of things resource allocation method based on federal marginal learning, which comprises the following steps: acquiring industrial equipment data; according to the obtained industrial equipment data, obtaining a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things; wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm. The invention adopts the Lyapunov optimization theory to realize the conversion from the long-term problem to the short-term problem, and simultaneously introduces the virtual energy queue to avoid the problem that the industrial equipment cannot work beyond the energy consumption requirement.
Description
Technical Field
The invention relates to the technical field of wireless communication, in particular to a resource allocation method and system of an industrial internet of things based on federal edge learning.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Currently, with the wide deployment of Industrial Internet of Things (IIOT) devices and the rapid development of Edge Artificial Intelligence (Edge AI) technology, the Industrial Internet of Things technology has been applied in large quantities on Edge networks, and large-scale Industrial data is generated at the Edge. However, training a Machine Learning (ML) model by a conventional centralized training method requires aggregating raw data of all devices to a central server for calculation through wireless transmission. This not only greatly consumes infinite bandwidth resources, but also consumes more computing resources and causes a lot of delay due to the fact that data of all devices are computed in the central server, thereby bringing about huge cost. Therefore, the traditional centralized training method is not practical for the actual IIOT scenario.
To solve the above problem, a distributed model training framework, federal edge learning (FEEL), is proposed. The goal of the FL is to enable devices to collaboratively learn a shared machine learning (machine learning) model and collaborate with the central server while keeping all training data on the device, thereby separating the ability to perform ML from the need to upload/store data in the central server. By updating the model parameters locally, FEEL takes advantage of the data and computational power distributed across the device, thus reducing energy consumption of the central server, reducing model training delays and protecting device data privacy, reducing overall costs.
Although the FEEL can reduce delay and reduce a certain cost in the scenario of the IIOT wireless network, since the bandwidth resources of the wireless network are limited, in each round of training, user selection and bandwidth allocation need to be performed reasonably; meanwhile, in a long-term FEEL process, decisions between different rounds have dependency, and the energy of the user is limited. Therefore, in order to train the federally learned model in the edge IIOT better, the selection of the devices and the allocation of the bandwidth need to be performed reasonably, so that the long-term total cost of the devices is minimized and the FEEL can be applied to the IIOT better under the condition that the long-term energy consumption requirement of the devices is met.
Disclosure of Invention
The invention provides an industrial Internet of things resource allocation method and system based on federal marginal learning to solve the problems, and the method and system can improve the training accuracy performance of federal learning, so that FEEL can be better applied to IIOT.
According to some embodiments, the invention adopts the following technical scheme:
an industrial Internet of things resource allocation method based on federal edge learning comprises the following steps:
acquiring industrial equipment data;
according to the obtained industrial equipment data, obtaining a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
Further, the optimization goal of minimizing the total cost of all the devices is realized by optimizing device selection and bandwidth allocation under the condition of time delay constraint and long-term energy consumption requirement of the devices.
Further, the Lyapunov optimization method is used for realizing the conversion from a long-term problem to a short-term problem.
Further, the optimization objective of minimizing the total cost of all devices is obtained by using the wireless communication network model of federal edge learning under the industrial internet of things, and the optimization objective comprises analyzing the wireless communication network model of federal edge learning under the industrial internet of things, and constructing a corresponding model aiming at local calculation of devices and an uplink for transmitting model parameters to an edge base station.
Further, after the data set of the industrial equipment is obtained, the data set is trained by considering the application of the federal study for a long time.
Further, the federal learning includes giving global model parameters to industrial devices participating in federal learning, which update local models based on local data.
Further, the federal learning further includes aggregating the updated local model parameters by using the edge server to obtain new global model parameters.
An industry thing networking resource allocation system based on federal edge learning, includes:
a data acquisition module configured to acquire industrial device data;
the calculation module is configured to obtain a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things according to the acquired industrial equipment data;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
A computer readable storage medium, wherein a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the method for allocating resources of the industrial internet of things based on federal edge learning.
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the industrial internet of things resource allocation method based on the federal edge learning.
Compared with the prior art, the invention has the beneficial effects that:
the invention adopts the Lyapunov optimization theory to realize the conversion from the long-term problem to the short-term problem, and simultaneously introduces the virtual energy queue to avoid the problem that the industrial equipment cannot work beyond the energy consumption requirement; by utilizing an iterative algorithm to reasonably select equipment and allocate bandwidth, the accuracy performance of Federal learning on the edge industrial Internet of things can be improved while the cost of all industrial equipment is minimized, so that the FEEL can be better trained on the IIOT.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.
Fig. 1 is a model diagram of a wireless network according to embodiment 1;
FIG. 2 is a flowchart of the present embodiment 1;
FIG. 3 is a flowchart of the algorithm for optimizing industrial equipment selection and bandwidth allocation of this embodiment 1;
FIG. 4 is a graph comparing the accuracy and effectiveness of federated learning under a method of jointly optimizing device selection and bandwidth allocation with a method of random device selection and bandwidth allocation;
FIG. 5 is a graph of a loss function for federated learning under a method of jointly optimizing device selection and bandwidth allocation versus a method of random device selection and bandwidth allocation.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example 1
As shown in fig. 1, a federal edge learning-based industrial internet of things resource allocation method includes:
acquiring industrial equipment data;
according to the obtained industrial equipment data, obtaining a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
The optimization goal of minimizing the total cost of all the equipment is realized by optimizing equipment selection and bandwidth allocation under the conditions of time delay constraint and long-term energy consumption requirement of the equipment.
The Lyapunov optimization method is used for realizing the conversion from a long-term problem to a short-term problem.
The method comprises the steps of utilizing a wireless communication network model of federal edge learning under the industrial internet of things to obtain an optimization target of minimizing the total cost of all equipment, analyzing the wireless communication network model of federal edge learning under the industrial internet of things, and constructing a corresponding model aiming at equipment local calculation and an uplink for transmitting model parameters to an edge base station.
After the industrial equipment data set is obtained, the data set is trained by applying federal learning for a long time.
The federal learning includes giving global model parameters to industrial devices participating in federal learning, which update local models based on local data.
The federated learning further includes aggregating the updated local model parameters with the edge server to obtain new global model parameters.
Specifically, as shown in fig. 2, the method for allocating resources of the industrial internet of things based on federal edge learning includes the following steps:
step S01: constructing a system model, and establishing infrastructure and equipment of a wireless network with minimized cost under an industrial Internet of things based on FEEL;
step S02: constructing a corresponding model aiming at an uplink which is locally calculated by equipment and transmits model parameters to an edge base station;
step S03: problem modeling, namely, under the conditions of time delay constraint and long-term energy consumption requirement of equipment, realizing the optimization target of minimizing the total cost of all the equipment by optimizing equipment selection and bandwidth allocation;
step S04: the conversion from a long-term problem to a short-term problem is realized by adopting the Lyapunov optimization theory, and then reasonable equipment selection and bandwidth allocation are carried out by adopting an iterative algorithm.
In the step S01, consider an industrial internet of things wireless network system (as shown in fig. 1) including a plurality of industrial devices and a base station (with a server):
under an industrial internet of things wireless network, a Base Station (Base Station) with an edge server and a plurality of industrial devices are provided, and in order to achieve better training of an ML (maximum likelihood) model under the scene, long-term application of FEEL (best effort) to carry out data training is considered.
The industrial equipment is integrated into [ N ]]N, each device N has a local data set
Wherein xiRepresents the ith input sample, yiDenotes xiThe flag of (2) is output. These data can be generated through the use of devices and applied to various machine learning tasks such as smart manufacturing, smart logistics, and the like.
The objective of the FL training process is to find the optimal learning model parameters ω to minimize its loss function value:
wherein
fn(ω，xi，yi) Is a sample (x) of the device ni，yi) Error value of parameter ω.
The training iteration process of the FL is generally divided into three steps:
step 1: the server broadcasts the t-1(t is more than or equal to 1) th global model parameter to all local devices n which are selected to participate in the FL:
step 2: after receiving the global model parameters broadcast by the server, the device n updates the local model according to the local data until the local accuracy ξ is reached and uploads the updated model parameters:
step 3: the server aggregates the local model parameters uploaded by all the devices, and calculates and updates to obtain new global model parameters:
the above process is repeated until convergence or all iterations are completed. Each iteration is referred to as a round (round).
In step S02, the three steps of FEEL iteration are respectively subjected to modeling analysis, and the uplink for local calculation of the device and transmission of the model parameters to the edge base station is considered.
Due to the heterogeneity of devices, the FEEL process requires a reasonable choice of devices to achieve better performance. Within each roundt, K devices need to be selected to participate in the FL procedure. By an(t) e {0, 1} indicates whether device n is selected in round to upload model parameters. a (t) { a ═ a1(t)，..，aN(t) } represents the selection decision in the t round, and sets the user whose current round t is selected as [ G []＝{n∈N|an(t)＝1}。
If device n is selected at the t round, i.e. anIf (t) is 1, then the last round global model parameter is received
And then, training a local model according to the local data. Suppose that the number of CPU revolutions required to process a data sample is CnThen the number of CPU revolutions required to perform a local iteration is Cn|DnL. Defining the computing CPU frequency of device n as fnThen a local iteration is performed for a time of
Thereby obtaining a locally calculated delay of
Energy consumption of local calculation is
Wherein the content of the first and second substances,
representing the number of local iterations of a round,
are based on the CPU frequency chip architecture.
Selected device n (a)n(t) ═ 1) after completing the local computation iteration, it needs to make its local model parameter
And uploading the data to an edge server, and meanwhile, energy consumption and time delay of a transmission process are caused.
For the model upload link we consider frequency division multiple access, FDMA, with a total channel bandwidth of B. bn(t)∈[0，1]The bandwidth allocation proportion of the device n in the t round is shown. (b) ═ b1(t)，...，bN(t) } denotes the round intra-t bandwidth allocation decision. With simultaneous presence of constraints
PnIs the transmission power of device n, the transmission rate
Then the transmission delay of device n at roundt is
The transmission energy consumption is
Wherein L isnUploading parameters for each model
Data size of the model.
Thus, the time delay of device n in roundt is obtained as
The energy consumption of the device n at round t is
And after all the selected devices upload the updated model parameters, the edge server carries out the aggregation of the global model.
The global model parameters are
Where Dt ═ ΣnDnan(t) representing data samples of all devices scheduled within the t-th roundThe number of the cells.
Within a round, the computational cost of device n is
Wherein
Unit price required for one CPU revolution; the communication cost is
Wherein
The cost of bandwidth to connect edge servers and devices. So the cost in each round is
In said step S03, all relevant devices are required to start model training simultaneously at the beginning of each round, with simultaneous model aggregation in the proposed framework. Under the condition of given T training rounds, the equipment selection and bandwidth allocation are jointly optimized, and meanwhile, the time delay requirement and the long-term equipment energy budget are guaranteed, so that the cost of all long-term equipment is minimized.
Note: a (T) · a (0),. a., (T-1), b (T) · b (0),. b (T-1), a (T) · a (a) · b (T-1), and a (T) ·1(t)，...an(t)，...，aN(t)}，b(t)＝{b1(t)，...bn(t)，...，bN(t) }. Wherein constraint (1) is a selection decision of device n; constraint (2) (3) is the requirement for the bandwidth allocation proportion of the device; constraint (4) is a delay requirement for device n, devices not meeting the requirement cannot participate in the FL process; constraint (5) is a long-term power consumption limit for device n,
representing the total energy budget of the device n.
In the step S04, the transformation from the long-term problem to the short-term problem is realized by using the lyapunov optimization theory, and then an iterative algorithm is used to perform reasonable device selection and bandwidth allocation.
Due to the long-term optimization goals and constraints of problem P1, the existing approaches cannot be directly solved. The Lyapunov optimization method is considered to be adopted, and the problem is converted into a series of short-term deterministic problems to be solved; meanwhile, based on the virtual queue theory, the long-term energy consumption limiting condition (5) is converted into stable queue limitation.
Qn(t) represents the queue backlog of device n running low of energy in roundt:
wherein Qn(1)＝0。
The original problem P1 is rewritten into a single-turn problem by utilizing the Lyapunov optimization theory, and is defined as
Where V is a positive control parameter to adjust the trade-off between cost and energy consumption.
Substitution formula is then expanded into
Note a in the above formulan(t) is a binary variable, and bn(t) is in [0, 1 ]]The continuous variable in (b), and thus the equation is a mixed integer problem, which is often difficult to solve. Here, we divide it into two sub-problems, namely the device selection problem and the bandwidth allocation problem are solved separately and then iterated.
Sub-problem 1: device selection problem
Assuming that the bandwidth ratios of all devices are
Order to
The P2 problem translates into:
s.t.Constraints(1)(4)
according to the equation, given the bandwidth, the system will want to select the device with lower cost and lower energy consumption, so ηnThe smaller the device, the higher the priority, the more likely it is to be selected. Industrial device selection decisions can be made based on this priority.
Firstly, all the devices are arranged according to etanThe values of (a) are arranged in ascending order to form a set I; if the device n time delay does not meet the limiting condition (4), an(t) ═ 0; otherwise an(t) 1. Get η from the set inLow first K anThe (t) ═ 1 device gets the current device selection set G.
Sub-problem 2: bandwidth allocation problem
And selecting a set G for the equipment, and allocating the bandwidth. Then the current problem with device-based selection is:
namely, it is
Observing the above equation here, optimizing the above equation is equivalent to
According to the definition of the convex function, the above formula can be judged to be a convex function. Therefore, the CVX toolkit is introduced here for solution.
And after the two sub-problems are solved, iteration is carried out until all rounds are finished.
The solving algorithm comprises the following steps:
the first step is as follows: initializing an energy queue
The second step is that: in a T1.2, T inner loop, the following steps are performed:
step 1: initial setting
And calculating the priority eta, and sequencing all the industrial equipment from small to large according to the eta value.
Step 2: judging whether all the devices meet the time delay condition (4), if not, judging whether all the devices meet the time delay condition an(t) ═ 0; otherwise an(t)＝1。
Step 3: get η from small to largenFirst K of anThe (t) ═ 1 device gets the current device selection set G.
Step 4: aiming at the equipment selection set G, the bandwidth is distributed by a CVX tool to obtain the optimal bandwidth distribution b*(G)。
Step 5: find the current optimal device selection set G*To obtain the optimal device selection decision
And optimal bandwidth allocation decision b*(G*)。
Step 6: and updating the energy queue according to the formula (#).
Step 7: t → t +1, repeat Step2-Step6 until the end of the cycle.
Fig. 4 is a graph comparing the accuracy effect of the joint learning under the method of jointly optimizing device selection and bandwidth allocation (deployed _ optimization) and the method of random device selection and bandwidth allocation (organic _ random), and fig. 5 is a graph comparing the loss function effect of the joint learning under the method of jointly optimizing device selection and bandwidth allocation and the method of random device selection and bandwidth allocation. By comparing the accuracy and loss function effect graphs obtained by the two schemes, the method for jointly optimizing equipment selection and bandwidth allocation can obtain higher federal learning accuracy, smaller loss function and better convergence effect than the method for randomly selecting equipment and allocating bandwidth: for the case that the number of rounds is 20 and the number of users is 50, the accuracy of federal learning finally reaches nearly 97%, and the federal learning tends to converge, and the loss function is very smooth; the method of random device selection and bandwidth allocation does not consider the heterogeneity among devices, and the random selection of devices each time may cause large differences among devices, and the training result is poor: for the case of 20 rounds and 50 users, the accuracy of federal learning is nearly 85% at last, and the learning process fluctuates greatly and the effect is poor.
Example 2
An industry thing networking resource allocation system based on federal edge learning, includes:
a data acquisition module configured to acquire industrial device data;
the calculation module is configured to obtain a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things according to the acquired industrial equipment data;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
Example 3
A computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are adapted to be loaded by a processor of a terminal device and execute a federal edge learning-based industrial internet of things resource allocation method provided in embodiment 1.
Example 4
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer-readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the federal edge learning-based industrial internet of things resource allocation method provided in embodiment 1.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. An industrial Internet of things resource allocation method based on federal edge learning is characterized by comprising the following steps:
acquiring industrial equipment data;
according to the obtained industrial equipment data, obtaining a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
2. The method for allocating resources of the internet of things for the industry based on federal edge learning as claimed in claim 1, wherein the optimization goal of minimizing the total cost of all devices is achieved by optimizing device selection and bandwidth allocation under the conditions of time delay constraint and long-term energy consumption requirement of the devices.
3. The method for allocating resources of the internet of things for the industry based on federal edge learning as claimed in claim 2, wherein the lyapunov optimization method is used for realizing the conversion from a long-term problem to a short-term problem.
4. The method as claimed in claim 3, wherein the optimization goal of minimizing the total cost of all devices is obtained by using the wireless communication network model for federal edge learning in the industrial internet of things, and the optimization goal includes analyzing the wireless communication network model for federal edge learning in the industrial internet of things, and constructing a corresponding model for local calculation of devices and uplink transmission of model parameters to an edge base station.
5. The method for allocating resources of the internet of things for industry based on federal edge learning as claimed in claim 4, wherein after acquiring the data set of the industrial equipment, the data set is trained by considering the long-term application of federal learning.
6. The method of claim 5, wherein the federal learning includes giving global model parameters to industrial equipment participating in federal learning, and the industrial equipment updates a local model according to local data.
7. The method of claim 6, wherein the federal learning further comprises aggregating updated local model parameters with an edge server to obtain new global model parameters.
8. An industry thing networking resource allocation system based on federal edge learning, comprising:
a data acquisition module configured to acquire industrial device data;
the calculation module is configured to obtain a distribution result by utilizing a wireless communication network model of federal edge learning under the industrial Internet of things according to the acquired industrial equipment data;
wherein the wireless communication network model is globally optimized with an optimization objective that minimizes the total cost of all devices; the optimization goal of minimizing the total cost of all equipment is achieved through the Lyapunov optimization method and the iterative algorithm.
9. A computer-readable storage medium having stored thereon a plurality of instructions adapted to be loaded by a processor of a terminal device and execute a method for federal edge learning based industrial internet of things resource allocation as claimed in any one of claims 1 to 7.
10. A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer-readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the federal edge learning based industrial internet of things resource allocation method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111444856.4A CN114363923B (en) | 2021-11-30 | 2021-11-30 | Industrial Internet of things resource allocation method and system based on federal edge learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111444856.4A CN114363923B (en) | 2021-11-30 | 2021-11-30 | Industrial Internet of things resource allocation method and system based on federal edge learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114363923A true CN114363923A (en) | 2022-04-15 |
| CN114363923B CN114363923B (en) | 2024-03-26 |
Family
ID=81097327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111444856.4A Active CN114363923B (en) | 2021-11-30 | 2021-11-30 | Industrial Internet of things resource allocation method and system based on federal edge learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114363923B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112804107A (en) * | 2021-01-28 | 2021-05-14 | 南京邮电大学 | Layered federal learning method for energy consumption adaptive control of equipment of Internet of things |
| CN112817653A (en) * | 2021-01-22 | 2021-05-18 | 西安交通大学 | Cloud-side-based federated learning calculation unloading computing system and method |
| CN113139663A (en) * | 2021-04-23 | 2021-07-20 | 深圳市大数据研究院 | Federal edge learning configuration information acquisition method, device, equipment and medium |
| CN113194489A (en) * | 2021-04-01 | 2021-07-30 | 西安电子科技大学 | Minimum-maximum cost optimization method for effective federal learning in wireless edge network |
| CN113406974A (en) * | 2021-08-19 | 2021-09-17 | 南京航空航天大学 | Learning and resource joint optimization method for unmanned aerial vehicle cluster federal learning |
| CN113504999A (en) * | 2021-08-05 | 2021-10-15 | 重庆大学 | Scheduling and resource allocation method for high-performance hierarchical federated edge learning |
| CN113537514A (en) * | 2021-07-27 | 2021-10-22 | 北京邮电大学 | High-energy-efficiency federal learning framework based on digital twins |
-
2021
- 2021-11-30 CN CN202111444856.4A patent/CN114363923B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112817653A (en) * | 2021-01-22 | 2021-05-18 | 西安交通大学 | Cloud-side-based federated learning calculation unloading computing system and method |
| CN112804107A (en) * | 2021-01-28 | 2021-05-14 | 南京邮电大学 | Layered federal learning method for energy consumption adaptive control of equipment of Internet of things |
| CN113194489A (en) * | 2021-04-01 | 2021-07-30 | 西安电子科技大学 | Minimum-maximum cost optimization method for effective federal learning in wireless edge network |
| CN113139663A (en) * | 2021-04-23 | 2021-07-20 | 深圳市大数据研究院 | Federal edge learning configuration information acquisition method, device, equipment and medium |
| CN113537514A (en) * | 2021-07-27 | 2021-10-22 | 北京邮电大学 | High-energy-efficiency federal learning framework based on digital twins |
| CN113504999A (en) * | 2021-08-05 | 2021-10-15 | 重庆大学 | Scheduling and resource allocation method for high-performance hierarchical federated edge learning |
| CN113406974A (en) * | 2021-08-19 | 2021-09-17 | 南京航空航天大学 | Learning and resource joint optimization method for unmanned aerial vehicle cluster federal learning |
Non-Patent Citations (3)
| Title |
|---|
| SIYU WANG等: "Content-based Vehicle Selection and Resource Allocation for Federated Learning in IoV", 《2021 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE WORKSHOPS (WCNCW)》, 7 May 2021 (2021-05-07) * |
| 纪秀朝: "基于联邦学习的工业物联网资源分配方法研究", 《中国优秀硕士学位论文全文数据库(电子期刊)》, 1 June 2023 (2023-06-01) * |
| 雷凯;黄硕康;方俊杰;黄济乐;谢英英;彭波;: "智能生态网络：知识驱动的未来价值互联网基础设施", 应用科学学报, no. 01, 30 January 2020 (2020-01-30) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114363923B (en) | 2024-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Client selection and bandwidth allocation in wireless federated learning networks: A long-term perspective | |
| Paliwal et al. | Reinforced genetic algorithm learning for optimizing computation graphs | |
| CN110968426B (en) | Edge cloud collaborative k-means clustering model optimization method based on online learning | |
| CN113504999B (en) | Scheduling and resource allocation method for high-performance hierarchical federal edge learning | |
| Jiang et al. | Distributed resource scheduling for large-scale MEC systems: A multiagent ensemble deep reinforcement learning with imitation acceleration | |
| EP4350572A1 (en) | Method, apparatus and system for generating neural network model, devices, medium and program product | |
| CN112637883B (en) | Federal learning method with robustness to wireless environment change in electric power Internet of things | |
| CN112100155A (en) | Cloud edge cooperative digital twin model assembling and fusing method | |
| CN112784362A (en) | Hybrid optimization method and system for unmanned aerial vehicle-assisted edge calculation | |
| CN116523079A (en) | Reinforced learning-based federal learning optimization method and system | |
| CN111988787B (en) | Task network access and service placement position selection method and system | |
| CN113469325A (en) | Layered federated learning method, computer equipment and storage medium for edge aggregation interval adaptive control | |
| CN115358487A (en) | Federal learning aggregation optimization system and method for power data sharing | |
| CN113472597A (en) | Distributed convolutional neural network fine-grained parameter transmission scheduling method and device | |
| CN109445386A (en) | A kind of most short production time dispatching method of the cloud manufacturing operation based on ONBA | |
| CN115002123A (en) | Fast adaptive task unloading system and method based on mobile edge calculation | |
| Badri et al. | A sample average approximation-based parallel algorithm for application placement in edge computing systems | |
| Li et al. | Data analytics for fog computing by distributed online learning with asynchronous update | |
| Yan et al. | Deep reinforcement learning based offloading for mobile edge computing with general task graph | |
| CN111340192A (en) | Network path allocation model training method, path allocation method and device | |
| Pérez et al. | Parallel/distributed implementation of cellular training for generative adversarial neural networks | |
| Wang et al. | Solving system-level synthesis problem by a multi-objective estimation of distribution algorithm | |
| CN113094180A (en) | Wireless federal learning scheduling optimization method and device | |
| CN114363923B (en) | Industrial Internet of things resource allocation method and system based on federal edge learning | |
| Asheralieva et al. | Efficient dynamic distributed resource slicing in 6g multi-access edge computing networks with online admm and message passing graph neural networks |
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 |