CN116827393A - Honeycomb-free large-scale MIMO uplink receiving method and system based on federal learning - Google Patents
Honeycomb-free large-scale MIMO uplink receiving method and system based on federal learning Download PDFInfo
- Publication number
- CN116827393A CN116827393A CN202310795618.0A CN202310795618A CN116827393A CN 116827393 A CN116827393 A CN 116827393A CN 202310795618 A CN202310795618 A CN 202310795618A CN 116827393 A CN116827393 A CN 116827393A
- Authority
- CN
- China
- Prior art keywords
- access point
- federal learning
- matrix
- model
- training
- 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 46
- 239000011159 matrix material Substances 0.000 claims abstract description 49
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims abstract description 16
- 230000004931 aggregating effect Effects 0.000 claims abstract description 7
- 230000001413 cellular effect Effects 0.000 claims description 15
- 238000005562 fading Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 7
- 238000001228 spectrum Methods 0.000 abstract description 4
- 230000003595 spectral effect Effects 0.000 description 6
- 210000002569 neuron Anatomy 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 241001131927 Placea Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The application discloses a honeycomb-free large-scale MIMO uplink receiving method and system based on federal learning, wherein the honeycomb-free large-scale MIMO system is constructed; the access point randomly generates Rayleigh distributed channel state information as a data set input by federal learning training; the central processing unit constructs a centralized minimum mean square error receiver matrix according to the channel state information generated by the access points, and sends each row in the matrix to the corresponding access point to serve as a federal learning training tag; training the local model of each access point, and aggregating model parameters at a central processing unit until the objective function of the system is converged, so as to finally obtain the optimal receiver output; the application can obviously improve the frequency spectrum efficiency by constructing the federal learning auxiliary receiving matrix, and can promote the scalability of a cell-free large-scale MIMO system, and the matrix can not add extra effective load at the front end, so that the method has stronger practical applicability and is more suitable for actual communication scenes.
Description
Technical Field
The application relates to the technical field of wireless communication, in particular to a honeycomb-free large-scale MIMO uplink receiving method and system based on federal learning.
Background
The non-cellular massive MIMO serves all users in the network in a unified way through an access point connected to the central processor with the front end. It eliminates cell boundaries and achieves higher macro diversity gain from distributed antennas. In addition, the proximity between the access point and the user provides high spectral efficiency, low data delay, and high energy efficiency. However, the high payload of the front end remains a major challenge for cellular-free massive MIMO. As the number of users increases, a large amount of channel state information needs to be exchanged at the front end for reception, which hinders the scalability of a non-cellular massive MIMO system. Thus, many related works assume that an access point processes its received signal using a locally received approach. The use of local MRC reception in the access point, although at a lower cost, does not completely eliminate the interference between users; local ZF reception, while better performing than the local MRC scheme, can only suppress interference from one access point itself, but not from other access points.
Centralized ZF or MMSE can largely cancel interference between access points, but they can result in unmanaged front-end traffic because channel state information needs to be transmitted from all access points to the central processor. Thus, there is a conflict between the front-end payload and spectral efficiency performance.
Disclosure of Invention
This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application.
The present application has been made in view of the above-described problems occurring in the prior art. Therefore, the application provides a honeycomb-free large-scale MIMO uplink receiving method based on federal learning, which is used for solving the problem that in practical problems, performance contradiction exists between spectrum efficiency and effective load.
In order to solve the technical problems, the application provides the following technical scheme:
in a first aspect, the present application provides a non-cellular massive MIMO uplink receiving method based on federal learning, including:
constructing a honeycomb-free large-scale MIMO system, wherein access points are uniformly distributed in an area and are connected to a central processing unit through a backhaul link to serve all users in the area;
the access point randomly generates Rayleigh distributed channel state information as a data set input by federal learning training;
the central processing unit constructs a centralized minimum mean square error receiver matrix according to the channel state information generated by the access points, and sends each row in the matrix to the corresponding access point to serve as a federal learning training tag;
training the local model at each access point, and aggregating model parameters at the central processing unit until the objective function of the system is converged, and finally obtaining the optimal receiver output.
As a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: the construction of the honeycomb-free massive MIMO system comprises the following steps:
the method comprises the steps of dividing a square area with the side length of D kilometers, uniformly distributing L access points with the antenna number of M and uniformly distributing single-antenna users with the antenna number of K in the area.
As a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: the access point randomly generates Rayleigh distributed channel state information as a data set for federal learning training input, comprising:
setting all users to send signals to the access point in the same time and frequency resource, and modeling a channel between the first (l=1,., L) access point and the K (k=1,., K) users, where the formula is:
wherein ,is a fast fading vector, I M Is M x MIdentity matrix beta of (2) kl Is a large-scale fade consisting of path loss and shadow fading;
by randomly generating fast fading vectors h kl Then, let the fast fading matrix at the first access point beRespectively take [ H ] l ] i，j Is (H) l ] i，j Imaginary part Im { [ H ] l ] i，j [ H ] and modulus value ] l ] i，j I stitching is the input dataset of the federal learning local model +.>Namely:
[[X l ] ：，：，1 ] i，j ＝Re{[H l ] i，j }，[[X l ] ：，：，2 ] i，j ＝Im{[H l ] i，j }，[[X l ] ：，：，3 ] i，j ＝|[H l ] i，j |。
as a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: the central processing unit constructs a centralized minimum mean square error receiver matrix according to the channel state information generated by the access points, and sends each row in the matrix to the corresponding access point as a federal learning training tag, comprising:
the channel state information of all access points is uploaded to a central processing unit through a backhaul link, and the central processing unit generates a centralized MMSE receiving matrix by using the collected channel state information, wherein the receiving matrix expression is as follows:
wherein ,t represents a transpose, H represents a conjugateTranspose, I K Is a K x K identity matrix, G l Channel matrix representing the first access point, i.e. +.> p u The transmit power for the user;
taking a centralized MMSE receiving matrixData of (l-1) M+1 line to lM line is transmitted to the first access point as its receiving matrix +.>
According to fast fading matrix H l Constructing a data tag of a local model of the first access pointNamely: [ [ Y ] l ] ：，：，1 ] i，j ＝Re{[V l ] i，j }，[[Y l ] ：，：，2 ] i，j ＝Im{[V l ] i，j }；
From input dataset X l And data label Y l Constructing a federal learning data set to make the data set of the first access point be For dataset +.>Is a length of (c).
As a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: training the local model at each access point, comprising:
respectively constructing a five-layer local DNN network at each access point;
let the model output of the first access point be f (θ l ，X l ) The loss function of the current model is defined as:
as a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: performing, at the central processor, aggregation of model parameters until an objective function of the system reaches convergence, comprising:
solving the optimization problem using federal learning is represented as follows:
wherein f represents a functional relationship between the model input and output, θ l For the weight parameters of the model at the ith AP, ω is the model weight parameter at the central processor and s.t is expressed as a constraint.
As a preferable scheme of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, the application comprises the following steps: each access point local model adopts small batch gradient descent to learn;
training local models of each access point respectively, and updating model parameters of each local model when the training epoch is equal to epoch_numUploading to a central processing unit;
the CPU willThe model parameters of all access points are aggregated to obtain new global model parameters omega t The method comprises the following steps:and the updated global model parameter omega t Downloading to each access point;
each access point takes the received global model parameters as local model parameters, namely:and let t=t+1, repeating the training process of the local model of each access point until the system objective function converges.
In a second aspect, the present application provides a non-cellular massive MIMO uplink receiving system based on federal learning, comprising:
the neural network module comprises a local model and a global model; parameters for locally training and updating the data and aggregating the respective local models;
and the user cooperation module is responsible for coordinating the local model updating of each user, determining the users participating in training, and determining the times of updating the local model of each user.
In a third aspect, the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, wherein: the processor, when executing the computer program, implements any of the steps of the method described above.
In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon a computer program, wherein: which when executed by a processor performs any of the steps of the method described above.
Compared with the prior art, the application has the beneficial effects that: the application can obviously improve the frequency spectrum efficiency by constructing the federal learning auxiliary receiving matrix, and the matrix can not add extra effective load at the front end; meanwhile, the federal learning auxiliary receiving matrix can promote the scalability of a cell-free large-scale MIMO system, so that the method has stronger practical practicability and can be more applied to actual communication scenes in a fitting way.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
fig. 1 is a flowchart of a non-cellular massive MIMO uplink receiving method based on federal learning according to an embodiment of the present application;
fig. 2 is a schematic diagram illustrating model parameter transfer during federal learning of a honeycomb-free massive MIMO uplink receiving method according to an embodiment of the present application;
fig. 3 is a graph of spectral efficiency of a non-cellular massive MIMO uplink reception method based on federal learning according to an embodiment of the present application.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present application can be understood in detail, a more particular description of the application, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present application is not limited to the specific embodiments disclosed below.
Further, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic can be included in at least one implementation of the application. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
While the embodiments of the present application have been illustrated and described in detail in the drawings, the cross-sectional view of the device structure is not to scale in the general sense for ease of illustration, and the drawings are merely exemplary and should not be construed as limiting the scope of the application. In addition, the three-dimensional dimensions of length, width and depth should be included in actual fabrication.
Example 1
Referring to fig. 1 and 2, a first embodiment of the present application provides a non-cellular massive MIMO uplink receiving method based on federal learning, including:
s1, constructing a honeycomb-free large-scale MIMO system;
further, in a square area with a side length of d=1 km, the system uniformly distributes access points with a number of l=9 antennas of m=2 and uniformly distributes single-antenna users with a number of k=6, and the access points are connected with the central processing unit through a backhaul link and serve all users in the area;
s2, the access point randomly generates Rayleigh distributed channel state information as a data set input by federal learning training;
further, it is set that all users send signals to the access point in the same time and frequency resource, and the channel between the first (l=1,) and the K (k=1,) access point is modeled, where the formula is:
wherein ,is a fast fading vector, I M Is an identity matrix of MxM, beta kl Is a large-scale fade consisting of path loss and shadow fading;
by random meansGenerating fast fading vector h kl Then, let the fast fading matrix at the first access point beRespectively take [ H ] l ] i，j Is (H) l ] i，j Imaginary part Im { [ H ] l ] i，j [ H ] and modulus value ] l ] i，j I stitching is the input dataset of the federal learning local model +.>Namely:
[[X l ] ：，：，1 ] i，j ＝Re{[H l ] i，j }，[[X l ] ：，：，2 ] i，j ＝Im{H l ] i，j }，[[X l ] ：，：，3 ] i，j ＝|[H l ] i，j |；
s3, the central processing unit constructs a centralized minimum mean square error receiver matrix according to channel state information generated by the access points, and sends each row in the matrix to the corresponding access point to serve as a federal learning training tag;
further, the channel state information of all access points is uploaded to the central processing unit through the backhaul link, and the central processing unit generates a centralized MMSE receiving matrix by using the collected channel state information, wherein the receiving matrix expression is as follows:
wherein ,t represents transpose, H represents conjugate transpose, I K Is a K x K identity matrix, G l Channel matrix representing the first access point, i.e. +.> pu is the transmit power of the user;
taking a centralized MMSE receiving matrixData of (l-1) M+1 line to lM line is transmitted to the first access point as its receiving matrix +.>
According to fast fading matrix H l Constructing a data tag of a local model of the first access pointNamely: [ [ Y ] l ] ：，：，1 ] i，j ＝Re{[V l ] i，j }，[[Y l ] ：，：，2 ] i，j ＝Im|[V l ] i，j }；
From input dataset X l And data label Y l Constructing a federal learning data set to make the data set of the first access point beFor dataset +.>Is a length of (2);
it should be noted that the federal learning aided reception matrix can significantly improve spectral efficiency, especially in large cell-free massive MIMO systems, and it does not add additional payloads at the front end;
s4, training the local model at each access point, and aggregating model parameters at a central processing unit until an objective function of the system is converged, so that the optimal receiver output is finally obtained;
further, training the local model at each access point is performed as follows:
respectively constructing a five-layer local DNN network at each access point;
let the model output of the first access point be f (θ l ，X l ) The loss function of the current model is defined as:
further, the local DNN network structure is: the number of neurons of the input layer is 3MK, and the activation function is sigmoid (x) =1/(1+e) -x ) The method comprises the steps of carrying out a first treatment on the surface of the The number of neurons of the second layer to the fourth layer is 512, 256 and 512 respectively, the number of neurons of the output layer is 2MK, and the activation functions from the second layer to the output layer are ReLU (x) =max (0, x);
further, the aggregation of the model parameters is performed at the central processing unit until the objective function of the system reaches convergence, and the steps are as follows:
solving the optimization problem using federal learning is represented as follows:
wherein f represents a functional relationship between the model input and output, θ l For the weight parameters of the model at the ith AP, ω is the model weight parameter at the central processor, s.t is expressed as a constraint;
each access point local model adopts small batch gradient descent to learn;
training local models of each access point respectively, and updating model parameters of each local model when the training epoch is equal to epoch_numUploading to the central placeA processor;
the central processing unit aggregates the model parameters of each access point to obtain new global model parameters omega t The method comprises the following steps:and the updated global model parameter omega t Downloading to each access point;
each access point takes the received global model parameters as local model parameters, namely:and making t=t+1, repeating the training process of the local model of each access point until the system objective function converges;
it should be noted that the learning rate lr=0.1, batch_size=64, epoch_num=10, and t=1.
Further, the embodiment also provides a non-cellular massive MIMO uplink receiving system based on federal learning, which includes:
the neural network module comprises a local model and a global model; parameters for locally training and updating the data and aggregating the respective local models;
and the user cooperation module is responsible for coordinating the local model updating of each user, determining the users participating in training, and determining the times of updating the local model of each user.
The embodiment also provides a computer device, which is suitable for the situation of the honeycomb-free large-scale MIMO uplink receiving method based on federal learning, and comprises the following steps:
a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the non-cellular massive MIMO uplink receiving method based on federal learning as set forth in the above embodiment.
The computer device may be a terminal comprising a processor, a memory, a communication interface, a display screen and input means connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
The present embodiment also provides a storage medium having stored thereon a computer program which, when executed by a processor, implements a non-cellular massive MIMO uplink reception method based on federal learning as proposed in the above embodiment.
The storage medium according to the present embodiment belongs to the same inventive concept as the data storage method according to the above embodiment, and technical details not described in detail in the present embodiment can be seen in the above embodiment, and the present embodiment has the same advantageous effects as the above embodiment.
Example 2
Referring to fig. 3, a second embodiment of the present application provides a non-cellular massive MIMO uplink receiving method based on federal learning, including:
the abscissa axis represents the spectrum efficiency of the system, and the ordinate axis is a cumulative distribution value, and in fig. 3, local MRC, local MMSE, centered MMSE and FL-Aided MMSE represent a Local MRC receiving matrix, a Local MMSE receiving matrix, a Centralized MMSE receiving matrix and a receiving matrix outputted by the Local model after federal learning, respectively; from the figure, it can be seen that the spectral efficiency of federal learning assisted MMSE
The probability of more than or equal to 0.95 is improved by more than 35 percent compared with the local MMSE; this demonstrates that federal learning-assisted MMSE receive matrices can significantly improve system spectral efficiency without adding additional payloads at the front end, demonstrating the feasibility of the inventive method.
It should be noted that the above embodiments are only for illustrating the technical solution of the present application and not for limiting the same, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present application may be modified or substituted without departing from the spirit and scope of the technical solution of the present application, which is intended to be covered in the scope of the claims of the present application.
Claims (10)
1. The honeycomb-free large-scale MIMO uplink receiving method based on federal learning is characterized by comprising the following steps of:
constructing a honeycomb-free large-scale MIMO system, wherein access points are uniformly distributed in an area and are connected to a central processing unit through a backhaul link to serve all users in the area;
the access point randomly generates Rayleigh distributed channel state information as a data set input by federal learning training;
the central processing unit constructs a centralized minimum mean square error receiver matrix according to the channel state information generated by the access points, and sends each row in the matrix to the corresponding access point to serve as a federal learning training tag;
training the local model at each access point, and aggregating model parameters at the central processing unit until the objective function of the system is converged, and finally obtaining the optimal receiver output.
2. The federally learned non-cellular massive MIMO uplink receiving method according to claim 1, wherein the constructing the non-cellular massive MIMO system comprises:
the method comprises the steps of dividing a square area with the side length of D kilometers, uniformly distributing L access points with the antenna number of M and uniformly distributing single-antenna users with the antenna number of K in the area.
3. The federal learning-based honeycomb-free massive MIMO uplink receiving method according to claim 2, wherein the access point randomly generates channel state information of a rayleigh distribution as a data set of federal learning training input, comprising:
setting all users to send signals to the access point in the same time and frequency resource, and modeling a channel between the first (l=1,., L) access point and the K (k=1,., K) users, where the formula is:
wherein ,is a fast fading vector, I M Is an identity matrix of MxM, beta kl Is a large-scale fade consisting of path loss and shadow fading;
by randomly generating fast fading vectors h kl Then, let the fast fading matrix at the first access point beRespectively take [ H ] l ] i,j Is (H) l ] i,j Imaginary part Im { [ H ] l ] i,j [ H ] and modulus value ] l ] i,j I stitching is the input dataset of the federal learning local model +.>Namely: [ [ X ] l ] :,:,1 ] i,j ＝Re{[H l ] i,j }，[[X l ] :,:,2 ] i,j ＝Im{[H l ] i,j }，[[X l ] :,:,3 ] i,j ＝|[H l ] i,j |。
4. The method for non-cellular massive MIMO uplink reception based on federal learning according to claim 3, wherein the central processor constructs a centralized minimum mean square error receiver matrix according to the channel state information generated by the access points, and transmits each row in the matrix to the corresponding access point as a federal learning training tag, comprising:
the channel state information of all access points is uploaded to a central processing unit through a backhaul link, and the central processing unit generates a centralized MMSE receiving matrix by using the collected channel state information, wherein the receiving matrix expression is as follows:
wherein ,t represents transpose, H represents conjugate transpose, I K Is a K x K identity matrix, G l Channel matrix representing the first access point, i.e. +.> p u The transmit power for the user;
taking a centralized MMSE receiving matrixData of (l-1) M+1 line to lM line is transmitted to the first access point as its receiving matrix +.>
According to fast fading matrix H l Constructing a data tag of a local model of the first access pointNamely: [ [ Y ] l ] :,:,1 ] i,j ＝Re{[V l ] i,j }，[[Y l ] :,:,2 ] i,j ＝Im{[V l ] i,j }；
From input dataset X l And data label Y l Constructing a federal learning data set to make the data set of the first access point be For dataset +.>Is a length of (c).
5. The federally learning based non-cellular massive MIMO uplink reception method according to claim 3 or 4, wherein training the local model at each access point comprises:
respectively constructing a five-layer local DNN network at each access point;
let the model output of the first access point be f (θ l ,X l ) The loss function of the current model is defined as:
6. the federally learned honeycomb-free massive MIMO uplink receiving method according to claim 5, wherein the aggregation of the model parameters is performed at the central processor until the objective function of the system reaches convergence, comprising:
solving the optimization problem using federal learning is represented as follows:
wherein f represents a functional relationship between the model input and output, θ l For the weight parameters of the model at the ith AP, ω is the model weight parameter at the central processor and s.t is expressed as a constraint.
7. The federally learned cell-free massive MIMO uplink reception method according to claim 6, wherein the objective function achieving convergence further comprises:
each access point local model adopts small batch gradient descent to learn;
training local models of each access point respectively, and updating model parameters of each local model when the training epoch is equal to epoch_numUploading to a central processing unit;
the central processing unit aggregates the model parameters of each access point to obtain new global model parameters omega t The method comprises the following steps:and the updated global model parameter omega t Downloading to each access point;
each access point takes the received global model parameters as local model parameters, namely:and let t=t+1, repeating the training process of the local model of each access point until the system objective function converges.
8. The honeycomb-free large-scale MIMO uplink receiving system based on federal learning, the honeycomb-free large-scale MIMO uplink receiving method based on federal learning according to any one of claims 1 to 7, is characterized by comprising:
the neural network module comprises a local model and a global model; parameters for locally training and updating the data and aggregating the respective local models;
and the user cooperation module is responsible for coordinating the local model updating of each user, determining the users participating in training, and determining the times of updating the local model of each user.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: the processor, when executing the computer program, implements the steps of the method of any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program, when executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310795618.0A CN116827393B (en) | 2023-06-30 | 2023-06-30 | Honeycomb-free large-scale MIMO receiving method and system based on federal learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310795618.0A CN116827393B (en) | 2023-06-30 | 2023-06-30 | Honeycomb-free large-scale MIMO receiving method and system based on federal learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116827393A true CN116827393A (en) | 2023-09-29 |
| CN116827393B CN116827393B (en) | 2024-05-28 |
Family
ID=88127295
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310795618.0A Active CN116827393B (en) | 2023-06-30 | 2023-06-30 | Honeycomb-free large-scale MIMO receiving method and system based on federal learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116827393B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021179720A1 (en) * | 2020-10-12 | 2021-09-16 | 平安科技（深圳）有限公司 | Federated-learning-based user data classification method and apparatus, and device and medium |
| CN113791895A (en) * | 2021-08-20 | 2021-12-14 | 北京工业大学 | Edge calculation and resource optimization method based on federal learning |
| US20220052925A1 (en) * | 2018-12-07 | 2022-02-17 | Telefonaktiebolaget Lm Ericsson (Publ) | Predicting Network Communication Performance using Federated Learning |
| CN115099420A (en) * | 2022-08-26 | 2022-09-23 | 香港中文大学（深圳） | Model aggregation weight dynamic distribution method for wireless federal learning |
| CN115102592A (en) * | 2022-06-22 | 2022-09-23 | 电子科技大学 | Multi-user MIMO beam forming method based on federal learning |
| WO2022233511A2 (en) * | 2021-05-05 | 2022-11-10 | Nokia Technologies Oy | Efficient federated-learning model training in wireless communication system |
| CN115568025A (en) * | 2022-10-19 | 2023-01-03 | 广州技象科技有限公司 | Edge learning-oriented multi-user scheduling method, system and storage medium |
| CN116009677A (en) * | 2022-09-02 | 2023-04-25 | 南通大学 | Federal learning equipment end energy consumption optimization method based on Cell-Free mMIMO network |
-
2023
- 2023-06-30 CN CN202310795618.0A patent/CN116827393B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220052925A1 (en) * | 2018-12-07 | 2022-02-17 | Telefonaktiebolaget Lm Ericsson (Publ) | Predicting Network Communication Performance using Federated Learning |
| WO2021179720A1 (en) * | 2020-10-12 | 2021-09-16 | 平安科技（深圳）有限公司 | Federated-learning-based user data classification method and apparatus, and device and medium |
| WO2022233511A2 (en) * | 2021-05-05 | 2022-11-10 | Nokia Technologies Oy | Efficient federated-learning model training in wireless communication system |
| CN113791895A (en) * | 2021-08-20 | 2021-12-14 | 北京工业大学 | Edge calculation and resource optimization method based on federal learning |
| CN115102592A (en) * | 2022-06-22 | 2022-09-23 | 电子科技大学 | Multi-user MIMO beam forming method based on federal learning |
| CN115099420A (en) * | 2022-08-26 | 2022-09-23 | 香港中文大学（深圳） | Model aggregation weight dynamic distribution method for wireless federal learning |
| CN116009677A (en) * | 2022-09-02 | 2023-04-25 | 南通大学 | Federal learning equipment end energy consumption optimization method based on Cell-Free mMIMO network |
| CN115568025A (en) * | 2022-10-19 | 2023-01-03 | 广州技象科技有限公司 | Edge learning-oriented multi-user scheduling method, system and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 孙艳华, 乔兰, 王朱伟等: "智能反射面辅助MIMO系统混合波束赋形算法", 北京邮电大学学报, vol. 46, no. 3, 22 March 2023 (2023-03-22) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116827393B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Federated learning via over-the-air computation | |
| Yang et al. | Joint multi-user computation offloading and data caching for hybrid mobile cloud/edge computing | |
| CN109617584B (en) | MIMO system beam forming matrix design method based on deep learning | |
| CN110300417B (en) | Energy efficiency optimization method and device for unmanned aerial vehicle communication network | |
| KR102510513B1 (en) | Deep learning based beamforming method and apparatus for the same | |
| CN113791895A (en) | Edge calculation and resource optimization method based on federal learning | |
| US11742901B2 (en) | Deep learning based beamforming method and apparatus | |
| CN110535517B (en) | Beam forming calculation method and system for vehicle-to-vehicle communication | |
| Ren et al. | Computation offloading game in multiple unmanned aerial vehicle‐enabled mobile edge computing networks | |
| CN115412134A (en) | Off-line reinforcement learning-based user-centered non-cellular large-scale MIMO power distribution method | |
| Gupta et al. | LSTM-based energy-efficient wireless communication with reconfigurable intelligent surfaces | |
| Liu et al. | Cell-free XL-MIMO meets multi-agent reinforcement learning: Architectures, challenges, and future directions | |
| Wang et al. | ENGNN: A general edge-update empowered GNN architecture for radio resource management in wireless networks | |
| Liu et al. | Power allocation in ultra-dense networks through deep deterministic policy gradient | |
| CN104602279B (en) | Cooperative base station cluster-dividing method and device based on threshold value in a kind of heterogeneous network | |
| CN111405571B (en) | Spectrum sharing method for satellite unmanned aerial vehicle hybrid network | |
| CN103929224B (en) | Disturbance restraining method and device in cellular network | |
| CN116827393B (en) | Honeycomb-free large-scale MIMO receiving method and system based on federal learning | |
| CN106533524A (en) | Forming method for beam with maximum energy efficiency in distributed antenna system | |
| Sun et al. | Joint ddpg and unsupervised learning for channel allocation and power control in centralized wireless cellular networks | |
| CN105391511B (en) | Eliminate the method and device of heterogeneous network interlayer interference | |
| CN105409311A (en) | System and method for cooperative precoding in heterogenous two-tier wireless networks | |
| CN113411112A (en) | Method for constructing distributed robust multi-cell cooperative beam forming ADMM network | |
| Yuan et al. | Joint multi-ground-user edge caching resource allocation for cache-enabled high-low-altitude-platforms integrated network | |
| CN111669814A (en) | Power transmission optimization method and device of lunar surface wireless energy-carrying sensor network |
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 |