CN113518457A - Power distribution strategy based on one-dimensional deep convolutional neural network - Google Patents

Power distribution strategy based on one-dimensional deep convolutional neural network Download PDF

Info

Publication number
CN113518457A
CN113518457A CN202110437441.8A CN202110437441A CN113518457A CN 113518457 A CN113518457 A CN 113518457A CN 202110437441 A CN202110437441 A CN 202110437441A CN 113518457 A CN113518457 A CN 113518457A
Authority
CN
China
Prior art keywords
neural network
convolutional neural
base station
dimensional
deep convolutional
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
Application number
CN202110437441.8A
Other languages
Chinese (zh)
Other versions
CN113518457B (en
Inventor
李君�
朱明浩
仲星
张茜茜
沈国丽
王秀敏
李正权
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.)
Binjiang College of Nanjing University of Information Engineering
Original Assignee
Binjiang College of Nanjing University of Information Engineering
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 Binjiang College of Nanjing University of Information Engineering filed Critical Binjiang College of Nanjing University of Information Engineering
Priority to CN202110437441.8A priority Critical patent/CN113518457B/en
Publication of CN113518457A publication Critical patent/CN113518457A/en
Application granted granted Critical
Publication of CN113518457B publication Critical patent/CN113518457B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0473—Wireless resource allocation based on the type of the allocated resource the resource being transmission power
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • 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
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a power distribution strategy based on a one-dimensional deep convolutional neural network, belongs to the field of communication systems, and aims at overcoming the defects of the existing power distribution algorithm based on the deep neural network. The method researches a resource allocation strategy of replacing the traditional algorithm with the one-dimensional convolutional neural network, learns the power allocation effect obtained based on the traditional algorithm through a supervised learning mode, realizes quick and reliable online decision, overcomes the defect of limited learning capability compared with the traditional power allocation algorithm based on deep learning, and has higher prediction capability.

Description

Power distribution strategy based on one-dimensional deep convolutional neural network
Technical Field
The invention belongs to the field of communication systems, and particularly relates to a power distribution strategy based on a one-dimensional deep convolutional neural network.
Background
With the popularization and promotion of internet technology and base station densification concepts, increasing service requirements and requirements on capacity and rate of a cellular network gradually increase, how to allocate resources in a communication system becomes a problem to be considered, and more specifically, how to allocate power becomes a problem to be considered.
In recent years, deep learning techniques are increasingly used in wireless communication networks due to convenience and timeliness of channel sample collection in wireless communication networks, and neural networks in deep learning, which can achieve a satisfactory nonlinear approximation to conventional algorithms, are widely studied, and include Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN). In communication systems, many researches on realizing the maximum minimization of an objective function by parameter control by using a neural network are concerned, for example, the maximization of a spectrum efficiency or energy efficiency target is realized by approximating a sub-gradient algorithm by using a deep neural network, or the maximization of a system and a rate is realized by approximating an IPM algorithm; in the aspect of the convolutional neural network, a convolutional filter is used for extracting local features, and the maximization of spectral efficiency, energy efficiency or sum rate is realized in a supervised learning mode.
In general, the neural network has the advantages of nonlinear approximation to the performance of the traditional algorithm, and meanwhile, the complexity is low, which is very helpful for real-time decision in a communication system, however, the deep neural network has limited feature learning capability, and the prediction capability of the resource allocation strategy based on the deep neural network is insufficient, so that the research and the improvement of the neural network have great significance for communication resource allocation based on deep learning by improving the prediction capability.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a power distribution strategy based on a one-dimensional deep convolutional neural network, and more real-time and reliable online power distribution is realized in a communication system.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the following technical scheme: the power distribution strategy based on the one-dimensional deep convolutional neural network comprises the following steps:
step 1, communication modeling;
step 2, collecting data set, collecting channel state information h between base station and userkAnd optimal power value p under water filling algorithm*As a set of data sets;
step 3, constructing a one-dimensional depth convolution neural network, and initializing the weight of the neural network;
step 4, training the neural network, and constructing the predicted power value p and the optimal power value p of the neural network*The mean square error between the two is used as a loss function, and an optimization algorithm during training is determined;
and 5, finishing training and storing the neural network when the loss function is smaller than a preset value or meets the iteration times of the training.
Further, in the step 1, the communication modeling specifically includes: the method comprises the steps of establishing a single-cell multi-user communication environment, wherein the positions of a base station and users in the environment are uniformly distributed, and a downlink channel between the base station and the users is a Rayleigh fading channel.
Further, in the step 2, 10 ten thousand sets of the data sets are collected repeatedly.
Further, the step 1 specifically includes the following steps: assuming that a single-cell cellular network environment is considered, a base station is deployed in the center of a cell, and k users are randomly distributed around the cell, wherein the user set is as follows: k ·, K }, which are served by centrally located base stations.
Further, in step 1, a rayleigh fading channel environment with an average value of 1 is considered, where channel state information from user k to the base station is hkK belongs to K, and the channel state information is transmission power p between users and a base station in consideration of path loss between users and the base station in a Rayleigh fading environmentkRepresents; the SINR (signal to interference plus noise ratio) at user k in the downlink is then expressed as
Figure BDA0003033624460000021
Wherein
Figure BDA0003033624460000022
Is background noise, hkIs the channel state information between user k and the base station, pkIs the transmission power from the user base station to user k, the achievable rate is denoted as Rk＝log2(1+SINRk) The system and rate performance is expressed as
Figure BDA0003033624460000023
Further, in step 3, the one-dimensional convolutional neural network is responsible for learning a mapping relationship from the channel gain of the input signal to the optimal user association or power allocation of the output signal.
Further, the step 3 specifically includes the following steps:
step 1) collecting channel state information between users and base stations in the environment, and operating a water injection algorithm to obtain a corresponding optimal power value p*Repeating the steps for 10 ten thousand times to obtain a data set;
step 2) determining the segmentation ratio of the training data and the test data;
and 3) designing and constructing a framework structure of the one-dimensional convolutional neural network, and initializing a weight parameter w and a bias parameter b of the neural network.
Further, in step 4, constructing a neural network specifically includes a predicted power value p and an optimal power value p*Mean square error therebetween as a loss function
Figure BDA0003033624460000031
An Adam optimizer is determined to optimize the training neural network.
Further, the step 5 is specifically to satisfy the iteration number epochs is 300 or the loss function is less than the preset value loss is less than or equal to 0.1 × 10-3The network is saved.
The invention aims at the research focus, namely how to control the downlink transmission power of the base station and the user can quickly and reliably maximize the sum rate of the whole system. A resource allocation strategy of replacing a traditional algorithm with a one-dimensional convolutional neural network is researched, a power allocation effect obtained based on the traditional algorithm is learned through a supervised learning mode, rapid and reliable online decision is realized, compared with the traditional power allocation algorithm based on deep learning, the defect of limited learning capability is overcome, and the prediction capability is higher.
Has the advantages that: compared with the prior art, the power distribution strategy based on the one-dimensional depth convolution neural network is used for improving the network prediction capability; aiming at the existing neural network power distribution strategy based on the full-connection form, the power distribution strategy based on the one-dimensional convolutional neural network is provided, compared with the traditional method, the method is short in calculation time and low in complexity, compared with the method based on the full-connection neural network, the prediction capability is stronger, and more real-time and reliable online power distribution is realized in a communication system.
Drawings
FIG. 1 is a diagram of a one-dimensional convolutional neural network architecture;
FIG. 2 is a diagram of a deep neural network architecture;
fig. 3 is a system flow diagram.
Detailed Description
The present invention will be further described with reference to the following embodiments.
The resource allocation strategy based on the deep convolutional neural network has better prediction capability than the deep neural network based on the full-connection form, and on the basis of the conclusion, the improved power allocation strategy based on the one-dimensional deep convolutional neural network is researched for the one-dimensional array.
Assuming that a single-cell cellular network environment is considered, a base station is deployed in the center of a cell, and k users are randomly distributed around the cell, wherein the user set is as follows: k {1,2,3 …, K }, which are served by a centrally located base station, considering a rayleigh fading channel environment with an average value of 1, wherein channel state information of user K to base station is hkK belongs to K, and the channel state information is transmission power p between users and a base station in consideration of path loss between users and the base station in a Rayleigh fading environmentkAnd (4) showing. The SINR (signal to interference plus noise ratio) at user k in the downlink is then expressed as
Figure BDA0003033624460000041
Wherein
Figure BDA0003033624460000042
Is background noise, the achievable rate of which is denoted as Rk：RK＝log2(1+Sk) System and rate performance can be expressed as
Figure BDA0003033624460000043
Firstly, a utility function maximization problem is provided, a power allocation strategy based on deep learning is provided aiming at the utility function maximization problem, specifically, the power allocation strategy based on a one-dimensional convolutional neural network is provided, and the structural design of the power allocation strategy is specifically described in the figures and the detailed description of the specific implementation mode.
As shown in FIG. 3, the training process of the present invention comprises the following steps:
step 1, communication modeling is carried out, and a single-cell multi-user communication scene is established;
step 2, collecting a data set which comprises channel state information and an optimal power distribution label under a water injection algorithm;
step 3, designing and constructing a one-dimensional convolutional neural network, and initializing a weight coefficient;
step 4, constructing a loss function and determining an optimization algorithm to train a neural network;
and 5, saving the network when the iteration times or the loss function is less than a preset value.
The neural network structure used in the present invention is shown in fig. 1, and they are responsible for learning the mapping relationship between the channel gain of the input signal to the optimal user association or power allocation of the output signal, and the whole process includes two parts of data set collection and neural network training. The method comprises the following specific steps:
step one, collecting channel state information between users and base stations in the environment, and operating a water injection algorithm to obtain a corresponding optimal power distribution label p*Repeating the steps for 10 ten thousand times to obtain a data set;
determining the division ratio of the training data and the test data;
thirdly, constructing a one-dimensional depth convolution neural network and initializing the weight of the neural network;
step four, training a neural network, and constructing a network predicted value p and a label value p*The mean square error between the two is used as a loss function, and an optimization algorithm during training is determined;
and step five, finishing training and storing the neural network when the loss function is smaller than a preset value or meets the iteration times of the training.
The following describes a process of optimal resource allocation by a neural network as a specific example. Assuming that a base station is deployed in the center of the environment in a square area of 1km × 1km, 10 user terminals are randomly distributed around the base station, and the channel state information between the users and the base station is the path loss under the consideration of the rayleigh fading channel, and the value of the path loss is 128.1+37.6log10(d) Where d is the euclidean distance between the user and the base station. For the sum rate formula, a system and rate maximization problem is proposed:
Figure BDA0003033624460000051
the constraint condition is
Figure BDA0003033624460000052
Means that the sum of the powers allocated to its associated users by each base station cannot be greater than the total power of the base station, where the total transmission power p of the base station is knowntotal20 dBm. After a target problem is determined, a power distribution strategy based on a one-dimensional convolutional neural network is provided, firstly, a data set is collected, channel state information between a user and a base station is collected, a water injection power distribution algorithm is operated to obtain a corresponding optimal power distribution label, the operation is repeated for 10 ten thousand times to obtain the data set, and the division ratio of training test data is determined to be 9: 1, designing and constructing a one-dimensional convolutional neural network, wherein the neural network comprises 7 layers including an input layer, 3 layers of one-dimensional convolutional layers, a flat layer, a full-connection layer and an output layer, the size of convolutional cores in the convolutional layers is 3 multiplied by 1, the number of the cores in each layer is 8, 16 and 32 respectively, the number of neurons in the full-connection layer is 3000, and the number of the neurons in the output layer is 10 while the number of the neurons in the output layer is consistent with the number of users. The activation function in the hidden layer is ReLU, the activation function of the output layer is sigmoid function, the predicted power distribution scheme P is output through the output layer, and the mean square error between the target value and the predicted value is constructed to be used as a loss function
Figure BDA0003033624460000053
Wherein M is 25, the number of samples of each batch in the small-batch gradient descent algorithm, the training period epochs is 300, and the Adam algorithm is selected as the optimization algorithm.
In order to verify the performance of the designed one-dimensional convolutional neural network, the performance of the neural network in a test stage needs to be tested and observed, a test data set is generated as a training data set, and the reliability of the scheme is judged by inputting channel gain and observing the error between an output predicted value and a tag value. Meanwhile, as comparison, a training deep neural network is designed as a comparison object, and the structure of the deep neural network is shown in fig. 2, so as to verify that the prediction capability of the proposed network structure in the aspect of power distribution is superior to that of a deep neural network scheme.
The specific implementation steps of the power control strategy based on the one-dimensional convolutional neural network are as follows:
step one, communication modeling is carried out, a single-cell multi-user communication environment is established, and path loss between a user and a base station under a Rayleigh fading channel is considered in a scene to serve as channel state information.
And step two, collecting a data set, namely collecting 10 ten thousand groups of data sets, wherein the data sets comprise channel state information between a user and a base station and an optimal power allocation label under a corresponding water filling algorithm.
Determining the segmentation ratio 9 of the training data set and the test data set: 1.
designing a one-dimensional convolutional neural network framework and initializing the weight.
And fifthly, constructing a loss function and determining an optimization algorithm to train a neural network.
And step six, continuously adjusting the hyper-parameters during training, and saving the network when the iteration times are met.
And step seven, inputting the test set into the trained neural network, and verifying the reliability of the proposed scheme.
The above description is only a preferred embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and variations can be made without departing from the technical principles of the present invention, and these modifications and variations should also be construed as the scope of the present invention.

Claims (9)

1. The power distribution strategy based on the one-dimensional depth convolution neural network is characterized in that: the method comprises the following steps:
step 1, communication modeling;
step 2, collecting data set, collecting channel state information h between base station and userkAnd optimal power value p under water filling algorithm*As a set of data sets;
step 3, constructing a one-dimensional depth convolution neural network, and initializing the weight of the neural network;
step 4, training the neural network, and constructing the predicted power value p and the optimal power value p of the neural network*The mean square error between the two is used as a loss function, and an optimization algorithm during training is determined;
and 5, finishing training and storing the neural network when the loss function is smaller than a preset value or meets the iteration times of the training.
2. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 1, wherein: in the step 1, the communication modeling specifically includes: the method comprises the steps of establishing a single-cell multi-user communication environment, wherein the positions of a base station and users in the environment are uniformly distributed, and a downlink channel between the base station and the users is a Rayleigh fading channel.
3. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 1, wherein: in the step 2, 10 ten thousand groups of the data sets are collected repeatedly.
4. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 1, wherein: the step 1 specifically comprises the following steps: assuming a single-cell cellular network environment, a base station is deployed in the center of a cell, k users are randomly distributed around the cell, and a user set is as follows: k ·, K }, which are served by centrally located base stations.
5. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 4, wherein: in step 1, a rayleigh fading channel environment with an average value of 1 is considered, where channel state information from a user k to a base station is hkK belongs to K, and the channel state information is transmission power p between users and a base station in consideration of path loss between users and the base station in a Rayleigh fading environmentkRepresents; the SINR at user k in the downlink is expressed as
Figure FDA0003033624450000011
Wherein
Figure FDA0003033624450000012
Is background noise, hkIs the channel state information between user k and the base station, pkIs subscriber base station to subscriberk, the achievable rate is denoted Rk＝log2(1+SINRk) The system and rate performance is expressed as
Figure FDA0003033624450000013
6. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 1, wherein: in the step 3, the one-dimensional convolutional neural network is responsible for learning the mapping relationship from the channel gain of the input signal to the optimal user association or power allocation of the output signal.
7. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 1, wherein: the step 3 specifically comprises the following steps:
step 1) collecting channel state information between users and base stations in the environment, and operating a water injection algorithm to obtain a corresponding optimal power value p*Repeating the steps for 10 ten thousand times to obtain a data set;
step 2) determining the segmentation ratio of the training data and the test data;
and 3) designing and constructing a framework structure of the one-dimensional convolutional neural network, and initializing a weight parameter w and a bias parameter b of the neural network.
8. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 7, wherein: in step 4, a neural network is constructed, specifically, a predicted power value p and an optimal power value p*Mean square error therebetween as a loss function
Figure FDA0003033624450000021
An Adam optimizer is determined to optimize the training neural network.
9. The one-dimensional deep convolutional neural network-based power allocation strategy of claim 8, wherein: the steps areStep 5 is specifically to satisfy the requirement that the iteration number epochs is 300 or the loss function is less than the preset value loss which is less than or equal to 0.1 multiplied by 10-3The network is saved.
CN202110437441.8A 2021-04-22 2021-04-22 Power distribution strategy based on one-dimensional deep convolutional neural network Active CN113518457B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110437441.8A CN113518457B (en) 2021-04-22 2021-04-22 Power distribution strategy based on one-dimensional deep convolutional neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110437441.8A CN113518457B (en) 2021-04-22 2021-04-22 Power distribution strategy based on one-dimensional deep convolutional neural network

Publications (2)

Publication Number Publication Date
CN113518457A true CN113518457A (en) 2021-10-19
CN113518457B CN113518457B (en) 2024-06-07

Family

ID=78061598

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110437441.8A Active CN113518457B (en) 2021-04-22 2021-04-22 Power distribution strategy based on one-dimensional deep convolutional neural network

Country Status (1)

Country Link
CN (1) CN113518457B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115361731A (en) * 2022-08-18 2022-11-18 东南大学 Multi-user power distribution method based on one-dimensional convolutional neural network

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140046771A1 (en) * 2011-03-28 2014-02-13 Verizon Patent And Licensing Inc. Generating and using network data to provide a content customization service
US20170353260A1 (en) * 2016-06-02 2017-12-07 Samsung Electronics Co., Ltd Network configuration
CN112153617A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Terminal equipment transmission power control method based on integrated neural network
CN112153616A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Power control method in millimeter wave communication system based on deep learning
CN112153615A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Deep learning-based user association method in multi-cell cellular D2D equipment
CN112492686A (en) * 2020-11-13 2021-03-12 辽宁工程技术大学 Cellular network power distribution method based on deep double-Q network

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140046771A1 (en) * 2011-03-28 2014-02-13 Verizon Patent And Licensing Inc. Generating and using network data to provide a content customization service
US20170353260A1 (en) * 2016-06-02 2017-12-07 Samsung Electronics Co., Ltd Network configuration
CN112153617A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Terminal equipment transmission power control method based on integrated neural network
CN112153616A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Power control method in millimeter wave communication system based on deep learning
CN112153615A (en) * 2020-09-15 2020-12-29 南京信息工程大学滨江学院 Deep learning-based user association method in multi-cell cellular D2D equipment
CN112492686A (en) * 2020-11-13 2021-03-12 辽宁工程技术大学 Cellular network power distribution method based on deep double-Q network

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115361731A (en) * 2022-08-18 2022-11-18 东南大学 Multi-user power distribution method based on one-dimensional convolutional neural network

Also Published As

Publication number Publication date
CN113518457B (en) 2024-06-07

Similar Documents

Publication Publication Date Title
CN112153616B (en) Power control method in millimeter wave communication system based on deep learning
CN111800828B (en) Mobile edge computing resource allocation method for ultra-dense network
CN111372314A (en) Task unloading method and task unloading device based on mobile edge computing scene
CN113473580B (en) User association joint power distribution method based on deep learning in heterogeneous network
CN112153615B (en) Deep learning-based user association method in multi-cell cellular D2D equipment
CN109729528A (en) A kind of D2D resource allocation methods based on the study of multiple agent deeply
CN109862573B (en) LTE hybrid networking self-planning method based on multi-target particle swarm
CN113596785B (en) D2D-NOMA communication system resource allocation method based on deep Q network
CN112153617B (en) Terminal equipment transmission power control method based on integrated neural network
CN112203345B (en) D2D communication energy efficiency maximization power distribution method based on deep neural network
CN109195207B (en) Energy-collecting wireless relay network throughput maximization method based on deep reinforcement learning
CN109089307B (en) Energy-collecting wireless relay network throughput maximization method based on asynchronous dominant actor critic algorithm
CN112637907B (en) Combined optimization method for user multi-association and downlink power distribution in millimeter wave network
Chen et al. A deep learning based resource allocation scheme in vehicular communication systems
CN113242066B (en) Multi-cell large-scale MIMO communication intelligent power distribution method
Lei et al. Learning-based resource allocation: Efficient content delivery enabled by convolutional neural network
CN110191489A (en) Resource allocation methods and device based on intensified learning in a kind of super-intensive network
CN111010222B (en) Deep learning-based large-scale MIMO downlink user scheduling method
CN114615730A (en) Content coverage oriented power distribution method for backhaul limited dense wireless network
CN108990141B (en) Energy-collecting wireless relay network throughput maximization method based on deep multi-network learning
CN114158010B (en) Unmanned aerial vehicle communication system and resource allocation strategy prediction method based on neural network
CN113518457B (en) Power distribution strategy based on one-dimensional deep convolutional neural network
CN117119486B (en) Deep unsupervised learning resource allocation method for guaranteeing long-term user rate of multi-cell cellular network
CN116981090A (en) D2D communication resource allocation method based on improved D3QN under multiple base stations
CN114531726B (en) Power control method based on dense wireless 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