CN113301576B - Cellular network resource allocation method based on improved genetic algorithm - Google Patents
Cellular network resource allocation method based on improved genetic algorithm Download PDFInfo
- Publication number
- CN113301576B CN113301576B CN202110579983.9A CN202110579983A CN113301576B CN 113301576 B CN113301576 B CN 113301576B CN 202110579983 A CN202110579983 A CN 202110579983A CN 113301576 B CN113301576 B CN 113301576B
- Authority
- CN
- China
- Prior art keywords
- base station
- user
- representing
- population
- fitness
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
- H04W16/10—Dynamic resource partitioning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
A cellular network resource allocation method based on an improved genetic algorithm provides a cellular network architecture with cache enabled, and according to the fact that whether different transmission links exist in a base station of a user request resource or not, the resource allocation problem is converted into a solution constraint optimization problem by combining user association, the improved genetic algorithm is introduced to solve the optimization problem, and resource allocation is completed. In the improved genetic algorithm, an adaptive function, self-adaptive crossover and mutation probability, crossover and mutation operators are designed, the global search capability is better, the local optimization is not easy to be trapped, and the convergence speed is high. And the throughput of the whole system can be improved by combining the cache setting and a reasonable allocation strategy.
Description
Technical Field
The invention belongs to the technical field of cellular network resource allocation, and particularly relates to a cellular network resource allocation method based on an improved genetic algorithm.
Background
With the development of the internet, resources on the network are increasingly abundant, more and more people acquire the resources through mobile terminals, the corresponding requirements on mobile communication are higher and higher, and wireless resources are increasingly tense. By deploying a large number of small base stations with low energy consumption and low cost, the total system capacity and the spectrum utilization rate can be improved compared with a macro base station, and the problems of blind areas, weak coverage and the like are solved. Meanwhile, the dense deployment of the small base stations also brings the problem of interference among the base stations, and a reasonable resource allocation mode is needed to be adopted for solving the problem. In addition, the rapidly growing data service has replaced the traditional voice service to become a new mainstream service, and the requests of a large number of users for the data service are mainly concentrated on a few hot contents such as video and information. The cellular network with the cache is provided with the cache of the hot content at the small base station, the content requested by the user can be directly transmitted from the cache of the base station without the request of a server of a content provider, the burden of the network can be effectively reduced, the transmission efficiency of the system is improved, and the cellular network is a research hotspot. For the cellular network with cache, a new effective resource allocation method is also needed to exert the advantage of cache in a limited way.
For the resource allocation of cellular networks, a common approach is to translate it into an optimization problem. For the non-convex problem, if a better iteration initial value cannot be set, the algorithm is easy to fall into a local optimal solution and cannot converge to global optimal; the other is a heuristic evolutionary algorithm which mainly comprises a genetic algorithm, a simulated annealing algorithm, a particle swarm algorithm and the like, wherein the algorithm simulates certain laws in the nature, searches from random solutions, evaluates a population through an adaptive function and iteratively searches for an optimal solution.
The genetic algorithm mainly comprises the following steps of firstly randomly generating an initial population, carrying out adaptability evaluation on individuals in the population by using an adaptability function, describing the quality of the population, selecting the individuals from the population according to the probability to carry out cross and variation operations, reserving the individuals with good adaptability, eliminating the individuals with poor adaptability to form a new population, and continuously carrying out iterative optimization on the new population until the value meets a stop condition. The adaptive function of the genetic algorithm is not continuously and slightly constrained, has wide application range and better global search capability, also has the problem of premature convergence, and needs to be improved.
Disclosure of Invention
The invention provides a cellular network resource allocation method based on an improved genetic algorithm, which can effectively allocate resources to a cellular network and improve the throughput of a system, and specifically comprises the following steps:
a cellular network resource allocation method based on an improved genetic algorithm comprises the following steps:
step 1, establishing a cellular network system model with a cache, and analyzing the transmission rate of each user and the total transmission rate of the system;
step 2, associating the user with the small base station, and establishing a constraint optimization problem by taking the maximum total throughput of the system as a target;
step 3, designing an adaptability evaluation function by combining the penalty function, and carrying out adaptability evaluation on the population, wherein the individual adaptability of the objective function value is good, and the individual adaptability of the objective function value is poor;
step 4, calculating the cross and variation probability according to the iteration times;
and 5, performing cross and variation operation on the selected individuals in the population, and performing iteration until a termination condition is met, finally obtaining an optimal resource allocation scheme, and sequentially completing resource allocation.
Further, in step 1, the cellular network with the cache enabled is composed of a cloud pool and L base stations connected to Yun Chi, each base station and the cloud pool have a certain cache capacity, the network has U users in total, each base station has N sub-channels, each channel is allocated to only one user, the number of cache contents at each small base station is G, the number of cache contents at the cloud pool is M, wherein M is not greater than G;
setting a first transmission path, wherein when a user directly requests the cache content at the base station, the transmission rate is as follows:
wherein g is l,n In order to obtain the gain of the channel,to path loss, d l,n Representing the distance, σ, of a user from the base station 2 Is the power of Gaussian noise, h i,l,k For the indication function, when the user i occupies the subchannel n of the base station l, the value is 1, otherwise, the value is 0; p is a radical of l,n Representing the power, p, of the current user communicating with the small cell base station l using subchannel n l',n Representing the power when other users and the small base station l 'use the subchannel n for communication, wherein l' represents the base station except the current user connection base station in the system;
setting a second transmission path, when no user requests resources in the base station, transmitting the resources from the cloud pool to the small base station and then transmitting the resources to the user by using a backhaul link, wherein the speed of the backhaul link of each subchannel is v f If the data received by the user in the unit time is B, the following equation is used:
in the formula, R j Representing the actual rate, R, of the user through transmission path two i Representing the maximum transmission rate between the user and the base station, and obtaining the transmission rate of the user by arranging as follows:
the sum rate of each small base station is:
wherein I l ={i 1 .i 2 ,…,i l Denotes the set of users using transmission path one, I s = {j 1 ,j 2 ,…,j s Indicating a user set using a transmission path II;
the sum of the rates of all small base stations is:
further, the implementation process of step 2 is as follows:
each user connects to the base station closest to it and is scored using the following equation:
S(i)=ω 1 I i -ω 2 d i
in which I i To indicate a function, I i =1 denotes that the base station has cached the content required by user I, I i If =0 then no, d i Representing the distance, omega, of a user from the base station 1 、ω 2 Selecting N users with highest scores to access at the small base station as the weight coefficient;
after channel allocation is determined, power allocation is rewritten as a standard optimization problem with constraints:
g 2 (p)=p l.n -p c <0,l∈L,n∈N sub
constraint 1 indicates that the sum of the powers allocated to all the channels of the small cell is less than the maximum limit power of the small cell, and constraint 2 indicates that the power allocated to each channel is less than the maximum limit power of the channel.
Further, the adaptive function in step 3 is designed as follows:
an optimization problem with m inequality constraints and q equality constraints in total, the penalty function of which is constructed as follows:
wherein ω is j For different weight constraints, f j (x) The violation of the jth constraint by the data x is represented, and the higher the violation degree is, the greater the penalty strength is, namely, the greater the penalty function value is; h is j (x) Constraints representing the above inequality and equality, j representing the jth constraint;
normalizing the objective function and the penalty function:
wherein f is max And f min Respectively representing the maximum and minimum values, p, of the objective function in the current population max Is the maximum value of the penalty function, N pop For population size, the values of both the penalty function and the objective function are normalized to [0,N ] pop ]The following fitness function is constructed:
wherein T is the current evolutionary algebra, T is the maximum evolutionary algebra, r and b are parameters, and all the occupied proportions of the penalty function are controlled.
Further, the crossover and mutation probability in step 4 is calculated according to the following formula:
wherein P is c 、P m Respectively representing the cross probability and the mutation probability, P c_max 、P c_min Respectively representing the maximum and minimum cross probabilities, P m_max 、P m_min Respectively representing maximum and minimum mutation probabilities, f' representing the fitness of a parent, f representing the fitness of a mutated individual, f avg The average fitness of the population is represented, and the cross probability and the variation probability are adaptively changed along with the iteration.
Further, the crossover and mutation operation in step 5 is performed as follows:
with the difference in fitness between parents, the normalized search direction d is represented as:
where f is high And f low Representing the fitness of two parents in a cross pair;
selecting a population needing to execute the cross operation from the population according to the cross probability, and executing the cross operation according to the following formula:
wherein alpha is epsilon [0,1]Is a random number, and is a random number,respectively representing the more adaptive individuals and the less adaptive individuals in the parent performing the crossover operation,representing 4 offspring generated by the cross operation, selecting two individuals with high fitness to be added into the next generation population, and eliminating the rest two individuals with low fitness;
selecting a population needing to perform mutation operation according to the mutation probability for the crossed population, and performing the mutation operation according to the following formula:
v t+1 =ω t +c 1 (xbest t -x t )+c 2 (t o -x t )
wherein ω is t 、x o Calculated as follows:
wherein c is 1 、c 2 Is a random number between 0 and 1, xbest t Represents individuals with the highest fitness in the t-th generation population, omega ini 、ω end For a predetermined parameter, x t Is an individual to be mutated and is,is an individual after mutation.
Further, the termination condition in step 5 is that the algorithm reaches a set maximum number of iterations or the optimal solution is not changed any more.
The invention achieves the following beneficial effects: the invention provides a cellular network resource allocation method based on an improved genetic algorithm, which utilizes the cache capacity at a base station to design an adaptive evaluation function, self-adaptive cross mutation probability and cross and mutation operators, improves the prematurity phenomenon of the genetic algorithm, reduces the network load and improves the network throughput.
Drawings
Fig. 1 is a schematic diagram of a cellular network model with cache enabled according to an embodiment of the present invention.
Fig. 2 is a flowchart of a cellular network resource allocation method according to an embodiment of the present invention.
FIG. 3 is a flow chart of an improved genetic algorithm in an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the drawings in the specification.
As shown in fig. 1, a cache-enabled cellular network system model is established, where the model includes a cloud pool and L base stations connected to Yun Chi, where caches are both disposed at the cloud pool and at the base stations, and a central controller is disposed in Yun Chi, and resources are allocated.
The resources requested by the user have two transmission paths: (1) When the base station has the resource requested by the user, the resource is directly transmitted to the user by the base station; (2) And when the resource requested by the user does not exist at the base station, the resource is transmitted to the base station from the cache at the cloud pool and then transmitted to the user.
When the transmission path (1) is used, the transmission rate of the user is as follows:
wherein g is l,n In order to obtain the gain of the channel,for path loss, d l,n Representing the distance, σ, of a user from the base station 2 Is the power of Gaussian noise, h i,l,k For the indication function, it is 1 when user i occupies subchannel n of base station l, otherwise it is 0.
When the transmission path (2) is used, the transmission rate of the user is as follows:
the sum rate for each base station is:
wherein I l ={i 1 .i 2 ,…,i l Denotes a set of users using the transmission path (1), I s = {j 1 ,j 2 ,…,j s Denotes a set of users using the transmission path (2).
The sum of the rates of all small base stations is:
associating base stations with users, each user connecting to the base station closest thereto, and using S (i) = omega 1 I i -ω 2 d i Scoring, wherein I i To indicate a function, I i =1 denotes that the base station has cached the content required by user I, I i If =0 then no, d i Representing the distance, omega, of a user from the base station 1 ,ω 2 And selecting N users with highest scores to access at the small base station as the weight coefficients.
After channel allocation is determined, power allocation is rewritten to a standard constrained bundle optimization problem:
g 2 (p)=p l.n -p c <0,l∈L,n∈N sub
constraint 1 indicates that the sum of the powers allocated to all the channels of the small cell is less than the maximum limit power of the small cell, and constraint 2 indicates that the power allocated to each channel is less than the maximum limit power of the channel.
Designing an adaptive function, and constructing a penalty function for an optimization problem constrained by m inequalities and q equalities as follows:
wherein ω is j For different weight constraints, f j (x) And the violation of the jth constraint by the data x is represented, and the higher the violation degree is, the greater the penalty strength is, namely, the greater the penalty function value is.
Wherein f is max And f min Respectively representing the maximum and minimum values, p, of the objective function in the current population max Is the maximum value of the penalty function, N pop For population size, the values of the penalty function and the objective function are normalized to [0,N ] pop ]The following fitness function is constructed:
wherein T is the current evolutionary algebra, T is the maximum evolutionary algebra, and r and b are parameters.
The crossover and mutation probabilities are calculated as follows:
wherein P is c 、P m Respectively representing the cross probability and the mutation probability, P c_max 、P c_min Respectively representing the maximum and minimum cross probabilities, P m_max 、P m_min Respectively representing maximum and minimum mutation probabilities, f' representing the fitness of a parent, f representing the fitness of a mutated individual, f avg The average fitness of the population is represented, and the cross probability and the variation probability are adaptively changed along with the iteration.
Selecting a population needing to execute the cross operation from the population according to the cross probability, and executing the cross operation according to the following formula:
wherein d is calculated as:
wherein alpha is epsilon [0,1]Is a random number, and is a random number,respectively representing the more adaptive individuals and the less adaptive individuals in the parent performing the crossover operation,representing 4 offspring generated by the crossover operation, selecting two individuals with high fitness to be added into the next generation population, and eliminating the other two individuals with low fitness.
Selecting a population needing to perform mutation operation according to the mutation probability for the crossed population, and performing the mutation operation according to the following formula:
v t+1 =ω t +c 1 (xbest t -x t )+c 2 (x o -x t )
wherein ω is t ,x o Calculated as follows:
wherein c is 1 ,c 2 Is a random number between 0 and 1, xbest t Represents the individual with the highest fitness in the population of the t generation, omega ini 、ω end For a predetermined parameter, x t Is an individual to be mutated and is,is an individual after mutation.
The whole algorithm flow is as shown in fig. 3, firstly generating an initial population, calculating fitness of individuals in the population, calculating a cross probability by a cross probability formula, selecting a crossed parent from the population by the cross probability, executing cross operation according to the cross formula, replacing individuals with poor fitness in the population by offspring generated by cross to obtain a new population, calculating the fitness by a pair, calculating a variation probability by a variation probability formula, selecting individuals to be varied according to the variation probability, executing variation according to the variation formula, replacing the individuals with poor fitness in the population by the individuals generated by variation, completing one iteration, and continuing to execute operation until reaching the maximum iteration number.
The above description is only a preferred embodiment of the present invention, and the scope of the present invention is not limited to the above embodiment, but equivalent modifications or changes made by those skilled in the art according to the disclosure of the present invention should be included in the scope of the present invention as set forth in the appended claims.
Claims (4)
1. A cellular network resource allocation method based on improved genetic algorithm is characterized in that: the method comprises the following steps:
step 1, establishing a cellular network system model with a cache, and analyzing the transmission rate of each user and the total transmission rate of the system;
step 2, associating the user with the small base station, and establishing a constraint optimization problem by taking the maximum total throughput of the system as a target;
step 3, designing an adaptability evaluation function by combining the penalty function, and carrying out adaptability evaluation on the population, wherein the individual adaptability of the objective function value is good, and the individual adaptability of the objective function value is poor;
the adaptive evaluation function in step 3 is designed as follows:
an optimization problem with m inequality constraints and q equality constraints, the penalty function of which is constructed as follows:
wherein ω is j For different weight constraints, f j (x) The violation of the jth constraint by the data x is represented, and the higher the violation degree is, the greater the penalty strength is, namely, the greater the penalty function value is; h is j (x) Constraints representing the above inequality and equality, j representing the jth constraint;
normalizing the objective function and the penalty function:
wherein f is max And f min Respectively representing the maximum and minimum of the objective function in the current population, p max Is the maximum value of the penalty function, N pop For population size, the values of the penalty function and the objective function are normalized to[0,N pop ]The following fitness evaluation function is constructed:
wherein T is the current evolutionary algebra, T is the maximum evolutionary algebra, r and b are parameters, and the proportion occupied by the penalty function is controlled;
step 4, calculating the cross and mutation probability according to the iteration times;
the cross and mutation probability in the step 4 is calculated according to the following formula:
wherein P is c 、P m Respectively representing the cross probability and the mutation probability, P c_max 、P c_min Respectively representing the maximum and minimum cross probabilities, P m_max 、P m_min Respectively representing maximum and minimum mutation probabilities, f' representing the fitness of parent, f representing the fitness of mutated individual, f avg The average fitness of the population is represented, and the cross probability and the variation probability are adaptively changed along with the iteration;
step 5, performing cross and variation operation on the selected individuals in the population, and performing iteration until a termination condition is met, finally obtaining an optimal resource allocation scheme, and sequentially completing allocation of resources;
the crossing and mutation operation in the step 5 is carried out as follows:
with the difference in fitness between parents, the normalized search direction d is represented as:
here f high And f low Representing the fitness of two parents in a cross pair;
selecting a population needing to execute the cross operation from the population according to the cross probability, and executing the cross operation according to the following formula:
wherein alpha is epsilon [0,1]Is a random number, and is a random number,respectively representing the more adaptive individuals and the less adaptive individuals in the parent performing the crossover operation,representing 4 offspring generated by the cross operation, selecting two individuals with high fitness to be added into the next generation population, and eliminating the rest two individuals with low fitness;
selecting a population needing to perform mutation operation according to the mutation probability for the crossed population, and performing the mutation operation according to the following formula:
v t+1 =ω t +c 1 (xbest t -x t )+c 2 (x o -x t )
wherein ω is t 、x o Calculated as follows:
2. The method of claim 1, wherein the cellular network resource allocation method based on the improved genetic algorithm comprises: in the step 1, the cellular network with the cache enabled consists of a cloud pool and L base stations connected to Yun Chi, each base station and the cloud pool have a certain cache capacity, the network has U users in total, each base station has N sub-channels, each channel is only allocated to one user, the number of cache contents at each small base station is G, the number of cache contents at the cloud pool is M, and M is less than or equal to G;
setting a first transmission path, wherein when a user directly requests the cache content at the base station, the transmission rate is as follows:
wherein g is l,n In order to obtain the gain of the channel,to path loss, d l,n Representing the distance, σ, of a user from the base station 2 Is the power of Gaussian noise, h i,l,n For the indication function, when the user i occupies the subchannel n of the base station l, the value is 1, otherwise, the value is 0; p is a radical of l,n Representing the power, p, of the current user communicating with the small cell base station l using subchannel n l',n Representing the power when other users and the small base station l 'use the subchannel n for communication, wherein l' represents the base station except the current user connection base station in the system;
setting a second transmission path, when no user requests resources in the base station, transmitting the resources from the cloud pool to the small base station and then transmitting the resources to the user by using a backhaul link, wherein the speed of the backhaul link of each subchannel is v f If the data received by the user in the unit time is B, the following equation is given:
in the formula, R j Representing the actual rate, R, of the user through transmission path two i Representing the maximum transmission rate between the user and the base station, and obtaining the transmission rate of the user by sorting as follows:
the sum rate of each small base station is:
wherein I l ={i 1 .i 2 ,…,i l Denotes the set of users using transmission path one, I s ={j 1 ,j 2 ,…,j s Means forUsing the user set of the transmission path two;
the sum of the rates of all small base stations is:
3. the method of claim 2, wherein the cellular network resource allocation method based on the improved genetic algorithm comprises: the implementation process of the step 2 is as follows:
each user connects to the base station closest to it and is scored using the following equation:
S(i)=ω 1 I i -ω 2 d i
wherein I i To indicate a function, I i =1 denotes that the base station has cached the content required by user I, I i If =0 then no, d i Representing the distance, omega, of a user from the base station 1 、ω 2 Selecting N users with highest scores to access at the small base station as the weight coefficient;
after channel allocation is determined, power allocation is rewritten as a standard optimization problem with constraints:
g 2 (p)=p l,n -p c <0,l∈L,n∈N sub
constraint 1 indicates that the sum of the powers allocated to all channels of the small cell is less than the maximum limit power P of the small cell max Constraint 2 means that the power allocated to each channel is less than the maximum limit power p of the channel c 。
4. The method of claim 1, wherein the cellular network resource allocation method based on the improved genetic algorithm comprises: and 5, the termination condition in the step 5 is that the algorithm reaches the set maximum iteration number or the optimal solution is not changed any more.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110579983.9A CN113301576B (en) | 2021-05-26 | 2021-05-26 | Cellular network resource allocation method based on improved genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110579983.9A CN113301576B (en) | 2021-05-26 | 2021-05-26 | Cellular network resource allocation method based on improved genetic algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113301576A CN113301576A (en) | 2021-08-24 |
CN113301576B true CN113301576B (en) | 2023-03-21 |
Family
ID=77325205
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110579983.9A Active CN113301576B (en) | 2021-05-26 | 2021-05-26 | Cellular network resource allocation method based on improved genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113301576B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117785949B (en) * | 2024-02-28 | 2024-05-10 | 云南省地矿测绘院有限公司 | Data caching method, electronic equipment, storage medium and device |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102711266B (en) * | 2012-05-17 | 2014-08-13 | 北京邮电大学 | Scheduling and resource allocation joint optimization method based on genetic algorithm |
CN105320988A (en) * | 2014-07-14 | 2016-02-10 | 重庆邮电大学 | Parameter optimization method for wireless energy transmission system based on improved genetic algorithm |
CN108738048B (en) * | 2018-04-25 | 2021-02-26 | 杭州电子科技大学 | Active storage method of maximized fairness base station based on genetic algorithm |
CN108880663B (en) * | 2018-07-20 | 2020-09-22 | 大连大学 | Space-ground integrated network resource allocation method based on improved genetic algorithm |
-
2021
- 2021-05-26 CN CN202110579983.9A patent/CN113301576B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN113301576A (en) | 2021-08-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113296845B (en) | Multi-cell task unloading algorithm based on deep reinforcement learning in edge computing environment | |
CN111585816B (en) | Task unloading decision method based on adaptive genetic algorithm | |
Jiang et al. | Optimal cooperative content caching and delivery policy for heterogeneous cellular networks | |
CN105979553B (en) | A kind of hierarchical network handover decisions method based on fuzzy logic and TOPSIS algorithm | |
CN111970733A (en) | Deep reinforcement learning-based cooperative edge caching algorithm in ultra-dense network | |
He et al. | Resource allocation based on graph neural networks in vehicular communications | |
Wang et al. | Joint resource allocation and power control for D2D communication with deep reinforcement learning in MCC | |
Tang et al. | Collaborative cache-aided relaying networks: Performance evaluation and system optimization | |
Gao et al. | Reinforcement learning based cooperative coded caching under dynamic popularities in ultra-dense networks | |
CN110191489B (en) | Resource allocation method and device based on reinforcement learning in ultra-dense network | |
CN112637907B (en) | Combined optimization method for user multi-association and downlink power distribution in millimeter wave network | |
CN113473580B (en) | User association joint power distribution method based on deep learning in heterogeneous network | |
Liu et al. | Deep Dyna-reinforcement learning based on random access control in LEO satellite IoT networks | |
Cai et al. | Joint optimal multi-connectivity enabled user association and power allocation in mmWave networks | |
Qiu et al. | Subchannel assignment and power allocation for time-varying fog radio access network with NOMA | |
CN113301576B (en) | Cellular network resource allocation method based on improved genetic algorithm | |
CN106060876A (en) | Load balancing method for heterogeneous wireless network | |
Liu et al. | Deep reinforcement learning-based MEC offloading and resource allocation in uplink NOMA heterogeneous network | |
CN110602722A (en) | Design method for joint content pushing and transmission based on NOMA | |
CN108965034A (en) | Small-cell base station super-intensive deployment under user-association to network method | |
CN116828534B (en) | Intensive network large-scale terminal access and resource allocation method based on reinforcement learning | |
CN109547979B (en) | Content distribution method combining energy consumption and user fairness under D2D cache network | |
Li et al. | Joint access point selection and resource allocation in MEC-assisted network: A reinforcement learning based approach | |
Zhang et al. | Energy efficient resource allocation in millimeter-wave-based fog radio access networks | |
CN113518457B (en) | Power distribution strategy based on one-dimensional deep convolutional neural 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 | ||
GR01 | Patent grant |