WO2019213997A1 - 通信网络中用户终端与控制器的匹配方法、装置及系统 - Google Patents

通信网络中用户终端与控制器的匹配方法、装置及系统 Download PDF

Info

Publication number
WO2019213997A1
WO2019213997A1 PCT/CN2018/088625 CN2018088625W WO2019213997A1 WO 2019213997 A1 WO2019213997 A1 WO 2019213997A1 CN 2018088625 W CN2018088625 W CN 2018088625W WO 2019213997 A1 WO2019213997 A1 WO 2019213997A1
Authority
WO
WIPO (PCT)
Prior art keywords
controller
user terminal
preset
load information
user terminals
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.)
Ceased
Application number
PCT/CN2018/088625
Other languages
English (en)
French (fr)
Inventor
粟欣
葛璐
吴斌伟
曾捷
刘蓓
许希斌
赵明
肖立民
王京
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.)
Tsinghua University
Original Assignee
Tsinghua University
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 Tsinghua University filed Critical Tsinghua University
Publication of WO2019213997A1 publication Critical patent/WO2019213997A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W76/00—Connection management
    • H04W76/10—Connection setup
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition

Definitions

  • the present application relates to the field of communications technologies, and in particular, to a method, device, and system for matching a user terminal and a controller in a communication network.
  • the matching connection between the user terminal and the controller is a problem that is of particular concern in the field of communication technology.
  • the commonly used user terminal and controller matching mechanism is a user controller pairing method based on integer linear programming, or a matching method of a user terminal and a controller based on an alliance game.
  • the embodiment of the present invention provides a method, a device, and a system for matching a user terminal and a controller in a communication network, which are used to solve the above-mentioned defects in the prior art.
  • an embodiment of the present invention provides a method for matching a user terminal and a controller in a communication network, which is applied to a system including N user terminals and M controllers, where the N user terminals are limited to Q. And the user terminals in each of the packets are: the request generation rate is the same and the transmission delays of the same controller are the same; the Q is smaller than N, and the N, M, and Q are positive integers; the method includes:
  • the user terminal selects any one of the plurality of controllers to perform a matching connection
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • an embodiment of the present invention provides a method for matching a user terminal and a controller in a communication network, which is applied to a system including N user terminals and M controllers, where the N user terminals are limited to Q
  • the grouping, the user terminal in each group satisfies: the request generation rate is the same and the transmission delays of the same controller are the same; the Q is smaller than N, and the N, M, and Q are positive integers; the method includes:
  • controller After the controller matches the connection with the user terminal, determining the load information in the current matching connection state in the preset first period;
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • an embodiment of the present invention provides a user terminal, including:
  • An initialization module configured to select any one of the plurality of controllers to perform a matching connection
  • a receiving module configured to receive, in a preset first period, load information sent by the multiple controllers in a current matching connection state
  • a matching module configured to adjust the current matching connection according to the initialization information, the preset utility function, and the received load information, so that an average response delay of all user terminals in the network system reaches a minimum value
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • an embodiment of the present invention provides a controller, including:
  • a determining module configured to determine, in a preset first period, load information in a current matching connection state after the controller matches the connected user terminal;
  • a sending module configured to send the load information to the N user terminals, so that each user terminal of the N user terminals is configured according to initialization information, a preset utility function, and the received load information. Adjusting the current matching connection so that the average response delay of all user terminals in the network system reaches a minimum value;
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • an embodiment of the present invention provides a matching system between a user terminal and a controller in a communication network, including: N user terminals and M controllers; the N user terminals are limited to Q groups, each The user terminal in the packet satisfies: the request generation rate is the same and the transmission delays of the same controller are the same; the Q is smaller than N, and the N, M, and Q are positive integers;
  • the user terminal is configured to select any one of the multiple controllers to perform a matching connection; receive, in a preset first period, load information sent by the multiple controllers in a current matching connection state; The information, the preset utility function, and the received load information are adjusted to adjust the current matching connection so that the average response delay of all user terminals in the network system reaches a minimum value;
  • the controller is configured to: after the controller matches the connected user terminal, determine load information in a current matching connection state in a preset first period; and send the load information to the N user terminals to And causing each of the N user terminals to adjust the current matching connection according to the initialization information, the preset utility function, and the received load information, so that the average response delay of all user terminals in the network system is reached.
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the user's utility function is derived by using the potential function in the evolutionary game framework, and the matching connection relationship between the user terminal and the controller can be obtained in a distributed manner. Even if the pairing of the user terminal and the controller reaches a steady state, once the network condition changes, the user terminal automatically selects a controller with a higher utility value for matching, and the pairing relationship automatically converges to a new steady state. Moreover, by adopting the distributed method, the centralized control node is not needed, and the calculation tasks are shared to each user, which greatly reduces the complexity of the matching, improves the matching efficiency, and ensures the matching accuracy.
  • FIG. 1 is a schematic structural diagram of a communication network system of a user terminal and a controller according to an embodiment of the present disclosure
  • FIG. 2 is a schematic diagram of steps of a method for matching a user terminal and a controller provided by the present application;
  • FIG. 3 is a schematic diagram of detailed steps of a method for matching a user terminal and a controller provided by the present application;
  • 4a is a second schematic diagram of steps of a method for matching a user terminal and a controller provided by the present application;
  • 4b is a specific schematic diagram of step 41 of a method for matching a user terminal and a controller provided by the present application;
  • FIG. 5 is a schematic structural diagram of a user terminal provided by the present application.
  • FIG. 6 is a schematic structural diagram of a controller provided by the present application.
  • FIG. 1 is a schematic diagram of a communication network system architecture of a user terminal and a controller in network communication according to an embodiment of the present disclosure, the system mainly includes:
  • a plurality of user terminals C and a plurality of controllers S each of which can be matched and connected to a corresponding one of the controllers S, and each controller S can be matched with one or more user terminals C.
  • the user terminal C can be understood as a mobile terminal or other electronic device that can match the information transmission service connected to the controller.
  • N user terminals C and M controllers S can be introduced as an example.
  • the N user terminals C are defined as Q packets, and the user terminals C in each packet satisfy the condition that the request generation rates of different user terminals C in the same packet are the same, and different user terminals in the same group C has the same transmission delay for the same controller.
  • the first packet is exemplified, and the first packet includes the user terminal 1, the user terminal 2, the user terminal 4, and the user terminal 7.
  • the request generation rates of the four user terminals C are the same; and the transmission delay between the user terminal 1 and the controller 1 is ty11, and the transmission delay between the user terminal 2 and the controller 1 is ty21, and the user terminal 4
  • N can be greater than or far greater than M (in fact, N can also be less than M, which is not limited in this application)
  • Q is less than N
  • N, M, and Q are positive integers.
  • the matching method of the user terminal and the controller involved in the present application is introduced by a specific example.
  • FIG. 2 it is a schematic diagram of steps of a method for matching a user terminal and a controller provided by the present application.
  • the method mainly includes the following steps:
  • Step 21 The user terminal selects any one of the plurality of controllers to perform a matching connection.
  • one of the plurality of controllers is selected to perform a matching connection, wherein a plurality of user terminals may be connected to the same controller, but each User terminals can only be connected to one controller.
  • any matching connection can be performed on multiple user terminals to form an initial connection relationship, and The connection relationship is adjusted.
  • the N user terminals on the user side have been divided into Q packets.
  • the specific division criterion is: in the same group, the request generation rate of each user terminal is the same, and each user The transmission delay of the terminal to the same controller is the same.
  • the number of user terminals in the group q is m q ; the request generation rate of the user terminal is ⁇ q (units per second), and the parameter is determined by the attribute of the user terminal, indicating the unit
  • the number of requests generated by the user terminal (such as authorization request, bearer generation request, etc.); the request transmission delay from the user terminal to the M controllers is (in seconds), this parameter can measure the transmission delay required by the user terminal to different controllers by sending ping packets; the maximum processing capacity of the mth controller is ⁇ m (units per second)
  • This parameter can know the maximum number of requests that can be processed by the controller per unit time according to the hardware configuration of the controller, that is, the maximum processing capacity of the controller.
  • the user terminal can know the various parameters mentioned above.
  • the specific manner of obtaining may be that the controller sends the corresponding information to the user terminal for parsing, so that the corresponding parameters between the connected user terminal and the controller are obtained through the information interaction.
  • Step 22 Receive load information in the current matching connection state sent by the multiple controllers in a preset first period.
  • the user terminal can perform information interaction with the controllers that match the connection, so that the controller can monitor the request processed by the user in the preset first period.
  • the number of the load information obtained by the user at the current time (actually also in the current matching connection state); since each controller can obtain the load information, each user terminal in the current preset first period Both can receive load information sent by multiple controllers.
  • controller 1-controller M load information ⁇ 1 (t), ⁇ 2 (t), ..., ⁇ M (t).
  • Step 23 Adjust the current matching connection according to the initialization information, the preset utility function, and the received first load information, so that the average response delay of the user terminal reaches a minimum value.
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the corresponding preset utility function may be substituted according to the initialization information and the received load information, and the current matching connection between the user terminal and the controller is adjusted according to the obtained utility value relationship, so as to The global user terminal and the controller are optimally matched, and the optimal matching connection relationship can minimize the average response delay of all user terminals in the communication network.
  • the average response delay of the user terminal includes a processing delay of the controller and a request transmission delay of the user terminal to the controller.
  • the user's utility function is derived by using the potential function in the evolutionary game framework, and the matching connection relationship between the user terminal and the controller can be obtained in a distributed manner. Even if the pairing of the user terminal and the controller reaches a steady state, once the network condition changes, the user terminal automatically selects a controller with a higher utility value for matching, and the pairing relationship automatically converges to a new steady state. Moreover, the distributed method does not require centralized control nodes, and the computing tasks are shared among the users, which greatly reduces the complexity of matching, improves matching efficiency, and ensures matching accuracy.
  • the sum of the processing amounts of the M processors is sufficient to satisfy the sum of the requests sent by all the N user terminals.
  • step 23 adjusts the current matching connection according to the initialization information, the preset utility function, and the received first load information, so that the average response delay of the user terminal is reached.
  • the minimum value can be specifically implemented as:
  • the first vector is an M-dimensional vector including M utility values
  • controllers that match the connections are adjusted according to the M utility values in the first vector until the M utility values are equal in the first vector determined under the newly updated matching connection.
  • the adjustment method is not limited to the above, and other adjustment schemes based on the initialization information, the preset utility function, and the first load information are implemented, and therefore, the initialization information is preset.
  • the utility function and the matching understanding adjustment scheme performed by the first load information are all covered within the scope of the present application.
  • each user terminal is Initialize an M-dimensional vector by itself
  • 0 1 ⁇ M represents a zero vector of 1 row and M columns.
  • the initial initialization information and the first load of the corresponding controller may be used for any user terminal.
  • the information is substituted into the preset utility function for calculation, and the utility value of one of the controllers of the user terminal (that is, the corresponding controller) is obtained; the utility values of the other controllers are respectively calculated in a similar manner, and finally, the target value is calculated.
  • the M utility values in the first vector of the user terminal are equal, it is determined that the matching is successful, indicating that the matching connection at this time reaches equilibrium, that is, the average response delay reaches a minimum value. However, considering that the number of user terminals is large, it is unlikely to hit the optimal matching connection after the initial adjustment. Then, when the utility values of the M in the first vector are not equal, it is necessary to use the M utility values in the first vector. The controllers that match the connections are adjusted until the M utility values are equal in the first vector determined under the most recently updated matching connection.
  • the controllers that match the connection are adjusted according to the M utility values in the first vector until the M utility values are equal in the first vector determined under the newly updated matching connection, and other adjustment manners may be selected. , preferably implemented in the following cycle:
  • a controller corresponding to the utility value with the largest value is selected from the first vector to perform a matching connection.
  • the reason is that the controller corresponding to the utility value with the largest value is selected for the matching connection. This is because the application mainly adopts the evolutionary game framework, introduces the potential function, and sets a reasonable preset utility function according to the potential function. Thus, based on the convex function theory, The maximum value of the utility value obtained by using the preset utility function should be the best matching connection object for each user terminal, and if not, the fine tuning is repeated.
  • the controller corresponding to the utility value with the largest value may be selected from the first vector corresponding to each user terminal to perform a matching connection.
  • the controller corresponding to the utility value having the largest value in the first vector M1 of the user terminal 1 is the controller 1 and the user.
  • the controller corresponding to the utility value having the largest value among the first vector M2 of the terminal 2 is the controller 3.
  • the controller corresponding to the utility value having the largest value among the first vectors M3 of the user terminal 3 is the controller 1, and the user terminal 4
  • the controller corresponding to the utility value having the largest value in a vector M4 is the controller 2
  • the controller corresponding to the utility value having the largest value among the first vectors M5 of the user terminal 5 is the controller 2.
  • the new matching connection relationship is that the user terminal 1 and the user terminal 3 temporarily match the connection controller 1, respectively, and the user terminal 2 temporarily matches the connection controller 3, and the user terminal 4 and the user terminal 5 temporarily match the connection controller 2, respectively.
  • the load information sent by the multiple controllers in the current matching connection state is received in a preset second period.
  • the user terminal needs to receive the second load information of the controller in the current matching connection state in the preset second period.
  • the second load information is determined in the same manner as the first load information, but is based on different matching connection relationships and load information generated in different periods.
  • the first vector is updated according to the initialization information, the preset utility function, and the received second load information.
  • U m (t) is the utility value of the mth controller at time t
  • ⁇ q represents the request generation rate of the user terminal of the qth packet
  • m q represents the number of user terminals in the qth packet
  • ⁇ m Indicates the maximum processing power of the mth controller
  • ⁇ m (t) represents second load information of the mth controller in the current cycle
  • the calculated utility values are replaced by the respective elements of the first vector.
  • the fourth step it is determined whether the M utility values in the updated first vector are equal. If they are equal, it is determined that the matching is successful. Only when the utility values of all controllers are equal, it indicates that the game has reached the equilibrium point, and the system at this time In a steady state, this is a condition of equilibrium. Otherwise, jump to the first step and re-establish the matching connection relationship to adjust.
  • determining the first vector according to the initialization information, the preset utility function, and the received first load information may be specifically performed as:
  • the obtained M utility values are respectively determined as elements as the first vector.
  • the matching scheme can be implemented by the user terminal, and mainly includes the following steps:
  • Step 31 Select any controller m and initialize the M-dimensional vector U * .
  • the M-dimensional vector U * in this step 31 is a zero vector.
  • Step 32 After the preset time period T, the load information of the M controllers is respectively received.
  • the M controllers respectively monitor the number of requests processed by the current preset period, and obtain the load information ⁇ 1 (t), ⁇ 2 (t), ..., ⁇ M of the M controllers at the current time. t) and transmitting the load information to N user terminals in the communication network system.
  • the load information at this time is the load information in the current matching connection state
  • the current matching connection state may be when the matching connection is initialized, or may be the adjusted matching connection.
  • Step 33 Update the M-dimensional vector according to the initialization information, the preset utility function, and the received load information.
  • the utility value calculated by the utility value of each controller in the current period is calculated by the above-mentioned preset utility function formula (1), and the M-dimensional vector is updated respectively to obtain a new M-dimensional vector.
  • Step 34 It is judged whether the updated M-dimensional vectors are equal. If they are equal, the matching connection is successful; otherwise, step 35 is performed.
  • Step 35 Select a controller corresponding to the utility value with the largest value to perform a matching connection, and jump to step 32.
  • the present application further provides a method for matching a user terminal and a controller.
  • the execution body of the method is a controller. Referring to FIG. 4a, the method mainly includes:
  • Step 41 After the controller matches the connection with the user terminal, determine the load information in the current matching connection state in the preset second period;
  • Step 42 Send the load information to the N user terminals, so that each user terminal of the N user terminals performs current information according to initialization information, a preset utility function, and the received load information.
  • the matching connection is adjusted so that the average response delay of all user terminals in the network system reaches a minimum.
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the above matching scheme is also applied to a system including N user terminals and M controllers, the N user terminals being defined as Q packets, and the user terminals in each packet satisfy: the request generation rate is the same and The transmission delays corresponding to the same controller are the same; the Q is smaller than N, and the N, M, and Q are positive integers.
  • Step 411 Determine the number of requests sent by the connected opposite user terminal processed in the preset second period.
  • Step 412 Determine load information in the current matching connection state according to the number of requests.
  • the operation on the controller side actually corresponds to the operation on the user terminal side.
  • the controller side performs information with the user terminal side according to the new matching connection relationship. Interaction.
  • the embodiment of the present application further provides a user terminal.
  • the user terminal mainly includes:
  • the initialization module 51 is configured to select any one of the multiple controllers to perform a matching connection
  • the receiving module 52 is configured to receive, in a preset first period, load information sent by the multiple controllers in a current matching connection state;
  • the matching module 53 is configured to adjust the current matching connection according to the initialization information, the preset utility function, and the received load information, so that the average response delay of all user terminals in the network system reaches a minimum value;
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the embodiment of the present application further provides a controller.
  • the controller mainly includes:
  • a determining module 61 configured to determine, in a preset second period, load information in a current matching connection state after the controller matches the connected user terminal;
  • the sending module 62 is configured to send the load information to the N user terminals, so that each user terminal of the N user terminals is configured according to initialization information, a preset utility function, and the received load information. Adjusting the current matching connection so that the average response delay of all user terminals in the network system reaches a preset minimum value;
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the embodiment of the present application further provides a matching system between a user terminal and a controller.
  • the matching system shown in FIG. 1 mainly includes: N user terminals C and M controllers S; the N user terminals C are limited to Q packets, the user terminal C in each packet satisfies: the request generation rate is the same and the transmission delays of the same controller S are the same; the Q is smaller than N, and the N, M, and Q are positive integers;
  • the user terminal C is configured to select any one of the plurality of controllers S to perform a matching connection; and receive, in a preset first period, the first in the current matching connection state sent by the multiple controllers S. a load information; adjusting the current matching connection according to the initialization information, the preset utility function, and the received first load information, so that the average response delay of all user terminals C in the network system reaches a minimum value;
  • the controller S is configured to determine load information in a current matching connection state in a preset period after the controller S matches the user terminal C, and send the load information to the N user terminals C. So that each of the N user terminals C adjusts the current matching connection according to the initialization information, the preset utility function, and the received load information, so that all the user terminals C in the network system The average response delay reaches a minimum value;
  • the initialization information includes: a maximum processing threshold of each controller, a number of user terminals, a request generation rate, and a request transmission delay of the user terminal to each controller in the group where the user terminal is located.
  • the embodiment of the present invention further provides a computer readable storage medium.
  • the computer readable storage medium stores a computer program.
  • the computer program is executed by the processor, the processes of the foregoing matching method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, we will not repeat them here.
  • the computer readable storage medium such as a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device.
  • computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请涉及通信技术领域，尤其涉及一种通信网络中用户终端与控制器的匹配方法、装置及系统，用以克服现有技术中存在的缺陷。本申请主要包括：用户终端选择多个控制器中的任一控制器进行匹配连接；在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的第一负载信息；根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得所述用户终端的平均响应时延最终达到最小值。该方法极大的降低了匹配的复杂程度，提升匹配效率。

Description

通信网络中用户终端与控制器的匹配方法、装置及系统 技术领域
本申请涉及通信技术领域，尤其涉及一种通信网络中用户终端与控制器的匹配方法、装置及系统。
背景技术
用户终端与控制器的匹配连接，是通信技术领域较为关注的问题。目前常用的用户终端与控制器匹配机制是基于整数线性规划的用户控制器配对方法，或是，基于联盟博弈的用户终端与控制器的匹配方法。
然而，基于整数线性规则的匹配方法，随着用户终端或控制器数量增大时，问题的解空间指数性的增大，由于是穷举算法，因此复杂程度大，实时性不好。而基于联盟博弈的匹配方法，为减少问题的复杂程度，在第一阶段中引入了稳定匹配方法，而稳定匹配方法会直接忽略一部分解集合，使得最终的配对关系不是全局最优解。
因此，亟需找到一种准确且高效的匹配方案。
发明内容
本申请实施例提供一种通信网络中用户终端与控制器的匹配方法、装置及系统，用以解决现有技术中存在的上述缺陷。
为了解决上述技术问题，本申请实施例采用下述技术方案：
第一方面，本发明实施例提供了一种通信网络中用户终端与控制器的匹配方法，应用于包含N个用户终端和M个控制器的系统中，所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；所述方法包括：
用户终端选择多个控制器中的任一控制器进行匹配连接；
在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的第一负载信息；
根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
第二方面，本发明实施例提供一种通信网络中用户终端与控制器的匹配方法，应用于包含N个用户终端和M个控制器的系统中，所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；所述方法包括：
当控制器匹配连接有用户终端后，在预设第一周期内确定在当前匹配连接状态下的负载信息；
将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
第三方面，本发明实施例提供一种用户终端，包括：
初始化模块，用于选择多个控制器中的任一控制器进行匹配连接；
接收模块，用于在预设第一周期内接收所述多个控制器发送的在当前匹配 连接状态下的负载信息；
匹配模块，用于根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
第四方面，本发明实施例提供一种控制器，包括：
确定模块，用于当控制器匹配连接有用户终端后，在预设第一周期内确定在当前匹配连接状态下的负载信息；
发送模块，用于将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
第五方面，本发明实施例提供一种通信网络中用户终端与控制器的匹配系统，包括：N个用户终端和M个控制器；所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；其中，
所述用户终端，用于选择多个控制器中的任一控制器进行匹配连接；在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息；根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
所述控制器，用于当控制器匹配连接有用户终端后，在预设第一周期内确 定在当前匹配连接状态下的负载信息；将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
本申请实施例采用的上述至少一个技术方案能够达到以下有益效果：
通过上述技术方案，基于演化博弈的用户终端与控制器的匹配连接机制，通过使用演化博弈框架中的势函数推导得到用户的效用函数，可分布式获取用户终端与控制器的匹配连接关系。即使用户终端与控制器的配对达到了稳态，一旦网络情况发生了改变，用户终端自动选择效用值更高的控制器进行匹配，配对关系会自动收敛至新的稳态。而且，采用分布式方法，不需要集中的控制节点，计算任务被分担到各个用户上，极大的降低了匹配的复杂程度，提升了匹配效率，保证匹配准确性。
附图说明
此处所说明的附图用来提供对本申请的进一步理解，构成本申请的一部分，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。在附图中：
图1为本说明书实施例涉及的用户终端与控制器的通信网络系统架构示意图；
图2为本申请提供的用户终端与控制器的匹配方法的步骤示意图之一；
图3为本申请提供的用户终端与控制器的匹配方法的详细步骤示意图；
图4a为本申请提供的用户终端与控制器的匹配方法的步骤示意图之二；
图4b为本申请提供的用户终端与控制器的匹配方法的步骤41的具体示意 图；
图5为本申请提供的用户终端的结构示意图；
图6为本申请提供的控制器的结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚，下面将结合本申请具体实施例及相应的附图对本申请技术方案进行清楚、完整地描述。显然，所描述的实施例仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
以下结合附图，详细说明本申请各实施例提供的技术方案。
参照图1所示，为本说明书实施例涉及的网络通信中的用户终端与控制器的通信网络系统架构示意图，该系统主要包括：
多个用户终端C和多个控制器S，而每个用户终端C都可以匹配连接至相应的一个控制器S，而每个控制器S可以匹配连接有一个或多个用户终端C。用户终端C可以理解为移动终端或其他可匹配连接至控制器进行信息传输业务的电子设备。
其中，可以N个用户终端C和M个控制器S为例进行介绍。这N个用户终端C被限定为Q个分组，每个分组内的用户终端C均满足这一条件：同一分组内的不同用户终端C的请求生成速率相同，且，同一分组内的不同用户终端C对应相同控制器的传输时延相同。例如，以第1分组为例，该第1分组中包含用户终端1、用户终端2、用户终端4和用户终端7。这4个用户终端C的请求生成速率相同；且用户终端1与控制器1之间的传输时延记为ty11，用户终端2与控制器1之间的传输时延记为ty21，用户终端4与控制器1之间的传输时延记为ty41，用户终端7与控制器1之间的传输时延记为ty71，ty11＝ty21＝ty41＝ty71。其中，N可以大于或是远远大于M(其实N也可以小于 M，本申请并不对此进行限定)，Q小于N，而N、M、Q均为正整数。
下面通过具体的实例对本申请所涉及的用户终端与控制器的匹配方法进行介绍。
实施例一
参照图2所示，为本申请提供的用户终端与控制器的匹配方法的步骤示意图，该方法主要包括以下步骤：
步骤21：用户终端选择多个控制器中的任一控制器进行匹配连接。
在实际的系统架构中，针对用户侧的每个用户终端，分别从多个控制器中选择任一控制器进行匹配连接，其中，可以是多个用户终端连接相同的控制器，但是，每个用户终端只能连接至一个控制器。考虑到在初始状态下，无法获知每个用户终端应连接至哪个控制器才能达到匹配连接的均衡，因此，可对多个用户终端进行任意匹配连接，形成初始化的连接关系，并在后续对该连接关系进行调整。
其实，在进行初始化的匹配连接之前，用户侧的N个用户终端，已经被划分为Q个分组，具体的划分准则为：在同一分组中，每个用户终端的请求生成速率相同，每个用户终端到相同控制器的传输时延相同。其中，以第q个分组为例，该分组q中的用户终端数量为m q；用户终端的请求生成速率为λ q(单位为个/秒)，该参数由用户终端的属性决定，表示单位时间内用户终端生成请求(如授权请求、承载生成请求等)的个数；用户终端到M个控制器的请求传输时延为
Figure PCTCN2018088625-appb-000001
(单位为秒)，该参数可通过发送ping报文的方式测得用户终端到不同控制器所需的传输时延；第m个控制器的最大处理能力为α m(单位为个/秒)该参数可根据控制器的硬件配置获知控制器单位时间内可处理请求的最大个数，即控制器的最大处理能力。
应理解，用户终端与控制器在初始化匹配连接之后，用户终端即可获知上述提及的各个参数。具体的获知方式，可以是通过控制器向用户终端发送相应 信息以解析得到，这样，通过信息交互获知连接的用户终端与控制器之间的相应参数。
步骤22：在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息。
具体地，在N个用户终端分别匹配连接至相应控制器后，用户终端与匹配连接的控制器之间可以进行信息交互，这样，控制器可通过监测自身在预设第一周期内处理的请求个数，得到自身在当前时刻(其实也是在当前匹配连接状态下)的负载信息；由于每个控制器都可以得到这一负载信息，因此，在当前预设第一周期内，每个用户终端都可以接收到多个控制器发送的负载信息。例如，控制器1-控制器M的负载信息：θ 1(t),θ 2(t),…,θ M(t)。
步骤23：根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得所述用户终端的平均响应时延达到最小值。
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
应理解，在本申请中，可具体根据初始化信息以及接收到的负载信息，代入相应的预设效用函数，通过得到的效用值的关系，对用户终端与控制器的当前匹配连接进行调整，以获得全局的用户终端与控制器最优匹配连接关系，而且，这种最优匹配连接关系可以使得通信网络中所有用户终端的平均响应时延达到最小值。其中，用户终端的平均响应时延包含控制器的处理时延和用户终端到控制器的请求传输时延。
通过上述技术方案，基于演化博弈的用户终端与控制器的匹配连接机制，通过使用演化博弈框架中的势函数推导得到用户的效用函数，可分布式获取用户终端与控制器的匹配连接关系。即使用户终端与控制器的配对达到了稳态，一旦网络情况发生了改变，用户终端自动选择效用值更高的控制器进行匹配，配对关系会自动收敛至新的稳态。而且，采用分布式方法，不需要集中的控制 节点，计算任务被分担到各个用户上，极大的降低了匹配的复杂程度，提升匹配效率，保证匹配准确性。
需要说明的是，在本申请实施例中，M个处理器的处理数量之和要足以满足所有N个用户终端发送的请求之和。
可选地，在本申请中，步骤23在根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得所述用户终端的平均响应时延达到最小值时，可具体执行为：
根据初始化信息、预设效用函数以及接收到的所述第一负载信息确定第一向量，所述第一向量为包含M个效用值的M维向量；
若所述第一向量中M个效用值相等，则确定匹配成功；
否则，根据所述第一向量中M个效用值调整匹配连接的控制器，直至在最新更新的匹配连接下确定的第一向量中M个效用值相等。
应理解，在本申请中，并不限于采用上述调整方法，还包括其他根据初始化信息、预设效用函数以及第一负载信息所实施的匹配连接的调整方案，因此，凡是以初始化信息、预设效用函数以及第一负载信息所执行的匹配了解调整方案都涵盖在本申请保护范围内。
其实，本申请实施例中，在用户终端选择多个控制器中的任一控制器进行匹配连接之后，且在用户终端接收多个控制器发送的第一负载信息之前，每个用户终端都会为自身初始化一个M维向量
Figure PCTCN2018088625-appb-000002
此时初始化的M维向量U *＝0 1×M＝[0,0……,0]。其中，0 1×M表示1行M列的零向量。
相应地，当根据初始化信息、预设效用函数以及接收到的所述第一负载信息确定第一向量时，可针对任一用户终端，将其对应的初始化信息、以及相应控制器的第一负载信息代入预设效用函数中进行计算，得到所述用户终端的其中一个控制器(即为所述相应控制器)的效用值；按照类似方式分别计算其他控制器的效用值，最后计算出针对所述用户终端的M个效用值。若本申请中初始化有M维向量U *＝0 1×M，则将计算出的M个效用值分别替换更新相应控 制器对应的零值，得到M维第一向量。需要说明的是，在此步骤中，每个用户终端都会得到一个相应的M维第一向量。
仍以上述用户终端为例，若该用户终端的第一向量中M个效用值相等，则确定匹配成功，表示此时的匹配连接达到均衡，即平均响应时延达到最小值。然而，考虑到用户终端的数量较多，不太可能在初始化调整后就命中最优匹配连接，那么，当第一向量中M各效用值不相等时，需要根据第一向量中M个效用值调整匹配连接的控制器，直至在最新更新的匹配连接下确定的第一向量中M个效用值相等。
应理解，在本申请中，根据第一向量中M个效用值调整匹配连接的控制器，直至在最新更新的匹配连接下确定的第一向量中M个效用值相等，可以选择其他的调整方式，优选以下述循环方式实现：
第一步，从所述第一向量中选取数值最大的效用值对应的控制器进行匹配连接。
之所以选取数值最大的效用值对应的控制器进行匹配连接，这是因为，本申请主要采用演化博弈框架，引入势函数，根据势函数设置合理的预设效用函数，这样，基于凸函数理论，利用预设效用函数得到的效用值中最大值应是每个用户终端最佳的匹配连接对象，若不是，则反复进行微调。
具体实现时，可分别从每个用户终端对应的第一向量中选取数值最大的效用值对应的控制器进行匹配连接。以该通信系统中仅包含用户终端1-用户终端5，以及控制器1-控制器3为例，用户终端1的第一向量M1中数值最大的效用值对应的控制器为控制器1，用户终端2的第一向量M2中数值最大的效用值对应的控制器为控制器3，用户终端3的第一向量M3中数值最大的效用值对应的控制器为控制器1，用户终端4的第一向量M4中数值最大的效用值对应的控制器为控制器2，用户终端5的第一向量M5中数值最大的效用值对应的控制器为控制器2。进而，新的匹配连接关系为：用户终端1和用户终端3分别暂时匹配连接控制器1，用户终端2暂时匹配连接控制器3，用户终端4 和用户终端5分别暂时匹配连接控制器2。
第二步，在预设第二周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息。
基于上述第一步重新确定的匹配连接关系，用户终端需要在预设第二周期内接收控制器在当前匹配连接状态下的第二负载信息。其中，第二负载信息的确定方式与第一负载信息的确定方式相同，只是基于不同的匹配连接关系以及不同的周期内产生的负载信息。
第三步，根据初始化信息、预设效用函数以及接收到的所述第二负载信息更新所述第一向量。
具体地，可针对任一用户终端：根据初始化信息以及接收到的第二负载信息，通过预设效用函数公式(1)，计算在当前周期内每个控制器的效用值：
Figure PCTCN2018088625-appb-000003
其中，U m(t)为在时刻t第m个控制器的效用值，λ q表示第q个分组的用户终端的请求生成速率，m q表示第q个分组中用户终端的数量，α m表示第m个控制器的最大处理能力，
Figure PCTCN2018088625-appb-000004
表示第q个分组中任一用户终端到第m个控制器的请求传输时延，θ m(t)表示第m个控制器在当前周期的第二负载信息；
将计算得到的效用值，分别替换更新所述第一向量的各个元素。
第四步，判断更新后的第一向量中M个效用值是否相等，若相等，则确定匹配成功，只有当所有控制器的效用值相等时，才说明博弈已达到了均衡点，此时系统处于稳定状态，此为均衡的条件。否则，跳转至第一步，重新建立匹配连接关系进行调整。
可选地，在本申请中，根据初始化信息、预设效用函数以及接收到的所述第一负载信息确定第一向量，可具体执行为：
针对任一用户终端：根据初始化信息以及接收到的所述第一负载信息通过预设效用函数公式(1)，计算在当前周期内每个控制器的效用值；
将得到的M个效用值分别作为元素确定为第一向量。
下面通过具体的实例对本申请所涉及的匹配方案进行详细描述。
参照图3所示，该匹配方案可由用户终端作为执行主体，主要包含以下步骤流程：
步骤31：选择任一控制器m，并初始化M维向量U *。
应理解，该步骤31中M维向量U *为0向量。
步骤32：待预设时间段T后，分别接收M个控制器的负载信息。
具体地，M个控制器分别监测自身在当前预设周期内处理的请求个数，获得M个控制器在当前时刻的负载信息θ 1(t),θ 2(t),…,θ M(t)，并将该负载信息发送给通信网络系统中的N个用户终端。应理解，此时的负载信息，是当前匹配连接状态下的负载信息，所述当前匹配连接状态可以是初始化匹配连接时，也可以是调整后的匹配连接时。
步骤33：根据初始化信息、预设效用函数以及接收到的负载信息，更新M维向量。
具体可通过上述预设效用函数公式(1)，计算当前周期内每个控制器的效用值将计算得到的效用值，分别更新M维向量，即得到新的M维向量。
步骤34：判断更新后的M维向量是否相等，若相等，则匹配连接成功；否则，执行步骤35。
步骤35：选取数值最大的效用值对应的控制器进行匹配连接，并跳转至步骤32。
同时，本申请还提供了一种用户终端与控制器的匹配方法，该方法的执行主体为控制器，具体参照图4a所示，该方法主要包括：
步骤41：当控制器匹配连接有用户终端后，在预设第二周期内确定在当前匹配连接状态下的负载信息；
步骤42：将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响 应时延达到最小值。
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
应理解，上述匹配方案同样应用于包含N个用户终端和M个控制器的系统中，所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数。
可选地，在预设第二周期内确定在当前匹配连接状态下的负载信息时，参照图4b所示，具体执行为：
步骤411：确定在预设第二周期内处理的由连接的对侧用户终端发送来的请求个数。
步骤412：根据所述请求个数确定在当前匹配连接状态下的负载信息。
需要说明的是，控制器侧的操作其实与用户终端侧的操作对应的，每当用户终端重新确定新的匹配连接关系，控制器侧就会按照该新的匹配连接关系与用户终端侧进行信息交互。
实施例二
本申请实施例还提供一种用户终端，参照图5所示，该用户终端主要包括：
初始化模块51，用于选择多个控制器中的任一控制器进行匹配连接；
接收模块52，用于在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息；
匹配模块53，用于根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户 终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
本申请实施例还提供一种控制器，参照图6所示，该控制器主要包括：
确定模块61，用于当控制器匹配连接有用户终端后，在预设第二周期内确定在当前匹配连接状态下的负载信息；
发送模块62，用于将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网络系统中所有用户终端的平均响应时延达到预设最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
本申请实施例还提供一种用户终端与控制器的匹配系统，参照图1所示该匹配系统主要包括：N个用户终端C和M个控制器S；所述N个用户终端C被限定为Q个分组，每个分组内的用户终端C满足：请求生成速率相同且对应相同控制器S的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；其中，
所述用户终端C，用于选择多个控制器S中的任一控制器S进行匹配连接；在预设第一周期内接收所述多个控制器S发送的在当前匹配连接状态下的第一负载信息；根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端C的平均响应时延达到最小值；
所述控制器S，用于当控制器S匹配连接有用户终端C后，在预设周期内确定在当前匹配连接状态下的负载信息；将所述负载信息发送给所述N个用户终端C，以使得所述N个用户终端C中的每个用户终端C根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网 络系统中所有用户终端C的平均响应时延达到最小值；
其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
本发明实施例还提供一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，该计算机程序被处理器执行时实现上述匹配方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。其中，所述的计算机可读存储介质，如只读存储器(Read-Only Memory，简称ROM)、随机存取存储器(Random Access Memory，简称RAM)、磁碟或者光盘等。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使 得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中，计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
还需要说明的是，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
以上所述仅为本申请的实施例而已，并不用于限制本申请。对于本领域技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原理之内所 作的任何修改、等同替换、改进等，均应包含在本申请的权利要求范围之内。

Claims (10)

  1. 一种通信网络中用户终端与控制器的匹配方法，其特征在于，应用于包含N个用户终端和M个控制器的系统中，所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；
    所述方法包括：
    用户终端选择多个控制器中的任一控制器进行匹配连接；
    在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的第一负载信息；
    根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
    其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
  2. 如权利要求1所述的匹配方法，其特征在于，根据初始化信息、预设效用函数以及接收到的所述第一负载信息，对当前匹配连接进行调整，以使得所述用户终端的平均响应时延达到最小值，具体包括：
    根据初始化信息、预设效用函数以及接收到的所述第一负载信息确定第一向量，所述第一向量为包含M个效用值的M维向量；
    若所述第一向量中M个效用值相等，则确定匹配成功；
    否则，根据所述第一向量中M个效用值调整匹配连接的控制器，直至在最新更新的匹配连接下确定的第一向量中M个效用值相等。
  3. 如权利要求2所述的匹配方法，其特征在于，根据所述第一向量中M个效用值调整匹配连接的控制器，直至在最新更新的匹配连接下确定的第一向量中M个效用值相等，具体包括以下循环：
    第一步，从所述第一向量中选取数值最大的效用值对应的控制器进行匹配连接；
    第二步，在预设第二周期内接收所述多个控制器发送的在当前匹配连接状态下的第二负载信息；
    第三步，根据初始化信息、预设效用函数以及接收到的所述第二负载信息更新所述第一向量；
    第四步，判断更新后的第一向量中M个效用值是否相等，若相等，则确定匹配成功，否则，跳转至第一步。
  4. 如权利要求3所述的匹配方法，其特征在于，根据初始化信息、预设效用函数以及接收到的所述第二负载信息更新所述第一向量，具体包括：
    针对任一用户终端：通过预设效用函数公式(1)，计算在预设第二周期的每个控制器的效用值：
    Figure PCTCN2018088625-appb-100001
    其中，U m(t)为在时刻t第m个控制器的效用值，λ q表示第q个分组的用户终端的请求生成速率，m q表示第q个分组中用户终端数量，α m表示第m个控制器的最大处理阈值，
    Figure PCTCN2018088625-appb-100002
    表示第q个分组中任一用户终端到第m个控制器的请求传输时延，θ m(t)表示第m个控制器在预设第二周期的第二负载信息；
    将计算得到的效用值，分别替换更新所述第一向量的各个元素。
  5. 如权利要求2-4任一项所述的匹配方法，其特征在于，根据初始化信息、预设效用函数以及接收到的所述第一负载信息确定第一向量，具体包括：
    针对任一用户终端：通过预设效用函数公式(1)，计算在预设第一周期的每个控制器的效用值；
    将得到的M个效用值分别作为元素确定为第一向量。
  6. 一种通信网络中用户终端与控制器的匹配方法，其特征在于，应用于包含N个用户终端和M个控制器的系统中，所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传 输时延相同；所述Q小于N，所述N、M、Q均为正整数；
    所述方法包括：
    当控制器匹配连接有用户终端后，在预设第一周期内确定在当前匹配连接状态下的负载信息；
    将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
    其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
  7. 如权利要求6所述的匹配方法，其特征在于，在预设第一周期内控制器确定在当前匹配连接状态下的负载信息，具体包括：
    确定在预设第一周期内处理的由连接的对侧用户终端发送来的请求个数；
    根据所述请求个数确定在当前匹配连接状态下的负载信息。
  8. 一种用户终端，其特征在于，包括：
    初始化模块，用于选择多个控制器中的任一控制器进行匹配连接；
    接收模块，用于在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息；
    匹配模块，用于根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
    其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
  9. 一种控制器，其特征在于，包括：
    确定模块，用于当控制器匹配连接有用户终端后，在预设第一周期内确定在当前匹配连接状态下的负载信息；
    发送模块，用于将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网络系统中所有用户终端的平均响应时延达到最小值；
    其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
  10. 一种通信网络中用户终端与控制器的匹配系统，其特征在于，包括：N个用户终端和M个控制器；所述N个用户终端被限定为Q个分组，每个分组内的用户终端满足：请求生成速率相同且对应相同控制器的传输时延相同；所述Q小于N，所述N、M、Q均为正整数；其中，
    所述用户终端，用于选择多个控制器中的任一控制器进行匹配连接；在预设第一周期内接收所述多个控制器发送的在当前匹配连接状态下的负载信息；根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以使得网络系统中所有用户终端的平均响应时延达到最小值；
    所述控制器，用于当控制器匹配连接有用户终端后，在预设第一周期内确定在当前匹配连接状态下的负载信息；将所述负载信息发送给所述N个用户终端，以使得所述N个用户终端中的每个用户终端根据初始化信息、预设效用函数以及接收到的所述负载信息，对当前匹配连接进行调整，以便网络系统中所有用户终端的平均响应时延达到最小值；
    其中，所述初始化信息至少包含：每个控制器的最大处理阈值，所述用户终端所在分组的：用户终端数量、请求生成速率以及用户终端到所述每个控制器的请求传输时延。
PCT/CN2018/088625 2018-05-07 2018-05-28 通信网络中用户终端与控制器的匹配方法、装置及系统 Ceased WO2019213997A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810427837.2 2018-05-07
CN201810427837.2A CN108848513B (zh) 2018-05-07 2018-05-07 通信网络中用户终端与控制器的匹配方法、装置及系统

Publications (1)

Publication Number Publication Date
WO2019213997A1 true WO2019213997A1 (zh) 2019-11-14

Family

ID=64212692

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/088625 Ceased WO2019213997A1 (zh) 2018-05-07 2018-05-28 通信网络中用户终端与控制器的匹配方法、装置及系统

Country Status (2)

Country Link
CN (1) CN108848513B (zh)
WO (1) WO2019213997A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104094565A (zh) * 2012-02-02 2014-10-08 日本电气株式会社 控制器、用于分配负载的方法、存储程序的非瞬时计算机可读介质、计算机系统和控制设备
CN105425669A (zh) * 2015-11-16 2016-03-23 珠海格力电器股份有限公司 通信接口电路及其控制方法以及应用其的家用电器
CN105898961A (zh) * 2016-01-07 2016-08-24 深圳市朗科智能电气股份有限公司 无线照明配对控制系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003318914A (ja) * 2002-04-19 2003-11-07 Murata Mfg Co Ltd 無線ネットワークシステム
WO2015026272A1 (en) * 2013-08-21 2015-02-26 Telefonaktiebolaget L M Ericsson (Publ) Network assisted rate adaptation
CN103781118B (zh) * 2014-01-14 2016-10-19 西安电子科技大学 基于多业务的异构无线网络接入控制与资源分配联合方法
CN106941456B (zh) * 2017-05-17 2019-08-30 华中科技大学 一种软件定义网络中控制平面的负载均衡方法及系统
CN107612771B (zh) * 2017-09-07 2020-11-17 广东工业大学 一种基于动态迁移的sdn网络负载均衡方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104094565A (zh) * 2012-02-02 2014-10-08 日本电气株式会社 控制器、用于分配负载的方法、存储程序的非瞬时计算机可读介质、计算机系统和控制设备
CN105425669A (zh) * 2015-11-16 2016-03-23 珠海格力电器股份有限公司 通信接口电路及其控制方法以及应用其的家用电器
CN105898961A (zh) * 2016-01-07 2016-08-24 深圳市朗科智能电气股份有限公司 无线照明配对控制系统

Also Published As

Publication number Publication date
CN108848513A (zh) 2018-11-20
CN108848513B (zh) 2020-09-01

Similar Documents

Publication Publication Date Title
CN111953759B (zh) 基于强化学习的协同计算任务卸载和迁移的方法及装置
CN113568727A (zh) 一种基于深度强化学习的移动边缘计算任务分配方法
CN115511103B (zh) 用于联邦学习的方法、装置、设备和介质
CN118586474B (zh) 基于深度强化学习的自适应异步联邦学习方法及系统
Wan et al. Reinforcement learning based mobile offloading for cloud-based malware detection
CN104734985A (zh) 数据接收流量控制方法及其系统
US12464051B2 (en) Queuing control for distributed compute network orchestration
CN112732738B (zh) 基于多目标优化的自适应网络数据采集方法及相关设备
Fan et al. Dual-agent DRL-based service placement, task scheduling, and resource allocation for multi-sensor and multi-user edge computing networks
CN117714295A (zh) 网络链路生成方法、服务器及存储介质
CN117077811A (zh) 联邦学习优化方法及相关装置
CN110650195B (zh) 分布式负载均衡方法以及装置
CN111711931A (zh) 一种快速收敛的最优信道选择方法
CN111182509A (zh) 一种基于上下文感知学习的泛在电力物联网接入方法
Qin et al. Distributed threshold-based offloading for heterogeneous mobile edge computing
WO2023175381A1 (en) Iterative training of collaborative distributed coded artificial intelligence model
CN116911403B (zh) 联邦学习的服务器和客户端的一体化训练方法及相关设备
CN112308233A (zh) 用于处理数据的方法、设备和计算机程序产品
CN112601232A (zh) 基于最小费用最大流的负载均衡的多服务迁移方法及系统
Rikos et al. Distributed optimization for quadratic cost functions with quantized communication and finite-time convergence
CN118368259B (zh) 网络资源分配方法、装置、电子设备和存储介质
CN114911171A (zh) 一种设备控制的方法和装置
CN107580773B (zh) 进行跨节点网络的负载平衡的装置和方法
CN108848513B (zh) 通信网络中用户终端与控制器的匹配方法、装置及系统
Aung et al. Active STAR-RIS empowered edge system for enhanced energy efficiency and task management

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18917637

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18917637

Country of ref document: EP

Kind code of ref document: A1