A user resource scheduling method

Abstract

The present application relates to the technical field of wireless communication, and discloses a user resource scheduling method, which comprises the following steps: calculating the reachable rate range of each user based on the user joint channel state information of the current time slot at the MAC layer; determining the first to-be-transmitted data of each user in the current time slot; determining the target rate of each user according to the reachable rate range and the first to-be-transmitted data of each user at the application layer, wherein the target rate is within the reachable rate range; allocating transmission resources to each user based on the target rate of each user, so that the transmission rate of each user reaches the target rate of each user; transmitting the first to-be-transmitted data by each user using the transmission resources at the application layer, and determining the second to-be-transmitted data of each user in the next time slot; adjusting the user joint channel state information of the next time slot based on the second to-be-transmitted data at the MAC layer, and returning to the first step. The present application allocates power resources to users by combining the MAC layer and the application layer, so as to improve the effective transmission rate of the users.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, in particular to a user resource scheduling method. BACKGROUND

[0002] Real-time multimedia and other 5.5G new services have data integrity constraints and delay constraints, that is, the amount of data to be transmitted by a user must be transmitted within the constraint time, otherwise the previous transmission is invalid transmission. Traditional services need to optimize the transmission scheme in each transmission time slot (TTI), while real-time multimedia transmission services need to consider the data transmission scheme in multiple transmission time slots (TTI).

[0003] The effective transmission rate of the user has a long-term attribute, and maximizing the achievable rate of a specific user in a certain TTI can maximize the amount of data transmitted by the specific user in the TTI, but it does not mean that the total transmission data amount of all users in the multi-TTI scale is optimal, so the power resource allocation of the application layer user is unreasonable, resulting in unreasonable MAC layer resource scheduling, low effective transmission rate of the user, and the problem that the transmission cannot be completed within the agreed time. SUMMARY

[0004] Therefore, the present application provides a user resource scheduling method, which allocates power resources to users jointly in the MAC layer and the application layer to improve the effective transmission rate of the user.

[0005] The present application provides a user resource scheduling method, which comprises: calculating the achievable rate range of each user based on the user joint channel state information of the current time slot in the MAC layer; determining the first data to be transmitted of each user in the current time slot; determining the target rate of each user according to the achievable rate range and the first data to be transmitted of each user in the application layer, the target rate being within the achievable rate range; allocating transmission resources to each user based on the target rate of each user, so that the transmission rate of each user reaches the target rate of each user; in the application layer, each user transmits the first data to be transmitted using the transmission resources, and determines the second data to be transmitted of each user in the next time slot; in the MAC layer, adjusting the user joint channel state information of the next time slot based on the second data to be transmitted, and returning to the first step.

[0006] In this embodiment, the optimal data transmission rate can be determined by considering the data scheduling of the application layer and the specific implementation of the MAC layer, the appropriate power is allocated to the user, the channel allocation of the user is adjusted based on the data transmission, the spectrum access scheme and the user power allocation scheme that meet the data integrity and delay constraints can be obtained, and the effective transmission rate of the user is improved.

[0007] In an optional implementation, the determining, according to the reachable rate range of each user and the first to-be-transmitted data, of a target rate of each user at the application layer comprises: determining an initial rate of each user based on the reachable rate range, the initial rate being within the reachable rate range; calculating an initial data transmission amount corresponding to the initial rate according to the initial rate of each user and the first to-be-transmitted data; adjusting the initial rate by using a neural network algorithm, so that the initial data transmission amount is maximized; and taking the initial rate corresponding to the maximum initial data transmission amount as the target rate of each user.

[0008] In an optional implementation, the calculating of the initial data transmission amount corresponding to the initial rate according to the initial rate of each user and the first to-be-transmitted data comprises: constructing a state space, the state space comprising the first to-be-transmitted data of each user corresponding to different deadlines in a current time slot; constructing an action space, the action space comprising initial transmission resources of each user in the current time slot, the initial transmission resources being allocated based on the initial rate; and constructing a reward function by using the state space and the action space, the reward function being used to represent the initial data transmission amount.

[0009] In an optional implementation, the state space is [B1(t),…,B i (t),…,B N (t),c(t)], where, B i (t) represents the first to-be-transmitted data of user i at different deadlines at t time, and c(t) represents channel conditions at t time; and the action space is [e1(t),…,e N (t)], where, e i (t) represents the initial transmission resources allocated to user i at t time.

[0010] In an optional implementation, the adjusting of the initial rate by using the neural network algorithm, so that the initial data transmission amount is maximized, comprises: maximizing the reward function by using a recurrent neural network, the reward function being r(t) = D(t) - λE(t); where, E(t) represents transmission resource consumption at t time, D(t) represents a successfully transmitted data amount; and the taking of the initial rate corresponding to the maximum initial data transmission amount as the target rate of each user comprises: determining the initial rate at which the reward function is maximized as the target rate of each user.

[0011] In an alternative implementation, the recurrent neural network comprises an actor network and a critic network, and the constructing the action space comprises: the actor network obtaining initial rates of the users in the current time slot based on the achievable rate range; allocating initial transmission resources to the users based on the initial rates of the users; and the actor network adjusting the initial rates according to the reward function to re-allocate the initial transmission resources.

[0012] In an alternative implementation, the maximizing the reward function by using the recurrent neural network comprises: training a Q value function by using the critic network, the Q value function being:

[0013] In this implementation, the target rates of the users are determined by using the neural network algorithm based on the determined achievable rate range, which can ensure that the overall data transmission effect of all users is optimal and improve the data transmission success rate.

[0014] In an alternative implementation, the adjusting the user joint channel state information of the next time slot based on the second to-be-transmitted data at the MAC layer comprises: determining to-be-transmitted users of the next time slot based on the second to-be-transmitted data at the MAC layer; constructing a plurality of transmission resource scheduling tables of the to-be-transmitted users and transmission frequency bands, one transmission resource scheduling table representing one transmission resource scheduling manner; calculating data throughputs of each transmission resource scheduling manner by using the transmission resource scheduling tables, and taking the transmission resource scheduling manner with the largest data throughput as a target transmission resource scheduling manner; and determining the user joint channel state information corresponding to the target transmission resource scheduling manner.

[0015] In an alternative implementation, the calculating the data throughputs of each transmission resource scheduling manner by using the transmission resource scheduling tables comprises: wherein, wherein, A is obtained based on iteration; wherein,

[0016] In an alternative implementation, the taking the transmission resource scheduling manner with the largest data throughput as the target transmission resource scheduling manner comprises: sequentially calculating data throughputs corresponding to each transmission resource scheduling table; replacing a previous transmission resource scheduling table with a current transmission resource scheduling table when the data throughput of the current transmission resource scheduling table is greater than the data throughput of the previous transmission resource scheduling table; traversing all the transmission resource scheduling tables to obtain a transmission resource scheduling manner corresponding to a transmission resource scheduling table with the largest data throughput; and taking the transmission resource scheduling manner as the target transmission resource scheduling manner.

[0017] In the embodiment, the user combination optimization is completed at the MAC layer, the heuristic algorithm based on field search is used, the complexity of the algorithm is reduced by reducing the number of inverse operations in the calculation of the throughput function, the optimal spectrum access scheme of the user is realized, and the subsequent data transmission efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 is a flowchart of a user resource scheduling method according to an embodiment of the present application;

[0020] Figure 2 is a schematic diagram of a specific user resource scheduling method according to an embodiment of the present application;

[0021] Figure 3 is a structural block diagram of a user resource scheduling device according to an embodiment of the present application;

[0022] Figure 4 is a hardware structure schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0024] For a downlink of a MIMO-OFDM system deploying frequency division multiple access (FDMA), time division multiple access (TDMA) and space division multiple access (SDMA) technologies. A TDMA frame includes multiple TTIs, and a TTI includes multiple OFDM symbols.

[0025] The MIMO-OFDM system performs channel estimation at the beginning of each TTI, and assumes that the channel state information within a TTI is constant. There are K users in the MIMO-OFDM system, denoted as The base station is equipped with T antennas, and each user is equipped with M antennas. The complex channel matrix from the base station to the user g is The user combination on the time slot t and the frequency band rb is The users sharing the frequency band rb are denoted as G(t, rb), and the received signal of user g e G(t, rb) can be represented as:

[0026]

[0027] where, is the transmit signal vector of the base station, is the precoding matrix of the base station for user g, and n g,t,rb is the additive complex Gaussian noise generated by the user receiver. 2 I) is the additive complex Gaussian noise generated by the user receiver.

[0028] The signal-to-interference-and-noise ratio (SINR) of the lth data stream of user g is:

[0029] .

[0030] where, g,t,l,rb is the first column of W g,t,rb . is the interference of multiple data streams in the massive MIMO system, which can be eliminated by using the nonlinear equalization technology at the user receiver end. is the inter-user interference generated by users sharing the frequency band rb in G(t, rb), which can be eliminated by using the zero-forcing precoding. The zero-forcing precoding is obtained by finding W g,t,rb such that On this basis, the SINR can be simplified as:

[0031]

[0032] where, is the normalized precoding vector of w g,t,l,rb , and P g,t,rb is the power allocated to user g on rb in time slot t, and the sum of the powers of all frequency bands and all users is P t , i.e.,

[0033]

[0034] where, g,t,rb can be obtained by singular value decomposition of the joint channel matrix H (For the sake of simple representation of symbols, the subscripts t and rb are omitted) to obtain the above formula (1).

[0035] According to the Shannon formula, the data transmission rate of user g in time slot t can be represented as:

[0036]

[0037] 5.5G new services with data integrity constraints and latency constraints need to consider data transmission on the time scale of multiple TTIs, which not only pursues the maximum data transmission rate in each TTI, but more importantly requires data to be transmitted within the constraint time. For all user data to be transmitted, τ = {τ1,..., τ K} is the remaining transmission time of the user. For user g, if the data transmission task is completed within τ g , it is considered that the user has completed the data transmission task within τ g , the effective transmission rate R g is Otherwise, the rate is 0, that is:

[0038]

[0039] The goal of joint data and resource scheduling is to maximize the effective transmission rate of all users. Let τ max = maxτ1,..., τ K , τ max is the effective transmission rate of all users in τ slots. Let G(t) represent the user scheduling matrix at time t, G(t) is a matrix with elements of 0 or 1, G(t) (g,rb) = 1 indicates that user g accesses frequency band rb. Let G = {G(1),..., G(τ max )} represent the user scheduling matrix set in a period of time. Similarly, P(t) represents the user power allocation matrix at time t, P(t) (g,rb) = P g,t,rb indicates the power allocated to user g on frequency band rb. Let P = P(1),..., P(τ max ) represent the user power allocation matrix set in a period of time. The joint data and resource scheduling optimization problem with data integrity constraints and latency constraints can be formalized as follows:

[0040]

[0041] To solve the above problem, the application proposes a user resource scheduling method, which can determine the optimal data transmission rate by considering the data scheduling of the application layer and the specific implementation of the MAC layer, allocate appropriate power to the user, and adjust the channel allocation of the user based on the data transmission, to obtain a spectrum access scheme and user power allocation scheme that meet the data integrity and latency constraints, and improve the effective transmission rate of the user.

[0042] According to the embodiment of the present application, a user resource scheduling method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0043] In the embodiment, a user resource scheduling method is provided, Figure 1 is a schematic diagram of the user resource scheduling method according to the embodiment of the present application, as Figure 1 shown, the flow includes the following steps:

[0044] Step S101, calculating the achievable rate range of each user based on the user joint channel state information of the current time slot at the MAC layer.

[0045] The user joint channel state information is used to represent the channel resource allocation of each user under the current frequency band access scheme. The achievable rate range is obtained by calculating the rate of each user when transmitting data at the MAC layer according to the user joint channel state information.

[0046] It can be understood that under the current channel resource allocation, the actual transmission rate of each user cannot exceed the achievable rate range.

[0047] Step S102, determining the first to-be-transmitted data of each user in the current time slot.

[0048] According to the data transmission task, the first to-be-transmitted data of each user in the current time slot is allocated.

[0049] Step S103, determining the target rate of each user at the application layer according to the achievable rate range and the first to-be-transmitted data of each user.

[0050] The target rate is within the achievable rate range.

[0051] The data scheduling of the first to-be-transmitted data is modeled according to the optimal control theory at the application layer, and the optimal transmission rate of each user is determined as the target rate of each user based on the achievable rate range of each user.

[0052] Step S104, allocating transmission resources to each user based on the target rate of each user, so that the transmission rate of each user reaches the target rate of each user.

[0053] According to the target rate of each user, a power allocation scheme is determined, and transmission resources are allocated to each user based on the power allocation scheme, so that each user can transmit data at the target rate.

[0054] Step S105, the users at the application layer use the transmission resource to transmit the first to-be-transmitted data, and determine second to-be-transmitted data of each user in the next time slot.

[0055] The users transmit data according to the allocated transmission resource for the transmission task in the current time slot, and when the data transmission is completed, the second transmission data of each user in the next time slot is allocated according to the data transmission task.

[0056] The data transmission task of each time slot is different, and the number of transmission users of each time slot may increase or decrease. For example, when the user a has completed the transmission task in the current time slot, there is no second transmission data in the next time slot of the user a.

[0057] Step S106, the MAC layer adjusts the user joint channel state information in the next time slot based on the second to-be-transmitted data.

[0058] The MAC layer determines the to-be-transmitted user information of the second time slot according to the second to-be-transmitted data, and determines the frequency band access scheme of the second time slot according to the to-be-transmitted user information and the channel frequency band information, to obtain the user joint channel state information in the next time slot.

[0059] Further, the user joint channel state information in the next time slot is used to determine the power allocation scheme for the user. For details, please refer to step S101.

[0060] The user resource scheduling method provided in the embodiment can determine the optimal data transmission rate by considering the data scheduling at the application layer and the specific implementation at the MAC layer, allocate appropriate power to the user, adjust the channel allocation of the user based on the data transmission, obtain the spectrum access scheme and the user power allocation scheme that meet the data integrity and delay constraints, and improve the effective transmission rate of the user.

[0061] In the embodiment, a user resource scheduling method is provided, which is divided into two parts, a power allocation scheme and a frequency band access scheme. Figure 2 The user resource scheduling method according to the embodiment of the application is shown in the flowchart as Figure 2 At t time, the MAC layer Rate estimator estimates the achievable rate range of all users based on the user joint channel state information H t The MAC layer Rate estimator estimates the achievable rate range of all users based on the user joint channel state information H and solves the modeling of the application layer data scheduling based on the optimal control theory, to obtain the target rate of the MAC layer as the optimization target of the MAC layer Rate optimizer. The target rate Data transmission is performed for all users. A frequency band access relationship table of different user combinations is constructed according to users with transmission tasks, a user combination optimization method is used to select a frequency band access scheme with optimal throughput, and the joint channel state information of the users at the t+1 time is determined by using the frequency band access scheme. The specific implementation is as follows:

[0062] First, the power allocation scheme is determined based on reinforcement learning.

[0063] The initial rate of each user is determined based on the achievable rate range, wherein the initial rate is within the achievable rate range, the initial data transmission amount corresponding to the initial rate is calculated according to the initial rate of each user and the first to-be-transmitted data, the initial rate is adjusted by using a neural network algorithm to maximize the initial data transmission amount, and the initial rate corresponding to the maximum initial data transmission amount is taken as the target rate of each user.

[0064] Step S201, constructing a state space and an action space.

[0065] Specifically, the state space includes the first to-be-transmitted data of each user corresponding to different deadlines in the current time slot. The state space is represented as [B1(t),…,B i (t),…,B N (t),c(t)], wherein, B i (t) represents the first to-be-transmitted data of user i at t time with different deadlines, and c(t) represents the channel condition at t time.

[0066] Specifically, the action space includes the initial transmission resource of each user in the current time slot, and the initial transmission resource is allocated based on the initial rate. The action space is represented as [e1(t),…,e N (t)], wherein, e i (t) represents the initial transmission resource allocated to user i at t time.

[0067] Step S202, constructing a reward function by using the state space and the action space.

[0068] The reward function is used to represent the initial data transmission amount.

[0069] Specifically, the reward function is r(t)=D(t)-λE(t).

[0070] Wherein, E(t) represents the transmission resource consumption at t time, D(t) represents the amount of successfully transmitted data.

[0071] Step S203, the initial rate when the reward function is maximum is determined as the target rate of each user by using the recurrent neural network to maximize the reward function.

[0072] Specifically, the reward function is maximized by state updating of the state space and the action space. The state updating process is as follows: the data packet to be transmitted is placed in the buffer, and each buffer contains an infinite number of independent delay-sensitive queues corresponding to N users. denotes the state of each sensitive queue i at time slot t. For user i with initial transmission resource e and channel condition c, the probability of successful transmission is denoted as P i (e,c), where e=0 indicates that the data is not successfully transmitted, P i (0,c) = 0. At the same time, for all e>0, P i (e)>0, if the data to be transmitted is not successfully transmitted, and it is judged that the data to be transmitted does not exceed the transmission time limit, the data to be transmitted is still placed in the buffer.

[0073] The recurrent neural network is trained on the basis of the above state updating. The recurrent neural network includes an actor network and a critic network.

[0074] The actor network determines the initial rate of each user at the current time slot based on the achievable rate range, and allocates initial transmission resources to each user based on the initial rate of each user, and outputs [e1(t),…,e N (t)]. The critic network trains the Q value function in reinforcement learning according to the data in the Replay buffer, where the Q value function is:

[0075] The actor network adjusts the initial rate according to the reward function to re-allocate the initial transmission resources, so that the reward function is maximum. The initial rate corresponding to the maximum reward function is taken as the target rate of each user, and the transmission resources are allocated to each user by using the target rate.

[0076] In this embodiment, the target rate of each user is determined by using the neural network algorithm on the basis of determining the achievable rate range, which can ensure that the overall data transmission effect of all users is optimal and improve the data transmission success rate.

[0077] Step S204, each user transmits the data to be transmitted at the current time slot by using the transmission resources, and determines the data to be transmitted by each user at the next time slot.

[0078] Second, determine the frequency band access scheme based on user combination optimization.

[0079] Step S205, a plurality of transmission resource scheduling tables of the users to be transmitted and the transmission frequency bands are constructed.

[0080] determining, at the MAC layer, a user to be transmitted in a next time slot based on the second data to be transmitted, and constructing a plurality of transmission resource scheduling tables of the user to be transmitted and transmission frequency bands.

[0081] Each of the transmission resource scheduling tables represents a transmission resource scheduling mode.

[0082] Specifically, based on the l users to be transmitted and the g transmission frequency bands, an l×g 0-1 resource scheduling table is constructed for the l users to be transmitted and the g transmission frequency bands. In the table, an element 1 indicates that a corresponding user accesses a corresponding frequency band, and an element 0 indicates that a corresponding user does not access a corresponding frequency band.

[0083] In step S206, a target transmission resource scheduling mode is determined from the transmission resource scheduling tables.

[0084] Specifically, the data throughput of each transmission resource scheduling mode is calculated using the transmission resource scheduling tables. The data throughput is calculated as follows: wherein, wherein A is obtained based on iteration, wherein, Channel correlation matrix

[0085] The transmission resource mode with the largest data throughput is taken as the target transmission resource scheduling mode.

[0086] Specifically, the data throughput corresponding to each transmission resource scheduling table is calculated in turn. When the data throughput of a current transmission resource scheduling table is greater than that of a previous transmission resource scheduling table, the current transmission resource scheduling table replaces the previous transmission resource scheduling table. All transmission resource scheduling tables are traversed to obtain the transmission resource scheduling mode corresponding to the transmission resource scheduling table with the largest data throughput. The transmission resource scheduling mode is taken as the target transmission resource scheduling mode.

[0087] Further, the user joint channel state information corresponding to the target transmission resource scheduling mode is determined, and the achievable rate range at t+1 time is determined using the user joint channel state information. The user resource scheduling is performed in a loop.

[0088] In the embodiment, the user combination optimization is completed in the MAC layer, the heuristic algorithm based on the field search is used, the complexity of the algorithm is reduced by reducing the number of inverse operations in the calculation of the throughput function, and meanwhile, the optimal spectrum access scheme of the user is realized, and the subsequent data transmission efficiency is improved. By considering the data scheduling of the application layer and the specific implementation of the MAC layer, the optimal data transmission rate can be determined, the appropriate power can be allocated to the user, and meanwhile, the channel allocation of the user is adjusted based on the data transmission, so that the spectrum access scheme and the user power allocation scheme meeting the data integrity and delay constraints can be obtained, and the effective transmission rate of the user is improved.

[0089] In the embodiment, a user resource scheduling apparatus is also provided, which is used to implement the above-described embodiments and preferred embodiments, and will not be described herein again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware or a combination of software and hardware is also possible and contemplated.

[0090] The embodiment provides a user resource scheduling apparatus, as shown in the following Figure 3 The apparatus comprises:

[0091] A calculation module 301 is configured to calculate the achievable rate range of each user based on the user joint channel state information of the current time slot in the MAC layer.

[0092] A first determination module 302 is configured to determine the first to-be-transmitted data of each user in the current time slot.

[0093] A second determination module 303 is configured to determine the target rate of each user according to the achievable rate range and the first to-be-transmitted data of each user in the application layer.

[0094] An allocation module 304 is configured to allocate transmission resources to each user based on the target rate of each user, so that the transmission rate of each user reaches the target rate of each user.

[0095] A transmission module 305 is configured to transmit the first to-be-transmitted data of each user by using the transmission resources in the application layer, and determine the second to-be-transmitted data of each user in the next time slot.

[0096] An adjustment module 306 is configured to adjust the user joint channel state information of the next time slot based on the second to-be-transmitted data in the MAC layer.

[0097] The further function description of each module is the same as that of the corresponding embodiment, and will not be described herein again.

[0098] In this embodiment, the user resource scheduling device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0099] This invention also provides a computer device having the above-described features. Figure 3 The user resource scheduling device shown.

[0100] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 4 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0101] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0102] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0103] The memory 20 can include a program storage area and a data storage area, where the program storage area can store an operating system, application programs required for at least one function, and the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, which can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0104] The memory 20 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid state disk, and can also include a combination of the above-mentioned types of memories.

[0105] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected by a bus or other means, Figure 4 The connection by the bus is taken as an example.

[0106] The input device 30 can receive inputted digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The display device includes, but is not limited to, a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0107] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium through network, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0108] Part of the present application can be applied as a computer program product, for example, computer program instructions, when executed by a computer, the operation of the computer can invoke or provide the method and / or technical solutions according to the present application. Those skilled in the art should understand that the form of computer program instructions in computer readable medium includes but is not limited to source file, executable file, installation package file, etc. Correspondingly, the way of computer program instructions executed by computer includes but is not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer readable medium can be any available computer readable storage medium or communication medium accessible to the computer.

[0109] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A user resource scheduling method, characterized in that, The method includes: At the MAC layer, the reachable rate range for each user is calculated based on the user joint channel state information of the current time slot. Determine the first data to be transmitted for each user in the current time slot; At the application layer, the target rate of each user is determined based on the achievable rate range of each user and the first data to be transmitted, wherein the target rate is within the achievable rate range; Based on each user's target rate, transmission resources are allocated to each user so that each user's transmission rate can reach its respective target rate. In the application layer, each user uses the transmission resources to transmit the first data to be transmitted, and determines the second data to be transmitted by each user in the next time slot. The MAC layer adjusts the user joint channel state information for the next time slot based on the second data to be transmitted, and then returns to step one; The step of determining the target rate for each user at the application layer based on the achievable rate range of each user and the first data to be transmitted includes: The initial rate for each user is determined based on the achievable rate range, wherein the initial rate is within the achievable rate range; Calculate the initial data transmission volume corresponding to the initial rate based on the initial rate of each user and the first data to be transmitted; The initial rate is adjusted using a neural network algorithm to maximize the initial data transmission volume; The initial rate corresponding to the largest initial data transmission volume is taken as the target rate for each user. The step of calculating the initial data transmission volume corresponding to the initial rate based on the initial rate of each user and the first data to be transmitted includes: Construct a state space, which includes the first data to be transmitted for each user in the current time slot at different deadlines; Construct an action space, which includes the initial transmission resources for each user in the current time slot, and the initial transmission resources are allocated based on the initial rate; A reward function is constructed using the state space and the action space, and the reward function is used to characterize the initial data transmission amount.

2. The user resource scheduling method according to claim 1, characterized in that, The state space is ,in, , This represents the first data to be transmitted by user i at different deadlines at time t. This represents the channel condition at time t; The action space is ,in, , This represents the initial transmission resources allocated to user i at time t.

3. The user resource scheduling method according to claim 2, characterized in that, The step of adjusting the initial rate using a neural network algorithm to maximize the initial data transmission volume includes: The reward function is maximized using a recurrent neural network, and the reward function is: ； in, E(t) represents the amount of transmission resources consumed at time t. , Indicates the amount of data successfully transmitted; The step of using the initial rate corresponding to the largest initial data transmission volume as the target rate for each user includes: The initial rate at which the reward function is maximized is determined as the target rate for each user.

4. The user resource scheduling method according to claim 3, characterized in that, The recurrent neural network includes an actor network and a critic network, and the construction of the action space includes: The actor network obtains the initial rate of each user in the current time slot based on the achievable rate range; The initial transmission resources are allocated to each user based on their initial rate. The actor network adjusts the initial rate according to the reward function to reallocate the initial transmission resources.

5. The user resource scheduling method according to claim 4, characterized in that, The method of maximizing the reward function using a recurrent neural network includes: The Q-value function is trained using the aforementioned critic network, and the Q-value function is: 。 6. The user resource scheduling method according to claim 1, characterized in that, The step of adjusting the user joint channel state information for the next time slot at the MAC layer based on the second data to be transmitted includes: The MAC layer determines the user to be transmitted in the next time slot based on the second data to be transmitted; Construct multiple transmission resource scheduling tables for the users to be transmitted and the transmission frequency bands, where each transmission resource scheduling table represents a transmission resource scheduling method; The data throughput of each transmission resource scheduling method is calculated using the transmission resource scheduling table, and the transmission resource method with the highest data throughput is selected as the target transmission resource scheduling method. Determine the user joint channel state information corresponding to the target transmission resource scheduling method.

7. The user resource scheduling method according to claim 6, characterized in that, The expression for calculating the signal-to-interference-plus-noise ratio (SINR) in the calculation of data throughput for each transmission resource scheduling method using the transmission resource scheduling table is as follows: ； in, ; in, The It is based on iterative acquisition; in, , .

8. The user resource scheduling method according to claim 7, characterized in that, The step of using the transmission resource scheduling method with the highest data throughput as the target transmission resource scheduling method includes: Calculate the data throughput corresponding to each transmission resource scheduling table in turn; When the data throughput of the current transmission resource scheduling table is greater than the data throughput of the previous transmission resource scheduling table, the current transmission resource scheduling table replaces the previous transmission resource scheduling table. Traverse all transmission resource scheduling tables to obtain the transmission resource scheduling method corresponding to the transmission resource scheduling table with the largest data throughput. The aforementioned transmission resource scheduling method is taken as the target transmission resource scheduling method.

Images