Resource allocation method and device of CF-RAN system, medium and product

Abstract

The invention discloses a resource allocation method and device of a CF-RAN system, a medium and a product. The method comprises the following steps: acquiring a channel estimation matrix between user equipment and an associated access point; determining the uplink and downlink URLLC transmission rate of the limited block length of the user equipment according to the channel estimation matrix; according to the transmission rate, the preset power and the preset service quality, constructing a target function and a constraint condition set which aim at the weighting and rate maximization of the uplink and downlink users, and obtaining an optimization problem; and solving the optimization problem based on a nested loop optimization algorithm, and determining a duplex mode of each access point, transceiver parameters and user power of uplink user equipment. A duplex mode is dynamically selected for each AP through outer circulation, continuous interference in a fixed full duplex mode is avoided, optimal configuration parameters are determined through combination of inner circulation and an optimization problem, service interruption caused by insufficient power or rate is avoided, and a stable communication link is provided for URLLC service.

Description

A method, apparatus, medium, and product for resource allocation in a CF-RAN system. Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a resource allocation method, apparatus, medium, and product for a CF-RAN system. Background Technology

[0002] As wireless communication technology evolves towards Ultra-Reliable Low-Latency Communication (URLLC), various fields are placing higher demands on the reliability, latency, and spectral efficiency of communication systems. Cellular Freedom Radio Access Network (CF-RAN), with its distributed architecture, effectively solves the high complexity problems of inter-cell interference and massive MIMO in traditional cellular networks, becoming an important architectural choice to support URLLC services. Network-Assisted Free-Duplex (NA-FD) technology, by dynamically scheduling the duplex mode of access points (APs), achieves flexible allocation of time-frequency resources, providing a new technical path to improve spectral efficiency. The combination of these two technologies has become an important research direction in URLLC scenarios.

[0003] In the research of communication technologies in CF-RAN distributed architecture, full-duplex schemes typically include simultaneous, same-frequency full-duplex, half-duplex, and traditional distributed full-duplex. Some schemes optimize transceiver parameters through centralized algorithms to pursue optimal transmission performance; others adopt a fixed-duplex distributed deployment to simplify system coordination complexity.

[0004] However, traditional distributed full-duplex solutions still need to address cross-link interference and distributed coordination complexity in practical deployments. Furthermore, the high coupling of optimization variables brought about by ultra-reliable low-latency communication (URLLC) places higher demands on algorithm performance. Summary of the Invention

[0005] This invention provides a resource configuration method, apparatus, medium, and product for a CF-RAN system. By dynamically determining the configuration parameters for each EDU, the complexity of distributed coordination is reduced, continuous interference in fixed full-duplex mode is avoided, and a stable communication link is provided for URLLC services.

[0006] According to a first aspect of the present invention, a resource configuration method for a CF-RAN system is provided, the CF-RAN system comprising a plurality of distributed units (EDUs) and access points (APs) associated with the EDUs, each EDU being associated with at least one AP, the method being applied to the EDUs, the method comprising:

[0007] Obtain the channel estimation matrix between the user equipment and the associated access point;

[0008] Based on the channel estimation matrix, determine the uplink and downlink URLLC transmission rates of the user equipment with a finite block length;

[0009] Based on the uplink and downlink URLLC transmission rates, preset power, and preset service quality, an objective function and a set of constraints are constructed with the goal of maximizing the weighted sum and rate of uplink and downlink users, thus obtaining the optimization problem.

[0010] The optimization problem is solved based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of the uplink user equipment for each access point. The nested loop optimization algorithm includes an outer loop and an inner loop. The inner loop is used to optimize the transceiver parameters, and the outer loop is used to optimize the duplex mode.

[0011] According to a second aspect of the present invention, a resource configuration apparatus for a CF-RAN system is provided, the CF-RAN system including a plurality of distributed units (EDUs) and access points (APs) associated with the EDUs, each EDU being associated with at least one AP, the apparatus being applied to the EDUs, the apparatus comprising:

[0012] The information acquisition module is used to acquire the channel estimation matrix between the user equipment and the associated access point;

[0013] A rate determination module is used to determine the uplink and downlink URLLC transmission rates of the user equipment with a finite block length based on the channel estimation matrix.

[0014] The problem determination module is used to construct an objective function and a set of constraints with the goal of maximizing the weighted sum rate of uplink and downlink users based on the uplink and downlink URLLC transmission rates, preset power, and preset service quality, thereby obtaining the optimization problem.

[0015] The resource determination module is used to solve the optimization problem based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of the uplink user equipment for each access point; wherein, the nested loop optimization algorithm includes an outer loop and an inner loop, the inner loop is used to optimize the transceiver parameters, and the outer loop is used to optimize the duplex mode.

[0016] According to a third aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the resource configuration method of the CF-RAN system according to any embodiment of the present invention.

[0020] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the resource configuration method of the CF-RAN system according to any embodiment of the present invention.

[0021] According to a fifth aspect of the present invention, embodiments of the present invention also provide a computer program product, the computer program product including a computer program, which, when executed by a processor, implements the resource configuration method of the CF-RAN system according to any embodiment of the present invention.

[0022] The technical solution of this invention involves obtaining the channel estimation matrix between the user equipment and the associated access point; determining the uplink and downlink URLLC transmission rates of the user equipment with a finite block length based on the channel estimation matrix; constructing an objective function and a set of constraints with the goal of maximizing the weighted sum of uplink and downlink user rates based on the transmission rate, preset power, and preset quality of service, thus obtaining an optimization problem; solving the optimization problem based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of the uplink user equipment for each access point. By dynamically determining the configuration parameters for each EDU, the complexity of distributed coordination is reduced. The outer loop dynamically selects the duplex mode for each AP, avoiding continuous interference in a fixed full-duplex mode. The inner loop, combined with the optimization problem, determines the optimal configuration parameters, preventing service interruptions due to insufficient power or rate, and providing a stable communication link for URLLC services.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 is a flowchart of a resource allocation method for a CF-RAN system according to Embodiment 1 of the present invention;

[0026] Figure 2 is a system example diagram of a resource allocation method for a CF-RAN system provided according to Embodiment 1 of the present invention;

[0027] Figure 3 is a flowchart of a resource allocation method for a CF-RAN system according to Embodiment 2 of the present invention;

[0028] Figure 4 is a schematic diagram of a resource allocation device for a CF-RAN system according to Embodiment 3 of the present invention;

[0029] Figure 5 is a schematic diagram of the structure of an electronic device that implements the resource configuration method of the XCF-RAN system according to an embodiment of the present invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] Example 1

[0033] Figure 1 is a flowchart of a resource configuration method for a CF-RAN system provided in Embodiment 1 of the present invention. This embodiment is applicable to the duplex mode selection and parameter configuration of decentralized transceivers in a Network-Assisted Free-Duplex (URLLC) system under a CF-RAN system. This method can be executed by a resource configuration device of the CF-RAN system. The CF-RAN system includes multiple Distributed Units (EDUs) and Access Points (APs) associated with each EDU. Each EDU is associated with at least one AP. The method is applied to the EDUs. The resource configuration device of the CF-RAN system can be implemented in hardware and / or software, and can be configured in an electronic device. As shown in Figure 1, the method includes:

[0034] S110. Obtain the channel estimation matrix between the user equipment and the associated access point.

[0035] In this embodiment, the channel estimation matrix is ​​a matrix used to describe the characteristics of the signal transmission channel between the user equipment (UE) and the access point (AP). For example, it may include information such as large-scale fading, small-scale fading and path loss, and is the basis for calculating the signal-to-interference-to-noise ratio and transmission rate.

[0036] Specifically, the EDU obtains channel characteristic data between its associated AP and the corresponding UE through signal detection, constructs a channel estimation matrix, and considers the effects of large-scale fading, small-scale fading and path loss under flat fading channels, providing a basis for subsequent rate calculation.

[0037] For example, a specific example can be used to illustrate the Network-Assisted Free-Duplex (NA-FD) system for Cellular Radio Access Network (CF-RAN) for Ultra-Reliable Low-Latency Communication (URLLC) of the present invention. Figure 2 is a system example diagram of the resource configuration method of a CF-RAN system provided in Embodiment 1 of the present invention. The system includes a central processing unit (CPU), three user-centric distributed units (UCDUs), and three EDUs. The first EDU is associated with two access points (APs), the second EDU is associated with three APs, and the third EDU is associated with two APs. The APs in the figure are divided into three duplex modes: hybrid duplex mode, flexible duplex module, and full-duplex mode, demonstrating the dynamic mode switching capability of the APs. Red arrows represent self-interference. Solid blue lines represent uplink to downlink interference, and dashed blue lines represent downlink to uplink interference.

[0038] For example, One EDU, Each equipped with The AP has one antenna and is connected to the EDU via fiber optic cable to provide services together. Single-antenna downlink UE and Each UE has a single antenna for uplink. The indices for EDU, AP, downlink UE, and uplink UE in the system are represented as follows: , , and Leveraging the flexibility and scalability of CF-RAN, EDUs can be... and Each AP is associated, satisfying Considering flat fading channels, EDU With downlink users The channel between them can be represented as ,in Indicates AP With downlink users The channel vector between them. Indicates uplink user With downlink users The IUI channel between them. Specifically... , ,in , Represents AP and downstream users Large-scale decay between The representative size is a unit diagonal matrix, and They represent EDU With downlink users Large-scale fading and small-scale fading. and Representing upstream users With downlink users Large-scale fading and small-scale fading between EDU. With uplink users The channel between them can be represented as ,in Indicates AP With uplink users The channel vector between them. Specifically, ,in , Represents AP Large-scale decline between upstream and downstream users, and They represent EDU With uplink users This invention assumes that all small-scale fading is composed of zero-mean, cyclically symmetric, independent and identically distributed Gaussian random variables with unit variance. .also, , , , ,in represent and Path loss between them.

[0039] S120. Based on the channel estimation matrix, determine the uplink and downlink URLLC transmission rates of the user equipment with a finite block length.

[0040] In this embodiment, finite block length can be understood as the limited packet length for data transmission in URLLC scenarios. Unlike traditional long block transmission, transmission reliability constraints (such as maximum error decoding rate) under finite block length need to be considered. Uplink and downlink URLLC transmission rates can be understood as the maximum uplink (UE to AP) and downlink (AP to UE) transmission rates determined based on the signal-to-interference-plus-noise ratio calculated based on the channel estimation matrix and combined with the characteristics of finite block length.

[0041] Specifically, EDU calculates the signal-to-interference-plus-noise ratio (SINR) of uplink and downlink signals based on the channel estimation matrix; and determines the lower bound of uplink and downlink URLLC transmission rates by combining the finite block length characteristics of URLLC scenarios and the preset maximum error decoding rate, thus ensuring transmission reliability.

[0042] For example, the downlink signal transmission model assumes a dynamic TDD mode. In a network-assisted free-duplex system, the dynamic selection of the AP's duplex mode enables flexible duplex transmission within the same time-frequency resource block. For downlink transmission in each time slot, the downlink user... The received signal is as follows:

[0043]

[0044] in, Representing EDU The transmitted signal, With a mean of 0 and a variance of Gaussian white noise, Representing upstream users The transmit power, v x,k EDU For downlink users The precoded vector. and Representing downlink users With uplink users Information symbols. Furthermore... ,in Representing EDU For downlink users , where Represents AP For downlink users The precoded vector, This represents the transpose operation. Assign a matrix to the downlink binary. Represents AP For sending AP (T-AP), Represents AP For receiving AP (R-AP). Therefore, downlink users The received signal-to-interference-plus-noise ratio is:

[0045]

[0046] By treating other terms as noise in a long block state, the downlink signal is decoded to obtain the Shannon rate function: To ensure downlink users Maximum error decoding rate at Given a block length of In the case of ( Represents bandwidth. The lower bound of the maximum downlink URLLC transmission rate (representing the duration of the signal) can be modeled as:

[0047]

[0048] in, .

[0049] For each time slot, the uplink signal transmission model uses EDU (Electronic Data Unit) for uplink transmission. The received signal is as follows:

[0050]

[0051] in, Represents a matrix with zero mean and covariance. Additive white Gaussian noise, This represents noise power.

[0052] The inter-AP IAI interference channel is modeled as follows: ,in and They represent EDU With EDU Estimate the channel and error channel between them. Further, assume... ,in Represents the residual error gain resulting from incomplete IAI cancellation in the digital or analog domain. After effective interference cancellation, the EDU... The uplink baseband signal at this location can be represented as:

[0053]

[0054] make Representing EDU Used for demodulating uplink users Information symbols The receiver, of which Representative and EDU All associated APs for uplink users The joint receiver vector, Represents AP For uplink users The receiver vector. This is the uplink binary allocation matrix. The signal-to-interference-plus-noise ratio (SINNR) at the CPU can be expressed as:

[0055]

[0056] in, Representing upstream users The sum of the power of interference and noise at the location.

[0057] By treating other terms as noise in a long block state, the uplink signal is decoded to obtain the Shannon rate function: To ensure uplink users Maximum error decoding rate at Given a block length of In this case, the lower bound of the maximum uplink URLLC transmission rate can be modeled as:

[0058]

[0059] S130. Based on the uplink and downlink URLLC transmission rates, preset power, and preset service quality, construct an objective function and a set of constraints with the goal of maximizing the weighted sum and rate of uplink and downlink users, and obtain the optimization problem.

[0060] In this embodiment, the preset power can be understood as the upper limit of the transmit power set by the system for the edge distributed unit (EDU) and uplink UE, which is one of the constraints of the optimization problem. The preset quality of service (QoS) can be understood as the minimum requirement for user data rate set in the URLLC scenario. Maximizing the weighted sum rate of uplink and downlink users can be understood as the core objective of system optimization, which is to maximize the overall transmission rate by assigning weights to uplink and downlink users. The objective function is a mathematical expression with the weighted sum rate of uplink and downlink users as its core, which is the direction for solving the optimization problem. The constraint set is used to limit the range of feasible solutions to the optimization problem. The optimization problem can be understood as a complete mathematical problem formed by combining the objective function and the constraint set, with the core being to maximize the weighted sum rate of uplink and downlink users while satisfying all constraints. The nested loop optimization algorithm can be understood as an algorithm that solves the optimization problem in two layers of loops: the outer loop optimizes the AP duplex mode, and the inner loop optimizes the transceiver parameters.

[0061] Specifically, the EDU can construct an objective function with the goal of maximizing the weighted sum of uplink and downlink user rates. A set of constraints can be determined through preset power and preset service quality, such as ensuring that the transmit power of the EDU and uplink UE does not exceed a preset power, that the uplink and downlink UE transmission rates meet preset QoS requirements, and binary constraints of the AP duplex mode.

[0062] S140. Solve the optimization problem based on the nested loop optimization algorithm to determine the duplex mode, transceiver parameters and user power of the uplink user equipment for each access point.

[0063] In this embodiment, the duplex mode of the access point can be understood as the duplex transmission mode of the AP, which can dynamically select full-duplex, half-duplex, or other flexible duplex modes. Transceiver parameters can be understood as parameters related to the operation of the transceiver, such as uplink receiver coefficients and downlink precoding vectors. The user power of the uplink user equipment can be understood as the transmit power of each uplink user equipment (UE). The nested loop optimization algorithm includes an outer loop and an inner loop; the inner loop is used for the transceiver parameters, and the outer loop is used to optimize the duplex mode.

[0064] Specifically, the outer loop can use an algorithm to dynamically select the optimal duplex mode for each AP, such as the Hybrid Multi-Strategy Enhanced Quantum Differential Evolution (HMSEQDE) algorithm. The inner loop can optimize transceiver parameters based on the fixed duplex mode determined by the outer loop, using corresponding algorithms, such as continuous convex approximation algorithms or low-complexity SCA hybrid algorithms combining maximum ratio transmission (MRT) and zero-forcing (ZF) precoding, ultimately outputting the optimal AP duplex mode, transceiver parameters, and uplink UE power.

[0065] The technical solution of this invention calculates the signal-to-interference-plus-noise ratio (SINR) of uplink and downlink signals based on the channel estimation matrix, determines the lower bound of URLLC transmission rate by combining the finite block length characteristic, ensures that the transmission error rate meets the requirements, and constructs an optimization problem based on this to adapt to the high reliability requirements of URLLC. An outer loop dynamically selects the duplex mode for each AP to avoid continuous interference in the fixed full-duplex mode, while an inner loop, combined with the optimization problem, determines the optimal configuration parameters to avoid service interruptions due to insufficient power or rate, thus providing a stable communication link for URLLC services.

[0066] As a first optional embodiment of this embodiment, based on the above embodiment, it further includes:

[0067] Before the next iteration begins after each iteration of the inner loop, the candidate uplink user power of other EDUs is obtained, and the candidate uplink user power with the highest power is selected as the user power of the uplink user device in this iteration. .

[0068] In this embodiment, the candidate uplink user power can be understood as the uplink user equipment (UE) transmit power scheme determined by each Edge Distributed Unit (EDU) in the inner optimization of this iteration. Other EDUs can be understood as other Edge Distributed Units that are in the same Cellular Radio Access Network (CF-RAN) architecture as the current EDU, associated with different Access Points (APs), and undertaking distributed computing tasks. The uplink user power in this iteration can be understood as the unified uplink UE transmit power finally determined after information exchange and filtering among EDUs, applicable to the subsequent process of this iteration and the initial state of the next iteration.

[0069] Specifically, each EDU can determine an uplink user power scheme that satisfies its own local constraints, i.e., a candidate uplink user power, based on the current iteration of its inner loop. Before the current iteration ends and the next iteration begins, a power information exchange strategy between EDUs is triggered. Each EDU sends its calculated candidate uplink user power to other associated EDUs via backhaul signaling, while simultaneously receiving candidate uplink user power from other EDUs. After collecting all candidate uplink user power, each EDU selects the maximum value according to preset rules and determines this maximum value as the globally unified uplink user power, serving as the final power result for this iteration. All EDUs synchronously adopt this maximum candidate uplink user power, not only for summarizing and verifying the optimization results of this iteration but also as the initial uplink power parameter for the next iteration.

[0070] In the first optional embodiment of this embodiment, during the interval between adjacent iterations, a globally unified power allocation scheme is determined by synchronizing the uplink user power used by all EDUs, thereby avoiding power allocation conflicts among EDUs in a distributed scenario and ensuring consistent inner loop constraints.

[0071] Example 2

[0072] Figure 3 is a flowchart of a resource configuration method for a CF-RAN system provided in Embodiment 2 of the present invention. This embodiment is a further refinement of the above embodiment. As shown in Figure 3, the method includes:

[0073] S201. Obtain the channel estimation matrix between the user equipment and the associated access point.

[0074] S202. Based on the channel estimation matrix, determine the uplink and downlink URLLC transmission rates of the user equipment with a finite block length.

[0075] S203. Construct a channel error model based on the channel state information at the EDU.

[0076] In this embodiment, channel state information can be understood as channel characteristic data between its associated access point (AP) and user equipment (UE) obtained by the edge distributed unit (EDU), including channel estimates and channel uncertainty-related information, which forms the basis for constructing the channel error model. The channel error model is a mathematical model used to describe the deviation between the estimated channel and the actual channel at the EDU, and is used to quantify the impact of incomplete channel state information on transmission performance.

[0077] Specifically, the EDU can obtain the channel estimate between its associated AP and UE, and then construct a channel error model based on the imperfections of the channel estimate. This model splits the real channel into two parts: the estimated channel and the error channel. It assumes that the variance attenuation rate of the error channel is proportional to the channel variance, and that the EDU knows the conditional probability information of the estimated channel, thereby quantifying the impact of channel uncertainty on transmission.

[0078] For example, the channel error model can be expressed as: ,in and They represent EDU Estimated channel and error channel, Furthermore, assuming EDU Having the conditional probability knowledge of estimating the channel, i.e., EDU Having conditional probability The information. The attenuation rate of the error variance of the channel state information is proportional to the channel variance, that is... ,in This represents relative uncertainty.

[0079] S204. Based on the uplink and downlink URLLC transmission rate and channel error model, construct an objective function with the goal of maximizing the uplink and downlink user weighted sum rate.

[0080] Specifically, EDU can use the sample averaging approximation method to transform the stochastic optimization problem brought about by the channel error model into a deterministic problem; then, combined with the already determined uplink and downlink URLLC transmission rates, it constructs an objective function with the goal of maximizing the weighted sum rate of uplink and downlink users in the system, and clarifies that the optimization direction is to improve the overall transmission rate by adjusting relevant parameters.

[0081] For example, the Sample Average Approximation (SAA) method can approximate a stochastic problem as a deterministic problem. Specifically, given... ,gather Includes A conditional distribution The sampling results in an independent and identically distributed channel.

[0082]

[0083] With a sufficiently large sample size, the random rate can be approximated by the sample mean. Therefore, the sample mean can be defined as:

[0084]

[0085] in, and Representing downlink users and uplink users Corresponding to channel implementation The signal-to-interference-plus-noise ratio (SINR). Here, "sample" refers to the lower bound of the transmission rate corresponding to each channel implementation. The sample average is... The average value of the lower bound of the transmission rate under secondary channel implementation.

[0086] Ultimately, the objective function was determined as follows: ,in and Representing downlink users and uplink users The weight.

[0087] S205. Based on the EDU transmit power budget in the preset power, construct the first constraint condition.

[0088] In this embodiment, the EDU transmit power budget is the upper limit of transmit power set for each EDU, which is a key parameter constraining the EDU signal transmission strength. The first constraint can be understood as a constraint constructed based on the EDU transmit power budget, limiting the EDU's transmit power to not exceed the preset budget.

[0089] Specifically, the EDU can refer to the preset power EDU transmit power budget to construct the first constraint condition, which clearly limits the transmit power of each EDU to not exceed the budget, so as to avoid the EDU from exceeding the power limit or aggravating interference due to excessive power.

[0090] S206. Based on the uplink UE transmit power constraint value in the preset power, construct the second constraint condition.

[0091] In this embodiment, the uplink UE transmit power constraint value is the upper limit of transmit power set for each uplink UE. The second constraint condition can be understood as a constraint constructed based on the uplink UE transmit power constraint value, which limits the transmit power of the uplink UE to not exceeding the preset constraint value.

[0092] Specifically, the EDU can construct a second constraint based on the uplink UE transmit power constraint value in the preset power, limiting the transmit power of each uplink UE to not exceed the constraint value, and ensuring that the uplink transmission power is within the system's allowable range.

[0093] S207. Based on the preset service quality, construct a third constraint condition.

[0094] In this embodiment, the third constraint can be understood as a constraint built based on a preset quality of service, which limits the transmission rate of the uplink and downlink UEs to no less than the preset QoS requirement.

[0095] Specifically, EDU can construct a third constraint based on preset service quality requirements, which clarifies that the transmission rates of uplink and downlink UEs must meet the corresponding minimum QoS standards, thus ensuring the low latency and high reliability transmission requirements of URLLC services.

[0096] S208. Based on the first constraint, the second constraint, and the third constraint, determine the set of constraints to obtain the optimization problem.

[0097] For example, EDU will jointly optimize ,in, Representing upstream users The transmission power, These represent the AP duplex mode selection vectors for EDUx uplink and downlink, respectively. This represents EDUx for downlink users. The precoded vector, Representing EDU Used for demodulating uplink users Information symbols Given a receiver, the optimization problem P1 aims to maximize the weighted sum rate of the URLLC system while comprehensively considering the uplink user and EDU transmit power and QoS constraints.

[0098]

[0099] The first line is the objective function. = 1 means AP l is R-AP, and 0 means AP l is T-AP. This is the first constraint condition. Representing EDU Power budget. For the second constraint Representing upstream users Power budget. This is the third constraint. and Representing downlink users and uplink users QoS (Quality of Service) constraints for communication are critical in URLLC scenarios because the data rate affects the duration required for the AP to send a specified packet length. (Binary variables) and The existence of makes the above optimization problem a mixed integer problem, which can theoretically be solved by exhaustive search, but has extremely high computational complexity.

[0100] S209. In the outer loop, determine the duplex mode of each access point in each loop.

[0101] Specifically, after starting the nested loop optimization algorithm, the HMSEQDE algorithm can be used to iterate in the outer loop, select the duplex mode of the current iteration round for each AP, and output the mode selection result of the round.

[0102] Furthermore, based on the above embodiments, the steps for determining the duplex mode of each access point in each loop in the outer loop can be refined as follows:

[0103] An AP mode matrix is ​​generated based on the initial qubit matrix. In each round of iteration, the duplex mode is selected based on the enhanced quantum differential evolution algorithm and the AP mode matrix to obtain the duplex mode of each access point in each round of iteration.

[0104] In this embodiment, the qubit matrix is ​​a matrix composed of qubits. A qubit is the basic unit of information in quantum computing. Each qubit exists in a superposition of "0" and "1" states. Each element in the matrix represents the quantum superposition probability of the duplex mode of the corresponding access point (AP), and is the core data structure of the Hybrid Multi-Strategy Enhanced Quantum Differential Evolution (HMSEQDE) algorithm. The AP mode matrix can be understood as a binary matrix recording the duplex modes of each AP. Each element in the matrix corresponds to an AP, taking a value of 0 or 1 (or other mode encoding), representing different duplex modes of the AP (such as half-duplex, full-duplex, and flexible duplex), and is the result carrier of duplex mode selection. The Enhanced Quantum Differential Evolution (HMSEQDE) algorithm can be understood as an optimization algorithm that integrates quantum computing and differential evolution. Through operations such as qubit encoding, multi-strategy mutation, quantum crossover, and adaptive quantum state updates, it achieves efficient search and selection of AP duplex modes, improving search diversity and convergence speed compared to traditional algorithms.

[0105] Specifically, EDU can set the dimension of the qubit matrix (corresponding to the number of APs), and the state of each qubit is randomly generated and satisfies the normalization condition (the sum of the probabilities of the "0" state and the "1" state is 1). Based on this qubit matrix, an initial AP mode matrix is ​​generated according to preset rules, and each element in the matrix is ​​randomly assigned a binary value (corresponding to the initial duplex mode), completing the parameter initialization before the algorithm starts. After each iteration, a corresponding binary population (i.e., the AP duplex mode candidate set) is generated based on the current qubit matrix, and each individual in the population corresponds to a set of AP duplex mode combinations. According to the fitness value of the previous iteration (based on the weighted sum rate of uplink and downlink users in the system), the quantum population is divided into three subpopulations with the best, medium, and lowest fitness, and different differential mutation strategies are adopted for each: the best subpopulation adopts the DE / bestassrand / 2 strategy (combined with the Mesh wavelet function to improve the mutation coefficient), the medium subpopulation adopts the DE / randtobest / 1 strategy, and the lowest subpopulation adopts the DE / rand / 1 strategy, ensuring population diversity and convergence speed. Through quantum crossover operations, parent individuals are combined with mutated individuals to generate new individuals, ensuring that each new individual has at least one qubit different from the original individual. A greedy strategy is used to evaluate the fitness of new individuals, retaining those with better performance for the next generation and eliminating those with poor performance. Based on the binary observations of the current individual and the best individual, the probability amplitude of the qubits is adjusted through quantum rotation gates to guide the population towards the optimal solution. Simultaneously, mutation probabilities are set, and the optimal position in memory is maintained through NOT gate mutation, preventing the algorithm from converging prematurely. After completing the above evolutionary operations, the corresponding binary code is extracted from the current best individual and mapped to the duplex mode of each AP. The AP mode matrix is ​​updated to obtain the AP duplex mode results for this round of iteration, providing fixed mode constraints for the transceiver parameter optimization of the inner loop.

[0106] For example, in classical computers, the smallest unit of information is called a qubit, defined as... ,in and Let represent the probabilities of a qubit being in the "0" or "1" state, respectively, and satisfy . .therefore, and It can also be expressed as: and , here It doesn't have a specific physical meaning; it can be considered a parametric tool. Its core purpose is to utilize trigonometric identities. It automatically satisfies the normalization condition of the quantum state.

[0107] It can contain The first bit individual Represented as

[0108]

[0109] in, , It is a randomly generated value between 0 and 1, and .

[0110] In each generation, binary strings are generated using qubits according to the following criteria.

[0111]

[0112] Based on the fitness values ​​obtained in the previous iteration, the quantum population is divided into three subpopulations, and different mutation schemes are implemented for each subpopulation. Assume the subpopulations... Include Individually, the specific implementation process is as follows: 1) Multiple population mutation evolution mechanisms: the subpopulation with the best fitness Differential mutation method (DE / bestassrand / 2): ,in It is in the interval Randomly generated and distinct index numbers are used within the index. as well as It is a random number in the range (0,1) and satisfies , Indicates the first In the nth iteration, the 1st The qubit space angle of an individual, where t represents the iteration round; m is a symbol, usually indicating the "mutated" state. Coefficient of variation. This is an important parameter in the current evolutionary process. To ensure the diversity of the quantum population and improve the convergence speed, this invention uses a Mesh wavelet function to improve the coefficient of variation: Subpopulations with medium fitness Differential mutation strategy (DE / randtobest / 1): ,in Indicates the first In each iteration, the qubit space angle corresponding to the individual with the best fitness in the entire population; the subpopulation with the lowest fitness. A DE / rand / 1 strategy is employed. This mutation expands and improves the search scope through random individuals: 2) Quantum crossover operation: The crossover process obtains new individuals by maximizing the combination of preset parent individuals and mutated individuals, thereby improving optimization capabilities.

[0113]

[0114] in, From The index number is randomly selected from the set, ensuring that at least one qubit in each individual set is different from the original set. The crossover probability can be expressed as... . Indicates the first In the nth iteration, the 1st The first individual The qubit space angle in one dimension.

[0115] Quantum selection operation: This invention employs a greedy strategy to evaluate the fitness of the test vector, ensuring that offspring are superior to previous generations.

[0116]

[0117] in, Representing the The th iteration in the Individual, This represents the candidate individuals generated after the crossover operation. This represents the fitness function, used to determine the quality of a solution. This represents the probability of mutation.

[0118] Adaptive quantum state update: Quantum rotation gates are a common strategy in population evolution. This invention uses the rotation scheme shown in Table 1 to update the quantum state from... Updated to (represented as) ),in and These represent the binary observations of the current individual and the optimal individual, respectively. For the fitness function, and This represents the probability amplitude of the current qubit. A value of 1 indicates clockwise rotation, -1 indicates counterclockwise rotation, 0 indicates no change, and ±1 indicates random rotation. The quantum rotation gate can be represented by the following table:

[0119] Table 1 Quantum Revolving Door Scheme

[0120]

[0121] Where T represents true, that is, the judgment condition is met. F stands for false, meaning the condition is not met.

[0122] NOT gate mutation: This invention achieves mutation by setting the mutation probability. And use NOT gates to maintain the optimal position in memory:

[0123]

[0124] Following the corresponding steps of the quantum algorithm described above, the corresponding binary code is extracted from the current best population individual using the HMSEQDE method described above, mapped to the duplex mode of each AP, and the AP mode matrix is ​​updated to obtain the AP duplex mode result under this round of iteration.

[0125] S210. In the inner loop, based on the duplex mode and optimization problem, determine the transceiver parameters for each loop.

[0126] Specifically, the EDU can start the optimization process based on the fixed duplex mode determined by the outer loop and the inner loop. First, the original optimization problem can be restated as a convex optimization problem with auxiliary variables. The downlink and uplink constraints are approximated separately using the continuous convex approximation method. Then, a distributed algorithm based on SCA is used, and each EDU independently calculates its local transceiver parameters (precoding vector and receiver coefficients). If it is necessary to reduce the computational complexity, it can be replaced with a low-complexity SCA hybrid algorithm that combines maximum ratio transmission (MRT) and zero-forcing (ZF) precoding to output the transceiver parameters of the current iteration round.

[0127] Furthermore, based on the above embodiments, using the continuous convex approximation algorithm as the solution algorithm for the inner loop, the steps for determining the transceiver parameters in each loop based on the duplex mode and optimization problem can be refined as follows:

[0128] Based on the auxiliary variables of the continuous convex approximation algorithm, the objective function is transformed into a convex objective function; based on the auxiliary variables, the uplink and downlink constraints in the constraint set are transformed into convex uplink and downlink constraints, resulting in the first set of constraints after transformation; the convex objective function and the first set of constraints are taken as a convex optimization problem; the convex optimization problem is solved based on the continuous convex approximation algorithm to determine the transceiver parameters in each cycle.

[0129] In this embodiment, the Continuous Convex Approximation (SCA) algorithm can be understood as an algorithm that transforms a non-convex optimization problem into a series of convex optimization subproblems for iterative solution. Auxiliary variables are intermediate variables introduced to achieve a convex approximation of the non-convex function, used to replace non-convex terms in the original objective function or constraints. The convex objective function can be understood as an objective function that satisfies the definition of a convex function after auxiliary variable substitution and convex approximation. Uplink and downlink constraints can be understood as constraints limiting the uplink and downlink signal transmission quality. Convex uplink and downlink constraints can be understood as transforming the original non-convex uplink and downlink constraints into convex constraints through auxiliary variable substitution and inequality approximation. The convex optimization problem can be understood as a complete optimization problem consisting of a convex objective function and a first set of constraints, which can be solved using a standard convex optimization algorithm. The first set of constraints can be understood as a set of convex constraints.

[0130] Specifically, for the non-convex terms in the original non-convex objective function caused by channel interference and power coupling, pre-defined auxiliary variables are introduced. Through variable substitution, the non-convex terms in the original objective function are transformed into a linear combination or convex function form of the auxiliary variables, ultimately yielding a convex objective function that satisfies the requirements of convex optimization.

[0131] The following is the set of inequalities used in the convex approximation process:

[0132]

[0133]

[0134]

[0135] in Represents two scalar variables, Represents vector variables, Represents a matrix, and They represent the first In the next iteration and The real-time value, and functions respectively and The lower bound.

[0136] For example, auxiliary variables can be introduced. and The optimization problem P1 above will be restated as optimization problem P2:

[0137]

[0138] in, , and Representative at EDU Auxiliary variables calculated at the location, Corresponding to the Secondary channel implementation.

[0139] Specifically, for the downlink constraints in the constraint set, the signal interference-related expressions are first simplified by treating the interference information of other EDUs as constants. Auxiliary variables are introduced to approximate the non-convex terms, and then the constraints are transformed into linear constraints through inequality transformations. For the uplink constraints, for the SINR function containing complex fourth-order terms, multiple auxiliary variables are introduced to decompose the non-convex structure. Convex upper bound approximation and inequality transformations are used to transform the non-convex constraints into convex constraints. Finally, the first constraint set consisting of convex uplink and downlink constraints is obtained (i.e., in the formula of optimization problem P2, the constraints below, excluding the objective function in the first row, constitute the first constraint set). The transformed convex objective function is combined with the first constraint set to form a complete convex optimization problem. The continuous convex approximation algorithm is iterated. In each iteration, the convex optimization problem is solved based on the parameter values ​​at the current iteration point (such as the channel estimation matrix and the current values ​​of auxiliary variables) to obtain the transceiver parameters (precoding vector) for this round. and receiver coefficient Update the auxiliary variables and iteration points, repeat the iteration process until the objective function value no longer changes (convergence condition is met), output the optimal transceiver parameters for the current round, and provide the basis for the next round of iteration of the nested loop algorithm.

[0140] For example, due to the high coupling of variables in the constraints, approximating uplink constraints is more difficult than approximating downlink constraints, especially when they contain complex quartic terms (i.e., The uplink SINR function is used. The approximation approach for uplink constraints is similar to that for downlink constraints: first, approximate sub-equation 8 in P2, and then approximate sub-equation 6.

[0141] For sub-expression 8, firstly... Preliminary derivation yields the following constraint relationships.

[0142]

[0143] in, Therefore, sub-equation 8 can be further transformed into

[0144]

[0145] Next, we introduce a series of auxiliary variables. Approximating the above equation as:

[0146]

[0147] The last two minors of the above equation can be transformed using the set of inequalities mentioned earlier.

[0148]

[0149] in, .

[0150] Next, let's focus on sub-expression 1 in the above equation. After a simple algebraic transformation, we obtain the following expression.

[0151]

[0152] The left and right sides of the above equation are neither non-convex nor concave functions. Fortunately, they can be bounded by the following convex upper bound. To approximate the non-convex right-hand side term

[0153]

[0154] The condition for the above equation to be satisfied is the auxiliary variable. .when At that time, the following two important properties can be easily verified: , .in, represent The gradient of . Let , , You can get .final, It can be approximated as

[0155]

[0156] Next, consider non-convex constraints. and By approximating the constraints using the set of inequalities mentioned earlier, these two constraints can be approximated as follows:

[0157]

[0158] Introducing auxiliary variables The minor expression 6 of problem P2 can be transformed into

[0159]

[0160] Further introduce auxiliary variables Sub-expression 1 in the above formula can be restated as

[0161]

[0162] Finally, the optimization problem P2 is solved using the SCA technique. The iterative algorithm (where i represents the number of iterations in the inner loop) can be expressed as:

[0163]

[0164]

[0165] in, .

[0166]

[0167] Furthermore, based on the above embodiments, in order to further reduce the computational complexity of the inner loop, a hybrid ZF-MRT beamforming algorithm can be used as the optimization algorithm for the inner loop. The steps for determining the transceiver parameters in each loop based on the duplex mode and the objective function in the inner loop can be refined as follows:

[0168] Based on the hybrid maximum ratio transmission and zero-forcing beamformer and the combination coefficient variables, the optimization problem is transformed into a combination coefficient optimization problem. The constraint set in the combination coefficient optimization problem is processed by the continuous convex approximation algorithm to obtain the second constraint set in convex form, thus forming a convex combination coefficient optimization problem. The convex combination coefficient optimization problem is solved to determine the transceiver parameters in each cycle.

[0169] In this embodiment, the hybrid maximum ratio transmittance and zero-forcing beamformer can be understood as a beamforming device that fuses maximum ratio transmittance (MRT) and zero-forcing (ZF) precoding techniques. Combination coefficient variables: Two variables used for the linear fusion of MRT and ZF precoding (hereinafter referred to as...). and (Represented). The combination coefficient optimization problem can be understood as a transformed problem where transceiver parameter optimization is replaced by combination coefficient optimization. The convex form combination coefficient optimization problem can be understood as a complete optimization problem consisting of a convex objective function and a second set of constraints, possessing the characteristics of "convex objective + convex constraints," and can be efficiently solved using standard convex optimization tools. The second set of constraints can be understood as a convex constraint set after processing by a hybrid maximum ratio transmission, zero-forcing beamformer, and SCA algorithm.

[0170] Specifically, the EDU can introduce a hybrid ZF-MRT beamformer to linearly combine MRT precoding and ZF precoding, where the combination coefficients determine the weight ratio of the two precoding methods. This structure allows the complex transceiver parameters (precoding vector, receiver coefficients) in the original optimization problem to be expressed as functions of the combination coefficients and the channel matrix, thus transforming the original transceiver parameter optimization problem into an optimization problem focusing solely on the combination coefficients, significantly reducing computational complexity. For the non-convex constraints in the combination coefficient optimization problem, auxiliary variables are introduced to decompose the non-convex structure, transforming constraints containing complex terms into convex constraints. The transformed convex objective function is then combined with the second set of constraints (i.e., in the formula of optimization problem P4, the constraints below the objective function in the first row are the second set of constraints) to form a complete convex combination coefficient optimization problem. The EDU can initiate SCA algorithm iterations. In each iteration, based on the current combination coefficient iteration point, auxiliary variable values, and the channel estimation matrix, the convex combination coefficient optimization problem is solved to obtain the optimal combination coefficients for that round. Based on the structural relationship of the hybrid ZF-MRT beamformer, the optimal combination coefficients are substituted into the expressions of the precoding vector and receiver coefficients to derive the transceiver parameters for the current round. This iterative process is repeated until the objective function value no longer changes (convergence condition is met), and the optimal transceiver parameters for this round are output.

[0171] For example, EDU can introduce two combined coefficient variables. and You can get ,in , , Representative channel matrix The null space orthogonal basis. Therefore, the link gain generated by the hybrid MRT-ZF beamformer can be obtained as:

[0172]

[0173] in, , , , n corresponds to the nth channel implementation, and H represents the conjugate transpose; v z,k v represents the precoding vector between EDU z and user k. x,k This represents the pre-encoding vector between EDU x and user k; that is, x and z are simply used to identify different EDUs.

[0174] When using hybrid beamforming, EDU Downstream users The transmitted power can be expressed as

[0175]

[0176] in, , , .

[0177] Subsequently, by and Substituting into the optimization problem P1, the present invention restates the above optimization problem P1 using a hybrid ZF-MRT beamformer, and represents it as optimization problem P3.

[0178]

[0179] in, , , , , and The detailed expansion is as follows

[0180]

[0181] in, This represents the Q-gain between user k and user k, which is the self-gain of user k. This represents the interference power between EDUz and EDUz'.

[0182] Clearly, the optimization problem P3 is a non-convex NP-hard problem. Therefore, a new variable is introduced. The variable satisfies And can be relaxed into .

[0183] therefore, and It can be represented as

[0184]

[0185]

[0186] in, .

[0187] Optimization problem P3 can be further transformed into optimization problem P4:

[0188]

[0189] in, , and These are auxiliary variables. The optimization problem P4 is clearly still non-convex. This invention employs a continuous convex approximation method to process sub-equations 3, 4, 5, and 6 in the constrained optimization problem P4 to obtain their convex approximate forms.

[0190] 1) Downward Constraint Approximation: This section focuses on sub-expressions 3 and 5. For sub-expression 3, the following transformation relationship exists: ,in, , . It is a concave function. Furthermore, because the function is concave... Since the function is concave, its first-order approximation can be used as an upper bound. Specifically, given any... ,use You can get The first-order approximation is

[0191]

[0192] Therefore, it can be Approximately

[0193]

[0194] in, .

[0195] Therefore, sub-equation 3 in optimization problem P4 can be approximated as:

[0196]

[0197] Next, we will deal with the non-convex constraint sub-equation 5. Through appropriate simplification, sub-equation 5 can be equivalently reconstructed into the following form:

[0198]

[0199] Introducing auxiliary variables As The upper bound, that is

[0200]

[0201] The above formula is equivalent to:

[0202]

[0203] Based on this, the following inequality can be derived.

[0204]

[0205]

[0206] 2) Uplink Constraint Approximation: This section focuses on uplink constraints (sub-equations 4 and 6). First, we address the non-convex constraint sub-equation 4. We first obtain: .in, and Can be respectively by and get.

[0207]

[0208]

[0209] The detailed expansion is as follows:

[0210]

[0211] Next, we will deal with constraint condition sub-equation 6, which can be transformed as follows:

[0212]

[0213] Introducing auxiliary variables The above formula can be further approximated as follows:

[0214]

[0215]

[0216] in, , according to Iterative updates will be performed.

[0217] 3) SCA-based hybrid beamforming algorithm for solving optimization problem P4: To date, all non-convex constraints in optimization problem P4 have been addressed. In each iteration, the following optimization problem needs to be solved:

[0218]

[0219]

[0220] in, .

[0221] S211. Continue until the value of the objective function no longer changes, and obtain the optimal duplex mode for each access point. Transceiver parameters and the power of uplink user equipment .

[0222] The technical solution of this invention constructs a channel error model based on channel state information, quantifies the impact of incomplete channel state information at the EDU, and transforms the stochastic optimization problem into a deterministic problem using the Sample Average Approximation (SAA) method. This improves the robustness of interference suppression, reduces interference amplification caused by channel estimation errors, and ensures high reliability of URLLC transmission as residual interference power decreases. The maximum error decoding rate is controlled within a preset range to meet critical service requirements. A complete constraint system is constructed by integrating finite block length characteristics and quality of service to ensure that the optimization results meet the low latency and high reliability requirements of URLLC. An enhanced quantum differential evolution algorithm is used to solve the outer loop, dynamically determining the duplex mode of each access point, improving search diversity, and efficiently searching for the optimal duplex mode of the AP. Compared to the greedy search algorithm, this improves convergence speed while maintaining similar spectral efficiency, providing precise mode constraints for inner transceiver optimization. The inner loop solves for the optimal parameters of the distributed transceiver based on SCA, effectively suppressing uplink-downlink interference and inter-AP interference. The inner loop can also solve for the optimal parameters of the transceiver by using a hybrid ZF-MRT beamforming algorithm, which reduces the computational complexity of the inner loop, greatly improves the real-time performance of the algorithm, and meets the low latency requirements of URLLC.

[0223] Example 3

[0224] Figure 4 is a schematic diagram of a resource configuration device for a CF-RAN system provided in Embodiment 3 of the present invention. The CF-RAN system includes multiple distributed units (EDUs) and access points (APs) associated with each EDU. Each EDU is associated with at least one AP. The device is applied to the EDUs. As shown in Figure 4, the device includes:

[0225] Information acquisition module 31 is used to acquire the channel estimation matrix between the user equipment and the associated access point;

[0226] Rate determination module 32 is used to determine the uplink and downlink URLLC transmission rates of the user equipment with a finite block length based on the channel estimation matrix;

[0227] Problem determination module 33 is used to construct an objective function and a set of constraints with the goal of maximizing the weighted sum rate of uplink and downlink users based on the uplink and downlink URLLC transmission rates, preset power and preset service quality, so as to obtain the optimization problem;

[0228] The resource determination module 34 is used to solve the optimization problem based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of the uplink user equipment for each access point; wherein the nested loop optimization algorithm includes an outer loop and an inner loop, the inner loop is used to optimize the transceiver parameters, and the outer loop is used to optimize the duplex mode.

[0229] The technical solution of this invention reduces the complexity of distributed coordination by dynamically determining the configuration parameters for each EDU, avoids continuous interference in the fixed full-duplex mode by dynamically selecting the duplex mode for each AP through the outer loop, and determines the optimal configuration parameters by combining the inner loop with optimization problems to avoid service interruption due to insufficient power or rate, thus providing a stable communication link for URLLC services.

[0230] Furthermore, the problem determination module 33 is specifically used for:

[0231] Based on the channel state information at the EDU, a channel error model is constructed;

[0232] Based on the uplink and downlink URLLC transmission rates and the channel error model, an objective function is constructed with the goal of maximizing the uplink and downlink user weighted sum rate.

[0233] Based on the EDU transmit power budget in the preset power, construct the first constraint condition;

[0234] Based on the uplink UE transmit power constraint value in the preset power, a second constraint condition is constructed;

[0235] Based on the preset service quality, a third constraint condition is constructed;

[0236] Based on the first constraint, the second constraint, and the third constraint, a set of constraints is determined, and the optimization problem is obtained.

[0237] Furthermore, the resource determination module 34 includes:

[0238] The first determining unit is used to determine the duplex mode of each access point in each round of the outer loop.

[0239] The second determining unit is used to determine the transceiver parameters in each loop based on the duplex mode and the optimization problem in the inner loop.

[0240] The third determining unit is used to obtain the optimal duplex mode, transceiver parameters, and uplink UE user power of each access point until the value of the objective function no longer changes.

[0241] Specifically, the first determining unit is used for:

[0242] Generate AP mode matrix based on initialization of qubit matrix;

[0243] In each round of iteration, the duplex mode is selected based on the enhanced quantum differential evolution algorithm and the AP mode matrix to obtain the duplex mode of each access point in each round of iteration.

[0244] Specifically, the second determining unit is used for:

[0245] The objective function is transformed into a convex form objective function based on the auxiliary variables of the continuous convex approximation algorithm;

[0246] Based on the auxiliary variables, the uplink and downlink constraints in the constraint set are transformed into convex uplink and downlink constraints to obtain the transformed first constraint set.

[0247] The convex objective function and the first set of constraints are treated as a convex optimization problem.

[0248] The convex optimization problem is solved using a continuous convex approximation algorithm to determine the transceiver parameters in each cycle.

[0249] Specifically, the second determining unit is used for:

[0250] Based on the hybrid maximum ratio transmission and zero-forcing beamformer and the combination coefficient variables, the optimization problem is transformed into a combination coefficient optimization problem;

[0251] The constraint set in the combined coefficient optimization problem is processed by the continuous convex approximation algorithm to obtain the second constraint set in convex form, thus forming a convex combined coefficient optimization problem.

[0252] The convex combination coefficient optimization problem is solved to determine the transceiver parameters in each cycle.

[0253] Optionally, the device further includes:

[0254] The power synchronization module is used to acquire the candidate uplink user power of other EDUs before the end of each iteration of the inner loop and the start of the next iteration, and to select the largest candidate uplink user power as the user power of the uplink user equipment in this iteration.

[0255] The resource configuration device for the CF-RAN system provided in this embodiment of the invention can execute the resource configuration method for the CF-RAN system provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0256] Example 4

[0257] Figure 5 illustrates a schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0258] As shown in Figure 5, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded from storage unit 48 into the RAM 43. The RAM 43 can also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0259] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0260] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as the resource configuration methods of a CF-RAN system.

[0261] In some embodiments, the resource configuration method of the CF-RAN system may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the resource configuration method of the CF-RAN system described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to execute the resource configuration method of the CF-RAN system by any other suitable means (e.g., by means of firmware).

[0262] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0263] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0264] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0265] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0266] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0267] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0268] In one embodiment, the present invention further includes a computer program product, which includes a computer program that, when executed by a processor, implements the resource configuration method of the CF-RAN system according to any embodiment of the present invention.

[0269] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0270] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0271] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A resource allocation method for a CF-RAN system, characterized in that, The CF-RAN system includes multiple distributed units (EDUs) and access points (APs) associated with each EDU. Each EDU is associated with at least one AP. The method is applied to the EDU and includes: obtaining a channel estimation matrix between the user equipment and the associated access point; determining the uplink and downlink URLLC transmission rates of the user equipment with a finite block length based on the channel estimation matrix; constructing an objective function and a set of constraints with the goal of maximizing the uplink and downlink user weighted sum rate based on the uplink and downlink URLLC transmission rates, preset power, and preset quality of service, thus obtaining an optimization problem; solving the optimization problem based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of the uplink user equipment for each access point; wherein the nested loop optimization algorithm includes an outer loop and an inner loop, the inner loop being used to optimize the transceiver parameters, and the outer loop being used to optimize the duplex mode.

2. The method according to claim 1, characterized in that, The process of constructing an objective function and a set of constraints with the goal of maximizing the weighted sum rate of uplink and downlink users based on the uplink and downlink URLLC transmission rates, preset power, and preset quality of service, to obtain the optimization problem, includes: constructing a channel error model based on the channel state information at the EDU; constructing an objective function with the goal of maximizing the weighted sum rate of uplink and downlink users based on the uplink and downlink URLLC transmission rates and the channel error model; constructing a first constraint based on the EDU transmit power budget in the preset power; constructing a second constraint based on the uplink UE transmit power constraint value in the preset power; constructing a third constraint based on the preset quality of service; and determining the set of constraints based on the first constraint, the second constraint, and the third constraint to obtain the optimization problem.

3. The method according to claim 1, characterized in that, The nested loop optimization algorithm is used to solve the optimization problem and determine the duplex mode, transceiver parameters, and uplink UE user power for each access point. This includes: in the outer loop, determining the duplex mode of each access point in each loop iteration; in the inner loop, determining the transceiver parameters in each loop iteration based on the duplex mode and the optimization problem; until the value of the objective function no longer changes, thus obtaining the optimal duplex mode, transceiver parameters, and uplink UE user power for each access point.

4. The method according to claim 3, characterized in that, The step of determining the duplex mode of each access point in each loop in the outer loop includes: generating an AP mode matrix based on an initialized qubit matrix; and selecting the duplex mode based on the enhanced quantum differential evolution algorithm and the AP mode matrix in each loop iteration to obtain the duplex mode of each access point in each loop.

5. The method according to claim 3, characterized in that, In the inner loop, based on the duplex mode and the optimization problem, the transceiver parameters for each loop are determined, including: converting the objective function into a convex objective function based on auxiliary variables of a continuous convex approximation algorithm; converting the uplink and downlink constraints in the constraint set into convex uplink and downlink constraints based on the auxiliary variables, obtaining a first set of transformed constraints; using the convex objective function and the first set of constraints as a convex optimization problem; and solving the convex optimization problem based on the continuous convex approximation algorithm to determine the transceiver parameters for each loop.

6. The method according to claim 3, characterized in that, The process of determining the transceiver parameters for each loop based on the duplex mode and the objective function in the inner loop further includes: transforming the optimization problem into a combination coefficient optimization problem based on the hybrid maximum ratio transmission and zero-forcing beamformer and combination coefficient variables; processing the constraint set in the combination coefficient optimization problem using a continuous convex approximation algorithm to obtain a second convex constraint set, forming a convex combination coefficient optimization problem; and solving the convex combination coefficient optimization problem to determine the transceiver parameters for each loop.

7. The method according to claim 1, characterized in that, Also includes: Before the end of each iteration of the inner loop and the start of the next iteration, the candidate uplink user power of other EDUs is obtained, and the candidate uplink user power is selected as the user power of the uplink user device in this iteration.

8. A resource allocation device for a CF-RAN system, characterized in that, The CF-RAN system includes multiple distributed units (EDUs) and access points (APs) associated with each EDU. Each EDU is associated with at least one AP. The device is applied to the EDU and includes: an information acquisition module for acquiring a channel estimation matrix between the user equipment and the associated access point; a rate determination module for determining the uplink and downlink URLLC transmission rates of the user equipment with a finite block length based on the channel estimation matrix; a problem determination module for constructing an objective function and a set of constraints to maximize the uplink and downlink user weighted sum rate based on the uplink and downlink URLLC transmission rates, preset power, and preset service quality, thereby obtaining an optimization problem; and a resource determination module for solving the optimization problem based on a nested loop optimization algorithm to determine the duplex mode, transceiver parameters, and user power of each access point and uplink user equipment. The nested loop optimization algorithm includes an outer loop and an inner loop, where the inner loop optimizes the transceiver parameters and the outer loop optimizes the duplex mode.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the resource allocation method of the CF-RAN system according to any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the resource configuration method for the CF-RAN system according to any one of claims 1-7.

Images