Communication method, device and medium based on scalable cell-free mimo system

Abstract

This application discloses a communication method, device, and medium based on a scalable decellularized MIMO system, relating to the field of communication technology. The communication method based on a scalable decellularized MIMO system includes: determining a first target access point based on large-scale fading information between a user terminal and multiple access points; then, establishing a service association between the user terminal and the first target access point, whereby the user terminal is determined by the first target access point to be a user whose association cannot be undone; furthermore, the user terminal, based on its own service demand information, determines a second target access point; then, establishing a service association between the user terminal and the second target access point, whereby the user terminal is determined by the second target access point to be a user whose association can be undone. The number of first target access points is one, and the number of second target access points is one or more. By performing the above operations, the user's service quality is improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method, device and medium based on a scalable decellularized MIMO system. Background Technology

[0002] Scalable decellularized massive multiple-input multiple-output (MIMO) systems are a novel wireless communication architecture that, thanks to their high macro diversity gain, ideal propagation conditions, and channel hardening effect, enables them to achieve more reliable coverage and provide users with stable quality of service.

[0003] In decellularized MIMO systems in related technologies, all access points are connected to a single central processing unit, and all access points jointly serve all users. This causes the signaling overhead and computational complexity at the front end to increase linearly with the number of users. At the same time, it cannot be guaranteed that every user accessing the network can continuously receive services from the access point, and there is a risk that users will leave the network midway. Summary of the Invention

[0004] The main objective of this application is to provide a communication method, device, and medium based on a scalable decellularized MIMO system, which aims to dynamically select the most suitable access point for each user to provide services, thereby improving the quality of service for users.

[0005] To achieve the above objectives, one aspect of this application proposes a communication method based on a scalable decellularized MIMO system, comprising the following steps:

[0006] The user terminal determines the first target access point based on the large-scale fading information between it and multiple access points;

[0007] The user terminal establishes a service association with the first target access terminal, and the user terminal is determined by the first target access terminal to be a user whose association cannot be canceled.

[0008] The user determines the second target access point based on its own service needs.

[0009] The user terminal establishes a service association with the second target access terminal, and the user terminal is determined by the second target access terminal to be a user whose association can be canceled.

[0010] The number of the first target access terminals is 1, and the number of the second target access terminals is 1 or more.

[0011] In some embodiments, the method further includes:

[0012] Based on the large-scale fading information obtained between the access end and the user end it serves, the access end jointly optimizes its total transmission power and downlink power allocation through a power optimization model.

[0013] The downlink transmission power of the access point is determined based on the total transmission power and the downlink power allocation;

[0014] The access terminal includes the first target access terminal and / or the second target access terminal, and the power optimization model is obtained by training based on a convolutional neural network.

[0015] In some embodiments, the power optimization model includes a stackable feature extraction network based on a residual network and two parallel output networks. The training steps of the power optimization model include:

[0016] The input tensor is determined by acquiring the large-scale fading information, and the dimension of the input tensor is determined by the maximum number of service users of the communication system and the number of access terminals connected to the edge processor.

[0017] The input tensor is input into the feature extraction network to obtain large-scale fading information features;

[0018] The large-scale fading information features are input into the convolutional neural network, and the total transmission power decision tensor and downlink power allocation decision tensor of the access end are obtained through the parallel output network.

[0019] Based on the large-scale fading information features, the total transmission power decision tensor, and the downlink power allocation decision tensor, the model parameters are optimized and training is completed.

[0020] In some embodiments, determining the downlink transmission power of the access point based on the total transmission power and the downlink power allocation includes:

[0021] The downlink transmission power of the access point is determined by the total transmission power decision tensor, the downlink power allocation decision tensor, and the maximum transmission power of the access point.

[0022] In some embodiments, the access terminal includes a user that can be unassociated, and the user terminal determines a first target access terminal based on large-scale fading information between itself and multiple access terminals, including:

[0023] The user terminal determines the large-scale fading coefficient between itself and the multiple access terminals, identifies the access terminal with the largest large-scale fading coefficient as the first access terminal, and sends an association request.

[0024] If the number of user terminals served by the first access terminal is less than the maximum number of users that can be served, then the association request sent by the user terminal is received, and the first access terminal is determined as the first target access terminal.

[0025] If the number of user terminals served by the first access terminal is equal to the maximum number of users that can be served and the first access terminal has users that can be unassociated, then if the large-scale fading coefficient between the user terminal and the first access terminal is greater than the smallest large-scale fading coefficient among the users that can be unassociated, the first access terminal is controlled to cancel the association with the user corresponding to the smallest large-scale fading coefficient, and the association request sent by the user terminal is received, and the first access terminal is determined as the first target access terminal.

[0026] If the first access terminal does not have any users that can be unassociated, the user terminal will identify the access terminal with the second largest large-scale fading coefficient as the new first access terminal and send an association request.

[0027] In some embodiments, the user terminal determines the second target access terminal based on its own service demand information, including:

[0028] The user terminal determines the large-scale fading coefficient between itself and multiple access terminals, and determines the access terminal whose large-scale fading coefficient is greater than or equal to its own service demand information as the second access terminal;

[0029] The user terminal sends an association request to all the second access terminals;

[0030] If the number of user terminals served by the second access terminal is less than the maximum number of users that can be served, then the association request sent by the user terminal is received, and the second access terminal is determined as the second target access terminal.

[0031] If the number of user terminals served by the second access terminal is equal to the maximum number of users that can be served and there are users whose association can be canceled at the second access terminal, then if the large-scale fading coefficient between the user terminal and the second access terminal is greater than the smallest large-scale fading coefficient among the users whose association can be canceled, the second access terminal is controlled to cancel the association with the user whose association corresponds to the smallest large-scale fading coefficient, and the association request sent by the user terminal is received, and the second access terminal is determined as the second target access terminal.

[0032] To achieve the above objectives, another aspect of this application proposes a communication method based on a scalable decellularized MIMO system, applied to a user terminal. The communication method based on a scalable decellularized MIMO system includes the following steps:

[0033] Determine the large-scale fading coefficient between itself and multiple access terminals, identify the access terminal with the largest large-scale fading coefficient as the first access terminal, and send an association request;

[0034] Upon receiving a message from the first access terminal allowing association, the first access terminal is identified as the first target access terminal. The first access terminal determines whether to provide association services to the user terminal based on the maximum number of users that can be served, the number of users that can cancel association, and a competition mechanism.

[0035] A user who establishes a service association with the first target access point and is identified by the first target access point as a user whose association cannot be canceled;

[0036] The access point whose large-scale fading coefficient is greater than or equal to its own service demand information is determined as the second access point;

[0037] Send association requests to all second access terminals;

[0038] Upon receiving a message from the second access terminal allowing association, the second access terminal is identified as the second target access terminal. The second access terminal determines whether to provide association services to the user terminal based on the maximum number of users that can be served, the number of users that can cancel association, and a competition mechanism.

[0039] Establish a service association with the second target access point and be identified by the second target access point as a user whose association can be undone;

[0040] The number of the first target access terminals is 1, and the number of the second target access terminals is 1 or more.

[0041] To achieve the above objectives, another aspect of this application proposes a communication method based on a scalable decellularized MIMO system, applied at an access end. The communication method based on a scalable decellularized MIMO system includes the following steps:

[0042] Receive association request sent by user terminal, the association request including the corresponding large-scale fading coefficient;

[0043] The decision on whether to provide association services to the user terminal is determined based on the large-scale fading coefficient, the maximum number of users that can be served, the number of users that can be unassociated, and the competition mechanism.

[0044] Based on the large-scale fading information obtained from associated user terminals, the total transmission power and downlink power allocation are jointly optimized through a power optimization model.

[0045] The downlink transmission power is determined based on the total transmission power and the downlink power allocation.

[0046] The access terminal includes a first target access terminal and / or a second target access terminal. The number of the first target access terminal is one, and the number of the second target access terminals is one or more. The power optimization model is obtained by training based on a convolutional neural network.

[0047] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0048] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0049] The embodiments of this application include at least the following beneficial effects:

[0050] This application provides a communication method, device, and medium based on a scalable decellularized MIMO system. This scheme determines a first target access point by utilizing large-scale fading information between a user terminal and multiple access points. Then, the user terminal establishes a service association with the first target access point, and the user terminal is determined by the first target access point as a user whose association cannot be undone. Based on this, the user terminal determines a second target access point according to its own service needs information. Then, the user terminal establishes a service association with the second target access point, and the user terminal is determined by the second target access point as a user whose association can be undone. The number of first target access points is one, and the number of second target access points is one or more. Thus, the user terminal determines the first target access point using large-scale fading information, enabling... By selecting a service access point based on the macroscopic attenuation characteristics of the signal, the stability of the user's service is ensured. Once the user establishes a service association with the first target access point and is identified as an unbreakable link, the user is guaranteed continuous and stable service from an access point, avoiding the risk of the access point leaving the network midway. Based on this, the user determines a second target access point according to its service requirements. This allows the user to expand its service network to include access points based on its desired service quality, forming its own access point service cluster. The number of second target access points is one or more, ensuring that one or more suitable access points are dynamically selected for each user to provide service, thereby improving the user's service quality.

[0051] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0052] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0053] Figure 1 These are schematic diagrams of communication systems based on scalable decellularized MIMO systems provided in some embodiments of this application;

[0054] Figure 2 This is a flowchart of a communication method based on a scalable decellularized MIMO system provided in some embodiments of this application;

[0055] Figure 3 This is a schematic diagram of the first-stage association system architecture based on a scalable decellularized MIMO system provided in some embodiments of this application;

[0056] Figure 4 This is a schematic diagram of the second-stage association system architecture based on a scalable decellularized MIMO system provided by some embodiments of this application;

[0057] Figure 5 These are schematic diagrams illustrating the communication process of a scalable decellularized MIMO system provided in some embodiments of this application;

[0058] Figure 6 These are schematic block diagrams of a communication device based on a scalable decellularized MIMO system provided in some embodiments of this application;

[0059] Figure 7 This is a schematic block diagram of a communication device based on a scalable decellularized MIMO system provided in some embodiments of this application;

[0060] Figure 8 These are schematic diagrams of the hardware structure of a communication system based on a scalable decellularized MIMO system provided in some embodiments of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the reference to "embodiment" herein means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. The implementation methods described in the following exemplary embodiments do not represent all implementation methods consistent with the embodiments of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0062] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0063] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0065] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0066] To facilitate understanding of the inventive concept of this application, before providing a detailed description of the embodiments of this application, the English abbreviations (terms) / related concepts involved in the embodiments of this application will be explained first. The English abbreviations (terms) / related concepts involved in the embodiments of this application are subject to the following interpretation.

[0067] MIMO (Multiple Input Multiple Output) is a wireless communication technology that improves communication performance and data transmission rates by using multiple antennas at both the transmitting and receiving ends. MIMO technology can transmit multiple data streams on the same frequency band, thereby effectively improving spectral efficiency and the capacity of the communication system.

[0068] Large-scale fading coefficient: In wireless communication, signals attenuate during propagation due to factors such as increased distance and obstruction from obstacles. This attenuation is called path loss, which can be quantified by a coefficient. The large-scale fading coefficient is typically used to describe the average power attenuation of a signal on a macroscopic scale. This coefficient is related to factors such as the distance between the user and the access point, and the environment.

[0069] Access point: In a wireless communication system, an access point (such as a base station, Wi-Fi hotspot, etc.) is a fixed device in the network used to communicate with mobile user equipment.

[0070] Decellularized massive MIMO (Multiple-Input Multiple-Output) systems represent a novel wireless communication architecture. Benefiting from high macro-diversity gain, ideal propagation conditions, and channel hardening, they enable more reliable coverage and provide stable quality of service to users, making them a promising fundamental component of sixth-generation (6G) wireless networks. However, in traditional decellularized MIMO systems, all access points are connected to a single central processing unit, and all access points collectively serve all users. This is impractical because the signaling overhead and computational complexity at the front end increase linearly with the number of users. Furthermore, the dense distribution of access points and users in decellularized MIMO systems makes the system environment highly susceptible to interference. Therefore, effective power allocation is crucial for suppressing interference and improving overall system performance.

[0071] Among related technologies, there is a user association technique that combines a user-centric approach with decellularized massive MIMO systems, where each access point selects a limited number of users to provide services. Additionally, there is a downlink power control method based on fully connected neural networks, where each model is deployed in an edge processor, enabling power control tasks using only locally collected large-scale fading information.

[0072] However, the related communication architecture, because access points only select users with the best channel environment to provide services, some users with poor channel quality may not receive services from the access point. Therefore, it cannot be guaranteed that every user accessing the network can continuously receive services from the access point, which may lead to some users losing connection during communication and posing a risk of users leaving the network midway. In addition, power control using machine learning methods requires a large amount of computing resources for preparing training data and training the model. At the same time, due to network design limitations, the system is unlikely to achieve an optimal solution for downlink power allocation. Furthermore, the single model of this method cannot adapt to scenarios where the number of associated users in the system changes dynamically.

[0073] In view of this, this application proposes a communication method, device, and medium based on a scalable decellularized MIMO system. In the embodiments of this application, a first target access point is determined by large-scale fading information between the user terminal and multiple access points. Then, the user terminal establishes a service association with the first target access point, and the user terminal is determined by the first target access point as a user whose association cannot be canceled. Based on this, the user terminal determines a second target access point according to its own service demand information. Then, the user terminal establishes a service association with the second target access point, and the user terminal is determined by the second target access point as a user whose association can be canceled. The number of first target access points is one, and the number of second target access points is one or more. Thus, the user terminal determines the first target access point through large-scale fading information. The access point can select a service access point based on the macroscopic attenuation characteristics of the signal, thereby ensuring the service stability of the user end. Once the user end establishes a service association with the first target access point and is identified as an unbreakable user by the first target access point, it ensures that the user end receives continuous and stable service from an access point, avoiding the risk of the access point leaving the network midway. Based on this, the user end determines a second target access point according to its own service needs. This allows it to expand its service to other access points based on its required service quality, thus forming its own access point service cluster. That is, the number of second target access points is one or more, ensuring that one or more suitable access points are dynamically selected to provide service to each user end, thereby improving the user's service quality.

[0074] The communication method based on a scalable decellularized MIMO system provided in this application relates to the field of communication technology. It can be applied to the electronic device provided in this application. The electronic device can be a user terminal, an access point, or an edge server.

[0075] In some embodiments, the terminal may be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited thereto.

[0076] Access points can be macro base stations, micro base stations, pico base stations, pico cells, distributed access points, massive MIMO, multi-user MIMO, or beamforming MIMO, but are not limited to these.

[0077] Edge servers can be configured as independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. A server can also be a node server in a blockchain network.

[0078] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0079] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0080] Before introducing the communication method for the scalable decellularized MIMO system proposed in this application, a brief introduction will first be given to the communication system on which the implementation of the communication method proposed in this application depends.

[0081] Please refer to Figure 1 , Figure 1 These are schematic diagrams of communication systems based on scalable decellularized MIMO systems provided in some embodiments of this application. Figure 1 The example shows two users, each with their own service cluster. Each service cluster includes three access points. The access points can provide services to user 1 or user 2 separately, or to both users simultaneously. The access points are associated with edge processors, which execute corresponding computing tasks.

[0082] The implementation steps of a communication method based on a scalable decellularized MIMO system provided in this application will be described in detail below with reference to the accompanying drawings.

[0083] Please refer to Figure 2, Figure 2 The flowcharts provide some embodiments of a communication method based on a scalable decellularized MIMO system. It should be noted that the steps shown in the flowcharts can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than that shown here.

[0084] The method of the embodiments of this application includes the following steps:

[0085] Step 201: The user terminal determines the first target access terminal based on the large-scale fading information between the user terminal and multiple access terminals;

[0086] Step 202: The user terminal establishes a service association with the first target access terminal, and the user terminal is determined by the first target access terminal to be a user whose association cannot be canceled;

[0087] Step 203: The user determines the second target access point based on its own service needs information;

[0088] Step 204: The user terminal establishes a service association with the second target access terminal, and the user terminal is identified by the second target access terminal as a user whose association can be canceled;

[0089] The number of first target access terminals is 1, and the number of second target access terminals is 1 or more.

[0090] Steps 201 to 204, as illustrated in this embodiment, determine the first target access point using large-scale fading information. This allows for the selection of a service access point based on the macroscopic attenuation characteristics of the signal, thereby ensuring the service stability of the user terminal. Based on the user terminal establishing a service association with the first target access point and being identified by the first target access point as a user whose association cannot be canceled, this ensures the user terminal receives continuous and stable service from an access point, avoiding the risk of the access point leaving the network midway. Furthermore, the user terminal determines a second target access point based on its own service requirements. This allows for the expansion of access points to serve the user based on the required service quality, thus forming its own access point service cluster. The number of second target access points is one or more, ensuring that one or more suitable access points are dynamically selected to provide service to each user terminal, thereby improving the user's service quality.

[0091] The specific implementation methods for each of the above steps are described below.

[0092] In step 201, the user terminal determines the first target access terminal based on the large-scale fading information between the user terminal and multiple access terminals.

[0093] Please refer to Figure 3In a real-world system, users (user terminals) periodically transmit synchronization signals. Access points (access terminals) can calculate the large-scale fading coefficient for each user locally based on the received signal strength (it should be understood that the user terminal can also calculate the large-scale fading coefficient for the corresponding access point based on the received broadcast information). Let the number of access points in the communication system be M, and the set of access points be... For the m-th access point, the set of users it serves is defined as follows: in Whether a user is served by an access point can be determined by large-scale fading information. The method proposed in this application can consist of two stages. Here, we first introduce the first stage: In the first stage, at least one access point is allocated to each user in the system to ensure that their basic service needs are met.

[0094] Optionally, large-scale fading information may include, but is not limited to, large-scale fading coefficients (path loss), shadowing fading, penetration loss, or received signal strength.

[0095] For example, the user terminal can measure the large-scale fading coefficient based on the periodically broadcast synchronization signals in the communication system. These periodically broadcast synchronization signals exist in cellular network standards, such as the primary and secondary synchronization signals in 5G. It is understood that the user terminal receives broadcast signals from multiple access points; therefore, the large-scale fading information between the user terminal and these access points can be calculated. Based on this, the user terminal selects the access point m with the largest large-scale fading coefficient. k Send an association request, upon receiving the request from access point m k If the returned message allows association, then access point m k It has been identified as the primary target access point.

[0096] In some implementations, the access point includes users that can be unassociated. The user terminal can determine a first target access point based on large-scale fading information between itself and multiple access points. For example, the user terminal determines the large-scale fading coefficient between itself and multiple access points, identifies the access point with the largest large-scale fading coefficient as the first access point, and sends an association request. If the number of user terminals served by the first access point is less than the maximum number of users that can be served, the user terminal receives the association request and identifies the first access point as the first target access point. If the number of user terminals served by the first access point is equal to the maximum number of users that can be served, and the first access point has users that can be unassociated, then if the large-scale fading coefficient between the user terminal and the first access point is greater than the smallest large-scale fading coefficient among the users that can be unassociated, the user terminal controls the first access point to cancel the association with the user corresponding to the smallest large-scale fading coefficient, receives the association request from the user terminal, and identifies the first access point as the first target access point. If the first access point does not have users that can be unassociated, the user terminal identifies the access point with the second largest large-scale fading coefficient as the new first access point and sends an association request.

[0097] Optionally, in the first phase, for the k-th user newly joining the system, the large-scale fading coefficient β between the user and its neighboring access points is first measured. m，k Subsequently, user k directs to access point m, which has the largest large-scale fading coefficient. k Send an association request, where m k The following calculations were performed:

[0098]

[0099] When access point m k If the number of users being served is less than the maximum number of users that can be served, then the association request from user k is accepted and association permission information is returned to user k; if access point m k The number of users served is equal to the maximum number of users that can be served, and the access point m k When there are users whose associations can be undone, a contention mechanism is triggered to ensure that every new user joining the network receives service. The principle of this mechanism is that the access point undone its association with the user with the smallest large-scale fading coefficient among the undoneable users, and then associated with user k. (It should be understood that the access point compares the large-scale fading coefficient of the requesting user with that of the user with the smallest large-scale fading coefficient; only if the large-scale fading coefficient of the requesting user is greater than the smallest undoneable large-scale fading coefficient in the access point's user association list will the access point choose to associate service with the requesting user). If access point m... k If there are no users whose association can be undeleted, then user k will be associated with any of the neighboring access points except access point m. kThe access point with the largest large-scale fading coefficient sends an association request and repeats the above process.

[0100] This application embodiment determines the first target access point by using large-scale fading information between multiple access points, which can ensure that the user has at least one stable access point to serve them and avoid the risk of the user leaving the network midway. In addition, determining the access point to serve the user based on large-scale fading information can fully protect the user's access channel environment and improve service quality.

[0101] In step 202, the user terminal establishes a service association with the first target access terminal, and the user terminal is determined by the first target access terminal to be a user whose association cannot be canceled.

[0102] For example, after the first target access point is determined, the user terminal can establish a service association with the first target access point, at which time the first target access point provides communication services to the user.

[0103] It should be understood that in the technical solution provided in this application, the users of the access terminal service include two types: users whose service cannot be cancelled and users whose service can be cancelled.

[0104] After a user selects the primary access point, that primary access point will then designate the requesting user as an unremovable user. This ensures that at least one stable access point will provide service to the user and avoids the risk of the user leaving the network midway.

[0105] Figure 3 In this scenario, User 1 has established a service association with one access point, while User 2 has established a service association with another access point. At this point, the service clusters of User 1 and User 2 each include one access point. In this situation, the user client will further expand its service cluster.

[0106] In step 203, the user terminal can determine the second target access terminal based on its own service needs information.

[0107] You can refer to Figure 4 For example, the service cluster can be further expanded based on the service needs of each user to optimize system performance. Each user's service cluster consists of two disjoint subsets. The first subset is determined in the first phase of the method, and the users in this subset cannot be unassociated. The second subset is determined in the second phase of the method, and the users served in this subset can be unassociated. The method assumes that the processing capacity of each access point is limited, meaning that each access point can only serve a limited number of users; therefore, there is a maximum limit to the number of users that can be served.

[0108] It should be understood that there may be multiple access points around the user terminal. By measuring large-scale fading information, multiple access points can be identified as meeting the conditions for providing services. Therefore, the user terminal can identify the access points that meet the conditions for providing services as the second target access points and send association requests to all the second target access points.

[0109] In some implementations, the user terminal first determines the large-scale fading coefficient between itself and multiple access terminals, and identifies access terminals with a large-scale fading coefficient greater than or equal to its own service demand information as second access terminals. The user terminal sends association requests to all second access terminals. If the number of user terminals served by a second access terminal is less than the maximum number of users that can be served, the user terminal receives the association request sent by the user terminal and identifies the second access terminal as the second target access terminal. If the number of user terminals served by a second access terminal is equal to the maximum number of users that can be served and there are users on the second access terminal whose association can be cancelled, then if the large-scale fading coefficient between the user terminal and the second access terminal is greater than the smallest large-scale fading coefficient among the users whose association can be cancelled, the user terminal controls the second access terminal to cancel the association with the user whose association corresponds to the smallest large-scale fading coefficient, and receives the association request sent by the user terminal and identifies the second access terminal as the second target access terminal.

[0110] Optionally, in the second phase, user k will expand its access point service cluster based on its own service needs. User k attempts to select all services that satisfy β. m，k Access point m, which is greater than ε, serves users whose service needs are represented by ε. If the number of users served by access point m has not reached the maximum number of users it can serve, then it accepts the association request from user k. If the number of users served by access point m has reached the maximum number of users it can serve, and there are users whose association can be canceled, then the contention mechanism is triggered.

[0111] For example, the competition mechanism could be that when the large-scale fading coefficient of user k and access point m is greater than that of the user with the smallest large-scale fading coefficient among the users whose association can be cancelled, the access point will cancel its association with the user with the smallest large-scale fading coefficient among the users whose association can be cancelled and establish an association with user k. The second-stage competition mechanism ensures that each access point selects users with better channel conditions to provide services.

[0112] Figure 4In this example, User 1 and User 2 have established associated services with three access points. At this point, the service clusters for both User 1 and User 2 each include three access points, which provide communication services to User 1 and User 2 respectively. It should be understood that in the embodiments provided in this application, there may be access points that provide services to both User 1 and User 2. In this case, having multiple access points serving the user can further improve the quality of service, thereby enhancing the user's communication experience and user satisfaction.

[0113] It should be understood that the first and second phases of the method provided in this application are performed separately, with the first phase performed first and the second phase performed later. The first phase completes the user's initial access to the network, while the second phase is for expanding its service cluster. Furthermore, the selection between the user and the access point is bidirectional; the user will send association requests to all access points with large-scale fading coefficients greater than their service requirements; the access point receiving the association request will decide locally whether to accept the user's association request.

[0114] In this embodiment, the first stage is to complete the initial user access to the system, and the second stage is to further optimize the user's service quality. The operation cycle of the second stage can be determined by the complexity of the method. The time complexity is related to the number of access points M in the communication system and the maximum number of users that can be accessed by each access point. Since the maximum number of users that can be accessed is usually a fixed value, the operation cycle is related to the number of access points M in the system. The more access points there are, the longer the operation cycle will be.

[0115] This application embodiment determines the second target access terminal by using the service demand information of the user terminal itself. It can further expand its service cluster according to the service needs of each user, improve the service quality of users, and thus optimize the overall performance of the communication system.

[0116] This application's embodiments introduce a competition mechanism into the user association method, comprehensively considering user service needs and access point workload, enabling the dynamic selection of the most suitable access point to provide services for each user; and throughout the entire association process, the local terminal only needs to select whether to associate with the requesting user based on its own access user situation, further simplifying the calculation process, reducing signaling overhead, and saving computing resources.

[0117] In step 204, the user terminal establishes a service association with the second target access terminal, and the user terminal is determined by the second target access terminal as a user whose association can be canceled. The number of first target access terminals is 1, and the number of second target access terminals is 1 or more.

[0118] For example, after the second target access point is determined, the user terminal can establish a service association with the second target access point, at which time the second target access point provides communication services to the user.

[0119] There are usually multiple second target access points, such as 2, 3 or 5, and this application does not limit this.

[0120] By establishing service associations with multiple second target access points, the user can expand the range of access points that serve it, which is equivalent to establishing a cluster of access points that serve it, where each access point in the cluster serves the user.

[0121] Multiple access points serving a single user can be achieved, for example, through joint transmission. Alternatively, it can be achieved through cross-site carrier aggregation, cooperative beamforming, or cooperative multipoint access, and this application does not limit this.

[0122] It should be understood that in the technical solution provided in this application, the users of the access terminal service include two types: users whose service cannot be cancelled and users whose service can be cancelled.

[0123] After a user designates an access point as the second target access point, the second target access point will then designate that user as a user whose association can be undeleted. This ensures that more users in the system receive better service, further improving the performance of the communication system.

[0124] The above method allows a user's service cluster to include an access point with the greatest large-scale fading selected for the user in the first phase, and an access point that meets the user's service needs in the second phase. When the user's location is fixed, the cluster in the system is also fixed and does not change dynamically. However, when the user's location moves, the user's wireless channel environment will also change, and therefore the large-scale fading coefficient will also change. Based on this, when the change exceeds a preset threshold, the association operation of the first phase and / or the second phase can be retried. Which phase is triggered can be determined based on the preset threshold. It should be understood that after the association operation of the first phase is retried, the association operation of the second phase will often also be retried, unless a comprehensive evaluation shows that the service effect provided by the user's service cluster can meet the user's current needs.

[0125] In some embodiments of this application, the communication method based on a scalable decellularized MIMO system further includes: the access end jointly optimizing its total transmission power and downlink power allocation through a power optimization model based on the large-scale fading information obtained between the access end and the user end it serves; determining the downlink transmission power of the access end based on the total transmission power and downlink power allocation; wherein the access end includes a first target access end and / or a second target access end, and the power optimization model is obtained by training based on a convolutional neural network.

[0126] Based on the service relationships between access points and users determined by the aforementioned user association algorithm, each edge processor can complete the power allocation task for all users in the system by relying solely on the large-scale fading information it collects locally. Simultaneously, each access point only performs power allocation operations on the users it serves. This minimizes signaling overhead and conserves computational resources. Furthermore, the power control algorithm aims to maximize downlink communication rate, possessing real-time performance and robustness, and can flexibly adapt to system scenarios with varying numbers of associated users.

[0127] Power optimization methods can be based on convolutional neural network design to jointly optimize the total transmission power of the access point and the downlink power allocation strategy. The power optimization model is deployed in an edge processor and aims to maximize the long-term downlink data rate across the entire communication network through learning and optimization, thereby performing efficient power allocation tasks.

[0128] Optionally, the power optimization model includes a stackable feature extraction network based on a residual network and two parallel output networks. The training steps for the power optimization model can be:

[0129] The input tensor is determined by acquiring large-scale fading information, and the dimension of the input tensor is determined by the maximum number of service users of the communication system and the number of access terminals connected to the edge processor.

[0130] Input the input tensor into the feature extraction network to obtain large-scale fading information features;

[0131] Large-scale fading information features are input into a convolutional neural network, and the total transmission power decision tensor and downlink power allocation decision tensor of the access end are obtained through a parallel output network.

[0132] Based on the large-scale fading information features, the total transmission power decision tensor, and the downlink power allocation decision tensor, the model parameters are optimized and training is completed.

[0133] For example, large-scale fading information locally collected in an edge processor can be processed using a stackable feature extraction network based on residual networks. It should be understood that a communication system includes multiple edge processors, and the large-scale fading information collected by the nth edge processor can be modeled as a tensor. Where M n T and T represent the number of access points connected to edge processor n and the number of input samples, respectively. The operation of the feature extraction network is shown in the following equation:

[0134] C = ConvBlock(B n ;{W,b});

[0135] Where ConvBlock represents the feature extraction network, W and b represent the weights and bias parameters in the feature extraction network, respectively, and C represents the output value of the feature extraction network, which can be regarded as the feature of the extracted large-scale fading information. It should be understood that the network proposed in this application is constructed based on the maximum number of users in the system. The maximum number of users K can be flexibly set according to the actual system configuration. Generally, once the system configuration is determined, the maximum number of users is also determined; therefore, the configuration of the neural network in this application is also determined.

[0136] In the power optimization model, the data dimension processed each time is fixed at M. n ×K, where M n K is usually a constant when the network system structure is fixed.

[0137] In practical communication system architectures, this embodiment employs zero-padding operations in convolutional neural networks during computation, enabling the network's input and output to adapt to the dynamically changing number of associated users in the system. This improves the flexibility of distributed downlink power allocation. Therefore, based on this, scenarios with varying numbers of associated users can be efficiently handled with only a single training iteration, meeting the power allocation requirements in dynamic network environments.

[0138] In other words, in the actual model training process, it is not necessary to train for all situations, but only for the largest number of users that the system can accommodate. Such training can handle scenarios with different numbers of associated users, thereby saving data resources. It is not necessary to collect data under various user association scenarios, and the number of training times is reduced, further saving computing resources.

[0139] Based on this, embodiments of this application use two parallel output networks to learn two decision variables, which are used to optimize the total transmit power and downlink power allocation of the access point, respectively. The operation of the output network can be represented as follows:

[0140] Y P =δ(FCB(C); θP )

[0141] Y η =δ(FCB(C); θ η );

[0142] Where FCB represents the output layer; δ is the activation function, which aims to normalize the output of the neural network to meet the power control constraints, so that the total power of each power allocation will not exceed the maximum transmission power of each access point; θ represents the set of weights and biases in the network; C represents the output value of the feature extraction network; the subscripts P and η are used to refer to the two types of parameters in the network, respectively; and Y represents the output tensor of the output network.

[0143] The dimensions of the output tensors are respectively and In other words, the output tensor dimension of the output layer is also designed to be constant. This means that the neural network proposed in this application can effectively address the problem of variable model input and output dimensions caused by changes in the number of associated users in the system. Through this design, the neural network can flexibly adapt to dynamically changing user association scenarios while keeping the input and output dimensions fixed.

[0144] The model parameters are optimized based on large-scale fading information characteristics, total transmission power decision tensor, and downlink power allocation decision tensor.

[0145] For example, a neural network can be trained using an unsupervised learning strategy with the aim of maximizing the total downlink transmission rate of the entire network. The loss function can be expressed as:

[0146]

[0147] In the formula, Θ represents the total parameters learned by the neural network; R k Let k be the downlink transmission rate for user k. During training, a gradient descent strategy is used for parameter updates.

[0148] This completes the training of the power optimization model.

[0149] In some implementations, the downlink transmission power of the access point can be determined by the total transmission power decision tensor, the downlink power allocation decision tensor, and the maximum transmission power of the access point.

[0150] For example, the downlink transmission power at the access point can be expressed as:

[0151] P = diag(P) max Y P )·Y η ;

[0152] In the above formula, P maxRepresenting the maximum transmission power of each access point, diag(·) can generate a diagonal matrix with the input vector as its diagonal elements; the power control task is completed by combining the maximum transmission power of each access point with the decision variables of downlink power allocation.

[0153] This application's embodiments employ a distributed computing approach, relying solely on information collected locally at the access point to complete operations without centralized processing, simplifying the computation process. It collaboratively optimizes the total transmission power and downlink power allocation of the access point, obtaining the optimal solution to the power control problem and improving network performance. Furthermore, by employing a neural network supporting variable input / output dimensions (through targeted design of the neural network's input and output layers, a single model can flexibly adapt to scenarios with different numbers of associated users), it requires only a single training iteration to adapt to scenarios with varying numbers of associated users, significantly enhancing the system's versatility and scalability.

[0154] This application also proposes a communication method based on a scalable decellularized MIMO system for use at the user end.

[0155] Communication methods based on scalable decellularized MIMO systems applied at the user end include:

[0156] First, determine the large-scale fading coefficient between itself and multiple access points, identify the access point with the largest large-scale fading coefficient as the first access point, and send an association request.

[0157] When the user receives a message from the first access terminal accepting association, the first access terminal is identified as the first target access terminal. The first access terminal can determine whether to provide association services to the user based on the maximum number of users that can be served, the number of users that can cancel association, and the competition mechanism (as illustrated in the above embodiment, it may include the triggering operation of the competition mechanism).

[0158] Subsequently, the user client establishes a service association with the first target access point and is identified by the first target access point as a user whose association cannot be canceled;

[0159] The access point with a large-scale fading coefficient greater than or equal to its own service demand information is identified as the second access point.

[0160] Send association requests to all second access points;

[0161] Upon receiving a message from the second access terminal allowing association, the second access terminal is identified as the second target access terminal. The second access terminal determines whether to provide association services to the user terminal based on the maximum number of users that can be served, the number of users that can cancel association, and the competition mechanism (as illustrated in the above embodiments, this may include triggering the competition mechanism).

[0162] Establish a service association with the second target access point and be identified by the second target access point as a user whose association can be undone;

[0163] The number of first target access terminals is 1, and the number of second target access terminals is 1 or more.

[0164] For example, when a user joins a new communication network, the user first measures the large-scale fading information between themselves and surrounding access points, and then sends an association request to the access point with the largest large-scale fading information. After confirming the association, the user further expands the number of access points they serve, that is, they send association requests to multiple sites that meet their service needs. If the user receives an agreement to associate, multiple access points will provide communication services to the user, thereby improving the user's service quality.

[0165] The embodiments provided in this application construct a service cluster consisting of a limited number of access points for each user through a user association method, thereby effectively reducing the system's fronthaul requirements and computational complexity, and constructing a scalable decellularized system.

[0166] This application also proposes a communication method based on a scalable decellularized MIMO system for use at the access end.

[0167] Communication methods based on scalable decellularized MIMO systems applied at the access end include:

[0168] First, receive association requests sent by the user, which include the corresponding large-scale fading coefficients;

[0169] The decision to provide association services to users is determined based on the large-scale fading coefficient, the maximum number of users that can be served, the number of users that can be unlinked, and the competition mechanism.

[0170] Based on the large-scale fading information obtained from associated user terminals, the total transmission power and downlink power allocation are jointly optimized through a power optimization model.

[0171] The downlink transmission power is determined based on the total transmission power and downlink power allocation.

[0172] The access point includes a first target access point and / or a second target access point. There is one first target access point and one or more second target access points. The power optimization model is obtained by training a convolutional neural network.

[0173] In practical communication systems, the access point calculates the large-scale fading coefficient for each user locally based on the received signal strength. The users served by the access point can be divided into two types: those for whom service cannot be cancelled and those for whom service can be cancelled.

[0174] When an access point receives an association request from a user, it first checks whether the number of users it serves has reached the maximum number of users. If the maximum number of users has not yet been reached, the access point accepts the user's association request. If this is the user's first association request after accessing the network, the access point classifies the user as a non-cancellable service type. If this is not the user's first association request after accessing the network, the access point classifies the user as a cancelable service type.

[0175] When an access point receives an association request from a user, it first checks if the number of users it serves has reached the maximum. If the maximum number of users has been reached and the access point has users whose associations can be cancelled, a contention mechanism will be initiated or triggered. When the large-scale fading information (which can be the large-scale fading coefficient) between the requesting user and the access point is greater than that of the user with the smallest large-scale fading coefficient among the users whose associations can be cancelled, the access point will cancel the association with the user with the smallest large-scale fading coefficient among the users whose associations can be cancelled and establish an association with the requesting user. At this time, this contention mechanism can ensure that each access point will select users with better channel conditions to provide services. If this is the user's first association request after accessing the network, the access point classifies the user into the non-cancellable service type; if this is not the user's first association request after accessing the network, the access point classifies the user into the cancelable service type.

[0176] When an access point receives an association request from a user, it first checks whether the number of users it serves has reached the maximum number of users. If the maximum number of users has been reached and the access point does not have any users that can be unassociated, then the access point will reject the user's association request.

[0177] Access points can also jointly optimize their total transmission power and downlink power allocation using a power optimization model based on the large-scale fading information obtained from associated user terminals. It should be understood that, to conserve and balance computing resources, power optimization is typically performed by edge processors on the access point side to achieve distributed computing.

[0178] Each edge processor can complete the power allocation task for the access terminals it connects to, relying solely on the large-scale fading information it collects locally. Simultaneously, each access terminal only performs power allocation operations for the users it serves. The power control algorithm optimizes downlink communication rate, possesses real-time performance and robustness, and can flexibly adapt to system scenarios with varying numbers of connected users.

[0179] Power control is mainly achieved through a power optimization model. The training method for the power optimization model can be found above and will not be repeated here.

[0180] After obtaining the total transmission power and downlink power allocation output by the power optimization model, the downlink transmission power can be determined by combining the maximum transmission power of each access point. The specific determination method can be referred to the above, and will not be repeated here.

[0181] The embodiments provided in this application innovatively introduce a competition mechanism and user service needs into the user association method, while combining the workload constraints of the access point, comprehensively analyzing user needs and access point resource conditions, selecting the most suitable access point for each user, and building an efficient service cluster; through a power optimization model, the total transmission power and downlink power allocation of the access point are collaboratively optimized to achieve targeted power control; the trained power optimization model can be based entirely on locally collected large-scale fading information without additional communication interaction, and through localized processing, distributed nodes can independently complete joint user association and power allocation operations.

[0182] The solutions of the embodiments of the present invention will be described in detail and explained below with reference to specific application examples:

[0183] Please refer to Figure 5 , Figure 5 These are schematic diagrams illustrating the communication process of a scalable decellularized MIMO system provided in some embodiments of this application. Figure 5 In the context of user k newly joining the communication network, it first sends an association request to the nearest access point with the greatest fading information. The access point determines whether to provide association service to the requesting user based on its own user base (whether the number of users served has reached the maximum service limit, whether there are users whose service can be cancelled, etc.). If an access point provides service to the requesting user, the edge processor located on the access point's side will perform power control calculations for that access point. The access point then controls its downlink transmit power according to the power control strategy and performs channel estimation and coding to provide communication service to the requesting user. It should be understood that... Figure 5 The arrows in the diagram only indicate the direction of a single communication session and do not indicate the continuous communication flow between the two ends. In other words, even... Figure 5 The communication flow is clearly defined, but signals can also be transmitted in the completely opposite direction, depending on the sender of the signal in the actual communication process.

[0184] The implementation of the communication device based on a scalable decellularized MIMO system provided in this application will now be described in detail with reference to the accompanying drawings.

[0185] In addition to the communication method based on a scalable decellularized MIMO system provided in the above embodiments, this application also provides a communication device (which may be a user terminal) based on a scalable decellularized MIMO system for implementing the above method, such as... Figure 6As shown, Figure 6 This is a schematic block diagram of a communication device based on a scalable decellularized MIMO system according to an embodiment of this application. The communication device based on a scalable decellularized MIMO system includes:

[0186] The first determining module is used to determine the first target access point based on the large-scale fading information between the access points and multiple access points;

[0187] The first service association establishment module is used to establish a service association with the first target access terminal, and the user terminal is determined by the first target access terminal to be a user whose association cannot be canceled.

[0188] The second determination module is used to determine the second target access point based on its own service requirement information;

[0189] The second service association establishment module is used to establish a service association with the second target access terminal, and the user terminal is determined by the second target access terminal as a user whose association can be canceled.

[0190] The number of first target access terminals is 1, and the number of second target access terminals is 1 or more.

[0191] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0192] In addition to the communication method based on a scalable decellularized MIMO system provided in the above embodiments, this application also provides another communication device (which may be an access end) based on a scalable decellularized MIMO system for implementing the above method, such as... Figure 7 As shown, Figure 7 This is a schematic block diagram of another communication device based on a scalable decellularized MIMO system, according to an embodiment of this application. The other communication device based on a scalable decellularized MIMO system includes:

[0193] The receiving module is used to receive association requests sent by the user terminal. The association request includes the corresponding large-scale fading coefficient.

[0194] The determination module is used to determine whether to provide associated services to users based on the large-scale fading coefficient, the maximum number of users that can be served, the number of users that can be unassociated, and the competition mechanism.

[0195] The joint optimization module is used to jointly optimize its total transmission power and downlink power allocation based on the large-scale fading information obtained from associated user terminals and through a power optimization model.

[0196] The power control module is used to determine the downlink transmission power based on the total transmission power and downlink power allocation.

[0197] The access point includes a first target access point and / or a second target access point. There is one first target access point and one or more second target access points. The power optimization model is obtained by training a convolutional neural network.

[0198] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0199] Reference Figure 8 This application embodiment also provides an electronic device 800, which includes a memory 801 and one or more processors 802. Figure 8 Only one is shown in the diagram, along with a computer program stored in the memory and executable on the processor. The memory 801 stores the software program and its components. The processor 802 executes various functional applications and data processing by running the software program and its components stored in the memory to obtain resources corresponding to the aforementioned preset events. Optionally, the processor implements the aforementioned communication method based on a scalable decellularized MIMO system by running the aforementioned computer program stored in the memory.

[0200] Memory, as a non-transitory computer-readable medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network.

[0201] It is understood that the content of the above method embodiments is applicable to the embodiments of this electronic device. The specific functions implemented by the embodiments of this electronic device are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0202] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described communication method based on a scalable decellularized MIMO system.

[0203] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0204] This application also provides a computer program product, which includes a computer program that, when executed by one or more processors, can implement the steps of the communication method based on a scalable decellularized MIMO system described above.

[0205] It is understood that the content of the above method embodiments is applicable to this computer program product. The specific functions implemented by the embodiments of this computer program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0206] The communication method, apparatus, electronic device, medium, and computer program product based on a scalable decellularized MIMO system provided in this application determines a first target access point by using large-scale fading information between a user terminal and multiple access points. Then, the user terminal establishes a service association with the first target access point, and the user terminal is determined by the first target access point as a user whose association cannot be canceled. Based on this, the user terminal determines a second target access point according to its own service demand information. Then, the user terminal establishes a service association with the second target access point, and the user terminal is determined by the second target access point as a user whose association can be canceled. The number of first target access points is one, and the number of second target access points is one or more. Thus, the user terminal determines the first target access point through large-scale fading information. The access point can select a service access point based on the macroscopic attenuation characteristics of the signal, thereby ensuring the service stability of the user end. Once the user end establishes a service association with the first target access point and is identified as an unbreakable user by the first target access point, it ensures that the user end receives continuous and stable service from an access point, avoiding the risk of the access point leaving the network midway. Based on this, the user end determines a second target access point according to its own service needs. This allows it to expand its service to other access points based on its required service quality, thus forming its own access point service cluster. That is, the number of second target access points is one or more, ensuring that one or more suitable access points are dynamically selected to provide service to each user end, thereby improving the user's service quality.

[0207] The communication method, apparatus, electronic device, medium, and computer program product based on a scalable decellularized MIMO system provided in this application proposes a two-stage user association strategy by introducing a competition mechanism and comprehensively considering the access point workload and user service needs. In the first stage, a preferred access point is allocated to each user based on channel quality to ensure basic service. In the second stage, the service cluster is expanded based on the user's service needs, and the competition mechanism is used to dynamically select the most suitable access point for each user in both stages, ultimately building an efficient service cluster. This method effectively improves the user's service quality and the system's resource utilization efficiency. Furthermore, existing downlink power control methods fail to coordinate the optimization of the total transmission power of the access point and downlink power allocation, and cannot adapt to dynamic changes in the number of associated users in the system. Therefore, this application proposes a downlink power control algorithm based on a convolutional neural network. By designing two parallel output networks, this algorithm can simultaneously optimize the total transmission power of the access point and the downlink power allocation. More importantly, this invention innovatively designs a neural network structure with fixed input and output layer dimensions, enabling a single neural network to flexibly adapt to system scenarios with different numbers of associated users with only one training iteration. Furthermore, the method proposed in this application is entirely based on locally collected large-scale fading information, requiring no additional communication interaction. Through localized processing, distributed nodes can independently complete joint user association and power allocation operations.

[0208] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0209] Although specific embodiments are described herein, those skilled in the art will recognize that many other modifications or alternative embodiments are also within the scope of this disclosure. For example, any of the functions and / or processing capabilities described in connection with a particular device or component can be performed by any other device or component. Furthermore, while various exemplary embodiments and architectures have been described according to embodiments of this disclosure, those skilled in the art will recognize that many other modifications to the exemplary embodiments and architectures described herein are also within the scope of this disclosure.

[0210] The foregoing description, with reference to block diagrams and flowcharts of systems, methods, systems, and / or computer program products according to exemplary embodiments, has described certain aspects of this disclosure. It should be understood that one or more blocks in the block diagrams and flowcharts, as well as combinations of blocks in the block diagrams and flowcharts, can be implemented by executing computer-executable program instructions, respectively. Similarly, according to some embodiments, some blocks in the block diagrams and flowcharts may not need to be executed in the order shown, or may not all need to be executed. Furthermore, additional components and / or operations beyond those shown in the blocks in the block diagrams and flowcharts may exist in some embodiments.

[0211] Therefore, blocks in block diagrams and flowcharts support combinations of means for performing a specified function, combinations of elements or steps for performing a specified function, and program instruction means for performing a specified function. It should also be understood that each block in a block diagram and flowchart, and combinations of blocks in block diagrams and flowcharts, can be implemented by a dedicated hardware computer system or a combination of dedicated hardware and computer instructions that performs a specific function, element, or step.

[0212] The program modules, applications, etc., described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, in response to execution, cause at least a portion of the functionality described herein (e.g., one or more operations of the exemplary methods described herein) to be performed.

[0213] Software components can be coded using any of a variety of programming languages. One exemplary programming language could be a low-level programming language, such as assembly language associated with a specific hardware architecture and / or operating system platform. Software components including assembly language instructions may need to be converted into executable machine code by an assembler before being executed by the hardware architecture and / or platform. Another exemplary programming language could be a higher-level programming language that is portable across multiple architectures. Software components including higher-level programming languages ​​may need to be converted into an intermediate representation by an interpreter or compiler before execution. Other examples of programming languages ​​include, but are not limited to, macro languages, shell or command languages, job control languages, scripting languages, database query or search languages, or report writing languages. In one or more exemplary embodiments, a software component containing instructions from one of the above programming language examples can be executed directly by the operating system or other software components without first being converted into another form.

[0214] Software components can be stored as files or other data storage structures. Software components of similar type or related function can be stored together in a specific directory, folder, or library. Software components can be static (e.g., pre-defined or fixed) or dynamic (e.g., created or modified at runtime).

[0215] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.

Claims

1. A communication method based on a scalable decellularized MIMO system, characterized in that, Includes the following steps: The user terminal determines the first target access point based on the large-scale fading information between it and multiple access points; The user terminal establishes a service association with the first target access terminal, and the user terminal is determined by the first target access terminal to be a user whose association cannot be canceled. The user terminal first determines the large-scale fading coefficient between itself and multiple access terminals, and determines the access terminals whose large-scale fading coefficient is greater than or equal to its own service demand information as the second target access terminal. The user terminal establishes a service association with the second target access terminal, and the user terminal is determined by the second target access terminal to be a user whose association can be canceled. The number of the first target access terminals is 1, and the number of the second target access terminals is 1 or more.

2. The communication method based on a scalable decellularized MIMO system according to claim 1, characterized in that, The method further includes: Based on the large-scale fading information obtained between the access end and the user end it serves, the access end jointly optimizes its total transmission power and downlink power allocation through a power optimization model. The downlink transmission power of the access point is determined based on the total transmission power and the downlink power allocation; The access terminal includes the first target access terminal and / or the second target access terminal, and the power optimization model is obtained by training based on a convolutional neural network.

3. The communication method based on a scalable decellularized MIMO system according to claim 2, characterized in that, The power optimization model includes a stackable feature extraction network based on a residual network and two parallel output networks. The training steps of the power optimization model include: The input tensor is determined by acquiring the large-scale fading information, and the dimension of the input tensor is determined by the maximum number of service users of the communication system and the number of access terminals connected to the edge processor. The input tensor is input into the feature extraction network to obtain large-scale fading information features; The large-scale fading information features are input into the convolutional neural network, and the total transmission power decision tensor and downlink power allocation decision tensor of the access end are obtained through the parallel output network. Based on the large-scale fading information features, the total transmission power decision tensor, and the downlink power allocation decision tensor, the model parameters are optimized and training is completed.

4. The communication method based on a scalable decellularized MIMO system according to claim 3, characterized in that, The step of determining the downlink transmission power of the access terminal based on the total transmission power and the downlink power allocation includes: The downlink transmission power of the access point is determined by the total transmission power decision tensor, the downlink power allocation decision tensor, and the maximum transmission power of the access point.

5. The communication method based on a scalable decellularized MIMO system according to claim 1, characterized in that, The access point includes users that can be unassociated. The user terminal determines the first target access point based on large-scale fading information between itself and multiple access points, including: The user terminal determines the large-scale fading coefficient between itself and the multiple access terminals, identifies the access terminal with the largest large-scale fading coefficient as the first access terminal, and sends an association request. If the number of user terminals served by the first access terminal is less than the maximum number of users that can be served, then the association request sent by the user terminal is received, and the first access terminal is determined as the first target access terminal. If the number of user terminals served by the first access terminal is equal to the maximum number of users that can be served and the first access terminal has users that can be unassociated, then if the large-scale fading coefficient between the user terminal and the first access terminal is greater than the smallest large-scale fading coefficient among the users that can be unassociated, the first access terminal is controlled to cancel the association with the user corresponding to the smallest large-scale fading coefficient, and the association request sent by the user terminal is received, and the first access terminal is determined as the first target access terminal. If the first access terminal does not have any users that can be unassociated, the user terminal will identify the access terminal with the second largest large-scale fading coefficient as the new first access terminal and send an association request.

6. The communication method based on a scalable decellularized MIMO system according to claim 1, characterized in that, The user terminal determines the second target access terminal based on its own service needs information, including: The user terminal determines the large-scale fading coefficient between itself and multiple access terminals, and determines the access terminal whose large-scale fading coefficient is greater than or equal to its own service demand information as the second access terminal; The user terminal sends an association request to all the second access terminals; If the number of user terminals served by the second access terminal is less than the maximum number of users that can be served, then the association request sent by the user terminal is received, and the second access terminal is determined as the second target access terminal. If the number of user terminals served by the second access terminal is equal to the maximum number of users that can be served and there are users whose association can be canceled at the second access terminal, then if the large-scale fading coefficient between the user terminal and the second access terminal is greater than the smallest large-scale fading coefficient among the users whose association can be canceled, the second access terminal is controlled to cancel the association with the user whose association corresponds to the smallest large-scale fading coefficient, and the association request sent by the user terminal is received, and the second access terminal is determined as the second target access terminal.

7. A communication method based on a scalable decellularized MIMO system, characterized in that, When applied to the user end, the communication method based on a scalable decellularized MIMO system includes the following steps: Determine the large-scale fading coefficient between itself and multiple access terminals, identify the access terminal with the largest large-scale fading coefficient as the first access terminal, and send an association request; Upon receiving a message from the first access terminal allowing association, the first access terminal is identified as the first target access terminal. The first access terminal determines whether to provide association services to the user terminal based on the maximum number of users that can be served, the number of users that can cancel association, and a competition mechanism. A user who establishes a service association with the first target access point and is identified by the first target access point as a user whose association cannot be canceled; The access point whose large-scale fading coefficient is greater than or equal to its own service demand information is determined as the second access point; Send association requests to all second access terminals; Upon receiving a message from the second access terminal allowing association, the second access terminal is identified as the second target access terminal. The second access terminal determines whether to provide association services to the user terminal based on the maximum number of users that can be served, the number of users that can cancel association, and a competition mechanism. Establish a service association with the second target access point and be identified by the second target access point as a user whose association can be undone; The number of the first target access terminals is 1, and the number of the second target access terminals is 1 or more.

8. A communication method based on a scalable decellularized MIMO system, characterized in that, Applied to the access end, the communication method based on a scalable decellularized MIMO system includes the following steps: Receive association request sent by user terminal, the association request including the corresponding large-scale fading coefficient; The decision on whether to provide association services to the user terminal is determined based on the large-scale fading coefficient, the maximum number of users that can be served, the number of users that can be unassociated, and the competition mechanism. Based on the large-scale fading information obtained from associated user terminals, the total transmission power and downlink power allocation are jointly optimized through a power optimization model. The downlink transmission power is determined based on the total transmission power and the downlink power allocation. The access terminal includes a first target access terminal and / or a second target access terminal. The number of the first target access terminal is one, and the number of the second target access terminals is one or more. The power optimization model is obtained by training based on a convolutional neural network.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the communication method based on a scalable decellularized MIMO system as described in any one of claims 1 to 6, 7 or 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the communication method based on a scalable decellularized MIMO system as described in any one of claims 1 to 6, 7 or 8.

Images