A cross-modal transmission resource scheduling method and system based on a cell-free MIMO network

Abstract

The present application relates to the technical field of wireless communication, specifically to a cross-modal transmission resource scheduling method and system based on a cell-free MIMO network, comprising: considering traditional continuous virtual reality audio-visual stream and intermittent generative artificial intelligence content stream for cross-modal transmission task; proposing a user conformity behavior model. According to the user conformity behavior probability, the total data amount required for user transmission is calculated, including the continuous virtual reality audio-visual stream data amount and the intermittent AIGC stream data amount driven by the conformity behavior; a resource scheduling optimization mathematical model is constructed to minimize the maximum immersive content transmission delay of the user, and a distributed power allocation algorithm based on the consensus alternating direction multiplier method is designed. The present application comprehensively considers the driving effect of user conformity behavior on AIGC service request, effectively reduces the maximum content transmission delay of the system through the distributed power allocation algorithm, and improves the fairness of multi-user immersive service.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, specifically to a method and system for cross-modal transmission resource scheduling based on a cellular-free MIMO network. Background Technology

[0002] With the development of sixth-generation (6G) communication technology, immersive communication has been identified as one of the typical scenarios in IMT-2030. Immersive communication provides users with a multi-sensory, low-latency, and high-fidelity interactive experience through a service format that blends virtual and reality. In this context, the integration of virtual reality audiovisual streaming and artificial intelligence-generated content (AIGC) has become a key technology for enhancing user immersion. AIGC technology can generate personalized virtual objects, virtual environments, and other visual content based on user prompts such as text and voice, combining with traditional audiovisual streaming to improve the user's immersive experience.

[0003] However, the introduction of AIGC services brings new challenges to wireless transmission. On the one hand, AIGC content generation is characterized by its suddenness and uncertainty, and its traffic characteristics differ significantly from traditional continuous audio-visual streams. On the other hand, user behavior in virtual spaces is often influenced by the group; when most users begin requesting AIGC services, other users may exhibit herd behavior, leading to a surge in AIGC requests within a short period, resulting in intermittent high-load traffic. This intermittent traffic characteristic driven by user behavior places higher demands on the resource scheduling capabilities of wireless networks. Cell-free Massive MIMO (MMIMO) networks are considered a potential technical solution to support immersive communication. This network serves all users collaboratively through a large number of distributed access points (APs), eliminating the cell boundary effect of traditional cellular networks and providing users with uniform quality of service. However, the distributed architecture of cell-free MIMO networks also brings high complexity to resource scheduling. How to comprehensively consider the heterogeneous transmission needs of traditional continuous audio-visual streams and intermittent AIGC streams in cell-free MIMO networks, and cope with the dynamic load changes caused by user herd behavior to achieve efficient resource scheduling remains an unsolved technical problem. Summary of the Invention

[0004] The purpose of this invention is to provide a cross-modal transmission resource scheduling method and system based on non-cellular MIMO networks. By jointly considering the transmission of continuous audio-visual streams and intermittent AIGC streams, a herd behavior model for user AIGC service requests is established, and a distributed power resource scheduling method for non-cellular MIMO networks is proposed to minimize the maximum immersive content transmission latency for all users and improve the fairness of multi-user immersive services.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A cross-modal transmission resource scheduling method based on a cellular-free MIMO network includes the following steps:

[0007] S1. Based on the user's cross-modal stream transmission task, construct a herd behavior model for the user's AIGC service request, and calculate the user's transmission rate and the total amount of cross-modal stream data that needs to be transmitted.

[0008] S2. Based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, establish an optimization mathematical model for resource scheduling, and solve the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints.

[0009] S3. Allocate power resources to multiple users based on the amount of resources required by each user.

[0010] Preferably, step S1 specifically includes:

[0011] S11. Construct a herding behavior model for user AIGC service requests and calculate the probability that a user becomes an AIGC service user:

[0012] For AIGC streams, the user's AIGC service adoption process is divided into a series of conformity intervals. Each conformity interval contains several transmission slots. The duration of the conformity interval is expressed as It depends on the progress of service diffusion; the duration of the transmission slot is expressed as Users decide whether to follow the crowd in each herding interval and schedule transmission resources in each transmission time slot. Then the user... In the The probability that someone is an AIGC service user within a certain conformity interval is: ;

[0013] in, Indicates user Total number of influencers The table shows the conformity threshold; if the number of influencers using AIGC reaches the conformity threshold, users will exhibit conformity behavior. This indicates the number of influencers who have become users;

[0014] S12. Establish transmission models for AIGC streams and audiovisual streams, and calculate user transmission rates and the total amount of cross-modal stream data to be transmitted:

[0015] During the uplink channel estimation phase, the access point is set. and users The channel between them is ;in This represents the small-scale fading coefficient. Represents the large-scale fading coefficient; sets the user... The pilot sequence used in the channel estimation stage is: Then, based on the minimum mean square error channel estimation, Channel estimation Represented as:

[0016] ;

[0017] ;

[0018] in, Indicates the uplink training length. Indicates user Uplink transmission power, Indicates to users Set of users using the same pilot sequence It is the noise variance of the uplink;

[0019] against ,have The variance Represented as:

[0020] ;

[0021] During the downlink cross-modal transmission phase, both traditional continuous audio-visual stream transmission and intermittent AIGC stream transmission are considered. If a user requests AIGC service, the central processing unit will generate cross-modal content based on the AIGC model and user prompts, and transmit the generated content along with the traditional audio-visual stream to the AP via the fronthaul link. Subsequently, the AP constructs a precoding vector based on locally estimated channel information and transmits the multimodal stream received from the CU to the user. If the user does not request AIGC service, the CU only transmits the traditional audio-visual stream to the AP, and then the AP transmits the received audio-visual stream to the user.

[0022] user The signal-to-interference-plus-noise ratio is expressed as: ;

[0023] in, Indicate each The number of antennas equipped, Indicates AP For users downlink transmit power, This represents the variance of downlink noise at the user end;

[0024] AP Service users In the time slot The transmission rate obtained within is expressed as: ;

[0025] in, It is the bandwidth of the spectrum resource block;

[0026] user In transmission time slot The size of the requested cross-modal flow is expressed as: ;

[0027] in, Indicates user In transmission time slot The size of the audio-visual stream, Indicates user In transmission time slot AIGC stream size and satisfy ,in For users The size of the image generated by each AIGC request. This represents the number of AIGC service requests made by each user within each conformity interval. Indicates the duration of the conformity interval. Indicates the duration of the transmission time slot.

[0028] Preferably, step S2 specifically includes:

[0029] S21. Calculate the user's immersive content transmission latency, take the power resource allocation strategy under non-cellular MIMO network as the optimization object, take minimizing the user's maximum immersive content transmission latency as the objective function, and take the user's audio-visual stream transmission latency limit and power resource limit as the constraint objects to construct a resource scheduling model for cross-modal stream transmission.

[0030] S22. Solve the resource scheduling model to obtain the number of resources that minimize the maximum immersive content transmission latency for users under the constraints of latency and power resources.

[0031] Preferably, step S21 specifically includes:

[0032] S211, According to the user In transmission time slot transmission rate and the total amount of cross-modal streaming data requested The calculated latency for immersive content transmission is: ;

[0033] S212. Establish an immersive cross-modal transmission resource scheduling model representation for non-cellular MIMO networks to address the optimization problem: ;

[0034] ; ; ; ;

[0035] in, Describes the set of APs. Indicates the number of APs. Represents a set of users. This indicates the number of users.

[0036] Preferably, step S22 specifically includes: first, transforming the original optimization problem P1 established in S212 into an equivalent form that is convenient for mathematical processing; then, for each access point, constructing local variables and global variables, further transforming the problem into a distributed consensus form; and finally, obtaining a stable power resource allocation strategy based on the distributed consensus ADMM algorithm.

[0037] The present invention also provides a cross-modal transmission resource scheduling system, comprising:

[0038] The transmission task acquisition module is used to acquire the user transmission rate and the total amount of cross-modal stream data to be transmitted for resource scheduling tasks based on non-cellular MIMO immersive cross-modal transmission.

[0039] The transmission strategy optimization module is used to establish an optimization mathematical model for resource scheduling based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, and solve the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints.

[0040] The transmission task allocation module is used to allocate power resources to multiple users based on the amount of resources required by each user.

[0041] The present invention also provides a non-transitory computer-readable storage medium storing computer instructions thereon, which cause a computer to execute the above-described method for cross-modal transmission resource scheduling based on a non-cellular MIMO network.

[0042] The present invention also provides an electronic device, including a processor, a communication interface, a memory, and a communication bus;

[0043] The processor, communication interface, and memory communicate with each other through a communication bus.

[0044] The processor is used to call logical instructions in memory to execute the aforementioned cross-modal transmission resource scheduling method based on a non-cellular MIMO network.

[0045] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0046] This invention comprehensively considers the needs of traditional continuous audiovisual streaming and intermittent AIGC streaming. By distributing downlink power resources in non-cellular MIMO networks, it effectively reduces the maximum immersive content transmission latency for users, enhancing their immersive experience in virtual-real converged services. Specifically, this invention aims to minimize the maximum immersive content transmission latency for users, using user audiovisual streaming latency constraints and power resource constraints as constraints to construct a resource scheduling model for cross-modal streaming. A distributed power allocation algorithm is designed based on the Alternating Directional Multiplier Method (ADMM). By introducing local and global variables to decouple signals from interference, each access point independently solves sub-problems locally and iteratively reaches consensus, obtaining the optimal power allocation strategy that minimizes the maximum transmission latency under power constraints. Utilizing the obtained resource allocation strategy, the maximum system transmission latency can be effectively reduced, improving the fairness of multi-user immersive services. Attached Figure Description

[0047] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0048] Figure 1 This is a schematic diagram of the immersive cross-modal downlink transmission system of the non-cellular MIMO network of the present invention;

[0049] Figure 2 This is a flowchart of a cross-modal transmission resource scheduling method based on a non-cellular MIMO network according to the present invention.

[0050] Figure 3 This is a schematic diagram of the relationship between user conformity interval and transmission time slot provided in Embodiment 1 of the present invention;

[0051] Figure 4 This is a graph showing the relationship between the maximum immersive content transmission latency and the number of user AIGC service requests under different schemes provided in Embodiment 4 of the present invention.

[0052] Figure 5 This is a graph showing the relationship between the maximum immersive content transmission latency and the number of users under different schemes provided in Embodiment 4 of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figures 1-5 The present invention provides the following technical solution:

[0055] Example 1: This example provides a cross-modal transmission resource scheduling method based on a cellular-free MIMO network. This method is used in applications such as... Figure 1 The transmission system shown here is a transmission task, which includes a user layer and an infrastructure layer.

[0056] The user layer consists of several users waiting for service. Users request multimodal content, including audio-visual streams and AIGC streams, from the access point via the wireless uplink.

[0057] The infrastructure layer consists of several access points and a central processing unit that provide users with audio-visual streams and cross-modal content generation. The central processing unit generates content based on the user's AIGC service request and transmits the audio-visual stream and AIGC stream to the access point through the fronthaul link. The access point then sends the audio-visual stream and AIGC stream to the user through the wireless downlink.

[0058] like Figure 2 As shown, a cross-modal transmission resource scheduling method based on a cellular-free MIMO network includes the following steps:

[0059] S1. For user cross-modal stream transmission tasks, construct a herd behavior model for user AIGC service requests to obtain user transmission rate and the total amount of cross-modal stream data to be transmitted.

[0060] S2. Based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, establish an optimization mathematical model for resource scheduling, and solve the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints.

[0061] S3. Allocate power resources to multiple users based on the amount of resources required by each user.

[0062] Preferably, step S1 specifically includes:

[0063] like Figure 3 As shown, the relationship between user conformity intervals and transmission time slots is illustrated, where each conformity interval contains multiple transmission time slots.

[0064] S11. Construct a herding behavior model for user AIGC service requests and calculate the probability that a user becomes an AIGC service user:

[0065] For AIGC streams, the user's AIGC service adoption process is divided into a series of conformity intervals. Each conformity interval contains several transmission slots. The duration of the conformity interval is expressed as It depends on the progress of service diffusion; the duration of the transmission slot is expressed as Users decide whether to follow the crowd in each herding interval and schedule transmission resources in each transmission time slot. Then the user... In the The probability that someone is an AIGC service user within a certain conformity interval is: ;

[0066] in, Indicates user Total number of influencers The table shows the conformity threshold; if the number of influencers using AIGC reaches the conformity threshold, users will exhibit conformity behavior. This indicates the number of influencers who have become users;

[0067] S12. Establish transmission models for AIGC streams and audiovisual streams, and calculate user transmission rates and the total amount of cross-modal stream data to be transmitted:

[0068] During the uplink channel estimation phase, the access point is set. and users The channel between them is ;in This represents the small-scale fading coefficient. Represents the large-scale fading coefficient; sets the user... The pilot sequence used in the channel estimation stage is: Then, based on the minimum mean square error channel estimation, Channel estimation Represented as:

[0069] ;

[0070] ;

[0071] in, Indicates the uplink training length. Indicates user Uplink transmission power, Indicates to users Set of users using the same pilot sequence It is the noise variance of the uplink;

[0072] against ,have The variance Represented as:

[0073] ;

[0074] During the downlink cross-modal transmission phase, both traditional continuous audio-visual stream transmission and intermittent AIGC stream transmission are considered. If a user requests AIGC service, the central processing unit will generate cross-modal content based on the AIGC model and user prompts, and transmit the generated content along with the traditional audio-visual stream to the AP via the fronthaul link. Subsequently, the AP constructs a precoding vector based on locally estimated channel information and transmits the multimodal stream received from the CU to the user. If the user does not request AIGC service, the CU only transmits the traditional audio-visual stream to the AP, and then the AP transmits the received audio-visual stream to the user.

[0075] user The signal-to-interference-plus-noise ratio is expressed as: ;

[0076] in, Indicate each The number of antennas equipped, Indicates AP For users downlink transmit power, This represents the variance of downlink noise at the user end;

[0077] AP Service users In the time slot The transmission rate obtained within is expressed as: ;

[0078] in, It is the bandwidth of the spectrum resource block;

[0079] user In transmission time slot The size of the requested cross-modal flow is expressed as: ;

[0080] in, Indicates user In transmission time slot The size of the audio-visual stream, Indicates user In transmission time slot AIGC stream size and satisfy ,in For users The size of the image generated by each AIGC request. This represents the number of AIGC service requests made by each user within each conformity interval. Indicates the duration of the conformity interval. Indicates the duration of the transmission time slot.

[0081] Preferably, step S2 specifically includes:

[0082] S21. Calculate the user's immersive content transmission latency, take the power resource allocation strategy under non-cellular MIMO network as the optimization object, take minimizing the user's maximum immersive content transmission latency as the objective function, and take the user's audio-visual stream transmission latency limit and power resource limit as the constraint objects to construct a resource scheduling model for cross-modal stream transmission.

[0083] Preferably, step S21 specifically includes:

[0084] S211, According to the user In transmission time slot transmission rate and the total amount of cross-modal streaming data requested The calculated latency for immersive content transmission is: ;

[0085] S212. Establish an immersive cross-modal transmission resource scheduling model representation for non-cellular MIMO networks to address the optimization problem: ;

[0086] ; ; ; ;

[0087] in, Describes the set of APs. Indicates the number of APs. Represents a set of users. This indicates the number of users.

[0088] S22. Solve the resource scheduling model to obtain the number of resources that minimize the maximum immersive content transmission latency for users under the constraints of latency and power resources.

[0089] Preferably, step S22 specifically includes:

[0090] The difficulty in solving the above maximization problem lies in the large number of optimization variables among multiple access points and users, and the coupling relationship between these optimization variables.

[0091] S221. By introducing auxiliary variables, the problem of minimizing the maximum immersive content transmission latency for users is transformed into an equivalent form, which is expressed as follows:

[0092]

[0093] in .

[0094] For AP The partially augmented Lagrangian function is expressed as:

[0095]

[0096] in, , as well as To scale the dual variable, For penalty parameters, For indicator functions, it is represented as follows:

[0097]

[0098] S222. To address the problem of minimizing the transmission latency of maximum immersive content, a distributed power allocation algorithm for cellular-free MIMO networks based on the consensus ADMM algorithm is proposed. It needs to be based on other APs (such as AP) , For users Signals and interference local computing users The received interference and signals, of which AP For users The signal is represented as Interference is represented as To address this challenge and achieve optimized distributed power allocation, an AP (Active Power Utilization) system is introduced. signal variables and interference variables The local copy of the global variable is represented as follows: and Global variables Used to force AP place and AP place Equal, global variables Used to force AP place and AP place They are equal. This can be achieved by setting consensus constraints on the signals. and and interference with consensus constraints and To achieve this. According to AP Local signal copy and interference from other AP transmissions, AP The latency of immersive content delivery for users can be calculated. The local and global variable sets for all APs are represented as follows: and For AP ,have as well as AP The feasible region defined by the local variable constraint at the location is represented as follows: The original problem is further transformed into a standard consensus optimization problem.

[0099]

[0100] S223. Based on the consensus-based ADMM algorithm, each AP independently updates its local and dual variables to drive consensus on local variables. Local and global variables are updated alternately, and dual variables are updated based on Lagrange multipliers. The update process of each ADMM iteration is explained as follows:

[0101] A. Local Variable Update: Each AP updates its local variables in parallel based on information exchanged with other APs to address its local minimization problem. The local optimization problem is represented as:

[0102]

[0103] B. Global Variable Update: Based on the augmented Lagrange formula, global variables are updated by fixing local variables. The update of global variables can be represented as the following subproblem:

[0104]

[0105]

[0106]

[0107] C. Dual Variable Update: After the global and local variables are updated, each AP updates its dual variable, as shown below:

[0108]

[0109]

[0110]

[0111] The power resource allocation algorithm for cellular MIMO networks based on the consensus ADMM algorithm is expressed as follows:

[0112] a. Given the initial value of a global variable , , and the initial values ​​of the dual variables , and ;

[0113] b. For each AP, when the following conditions are met... At that time, local variable updates, global variable updates, and dual variable updates are performed sequentially;

[0114] c. Continue until the loop ends.

[0115] Example 2: In order to apply the above-mentioned cross-modal transmission resource scheduling method based on non-cellular MIMO network, this example provides a cross-modal transmission resource scheduling system based on non-cellular MIMO network.

[0116] The transmission task acquisition module is used to acquire the user transmission rate and the total amount of cross-modal stream data to be transmitted for immersive cross-modal transmission resource scheduling tasks based on non-cellular MIMO networks.

[0117] The transmission strategy optimization module establishes an optimization mathematical model for resource scheduling based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, and solves the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints.

[0118] The transmission task allocation module is used to allocate power resources to multiple users based on the amount of resources required by each user.

[0119] The working process and working principle of each module are the same as in Example 1, and will not be repeated in this example.

[0120] Example 3: This example provides an electronic device that may include a processor, a communications interface, a memory 630, and a communication bus. The processor, communications interface, and memory communicate with each other via the communication bus. The processor can call logical instructions from the memory to execute the resource scheduling method for the multi-user, multi-modal stream integrity transmission task in Example 1.

[0121] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] Example 4: This example provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the immersive cross-modal transmission resource scheduling method based on a non-cellular MIMO network in Example 1.

[0123] Based on the above embodiments, in order to verify the immersive cross-modal transmission resource scheduling method and system based on non-cellular MIMO networks of the present invention, in the face of situations such as Figure 1 The effect of resource scheduling in the transmission task of the system shown is investigated through simulation experiments.

[0124] The simulation parameters are set as follows: Total number of APs The total number of users is 20. The number of antennas configured per AP is 10, the communication bandwidth is 20 MHz, and the number of antennas configured per AP is [number missing]. The coherence interval is 16. For 200 symbols, orthogonal pilot number The value is 6, and the noise power is Maximum transmission power The data volume of the user's audiovisual stream is 1 W. The data volume is selected from a set of {20, 30, 40, 50} Mbits, with a compression ratio of 150. The AIGC image data volume is obtained based on experiments using SD3 and SDXL models. The AIGC service request rate is... The interval is 10; conformity interval A 1-minute transmission time slot The simulation time is 1 second. The simulation area is a 500m × 500m square region, and a distance-based large-scale fading model (including shadow fading) is adopted. The proposed consensus-based ADMM-based distributed power allocation scheme is compared with the independent computation power allocation scheme (PAWIC) and the power allocation without aconsideration of pilot contamination scheme (PAWPC). In the PAWIC scheme, each AP performs local power allocation independently without considering consensus and information exchange with other APs, while the PAWPC scheme does not consider the impact of pilot contamination during power allocation.

[0125] like Figure 4 The diagram illustrates the relationship between the number of AIGC service requests and the maximum immersive content delivery latency. When the number of AIGC service requests... When determined, the maximum immersive content transmission latency varies with the maximum transmit power. It increases with the increase of, because An increase in this will increase interference between users. When When the value is fixed, the maximum immersive content transmission latency in all schemes increases with... It increases with the increase of [something]. When [something] increases. When the value is fixed at 10, the maximum immersive content delivery latency of the proposed scheme is reduced by 37.9% and 29.7% respectively compared with PAWIC and PAWPC.

[0126] like Figure 5 The diagram illustrates the relationship between the number of users and the maximum immersive content transmission delay. When the number of orthogonal pilot signals is fixed at 4, the maximum immersive content transmission delay first increases slightly from U=6 to U=8, then increases sharply from U=8 to U=9, and then varies slightly from U=9 to U=11. When the number of orthogonal pilot signals is fixed at 6, the maximum immersive content transmission delay first increases sharply with the increase of the number of users, and then increases steadily. The difference in maximum immersive content transmission delay among the three schemes gradually becomes significant with the increase of the number of users. When the number of users is set to 11, compared with PAWIC and PAWPC, the maximum immersive content transmission delay of the proposed scheme is reduced by 34% and 29.4%, respectively.

[0127] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cross-modal transmission resource scheduling method based on a cellular-free MIMO network, characterized in that, Includes the following steps: S1. Based on the user's cross-modal stream transmission task, construct a herd behavior model for the user's AIGC service request, and calculate the user's transmission rate and the total amount of cross-modal stream data that needs to be transmitted. S2. Based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, establish an optimization mathematical model for resource scheduling, and solve the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints. S3. Allocate power resources to multiple users based on the amount of resources required by each user.

2. The cross-modal transmission resource scheduling method based on a cellular-free MIMO network as described in claim 1, characterized in that, Step S1 specifically includes: S11. Construct a herding behavior model for user AIGC service requests and calculate the probability that a user becomes an AIGC service user: For AIGC streams, the user's AIGC service adoption process is divided into a series of conformity intervals. Each conformity interval contains several transmission slots. Then the user In the The probability that someone is an AIGC service user within a certain conformity interval is: ; in, Indicates user Total number of influencers The threshold for conformity is... This indicates the number of influencers who have become users; S12. Establish transmission models for AIGC streams and audiovisual streams, and calculate user transmission rates and the total amount of cross-modal stream data to be transmitted: During the uplink channel estimation phase, the access point is set. and users The channel between them is ;in This represents the small-scale fading coefficient. Represents the large-scale fading coefficient; sets the user... The pilot sequence used in the channel estimation stage is: Then, based on the minimum mean square error channel estimation, Channel estimation Represented as: ； ； in, Indicates the uplink training length. Indicates user Uplink transmission power, Indicates to users Set of users using the same pilot sequence It is the noise variance of the uplink; against ,have The variance Represented as: ； During the downlink cross-modal transmission phase, if the user requests AIGC service, the central processing unit will generate cross-modal content based on the AIGC model and user prompts, and transmit the generated content along with the traditional audio-visual stream to the AP via the fronthaul link; subsequently, the AP constructs a precoding vector based on locally estimated channel information and transmits the multimodal stream received from the CU to the user; if the user does not request AIGC service, the CU only transmits the traditional audio-visual stream to the AP, and then the AP transmits the received audio-visual stream to the user. user The signal-to-interference-plus-noise ratio is expressed as: ; in, Indicate each The number of antennas equipped, Indicates AP For users downlink transmit power, This represents the variance of downlink noise at the user end; AP Service users In the time slot The transmission rate obtained within is expressed as: ; in, It is the bandwidth of the spectrum resource block; user In transmission time slot The size of the requested cross-modal flow is expressed as: ; in, Indicates user In transmission time slot The size of the audio-visual stream, Indicates user In transmission time slot AIGC stream size and satisfy ,in For users The size of the image generated by each AIGC request. This represents the number of AIGC service requests made by each user within each conformity interval. Indicates the duration of the conformity interval. Indicates the duration of the transmission time slot.

3. The cross-modal transmission resource scheduling method based on a cellular-free MIMO network as described in claim 1, characterized in that, Step S2 specifically includes: S21. Calculate the user's immersive content transmission latency, take the power resource allocation strategy under non-cellular MIMO network as the optimization object, take minimizing the user's maximum immersive content transmission latency as the objective function, and take the user's audio-visual stream transmission latency limit and power resource limit as the constraint objects to construct a resource scheduling model for cross-modal stream transmission. S22. Solve the resource scheduling model to obtain the number of resources that minimize the maximum immersive content transmission latency for users under the constraints of latency and power resources.

4. The cross-modal transmission resource scheduling method based on a cellular-free MIMO network as described in claim 3, characterized in that, Step S21 specifically includes: S211, According to the user In transmission time slot transmission rate and the total amount of cross-modal streaming data requested The calculated latency for immersive content transmission is: ; S212. Establish an immersive cross-modal transmission resource scheduling model representation for non-cellular MIMO networks to address the optimization problem: ; ； ； ； ； in, Describes the set of APs. Indicates the number of APs. Represents a set of users. This indicates the number of users.

5. The cross-modal transmission resource scheduling method based on a cellular-free MIMO network as described in claim 3, characterized in that, Step S22 specifically includes: first, transforming the original optimization problem P1 established in S212 into an equivalent form that is convenient for mathematical processing; then, for each access point, constructing local variables and global variables, further transforming the problem into a distributed consensus form; and finally, obtaining a stable power resource allocation strategy based on the distributed consensus ADMM algorithm.

6. A cross-modal transmission resource scheduling system for implementing the cross-modal transmission resource scheduling method based on a non-cellular MIMO network according to any one of claims 1-5, characterized in that, The system includes: The transmission task acquisition module is used to acquire the user transmission rate and the total amount of cross-modal stream data to be transmitted for resource scheduling tasks based on non-cellular MIMO immersive cross-modal transmission. The transmission strategy optimization module is used to establish an optimization mathematical model for resource scheduling based on the total data volume and transmission rate of the cross-modal stream, with the goal of minimizing the maximum immersive content transmission latency for the user, and solve the model to obtain the amount of power resources that minimize the maximum immersive content transmission latency for the user under latency and power resource constraints. The transmission task allocation module is used to allocate power resources to multiple users based on the amount of resources required by each user.

7. A non-transitory computer-readable storage medium having stored thereon computer instructions that cause a computer to execute a cross-modal transmission resource scheduling method based on a non-cellular MIMO network as described in any one of claims 1-5.

8. An electronic device, comprising a processor, a communication interface, a memory, and a communication bus; in, The processor, communication interface, and memory communicate with each other through a communication bus; The processor is used to call logical instructions in memory to execute the cross-modal transmission resource scheduling method based on a non-cellular MIMO network as described in any one of claims 1 to 5.

Images