A flexible antenna-assisted common sense fusion NOMA network resource scheduling method and device

Abstract

The application provides a flexible antenna assisted common sense fusion NOMA network resource scheduling method and device, including: exploring a flexible antenna assisted common sense fusion NOMA network architecture, considering the scenario of multiple pinching antennas on the same waveguide, users in the form of clustering non-orthogonal multiple access network, and comprehensively considering the communication and sensing demand, according to the wave leakage characteristics of the pinching antenna, the line-of-sight transmission link is reconstructed, the pinching antenna link channel model and the transmission signal model are established; a decision variable set is constructed; the user cluster communication data rate and the detection signal power of the sensing target are comprehensively analyzed, the regularization coefficient is introduced according to the channel model and the transmission signal model and the decision variable set, the regularization clustering user data rate and the sensing target detection power model are constructed, and the common sense performance optimization target is solved by using an artificial intelligence algorithm, and the optimal correlation and power allocation set is obtained. The application can perform resource scheduling on the common sense fusion NOMA network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, in particular to a flexible antenna assisted integrated sensing and communication NOMA network resource scheduling method and device. BACKGROUND

[0002] Integrated sensing and communication (ISAC) technology integrates wireless communication and radar sensing functions into the same platform, uses shared resources such as hardware, spectrum and energy, improves operational efficiency, reduces costs and promotes sustainable development. Integrated sensing and communication networks use wireless communication signals to achieve identification, positioning, imaging and other sensing functions, and use sensing information to further enhance and exploit potential communication capabilities, so that wireless signals not only achieve the transmission of effective communication information, but also can sense, detect and characterize the physical world. In the research of ISAC, there are many studies on the integration of non-orthogonal multiple access (NOMA) technology. NOMA technology allows multiple users to share resources through different power levels or coding methods at the same time and frequency, and this technology uses superposition coding and serial interference cancellation technology to separate user signals at the receiving end, enhances system connectivity, reduces interference and improves spectrum efficiency.

[0003] Integrated sensing and communication non-orthogonal multiple access networks (ISAC-NOMA) give wireless communication stronger sensing capabilities and give birth to more abundant application scenarios. The utilization efficiency of shared resources and the adaptability of dynamic environments bring more challenges to integrated sensing and communication non-orthogonal access systems, especially their antenna systems. On the one hand, the resource sharing between sensing and communication functions requires the use of antennas with high directivity and flexibility to reduce signal interference and improve resource utilization efficiency. On the other hand, in a complex time-varying environment, the antenna must have the ability to dynamically adapt to environmental changes to ensure communication quality and sensing accuracy. However, traditional fixed-position antenna designs often fail to meet the requirements of shared resource utilization efficiency and dynamic environmental adaptability, so it is necessary to build an integrated sensing and communication non-orthogonal multiple access network architecture for flexible antenna systems.

[0004] In recent years, flexible antenna systems (such as pinching antennas) have received extensive attention. Pinching antenna systems create new line-of-sight links and / or enhance existing transceiving channels by applying low-cost dielectric materials at any position on the dielectric waveguide. Unlike traditional antennas, pinching antennas can be deployed flexibly, and increasing their number will not incur additional costs. Pinching antennas avoid the high costs and difficulty of combating large-scale path loss of other flexible antennas, and their flexible radiation patterns and strong adaptability in layout show broad application prospects. However, there is currently a lack of systematic research on resource scheduling for pinching antennas in sensory-fusion non-orthogonal access networks. SUMMARY

[0005] To solve the technical problems existing in the prior art, the present application provides a flexible antenna assisted sensory-fusion non-orthogonal multiple access (NOMA) network resource scheduling method and device, and the technical solution is as follows:

[0006] On the one hand, a flexible antenna assisted sensory-fusion non-orthogonal multiple access (NOMA) network resource scheduling method is provided, which comprises:

[0007] S1, a flexible antenna assisted sensory-fusion non-orthogonal multiple access (NOMA) network architecture is explored, which considers the scenario of multiple pinching antennas on the same waveguide. In this scenario, communication users non-orthogonally access the network in a clustered form, and the communication and sensing requirements are comprehensively considered. According to the wave leakage characteristics of pinching antennas, line-of-sight transmission links are reconstructed, and pinching antenna link channel models and transmission signal models are established.

[0008] S2, a decision variable set is constructed, which includes the correlation of flexible antennas and user clusters α, the power allocation parameter of intra-cluster users η, and the transmission power p of communication users and sensing targets.

[0009] S3, the communication data rate of user clusters and the detection signal power of sensing targets are comprehensively analyzed. According to the pinching antenna link channel model and the transmission signal model, and the decision variable set, a regularization coefficient is introduced to construct a regularized clustering user data rate and sensing target detection power model, as well as a sensory-fusion performance optimization target.

[0010] S4, an artificial intelligence algorithm is used to solve the sensory-fusion performance optimization target, which is a complex non-convex optimization problem of continuous-discrete mixed variables, and the optimal correlation and power allocation set is obtained.

[0011] Optionally, the pinching antenna link channel model established in S1 specifically includes:

[0012] Let the position of antenna l and user i in the mth cluster be ψ m,i = (x m,i , y m,i , 0), where d represents the height of the antenna, each cluster of users contains a strong user s and a weak user w, based on the spherical wave channel model, the channel gain of antenna l to user U(m, i) is represented as:

[0013]

[0014] where i ∈ {s, w}, τ represents the spherical wave parameter, e is the natural base, j represents the imaginary unit, π is an irrational number, λ represents the wavelength, and ||·|| represents the norm operation.

[0015] Optionally, the S1 establishes a pinching antenna link signal transmission model, specifically including:

[0016] Due to multiple pinching antennas on the same waveguide, the signals of all users are transmitted in a superimposed manner, and the signal of the Mth cluster of users is represented as:

[0017]

[0018] where M represents the total number of user clusters, the intra-cluster power allocation parameter x m,i , i ∈ {s, w} represents the communication signal sent to U(m, i).

[0019] Optionally, the S3 specifically includes:

[0020] According to the serial interference cancellation rule of non-orthogonal multiple access (NOMA), the data rates of the weak user U(m, w) and the strong user U(m, s) are respectively represented as:

[0021]

[0022] where the correlation α l,m ∈ {0, 1}, α = (α l,m ) l∈L,m∈M , p m is the allocated power of the user cluster m, the intra-cluster power allocation parameter h l,m,w ,h l,m,s are the channel gains of the weak user and the strong user respectively, θ l is the phase shift of the superimposed signal through the antenna l, and represent the intra-cluster interference and inter-cluster interference of the weak user respectively, represent the inter-cluster interference of the strong user, and σ 2denotes an additive white Gaussian noise;

[0023] Therefore, the data rate of the user cluster m is represented as R m = R m,w + R m,s ;

[0024] The detection signal power P k is represented as P H The channel vector h k is determined, which is represented as

[0025] The optimization objective is represented as maximizing the data rate of the clustered users and the detection signal power of the sensing targets:

[0026]

[0027] wherein, alpha represents the antenna and user cluster correlation vector, beta is a discrete variable, eta represents the intra-cluster power allocation parameter vector, which is a continuous variable, p represents the transmission power vector, which is a continuous variable, M represents the total number of user clusters, K represents the total number of sensing targets, gamma c > 0, gamma s > 0 represents the regularization coefficients of communication and sensing, by introducing the regularization coefficients, the complexity of the model is limited, and the generalization ability of the model is improved.

[0028] Optionally, the artificial intelligence algorithm in the S4 comprises a deep deterministic policy gradient algorithm and a reinforcement learning algorithm.

[0029] In another aspect, a flexible antenna-assisted communication and sensing fusion NOMA network resource scheduling device is provided, and the device comprises:

[0030] A building module is configured to explore a flexible antenna-assisted communication and sensing fusion non-orthogonal multiple access (NOMA) network architecture, wherein the network architecture considers the scenario of multiple pinching antennas on the same waveguide. In this scenario, communication users are non-orthogonal multiple access to the network in a clustered form, and the communication and sensing requirements are comprehensively considered. According to the wave leakage characteristics of the pinching antenna, the line-of-sight transmission link is reconstructed, and the pinching antenna link channel model and the transmission signal model are established.

[0031] A first building module is configured to build a decision variable set, wherein the decision variable set comprises: the correlation alpha of the flexible antenna and the user cluster, the power allocation parameter eta of the intra-cluster user, and the transmission power p of the communication user and the sensing target.

[0032] a second constructing module, configured to comprehensively analyze user cluster communication data rate and sensing target probe signal power, construct a regularized clustering user data rate and sensing target probe power model according to the pinching antenna link channel model and the transmission signal model, and a sensing performance optimization target, and introduce a regularization coefficient and a decision variable set;

[0033] a solving module, configured to solve the sensing performance optimization target by using an artificial intelligence algorithm, the sensing performance optimization target being a complex non-convex optimization problem of continuous-discrete mixed variables, and obtain an optimal correlation and power allocation set.

[0034] Optionally, the establishing module is specifically configured to:

[0035] Let the positions of the antenna l and the user i in the mth cluster be ψ m,i =(x m,i ,y m,i ,0), where d represents the height of the antenna, each cluster user includes a strong user s and a weak user w, based on a spherical wave channel model, the channel gain of the antenna l to the user U(m,i) is represented as:

[0036]

[0037] where i∈{s,w}, τ represents a spherical wave parameter, e is a natural base, j represents an imaginary unit, π is an irrational number, λ represents a wavelength, and ||·|| represents a norm operation.

[0038] Optionally, the establishing module is further configured to:

[0039] Since multiple pinching antennas are on the same waveguide, the signals of all users are transmitted in a superimposed manner, and the signal of the Mth cluster user is represented as:

[0040]

[0041] where M represents the total number of user clusters, the intra-cluster power allocation parameter x m,i , i∈{s,w} represents a communication signal sent to U(m,i).

[0042] Optionally, the second constructing module is specifically configured to:

[0043] According to a serial interference cancellation rule of a non-orthogonal multiple access (NOMA), the data rates of the weak user U(m,w) and the strong user U(m,s) are respectively represented as:

[0044]

[0045] Wherein, the correlation of flexible antenna l and user cluster m is alpha l,m ∈{0,1}, alpha=(alpha l,m ) l∈L,m∈M , p m is the allocation power of user cluster m, the intra-cluster power allocation parameter h l,m,w , h l,m,s are the channel gains of weak users and strong users respectively, theta l is the phase shift of superimposed signals through antenna l, and represent the intra-cluster interference and inter-cluster interference of weak users respectively, represent the inter-cluster interference of strong users, sigma 2 represents additive white Gaussian noise;

[0046] Therefore, the data rate of user cluster m is represented as R m =R m,w +R m,s ;

[0047] The probe signal power P k of the sensing target direction is determined by the covariance matrix M=xx H of the transmit signal vector, and the channel vector h k of the sensing target k to the pinching antenna, and is represented as

[0048] The optimization target is represented as maximizing the data rate of the clustered user and the probe signal power of the sensing target:

[0049]

[0050] Wherein, alpha represents the correlation vector of antenna and user cluster, beta is a discrete variable, eta represents the intra-cluster power allocation parameter vector, p represents the transmission power vector, M represents the total number of user clusters, K represents the total number of sensing targets, gamma c >=0, gamma s >=0 represent the regularization coefficients of communication and sensing, by introducing the regularization coefficients, the complexity of the model is limited, and the generalization ability of the model is improved.

[0051] Optionally, the artificial intelligence algorithm of the solving module comprises: a deep deterministic policy gradient algorithm, a reinforcement learning algorithm.

[0052] The technical scheme provided by the application has at least the following beneficial effects:

[0053] The application solves the resource scheduling problem in a sensing-fusion non-orthogonal multiple access network, maximizes the cluster user data rate and sensing target detection power (that is, maximizes the communication performance and sensing performance), and is suitable for various sensing-fusion non-orthogonal multiple access scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0055] Figure 1 is a flexible antenna assisted sensing-fusion NOMA network resource scheduling method flow chart provided by the embodiment of the application;

[0056] Figure 2 is a flexible antenna assisted sensing-fusion non-orthogonal multiple access (NOMA) network schematic diagram provided by the embodiment of the application;

[0057] Figure 3 is a flexible antenna assisted sensing-fusion NOMA network resource scheduling device block diagram provided by the embodiment of the application. DETAILED DESCRIPTION

[0058] In order to make the technical problems, technical solutions and advantages of the application more clear, the following will be described in detail in combination with the drawings and specific embodiments.

[0059] The embodiment of the application provides a flexible antenna assisted sensing-fusion NOMA network resource scheduling method, which can be realized by an electronic device, which can be a terminal or a server. Figure 1 The method flow chart is shown, and the processing flow can include the following steps:

[0060] S1, explore flexible antenna assisted sensing and communication fusion non-orthogonal multiple access (NOMA) network architecture, the network architecture considers the scenario of multiple pinching antennas on the same waveguide, in this scenario, the communication users are in the form of clustering non-orthogonal multiple access network (user clustering ensures that users with different channel conditions (such as strong users and weak users) can obtain service, through power domain multiplexing, the fairness between users is realized. Reasonable user clustering can reduce the interference between users, and the users in the cluster can effectively separate their respective signals through serial interference cancellation technology. In the cluster, more refined power allocation can be made according to the channel conditions of the users, so as to optimize the system performance. User clustering provides higher flexibility and scalability for the network, so that the network can dynamically adjust the size and number of clusters according to the user distribution and business needs), and comprehensively considers the communication and sensing requirements (such as Figure 2 Reconstruct the line-of-sight transmission link according to the wave leakage characteristics of the pinching antenna, establish a pinching antenna link channel model and a transmission signal model;

[0061] Optionally, the pinching antenna link channel model in S1 is established, specifically including:

[0062] Let the positions of antenna l and user i in the mth cluster be ψ m,i =(x m,i ,y m,i ,0), where d represents the height of the antenna, each cluster user includes a strong user s and a weak user w, based on the spherical wave channel model, the channel gain from antenna l to user U(m,i) is represented as:

[0063]

[0064] Wherein, i∈{s,w}, τ represents the spherical wave parameter, e is the natural base, j represents the imaginary unit, π is an irrational number, λ represents the wavelength, and ||·|| represents the norm operation.

[0065] Optionally, the pinching antenna link transmission signal model in S1 is established, specifically including:

[0066] Due to multiple pinching antennas on the same waveguide, the signals of all users are transmitted in a superimposed manner, and the signals of M cluster users are represented as:

[0067]

[0068] Wherein, M represents the total number of user clusters, and the power allocation parameter in the cluster is x m,i , i∈{s,w} represents the communication signal sent to U(m,i).

[0069] S2, constructing a decision variable set, the decision variable set comprising: a correlation a of a flexible antenna and a user cluster, a power allocation parameter η of an intra-cluster user, a transmission power p of a communication user and a sensing target;

[0070] S3, comprehensively analyzing a user cluster communication data rate and a sensing target probe signal power, according to the pinching antenna link channel model and the transmission signal model, and the decision variable set, introducing a regularization coefficient, constructing a regularized clustering user data rate and sensing target probe power model, and a sensing performance optimization target;

[0071] Optionally, the S3 specifically comprises:

[0072] According to a serial interference cancellation rule of non-orthogonal multiple access (NOMA), the data rates of the weak user U (m, w) and the strong user U (m, s) are respectively represented as:

[0073]

[0074] Wherein, the correlation a of the flexible antenna l and the user cluster m l,m ∈{0,1},α=(α l,m ) l∈L,m∈M , p m is the allocation power of the user cluster m, the intra-cluster power allocation parameter η h l,m,w ,h l,m,s are respectively the channel gains of the weak user and the strong user, θ l is the phase shift of the superimposed signal through the antenna l, and respectively represent the intra-cluster interference and the inter-cluster interference of the weak user, represent the inter-cluster interference of the strong user, σ 2 represents an additive white Gaussian noise;

[0075] Therefore, the data rate of the user cluster m is represented as R m =R m,w +R m,s ;

[0076] The probe signal power P k of the sensing target direction is determined by the covariance matrix M=xx H of the transmission signal vector and the channel vector h k of the sensing target k to the pinching antenna, and is represented as

[0077] The optimization target is represented as maximizing the clustering user data rate and the probe signal power of the sensing target:

[0078]

[0079] wherein, a represents an antenna and user cluster correlation vector, is a discrete variable, η represents a cluster inner power allocation parameter vector, is a continuous variable, p represents a transmission power vector, is a continuous variable, M represents a total number of user clusters, K represents a total number of sensing targets, γ c ≥ 0, γ s ≥ 0 represents a regularization coefficient of communication and sensing, by introducing the regularization coefficient, the complexity of the model is limited, and the generalization ability of the model is improved.

[0080] S4, solving the all-sensing performance optimization target by using an artificial intelligence algorithm, the all-sensing performance optimization target being a complex non-convex optimization problem of continuous-discrete mixed variables, and obtaining an optimal correlation and power allocation set.

[0081] Optionally, the artificial intelligence algorithm in S4 includes a deep deterministic policy gradient algorithm and a reinforcement learning algorithm (such algorithms have good adaptability and flexibility when processing complex non-convex optimization problems of continuous-discrete mixed variables).

[0082] As shown in Figure 3 the embodiment of the application further provides a flexible antenna assisted all-sensing fusion NOMA network resource scheduling device, the device comprises:

[0083] The establishment module 310 is configured to explore a flexible antenna assisted all-sensing fusion non-orthogonal multiple access (NOMA) network architecture, the network architecture considers the scenario of multiple pinching antennas on the same waveguide, in this scenario, communication users are in a clustered form of non-orthogonal multiple access to the network, and the communication and sensing requirements are comprehensively considered, the line-of-sight transmission link is reconstructed according to the wave leakage characteristics of the pinching antenna, and a pinching antenna link channel model and a transmission signal model are established.

[0084] The first construction module 320 is configured to construct a decision variable set, the decision variable set comprising: a correlation a of the flexible antenna and the user cluster, a power allocation parameter η of the user in the cluster, and a transmission power p of the communication user and the sensing target.

[0085] The second construction module 330 is configured to comprehensively analyze the user cluster communication data rate and the sensing target detection signal power, introduce a regularization coefficient according to the pinching antenna link channel model and the transmission signal model and the decision variable set, construct a regularized clustering user data rate and sensing target detection power model, and construct an all-sensing performance optimization target.

[0086] The solving module 340 is configured to solve the all-sensing performance optimization target by using an artificial intelligence algorithm, the all-sensing performance optimization target being a complex non-convex optimization problem of continuous-discrete mixed variables, and obtaining an optimal correlation and power allocation set.

[0087] Optionally, the establishing module is specifically configured to:

[0088] Let the positions of the antenna l and the user i in the mth cluster be ψ m,i = (x m,i , y m,i , 0), where d represents the height of the antenna, each cluster of users includes a strong user s and a weak user w, based on a spherical wave channel model, the channel gain of the antenna l to the user U(m, i) is represented as:

[0089]

[0090] where i ∈ {s, w}, τ represents a spherical wave parameter, e is a natural base, j represents an imaginary unit, π is an irrational number, λ represents a wavelength, and ||·|| represents a norm operation.

[0091] Optionally, the establishing module is specifically configured to:

[0092] Since multiple pinching antennas are on the same waveguide, the signals of all users are transmitted in a superimposed manner, and the signal of the Mth cluster of users is represented as:

[0093]

[0094] where M represents the total number of user clusters, and the power allocation parameter in the cluster is x m,i , i ∈ {s, w} represents a communication signal sent to U(m, i).

[0095] Optionally, the second constructing module is specifically configured to:

[0096] According to the serial interference cancellation rule of non-orthogonal multiple access (NOMA), the data rates of the weak user U(m, w) and the strong user U(m, s) are respectively represented as:

[0097]

[0098] where the correlation α of the flexible antenna l and the user cluster m l,m ∈ {0, 1}, α = (α l,m ) l∈L,m∈M , p m is the allocated power of the user cluster m, the power allocation parameter in the cluster is h l,m,w ,h l,m,s are the channel gains of the weak user and the strong user, respectively, θ l is the phase shift of the superimposed signal through the antenna l, and respectively represent the intra-cluster interference and inter-cluster interference of the weak users, represent the inter-cluster interference of the strong users, and 2 represents an additive white Gaussian noise;

[0099] Therefore, the data rate of the user cluster m is represented as R m = R m,w + R m,s ;

[0100] The probe signal power P k is perceived by the target direction, which is represented as P H is the channel vector of the kth sensing target to the pinching antenna, which is represented as h k is determined, which is represented as

[0101] The optimization target is represented as maximizing the data rate of the clustered users and the probe signal power of the sensing targets:

[0102]

[0103] wherein, alpha represents the antenna and user cluster correlation vector, beta is a discrete variable, eta represents the intra-cluster power allocation parameter vector, p represents the transmission power vector, M represents the total number of user clusters, K represents the total number of sensing targets, gamma c > 0, and gamma s > 0 represent the regularization coefficients of communication and sensing, which limit the complexity of the model and improve the generalization ability of the model by introducing the regularization coefficients.

[0104] Optionally, the artificial intelligence algorithm of the solving module comprises a deep deterministic policy gradient algorithm and a reinforcement learning algorithm.

[0105] The flexible antenna assisted communication and sensing integrated NOMA network resource scheduling device provided by the embodiment of the application corresponds to the flexible antenna assisted communication and sensing integrated NOMA network resource scheduling method provided by the embodiment of the application, and will not be described here.

[0106] The above only describes the preferred embodiments of the application and is not intended to limit the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A flexible antenna-assisted inductive fusion NOMA network resource scheduling method, characterized in that, The method includes: S1. Explore a flexible antenna-assisted sensing fusion non-orthogonal multiple access (NOMA) network architecture. The network architecture considers the scenario of multiple pinching antennas on the same waveguide. In this scenario, communication users access the network in a clustered manner using non-orthogonal multiple access. Taking into account both communication and sensing requirements, the line-of-sight transmission link is reconstructed based on the leakage characteristics of the pinching antennas, and a pinching antenna link channel model and transmission signal model are established. S2. Construct a set of decision variables, which includes: the correlation α between the flexible antenna and the user cluster, the power allocation parameter η of the users within the cluster, and the transmission power p of the communication users and the sensing targets; S3. Based on the comprehensive analysis of user cluster communication data rate and sensing target detection signal power, and according to the pinching antenna link channel model and transmission signal model, as well as the decision variable set, a regularization coefficient is introduced to construct a regularized clustered user data rate and sensing target detection power model, as well as a sensing performance optimization target. S4. Use artificial intelligence algorithms to solve the synesthetic performance optimization objective, which is a complex non-convex optimization problem of continuous-discrete mixed variables, to obtain the optimal set of correlation and power allocation.

2. The method according to claim 1, characterized in that, The pinching antenna link channel model established in S1 specifically includes: Let the positions of antenna l and user i in the m-th cluster be respectively ψ m,i =(x m,i ,y m,i ,0), where d represents the antenna height, and each user cluster contains strong users s and weak users w. Based on the spherical wave channel model, the channel gain from antenna l to user U(m,i) is expressed as: Where i∈{s,w}, τ represents the spherical wave parameter, e is the natural base, j represents the imaginary unit, π is an irrational number, λ represents the wavelength, and ||·|| represents the norm operation.

3. The method according to claim 2, characterized in that, The pinching antenna link transmission signal model established in S1 specifically includes: Because there are multiple pinching antennas on the same waveguide, the signals of all users are transmitted in a superimposed manner. The signal of the M-cluster users is represented as: Where M represents the total number of user clusters and the intra-cluster power allocation parameters. x m,i , i∈{s,w} represents the communication signal sent to U(m,i).

4. The method according to claim 3, characterized in that, S3 specifically includes: According to the serial interference cancellation rules of Non-Orthogonal Multiple Access (NOMA), the data rates of weak user U(m,w) and strong user U(m,s) are expressed as follows: Among them, the correlation α between the flexible antenna l and the user cluster m l,m ∈{0,1},α=(α l,m ) l∈L,m∈M p m Power allocation for user cluster m, intra-cluster power allocation parameters h l,m,w ,h l,m,s The channel gains for weak and strong users are θ, respectively. l To superimpose the phase shift of the signal after passing through antenna l, and These represent intra-cluster interference and inter-cluster interference for weak users, respectively. σ represents inter-cluster interference from strong users. 2 This represents additive white Gaussian noise; Therefore, the data rate of user cluster m is expressed as R m =R m,w +R m,s ; The power P of the detection signal in the direction of the target. k The covariance matrix M = xx of the transmitted signal vector H The channel vector h from the sensing target k to the pinching antenna k Decision, expressed as The optimization objective is expressed as maximizing the data rate of clustered users and the detection signal power of the sensed target: Where α represents the antenna-user cluster correlation vector, a discrete variable; η represents the intra-cluster power allocation parameter vector, a continuous variable; p represents the transmission power vector, a continuous variable; M represents the total number of user clusters; K represents the total number of sensing targets; and γ... c ≥0, γ s ≥0 represents the regularization coefficient for communication and sensing. By introducing the regularization coefficient, the complexity of the model is limited and the generalization ability of the model is improved.

5. The method according to claim 1, characterized in that, The artificial intelligence algorithms in S4 include: deep deterministic policy gradient algorithm and reinforcement learning algorithm.

6. A flexible antenna-assisted inductive fusion NOMA network resource scheduling device, characterized in that, The device includes: A module is established to explore a flexible antenna-assisted sensing fusion nonorthogonal multiple access (NOMA) network architecture. The network architecture considers scenarios with multiple pinching antennas on the same waveguide. In this scenario, communication users access the network in a clustered manner using nonorthogonal multiple access. Taking into account both communication and sensing requirements, the line-of-sight transmission link is reconstructed based on the leakage characteristics of the pinching antennas, and a pinching antenna link channel model and transmission signal model are established. The first construction module is used to construct a set of decision variables, which includes: the correlation α between the flexible antenna and the user cluster, the power allocation parameter η of the users within the cluster, and the transmission power p of the communication users and the sensing targets; The second construction module is used to comprehensively analyze the user cluster communication data rate and the detection signal power of the sensing target. Based on the pinching antenna link channel model and transmission signal model, as well as the decision variable set, a regularization coefficient is introduced to construct a regularized clustered user data rate and sensing target detection power model, as well as a sensing performance optimization target. The solution module is used to solve the synesthetic performance optimization objective using artificial intelligence algorithms. The synesthetic performance optimization objective is a complex non-convex optimization problem with continuous-discrete mixed variables, and obtains the optimal set of correlation and power allocation.

7. The apparatus according to claim 6, characterized in that, The establishment module is specifically used for: Let the positions of antenna l and user i in the m-th cluster be respectively ψ m,i =(x m,i ,y m,i ,0), where d represents the antenna height, and each user cluster contains strong users s and weak users w. Based on the spherical wave channel model, the channel gain from antenna l to user U(m,i) is expressed as: Where i∈{s,w}, τ represents the spherical wave parameter, e is the natural base, j represents the imaginary unit, π is an irrational number, λ represents the wavelength, and ||·|| represents the norm operation.

8. The apparatus according to claim 7, characterized in that, The establishment module is further used for: Because there are multiple pinching antennas on the same waveguide, the signals of all users are transmitted in a superimposed manner. The signal of the M-cluster users is represented as: Where M represents the total number of user clusters and the intra-cluster power allocation parameters. x m,i , i∈{s,w} represents the communication signal sent to U(m,i).

9. The apparatus according to claim 8, characterized in that, The second building module is specifically used for: According to the serial interference cancellation rules of Non-Orthogonal Multiple Access (NOMA), the data rates of weak user U(m,w) and strong user U(m,s) are expressed as follows: Among them, the correlation α between the flexible antenna l and the user cluster m l,m ∈{0,1},α=(α l,m ) l∈L,m∈M p m Power allocation for user cluster m, intra-cluster power allocation parameters h l,m,w ,h l,m,s The channel gains for weak and strong users are θ, respectively. l To superimpose the phase shift of the signal after passing through antenna l, and These represent intra-cluster interference and inter-cluster interference for weak users, respectively. σ represents inter-cluster interference from strong users. 2 This represents additive white Gaussian noise; Therefore, the data rate of user cluster m is expressed as R m =R m,w +R m,s ; The power P of the detection signal in the direction of the target. k The covariance matrix M = xx of the transmitted signal vector H The channel vector h from the sensing target k to the pinching antenna k Decision, expressed as The optimization objective is expressed as maximizing the data rate of clustered users and the detection signal power of the sensed target: Where α represents the antenna-user cluster correlation vector, a discrete variable; η represents the intra-cluster power allocation parameter vector, a continuous variable; p represents the transmission power vector, a continuous variable; M represents the total number of user clusters; K represents the total number of sensing targets; and γ... c ≥0, γ s ≥0 represents the regularization coefficient for communication and sensing. By introducing the regularization coefficient, the complexity of the model is limited and the generalization ability of the model is improved.

10. The apparatus according to claim 6, characterized in that, The artificial intelligence algorithms of the solution module include: deep deterministic policy gradient algorithm and reinforcement learning algorithm.

Images