IRS-assisted NOMA semantic transmission resource allocation method

Abstract

The invention belongs to a wireless communication technology, and particularly relates to an IRS-assisted NOMA semantic transmission resource allocation method, which comprises the step of maximizing the minimum semantic spectrum efficiency of a user in an IRS-assisted NOMA semantic transmission model by optimizing a semantic compression ratio, power allocation and phase shift at an IRS. Simulation results show that fairness of information transmission between users can be guaranteed, transmission reliability and transmission rate are improved compared with the prior art, and accurate mapping can be achieved based on the quality of signals and channels.

Description

Technical Field

[0001] The present invention belongs to wireless communication technology, and in particular to an IRS-assisted NOMA semantic transmission resource allocation method. Background Art

[0002] With the widespread use of various artificial intelligence (AI) applications such as autonomous driving and virtual reality, the number of connections of smart devices has shown an exponential growth. This large-scale connection and the increase in massive data traffic have posed unprecedented challenges to existing communication systems. To meet these challenges, semantic communication (SC) has emerged as an emerging communication technology, showing the potential to break through the bottleneck of traditional communications. Traditional communication systems mainly focus on the transmission of bit streams, while the purpose of semantic communication is to transmit semantic information in source data. By extracting and compressing the semantics of information, SC technology can significantly reduce the amount of data that needs to be transmitted in the network. This feature is due to the development of joint source channel coding technology, which shows better performance than separate coding, thereby significantly improving the reliability of semantic transmission.

[0003] In the field of semantic communication, combining the advantages of non-orthogonal multiple access technology (NOMA) and intelligent reflecting surface technology (IRS) can efficiently support the connection and data transmission of more users within the same spectrum resources, thereby achieving efficient semantic transmission. This combination not only optimizes communication performance, but also provides new ideas for the design of future wireless communication systems.

[0004] Compared with the traditional orthogonal multiple access technology (OMA), NOMA superimposes multiple user signals with different powers on the same resource block, and decodes the multi-user information stream through continuous interference elimination at the receiving end, thereby realizing power domain multiple access. NOMA technology performs well in user access. IRS integrates a large number of low-cost passive reflective elements to intelligently reconfigure the wireless propagation environment on the plane, significantly improving the performance of wireless communication networks. The core of IRS technology lies in its ability to optimize the signal propagation path. It can dynamically adjust the direction and phase of the reflected signal to overcome the problems of signal attenuation and short transmission distance in traditional wireless communications, and enhance the signal strength and transmission reliability of the system.

[0005] However, the existing technology still lacks an effective and more efficient application solution for semantic communication resource allocation in IRS-NOMA scenarios. Summary of the invention

[0006] In order to ensure the communication quality of each user without distinction and realize a more efficient semantic transmission system, the present invention proposes an IRS-assisted NOMA semantic transmission resource allocation method, including: in the IRS-assisted NOMA semantic transmission model, by optimizing the semantic compression rate, power allocation and phase shift at the IRS, the minimum semantic spectrum efficiency of the user is maximized.

[0007] Furthermore, the optimization problem of maximizing the user's minimum semantic spectrum efficiency is expressed as:

[0008] Optimization problem:

[0009] Constraints:

[0010] 0≤θ m ≤2π,m∈M

[0011]

[0012] Among them, α is the power allocation of all users, expressed as α = [α1,…,α K ] T , α k is the power allocated to user k, K is the number of users in the IRS-assisted NOMA semantic transmission model; Θ is the reflection matrix of IRS; τ k is the average number of semantic symbols in the word mapping of user k, B is the maximum value of the average number of semantic symbols in the single word mapping; Φ k is the semantic spectrum efficiency of user k; is the set of users; θ m is the reflection phase shift of the mth reflection element, M is the number of reflection elements; γ t→k represents the signal interference plus noise ratio of the kth user decoding the tth data stream; P is the total power allocated by the base station to all users; is the conjugate transpose of the channel gain from IRS to the tth user; f is the channel gain from the base station to IRS; v t is the channel gain from the base station to the t-th user; σ 2 is the noise power; γ t is the SINR of the t-th data stream; ξ k is the semantic similarity of user k; ξ th is the threshold of semantic similarity.

[0013] Furthermore, the semantic spectral efficiency of user k is expressed as:

[0014]

[0015] Among them, ξ(τk ,γ k ) represents the semantic similarity of user k, and I represents the text S requested by user k. k The semantic information of each text is represented by L, and L represents the average number of words in each text.

[0016] Furthermore, the semantic similarity ξ(τ k ,γ k ) is expressed as:

[0017]

[0018] in, are parameters determined using the nonlinear least squares criterion.

[0019] Furthermore, the process of solving the optimization problem includes:

[0020] The decoding order is determined by optimizing the IRS phase shift to maximize the maximum achievable combined channel strength of all K users;

[0021] For a given decoding order, the SIC decoding constraints in the optimization problem are simplified to construct an optimization problem that maximizes the minimum signal-to-noise ratio of all users;

[0022] The optimization problem of maximizing the minimum signal-to-noise ratio of all users is solved alternately, that is, the power allocation coefficient is first fixed to optimize the reflection phase shift, and then the power allocation coefficient is optimized under the fixed reflection phase shift;

[0023] After determining the minimum signal-to-noise ratio of all users, the optimization problem of maximizing the minimum semantic spectrum efficiency of the user is converted into solving the number of semantic symbols, and the final solution is obtained by looking up the table.

[0024] The present invention considers an IRS-assisted NOMA semantic transmission framework, in which the base station can effectively transmit semantic information to multiple users by sending superimposed downlink signals to these users. In order to further improve spectrum utilization and ensure user fairness, the present invention jointly optimizes the number of semantic symbols, the transmission power at the base station and the phase shift at the IRS to maximize the minimum semantic spectrum efficiency of all users; in order to solve this problem, the present invention proposes a new three-step resource allocation algorithm, and the simulation results clearly confirm the effectiveness of the resource allocation method proposed in the present invention, which promotes the coexistence of traditional communication and semantic communication in future networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of the IRS-assisted downlink NOMA semantic transmission framework of the present invention;

[0026] Figure 2 A schematic diagram of the convergence of the combined power allocation and phase shift design algorithm of the present invention;

[0027] Figure 3 The solution of the present invention and other prior art solutions are th Comparative graph of changes;

[0028] Figure 4 For the time being th =0.7, a comparison curve diagram of the scheme of the present invention and other prior art schemes as the power P changes. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] The present invention proposes an IRS-assisted NOMA semantic transmission resource allocation method, comprising: in an IRS-assisted NOMA semantic transmission model, by optimizing the semantic compression rate, power allocation and phase shift at the IRS, maximizing the user's minimum semantic spectrum efficiency.

[0031] In order to ensure the communication quality of each user without distinction and realize a more efficient semantic transmission system, the present invention considers a semantic transmission framework of IRS-assisted NOMA, in which the present invention establishes an optimization problem with the goal of maximizing the user's minimum semantic spectral efficiency (Semantic Spectral efficiency, S-SE), and optimizes the optimization problem by optimizing the semantic compression rate, power allocation, and phase shift at IRS. Due to the various coupled variables and non-convexity in the problem, it is challenging to solve the problem. Therefore, the present invention designs a three-step algorithm for the joint optimization problem of the transmission power, IRS phase shift, and number of semantic symbols of the semantic communication base station, which specifically includes:

[0032] Firstly, a low-complexity NOMA decoding scheme is proposed to decouple the decoding order design from the design of IRS reflection coefficients.

[0033] Then, a joint optimization algorithm is proposed to solve the joint power allocation and reflection coefficient design problem given the decoding order;

[0034] Finally, under the given phase shift and power allocation, the number of transmission semantic symbols is designed through exhaustive search method;

[0035] The resource allocation method of the present invention effectively optimizes S-SE and reduces the computational complexity.

[0036] like Figure 1 As shown, this embodiment considers an IRS-assisted downlink NOMA semantic transmission framework, which consists of a base station, a RIS with M independent passive reflective elements, and K users. This embodiment defines the reflective element set as The user collection is The DeepSC model is first trained at the base station to improve its ability to extract the underlying semantics. The trained DeepSC model is then broadcast to the user. The transmitter extracts features from the source information, followed by semantic compression coding and channel coding. During this process, the base station acts as a coordination center to process the extracted mission-critical information. The base station uses non-orthogonal multiple access (NOMA) technology to superimpose this information to ensure that multiple users can efficiently share channel resources. During transmission, the IRS is used to enhance and optimize the signal to ensure effective transmission of the signal in a complex propagation environment. The receiver uses channel decoding and semantic decoding to ensure that mission-critical information can be recovered to support the decision-making and execution of downstream applications.

[0037] 1. IRS-assisted NOMA semantic transmission model

[0038] User k requests to receive a text file. The text data can be represented as S k =[s1,s2,...s i ,...,s L ], where s i is the i-th word, and L is the average number of words per text. The file is mapped into a semantic symbol vector x by the DeepSC transmitter k , where x k Contains S k The semantic information of x k The length is τ k L, τ k represents the average number of semantic symbols per word mapping, and:

[0039]

[0040] Where B represents the maximum value that the average number of semantic symbols in word mapping can take.

[0041] A single-antenna base station sends superimposed signals to K users on the same time and frequency block. The smart controller connected to the IRS can intelligently adjust the phase shift to assist NOMA transmission, defining As the reflection matrix of IRS, where β m represents the reflection coefficient of the mth reflective element, θ m represents the reflection phase shift of the mth reflection element. In this case, a fixed amplitude reflection coefficient β is used.m =1, the phase shift at IRS needs to satisfy the following constraints:

[0042] 0≤θ m ≤2π,m∈M (2)

[0043] The baseband transmission signal at the base station can be expressed as where α k is the power allocation coefficient obtained by allocating part of the total power P to the kth user, that is, the power allocated to user k. In order to adapt to the power resource constraints at the base station, this embodiment allocates vector α k The following restrictions apply:

[0044]

[0045] Define the channel from IRS to user k as The channel from the base station to the IRS is The channel from the base station to user k is v k , represents the additive white Gaussian noise at user k. The signal received at user k can be expressed as:

[0046]

[0047] The successive interference cancellation (SIC) decoding order is an important issue in the NOMA system, where the optimal decoding order is determined by the channel gain. The channel gain of the user is Without loss of generality, this embodiment assumes that the user ranking is 0<|h1| 2 ≤|h2| 2 ≤...≤|h K | 2 .

[0048] The corresponding signal to interference and noise ratio (SINR) of the k-th user decoding the t-th data stream can be expressed as:

[0049]

[0050] In order to ensure that SIC can be performed correctly, the kth user decodes the tth data stream (i.e., γ t→k ) needs to be no less than the SINR of the t-th data stream. This constraint can be expressed as:

[0051]

[0052] The semantic spectral efficiency (S-SE) of user k can be defined as:

[0053]

[0054] Where I represents the text S k The semantic information of k ,γ k ) represents semantic similarity. Semantic similarity lacks a clear form. In this embodiment, the relationship between semantic similarity and SINR for a given semantic compression degree can be modeled as a generalized logical function, namely:

[0055]

[0056] in, are parameters determined using the nonlinear least squares criterion.

[0057] In order to achieve the optimization and coordination between effectiveness and transfer efficiency, the semantic similarity is constrained, which is expressed as:

[0058]

[0059] 2. Constructing the optimization problem

[0060] According to the above analysis, this embodiment optimizes the semantic compression rate (number of semantic symbols) τ k , power allocation (i.e., α = [α1, ..., α K ] T ) and the phase shift at IRS (i.e. Θ) to maximize the resource allocation problem of the user's minimum S-SE. The optimization problem of maximizing the user's minimum semantic spectrum efficiency is expressed as:

[0061] Optimization Problem (P0)

[0062] Constraints: (1), (2), (3), (6), (9)

[0063] The optimization problem P0 is a mixed integer nonlinear programming problem. Solving the problem P0 faces three major challenges: first, due to the existence of the integer variable k, the optimization problem P0 is an NP-hard problem; second, since the IRS reflection coefficient can control the decoding order, it is difficult for NOMA users to obtain the optimal decoding order; third, the transmission power and reflection coefficient are highly coupled, which makes the problem more challenging. In order to find a feasible solution, this embodiment proposes a three-step algorithm, which specifically includes:

[0064] (1) Determine the decoding order

[0065] The SIC decoding order is determined by the combined channel gain of the direct link and the reflected link. The decoding order is determined by optimizing the IRS phase shift to maximize the maximum achievable combined channel strength of all K users. Specifically, the maximum achievable strength of the combined channel of the kth user can be obtained by solving the following problem, which is expressed as:

[0066] Optimization Problem (P1)

[0067] Constraints: (2)

[0068] It is determined by ranking the combined channel strengths achievable by the users after solving K optimization problems.

[0069] (2) Joint power allocation and reflection coefficient design

[0070] For a given decoding order, the SIC decoding constraint (6) in the optimization problem P0 can be further simplified as:

[0071]

[0072] From formula (8), we know that ξ(τ k ,γ k ) with γ k The increase of is monotonically non-decreasing, so to solve the problem P0, we need to maximize the minimum signal-to-noise ratio of all users, that is:

[0073]

[0074] Problem P0 can be simplified to the following problem:

[0075] Optimization Problem (P2)

[0076] Constraints: (1), (2), (3), (9), (10), (11)

[0077] The present invention ensures that the signal-to-noise ratio of each user exceeds Q, where Q is a slack variable, indicating that the minimum SINR is to be maximized. Since there are two variable blocks (i.e., α and θ) coupled in the optimization problem (P2), this embodiment alternately optimizes the power allocation coefficient and the reflection phase shift, and P2 can be decomposed into two sub-problems P2.1 and P2.2 for solution. Specifically:

[0078] Given the power division coefficient, optimize the reflection phase shift, that is:

[0079] Optimization Problem (P2.1)

[0080] Constraints: (2), (10), (11)

[0081] Given the reflection phase shift, the power division coefficient is optimized, that is:

[0082] Optimization Problem (P2.2)

[0083] Constraints: (3), (11)

[0084] In the optimization problem P2.1, the solver CVX can be used to optimize and solve the process. Specifically: first, the binary search method is used to solve the non-convex constraints and the SDR (semi-definite relaxation) technology is used for approximate solution, but the phase shift solution obtained usually does not satisfy the rank-one constraint; then, a Gaussian randomization scheme is introduced to obtain a rank-one solution.

[0085] In the optimization problem P2.1, the SINR of each user must be equal at the optimal solution, resulting in a second-order nonlinear equation that can be effectively solved by existing software such as MATLAB.

[0086] (3) Semantic symbol digital design

[0087] When the SINR is known, the semantic spectrum efficiency of the user is maximized, and the optimization variable is the number of semantic symbols, resulting in the following optimization problem:

[0088] Optimization Problem (P3)

[0089] Constraints: (1), (9)

[0090] Among them, ξ(τ k ,γ k ) is affected by the number of semantic symbols τ k and signal-to-noise ratio γ k Impact; I / L is the average amount of semantic information contained in each word or phrase. In the present invention, I / L depends on the type of source. This item is a constant for a specific type of source and does not affect the optimization of resources. It can be obtained by looking up a table. After determining the minimum signal-to-noise ratio, maximizing the minimum semantic spectrum efficiency only requires maximizing The value of τ can be obtained k , so the exhaustive method is used to solve P3.

[0091] Next, the effectiveness of the present invention is verified by simulation. In the simulation, consider two users, that is, K = 2, the distance from BS to IRS is 10m, the distances from BS and IRS to user 1 are set to 40m and 20m respectively, and the distances from BS and IRS to user 2 are set to 35m and 25m respectively; IRS has two reflection array elements, and the noise power at the user is: σ 2=-80dBm. In addition, the link from the base station to the user is considered to be Rayleigh fading, the link from the base station to the IRS and the link from the IRS to the user is Rice fading, and the distance-dependent path loss model P is also considered. L =C0(d / d0) -η , where C0 = -20 dB represents the path loss at the reference distance d0 = 1 m, η represents the path loss index, and d represents the distance between the transmitter and the receiver. The path loss coefficients for the base station-user, base station-IRS, and IRS-user are 3.5, 2, and 2, respectively. k Make restrictions to ensure that the experimental results are meaningful, that is, Φ k ≥Φ th ,Φ th = 0.05, and the number of random vectors for Gaussian randomization is 400 times.

[0092] Figure 2 This shows the convergence of the joint power allocation and phase shift design algorithm of the present invention. Initially, the Max-minSINR value is low, and then through alternating iterations of phase and power, the optimal phase and power values ​​are continuously found. As the number of iterations increases, the growth value gradually stabilizes and stabilizes at 11.4, reflecting the guarantee of fairness of the algorithm to users.

[0093] Figure 3 Depicts the semantic similarity threshold ξ for different algorithms th As ξ th With the increase of , the max min S-SE of RMS and MMS schemes decreases slightly, while the proposed scheme (PS) and the least mapping scheme (LMS) of the present invention decrease more significantly from 0.8 to 0.9. For the random mapping scheme (RMS) and the maximum mapping scheme (MMS) schemes, a fixed average number of semantic symbols per word is adopted, so the max min S-SE depends only on SINR. The increased semantic similarity threshold can be achieved by increasing the average number of semantic symbols per word, but this also leads to a decrease in max min S-SE. In short, the improvement in transmission reliability comes at the expense of transmission rate. From Figure 3 It can be observed that the performance of the scheme of the present invention is significantly better than the other three schemes. The algorithm proposed in the present invention adopts an optimal mapping strategy, which the other three schemes lack, and can accurately map based on the quality of the signal and channel.

[0094] Figure 4 Describes when th=0.7, the comparison curve of different algorithms with the change of power P. It can be observed from the figure that the max min S-SE of each scheme increases with the increase of power. This is because the increase of power enhances the signal, thereby improving SINR, and a relatively low code rate can be used to ensure the task performance, thus reducing the average number of semantic symbols per word. Figure 4 It can be observed that the algorithm proposed in the present invention achieves near-optimal performance and is significantly better than other solutions.

[0095] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An IRS-assisted NOMA semantic transmission resource allocation method, characterized in that: include: In the IRS-assisted NOMA semantic transmission model, the user's minimum semantic spectrum efficiency is maximized by optimizing the semantic compression rate, power allocation, and phase shift at the IRS.

2. According to claim 1, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The optimization problem of maximizing the user's minimum semantic spectrum efficiency is expressed as: Optimization problem: Constraints: 0≤θ m ≤2π,m∈M Among them, α is the power allocation of all users, expressed as α = [α1, ..., α K ] T , α k is the power allocated to user k, K is the number of users in the IRS-assisted NOMA semantic transmission model; Θ is the reflection matrix of IRS; τ k is the average number of semantic symbols in the word mapping of user k, B is the maximum value of the average number of semantic symbols in the single word mapping; Φ k is the semantic spectrum efficiency of user k; is the set of users; θ m is the reflection phase shift of the mth reflection element, M is the number of reflection elements; γ t→k represents the signal interference plus noise ratio of the kth user decoding the tth data stream; P is the total power allocated by the base station to all users; is the conjugate transpose of the channel gain from IRS to the tth user; f is the channel gain from the base station to IRS; v t is the channel gain from the base station to the t-th user; σ 2 is the noise power; γ t is the SINR of the t-th data stream; ξ k is the semantic similarity of user k; ξ th is the threshold of semantic similarity.

3. According to claim 2, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The semantic spectral efficiency of user k is expressed as: Among them, ξ(τ k ,γ k ) represents the semantic similarity of user k, and I represents the text S requested by user k. k The semantic information of each text is represented by L, and L represents the average number of words in each text.

4. According to claim 3, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The semantic similarity ξ(τ k ,γ k ) is expressed as: in, are parameters determined using the nonlinear least squares criterion.

5. According to claim 2, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The process of solving an optimization problem includes: The decoding order is determined by optimizing the IRS phase shift to maximize the maximum achievable combined channel strength of all K users; For a given decoding order, the SIC decoding constraints in the optimization problem are simplified to construct an optimization problem that maximizes the minimum signal-to-noise ratio of all users; The optimization problem of maximizing the minimum signal-to-noise ratio of all users is solved alternately, that is, the power allocation coefficient is first fixed to optimize the reflection phase shift, and then the power allocation coefficient is optimized under the fixed reflection phase shift; After determining the minimum signal-to-noise ratio of all users, the optimization problem of maximizing the minimum semantic spectrum efficiency of the user is converted into solving the number of semantic symbols, and the final solution is obtained by looking up the table.

6. According to claim 5, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The maximum achievable strength of the combined channel is determined for each user, and the optimization problem of obtaining the maximum achievable strength of the combined channel of the kth user is expressed as: Optimization problem: Constraint: 0≤θ m ≤2π,m∈M in, represents the channel g from IRS to user k k The conjugate transpose of k Denotes the channel from the base station to user k, k∈{1,2,3,...,K}.

7. According to claim 5, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The optimization problem of maximizing the minimum signal-to-noise ratio of all users is expressed as: Optimization problem: Constraints: 0≤θ m ≤2π,m∈M Where Q represents the minimum signal-to-noise ratio of the user; represents the channel g from IRS to user k k The conjugate transpose of k Denotes the channel from the base station to user k, k∈{1,2,3,...,K}.

8. According to claim 7, an IRS-assisted NOMA semantic transmission resource allocation method is characterized in that: The process of alternately solving the optimization problem of maximizing the minimum signal-to-noise ratio of all users includes: First, the power distribution coefficient is fixed to optimize the reflection phase shift, that is: Optimization problem: Constraint: 0≤θ m ≤2π,m∈M The process is optimized and solved using the solver CVX, namely: firstly, a binary search method is used to solve the non-convex constraints and a semi-positive definite relaxation technique is used for approximate solution, and then a Gaussian randomization scheme is introduced to obtain a rank-one solution; The power division coefficient is optimized at a fixed reflection phase shift, that is: Optimization problem: Constraints: At the optimal solution, the signal-to-interference-to-noise ratio of each user must be equal, resulting in a second-order nonlinear equation, which is solved by MATLAB to obtain the power allocation coefficient.

9. The IRS-assisted NOMA semantic transmission resource allocation method according to claim 5, characterized in that: After determining the minimum signal-to-noise ratio of all users, the optimization problem of maximizing the minimum semantic spectrum efficiency of the user is converted into solving the number of semantic symbols, which is expressed as: Optimization problem: Constraints:

Images