A SIM-enabled massive MIMO communication system

Abstract

This invention belongs to the field of wireless communication technology, specifically relating to a SIM-enabled massive MIMO communication system. The system employs a single-RF link transmitter architecture, replacing traditional antenna arrays and single-layer reconfigurable smart surfaces with a SIM, enabling efficient signal transmission in multi-user communication scenarios. Specifically, the stacked smart metasurface applies controllable phase modulation to the incident single-frequency carrier signal, achieving signal modulation and precoding without the need for traditional baseband modulation. This allows different users to receive their desired signals with low mutual interference. Compared to traditional massive MIMO systems, this system architecture effectively alleviates the engineering challenges of high hardware complexity, high integration density, increased power consumption, and limited heat dissipation faced by high-frequency communication, while significantly reducing reliance on high-precision digital-to-analog converters and multiple RF links.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to a SIM-enabled massive MIMO communication system. Background Technology

[0002] With the development of sixth-generation (6G) mobile communication systems, communication networks are placing higher demands on ultra-high-speed data transmission, large-scale terminal access, and high-energy-efficiency operation. To meet these requirements, the operating frequency bands of communication systems are gradually expanding from Sub-6 GHz to millimeter-wave and terahertz bands. However, under high-frequency conditions, signal propagation loss increases significantly, posing a severe challenge to the system's efficiency in utilizing spectrum and energy resources. In existing technologies, Massive MIMO, by introducing a large number of antenna elements, can improve spectrum efficiency and suppress multi-user interference to some extent, but its engineering implementation in high-frequency scenarios has significant shortcomings. On the one hand, the system requires a large number of RF links, phase shifters, and high-resolution digital-to-analog converters (DACs), leading to a significant increase in overall power consumption; on the other hand, the integration of high-density RF devices further increases the complexity and cost of hardware design, limiting the practical application of this technology in high-frequency communication systems.

[0003] To address these issues, the academic community has proposed a novel wireless communication aid technology: Reconfigurable Intelligent Surface (RIS). RIS typically consists of a large number of subwavelength-scale tunable units, enabling the reconfiguration of the wireless propagation environment through programmable control of the phase, amplitude, and polarization characteristics of incident electromagnetic waves. Deploying RIS at the base station side and participating in precoding can reduce the number of radio frequency links to some extent, thereby reducing system hardware complexity. However, the controllability of existing single-layer RIS architectures is usually limited to phase control, resulting in limited ability to suppress inter-user interference in multi-user communication scenarios, making it difficult to further improve system capacity and communication quality.

[0004] The proposed Stacked Intelligent Metasurface (SIM) offers a feasible solution to alleviate the limitations of single-layer RIS architectures. SIM consists of multiple stacked metasurface structures, each layer composed of low-cost tunable units, enabling independent control of the amplitude and phase of electromagnetic waves. Leveraging this layer-by-layer controllable structure, SIM-assisted base stations can achieve high hardware efficiency with only a small number of RF links, reducing reliance on high-precision digital-to-analog converters in the RF links. Simultaneously, the additional degrees of control freedom provided by the multi-layer structure enhance beamforming capabilities, thereby reducing reliance on traditional digital beamforming techniques to some extent. Summary of the Invention

[0005] This invention proposes a massive MIMO communication system enabled by a stacked smart metasurface. The system employs a single-RF link transmitter architecture, replacing the traditional antenna array and single-layer reconfigurable smart surface with a SIM (Signal Module Identity), enabling efficient signal transmission in multi-user communication scenarios. Specifically, the stacked smart metasurface applies controllable phase modulation to the incident single-frequency carrier signal, achieving signal modulation and precoding without the need for traditional baseband modulation. This allows different users to receive their desired signals with low mutual interference. Compared to traditional massive MIMO systems, this system architecture effectively alleviates the engineering challenges of high hardware complexity, high integration density, increased power consumption, and limited heat dissipation in high-frequency communication, while significantly reducing reliance on high-precision digital-to-analog converters and multiple RF links.

[0006] The technical solution of this invention is as follows:

[0007] A SIM-enabled massive MIMO communication system, comprising: a carrier transmitter; and a [missing information - likely a component name or component name]... It consists of layers, and each layer contains A stacked smart metasurface with tunable reflective units; and A single-antenna user. A single-RF link transmitter outputs a carrier signal without baseband modulation, which is first incident on the first layer of a stacked smart metasurface. By adjusting the controllable coefficients of each reflecting element, the incident electromagnetic wave is phase-modulated and then transmitted to the next layer of the metasurface.

[0008] The design method for each phase layer of SIM is as follows: assuming time The expected received signal for each user is

[0009] (1)

[0010] SIM The phase offset matrix of the layer is , No. layer to the first The inter-layer transmission matrix is The electromagnetic wave incident process from the RF link to the first layer of the SIM can be equivalently represented as a transmission vector. At this time, the output signal of the last layer of the SIM for

[0011] (2)

[0012] After layer-by-layer phase modulation by the SIM, the equivalent transmitted signal passes through the downlink channel matrix. Transmit to There are one user. The received signal of a user can be represented as...

[0013] (3)

[0014] in for The power amplification factor of the transmitting antenna at any given time. This represents the power of the transmitting antenna.

[0015] To characterize the signal modulation and beamforming performance in multi-user scenarios, this invention uses the sum of the mean square error (MSE) between the received signal and the desired signal of each user as a performance metric, and minimizes this MSE as the optimization objective, constructing the following optimization problem.

[0016] (4)

[0017] The phase offset coefficients of each element in each layer can be obtained by solving the above optimization problem.

[0018] Utilizing a layer-by-layer modulation mechanism based on a multi-layer metasurface structure, the electromagnetic wave signal output by the single-RF link transmitter undergoes multiple phase modulations during propagation, thereby achieving signal modulation and precoding. Finally, the signal processed by the SIM is transmitted to... Individual users, enabling each user to receive only the signal they desire.

[0019] The beneficial effects of this invention are as follows: This invention proposes a novel communication system architecture based on stacked smart metasurfaces. This architecture requires only one radio frequency link. Electromagnetic waves are modulated and precoded through phase control of each layer of the SIM before being transmitted to the user, thereby significantly reducing the hardware complexity of the system. The solution is simple to implement and has strong application value. Attached Figure Description

[0020] Figure 1 : Schematic diagram of the system composition of this invention;

[0021] Figure 2 Comparison chart of distortion in different SIM systems;

[0022] Figure 3 Comparison of bit error rates for different SIM systems. Detailed Implementation

[0023] The present invention will now be described in detail with reference to the accompanying drawings and simulation examples.

[0024] This invention proposes a SIM-enabled massive MIMO communication system architecture. For example... Figure 1As shown, the system mainly comprises: a single-frequency signal generator oscillating at a carrier frequency, a multi-layer stacked smart metasurface, and... A single-antenna user terminal.

[0025] The system operates based on a joint optimization strategy: First, an optimization problem regarding the SIM phase offset is constructed with the minimum mean square error (MMSE) as the objective function, and the phase parameters are iteratively updated using the gradient descent algorithm. Simultaneously, the optimal power amplification gain of the RF link is dynamically solved using the least squares (LS) method. These two processes operate collaboratively through an alternating optimization mechanism until the system converges to the optimal transmission parameters. During signal transmission, the electromagnetic waves emitted by the RF source are incident on the first surface of the SIM. After cascaded phase modulation and wavefront reconstruction through multiple metasurfaces, efficient modulation and precoding of the signal are completed in the wave domain, ultimately achieving signal transmission. Precise beamforming and data transmission for single-antenna receiver users.

[0026] Based on the above description, the system serves... There are a set of single-antenna users, denoted as [a_n], [b ... The first The information of each user is encoded as a length of [length missing]. codeword sequence ,in Subsequently, for each time slot In a single-RF link transmitter architecture, the transmitter relies on known Channel State Information (CSI) and information symbol vectors.

[0027] (5)

[0028] Generation dimension is Transmitted signal vector .

[0029] like Figure 1 As shown, the SIM is composed of It is composed of stacked metasurfaces. Each layer is modeled as containing A uniform planar array (UPA) of adjustable elements, with the elements arranged in a square grid and the spacing between adjacent elements set to half the carrier wavelength. The total thickness of the SIM is denoted as... The spacing between two adjacent layers is At the hardware level, each unit of the SIM integrates an FPGA-based control module, capable of independently and dynamically adjusting the electromagnetic response characteristics at that location. Electromagnetic field propagation between adjacent SIM layers can be controlled by a transfer matrix. , To represent, where matrix elements Described from the first The first in the layer Unit 1 to the 1st The first in the layer The diffraction coupling relationship between the units can be precisely characterized by Rayleigh–Sommerfeld diffraction theory.

[0030] (6)

[0031] in, This represents the effective area of ​​a single unit. This represents the angle of incidence relative to the metasurface normal. For electromagnetic wavelengths, and Indicates from the first The first in the layer Unit 1 to the 1st The first in the layer Interlayer propagation distance between units.

[0032] Furthermore, signal propagation from the RF link to the SIM's first layer can be represented by the transmission vector. Characterized, among which the first element Indicates the transmitting antenna and the first layer of the SIM The channel coefficients corresponding to each unit are determined according to the diffraction relationship described in equation (6). Furthermore, the... The phase shift matrix of the layer can be expressed as

[0033] (7)

[0034] in Indicates the first The first in the layer The transmission coefficient of each unit, and its phase shift Values ​​are taken from The electromagnetic response of each metasurface layer can be determined by the transfer matrix. With phase shift matrix Both parts are characterized together. Therefore, after... After cascaded modulation of the metasurface, the output signal of the last layer of SIM It can be represented as

[0035] (8)

[0036] Subsequently, the signal reaches each user via the downlink channel. Indicates from the last layer of SIM to The channel matrix for each user. In the considered large-scale MIMO communication system, due to the compact deployment of SIM cells, the cells exhibit significant spatial correlation. In contrast, multi-user terminals typically have larger user spacing and operate in a scattering-rich environment, and their channels can generally be modeled as uncorrelated. Based on these characteristics, this invention adopts a spatially correlated channel model:

[0037] (9)

[0038] in, and Let represent the Rayleigh-log-normal fading channel matrix and the spatial correlation matrix, respectively.

[0039] Taking into account both Rayleigh fading and shadow fading, elements in (i.e., the first) Unit 1 to the 1st The channel coefficients of each user can be expressed as follows:

[0040] (10)

[0041] in, , Let be an independent and identically distributed (iid) complex Gaussian random variable with zero mean and variance of 1; This represents the shadow fading coefficient that follows a log-normal distribution. The path loss index; Denotes the normalized distance, where, The actual distance between the user and the transmitter. This is the minimum reference distance.

[0042] For the spatial correlation matrix , its first Line number Column element values ​​depend on the physical distance between cells. With wavelength The ratio of them follows the following relationship.

[0043] (11)

[0044] in, Indicates the first layer on the same metasurface The unit and the first The spacing between units.

[0045] During the signal transmission phase, it is assumed that the channel experiences flat fading, i.e., the channel matrix... It remains constant during the coherence time. Definition This refers to the single-frequency signal transmission power of the transmitter. This refers to the power amplification gain of the RF link. (During the transmission time slot) ,Include Received signal vector of each user It can be represented as

[0046] (12)

[0047] in, It is additive white Gaussian noise (AWGN) with a mean of zero and a variance of . That is, a noise sequence. They are independent and identically distributed, and for any All have .

[0048] The core objective of transmitter design is to synthesize the optimal transmit signal by jointly optimizing the phase parameters and RF gain of the SIM. This makes users Received signal As close as possible to the desired signal Therefore, the mean square error of the receiver is defined as follows:

[0049] (13)

[0050] Utilizing the statistical independence of AWGN, and in equation (7) Substituting into the above expression, we can obtain the simplified form of the receiver mean square error as follows:

[0051] (14)

[0052] in, This represents the total number of users in the system.

[0053] Therefore, it can be seen that in the transmission time slot Within this framework, the optimal phase shift parameters and amplification gain parameters can be obtained by solving the following optimization problem.

[0054] (15)

[0055] Because of the unit modulus constraint, the optimization problem described above is a nonconvex optimization problem. It should be noted that, for simplicity, symbols have been omitted in this problem description. and Time Index .

[0056] The following is a feasible solution to the above optimization problem, used to verify system performance.

[0057] The algorithm proposed in this invention calculates the partial derivatives of the loss function corresponding to the phase shift parameters and maintains the constant modulus structure of the variables throughout the iterative update process, thereby avoiding violation of the constraints. The iterative process of the algorithm is as follows:

[0058] Step 1: Calculate the loss function Regarding the first The first emission metasurface layer Phase shift parameters corresponding to each unit partial derivatives

[0059] , (16)

[0060] in, This indicates taking the imaginary part. for

[0061] (17)

[0062] Representation matrix The OK, Representation matrix etc. List, Representation matrix etc. OK;

[0063] Step 2: Normalize the above partial derivatives; the expression is as follows:

[0064] (18)

[0065] in, Indicates the first The maximum amplitude of the partial derivative of the phase shift of the emission metasurface layer.

[0066] Step 3: Update the phase shift parameters, the update formula is as follows:

[0067] (19)

[0068] in, This represents the learning rate, used to control the step size in each iteration. The learning rate is then updated.

[0069] (20)

[0070] This is a hyperparameter used to control the decay rate of the learning rate.

[0071] Step 4: After completing the gradient descent update above, the transmitted signal under the current iteration can be obtained. Based on this, the LS method is used to calculate the power amplification gain of the radio frequency link (RFC) according to equation (10).

[0072] ,(twenty one)

[0073] Through iterative updates of equations (16)- and (21), the loss function Gradually convergent, This indicates taking the real part.

[0074] The specific simulation parameters are set as follows: the output power is [single RF chain output power]. The operating wavelength is The corresponding carrier frequency is The thickness of the stacked smart metasurface is The distance between the RF link and the first metasurface layer is , Number of metasurface units per layer. Large-scale fading coefficient. The shadow fading components are modeled as zero-mean, independent and identically distributed log-normal shadow fading components, with a standard deviation of: The path loss index is set to... The reference distance is Furthermore, the first The actual distance between the user and the transmitter Follow the interval Uniform distribution on the surface. This invention employs... modulation method, noise power To evaluate the performance under the above system configuration, the distortion for each user is defined as follows:

[0075] ,(twenty two)

[0076] To ensure a fair comparison, we will use time slots Instantaneous transmit power within is defined as

[0077] .(twenty three)

[0078] Figure 2 Comparison of user distortion with user number The relationship between them, the number of units in each layer is The results show that increasing the number of SIM layers can significantly reduce system distortion, which fully demonstrates the effectiveness of the gains brought by inter-layer multiplexing in suppressing multi-user interference. Furthermore, compared with existing single-layer RIS systems, the gradient descent algorithm of this invention also achieves better results in single-layer systems.

[0079] Figure 3 This compares the SIM system with a user base of [number missing]. The effects of varying the number of layers and the number of units per layer on the bit error rate were investigated. The results showed that the bit error rate decreased significantly with increasing SIM layer number, and under the condition of a fixed total number of units, the two-layer SIM (M=49 per layer) also achieved a lower bit error rate than the single-layer SIM (M=100).

Claims

1. A SIM-enabled massive MIMO communication system, wherein the system adopts a single-radio link transmitter architecture, characterized in that, Stacked smart metasurfaces (SIM) including carrier transmitters and tunable reflective units Single-antenna users, where the SIM is provided by It consists of layers, each layer containing One adjustable reflection unit; The carrier transmitter outputs a carrier signal without baseband modulation to the SIM; The SIM sequentially performs phase modulation processing on the incident electromagnetic wave from the first layer to the last layer to obtain the transmitted signal, specifically: Define SIM The phase offset matrix of the layer is , , , No. layer to the first The inter-layer transmission matrix is The electromagnetic wave incident process from the carrier transmitter to the first layer of the SIM can be equivalently represented as a transmission vector. ,get Output signal of the last layer of the time SIM for: ， The user's received signal is: ， in, for The power amplification factor of the transmitting antenna at any given time. For transmitting antenna power, Indicates from the last layer of SIM to Channel matrix for each user It is additive white Gaussian noise; The method for obtaining the phase offset coefficients of each layer and unit in a SIM is to establish an optimization problem with the goal of minimizing the sum of the mean square error (MSE) between the received signal and the desired signal for each user. for: ， in, yes time The expected signal received by each user For the first The first in the layer Transmission coefficient of each unit For the first The first in the layer Phase shift of each unit, , , The phase offset coefficients of each unit in each layer can be obtained by solving the optimization problem.

2. The SIM-enabled massive MIMO communication system according to claim 1, characterized in that, The specific solution method for optimization problems is as follows: S1, Calculation Regarding the first The first emission metasurface layer Phase shift parameters corresponding to each unit partial derivatives : ， in, This indicates taking the imaginary part. , Representation matrix The OK, Representation matrix etc. List, Representation matrix etc. OK; S2. Normalize the above partial derivatives: ， in, Indicates the first The maximum magnitude of the partial derivative of the phase shift of the emission metasurface layer; S3, Update phase shift parameters: ， in, The learning rate is used to control the step size in each iteration; the learning rate is updated as follows: ， in, This is a hyperparameter used to control the decay rate of the learning rate; S4. After completing the gradient descent update, obtain the emission signal in the current iteration. ; Calculate the power amplification gain of the RF link: ； Through iterative updates from S1 to S4, The process gradually converges, yielding the phase offset coefficients of each unit in each layer of the SIM.

Images