A multi-layered metasurface device and a beamforming method

Abstract

The application provides a kind of multilayer intelligent metasurface device for modulating incident wireless signal and radiating to reflection end user and transmission end user simultaneously by reflection and transmission, comprising: two or more layers of intelligent metasurface.The application also provides a beam forming method of downlink multi-user communication system assisted by multilayer intelligent metasurface device, comprising the following steps: step 1, modeling the downlink multi-user communication system assisted by multilayer intelligent metasurface device, and modeling the optimization problem of multi-user and maximum rate;Step 2, convert the optimization problem of multi-user and maximum rate into the problem of minimizing weighted mean square error, for the problem of minimizing weighted mean square error, alternately optimize the receiving end coefficient matrix and the auxiliary weight coefficient matrix, the base station end digital beam forming design and the analog beam forming of different layers of intelligent metasurface, and finally obtain the base station end digital beam forming and the analog beam forming of different layers of intelligent metasurface.

Description

Technical Field

[0001] This invention belongs to the field of wireless technology, specifically relating to a multilayer intelligent metasurface device and a beamforming method. Background Technology

[0002] In recent years, with the global proliferation of commercial fifth-generation (5G) wireless communication networks, an increasing number of scientists are dedicated to researching upcoming fifth-generation beyond fifth-generation (B5G) and sixth-generation (6G) wireless communication network technologies, including higher carrier frequencies, lower power consumption, microsecond-level latency, and full-dimensional network coverage. To meet these stringent technical requirements, many new technologies have been proposed in recent years, including ultra-large-scale multiple-input multiple-output (UM-MIMO), ultra-dense networks (UDNs), and terahertz (THz) communication technologies. However, the increased number of base station antennas leads to higher power consumption and hardware costs, while the higher carrier frequencies result in higher path losses, further increasing power consumption. This makes the application of these technologies in next-generation wireless communication networks extremely challenging.

[0003] Due to the rapid development of metasurface-related technologies, reconfigurable smart surfaces (RIS) have attracted widespread attention from the scientific and industrial communities, becoming a promising and important solution for 6G wireless network technology. Smart metasurfaces consist of numerous passive subwavelength scattering units with pin junctions distributed on their surfaces. By controlling the on / off state of these pin junctions, the scattering units can control the amplitude and phase of the signal. Therefore, if the states of different scattering units can be properly controlled, the wireless environment can be reconfigured, improving system performance. Compared to traditional relay communication, the scattering units of smart reflectors are passive devices, eliminating the need for an RF chain, thus resulting in lower power consumption and cost.

[0004] In previous research, the most well-known RIS model is the Intelligent Reflector (IRS), which consists of three layers: a scattering unit layer, a metal base plate, and a control circuit board. It can only reflect incident signals to the corresponding user. Therefore, if the user and the base station are located on opposite sides of the IRS, the IRS will no longer be able to serve the user.

[0005] To address this issue, a new RIS model called Intelligent Omnidirectional Metasurface (IOS) has recently been proposed, which can serve users on both sides of the metasurface simultaneously. Specifically, taking the following example, signals incident on both sides of the IOS can be simultaneously reflected and refracted to users at both the reflecting and transmitting ends. This dual function of simultaneously reflecting and refracting signals allows the IOS to extend wireless coverage throughout the entire space, thereby significantly improving service quality.

[0006] Despite the advantages mentioned above, a major drawback of existing IOS designs is that the scattering unit cannot independently control the phase shift of the reflected and transmitted signals. Therefore, the beamforming of users on both sides of the smart reflector is inevitably coupled together, which significantly impacts the performance of the communication system. Thus, an IOS design that is simple in design and has independent degrees of freedom for controlling the transmitted and reflected signals is of great significance. Summary of the Invention

[0007] This invention is made to solve the above-mentioned problems, and aims to provide a multilayer intelligent metasurface device and a beamforming method.

[0008] This invention provides a multilayer smart metasurface device for modulating incident wireless signals and simultaneously radiating them to both a reflecting end user and a transmitting end user through reflection and transmission. The device comprises two or more smart metasurfaces, each consisting of multiple scattering units and a digital control module connected to the scattering units. Wireless signals emitted by a base station are reflected by the first smart metasurface to communicate with the reflecting end user. Wireless signals emitted by the base station are transmitted multiple times through all smart metasurfaces to communicate with the transmitting end user. The shaped beams of the reflecting end user and the transmitting end user are independently controlled by controlling the states of different scattering units.

[0009] This invention also provides a beamforming method for a downlink multi-user communication system assisted by a multilayer intelligent metasurface device, characterized by the following steps:

[0010] Step 1: Model the downlink multi-user communication system assisted by the multi-layer intelligent metasurface device, and model the optimization problem of maximizing multi-user and speed.

[0011] Step 2 transforms the optimization problem of maximizing the number of users and the maximum data rate into a problem of minimizing the weighted mean square error. For minimizing the weighted mean square error, the receiver coefficient matrix and auxiliary weight coefficient matrix, the base station digital beamforming design, and the simulated beamforming of different layers of intelligent metasurfaces are alternately optimized. This ultimately yields the base station digital beamforming and the simulated beamforming of different layers of intelligent metasurfaces.

[0012] The downlink multi-user communication system assisted by the multi-layer intelligent metasurface device includes a base station with multiple antennas, a multi-layer intelligent metasurface device for receiving wireless signals from the base station and simultaneously reflecting and transmitting them, and multiple users, including reflective end users and transmission end users. The optimization of simulated beamforming is based on the coordinate iterative descent algorithm.

[0013] The beamforming method for a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device provided by this invention may also have the following feature: Step 1 includes the following sub-steps:

[0014] Step 1-1, the coefficient matrix of a single-layer smart metasurface is expressed as follows:

[0015]

[0016] Steps 1-2: Based on the coefficient matrix of a single-layer smart metasurface, the equivalent coefficient matrix expression of a multi-layer smart metasurface is obtained as follows:

[0017]

[0018] Steps 1-3, the mathematical expression of the base station's transmitted signal is as follows:

[0019] x = Vs (3);

[0020] Steps 1-4, the mathematical expression of the user's received signal is as follows:

[0021]

[0022] Steps 1-5, the mathematical form of the user-side speed is as follows:

[0023]

[0024] Steps 1-6, the mathematical form of the optimization problem of maximizing the sum of multiple users and the rate is as follows:

[0025]

[0026] subject to Tr(VV H )≤P b

[0027]

[0028]

[0029] In formula (1), M represents the number of scattering units in a single-layer smart metasurface, ∈ μ This represents the intensity coefficients of the reflected and transmitted signals of a single-layer smart metasurface. When μ = r, it represents reflection; when μ = t, it represents transmission. t +∈ r =1, φ m This represents the phase shift of the incident signal produced by the m-th scattering unit.

[0030] In formula (2), These are the diagonal coefficient matrices corresponding to the first and second layer smart metasurfaces, respectively. μ ξ μThese are the intensity coefficients of the reflected and transmitted signals of the first and second smart metasurfaces, respectively. The channel matrix between the two smart metasurface layers.

[0031] In formula (3), Represents the precoding matrix, This represents the information symbols sent to each user, where N represents the number of base station antennas and K represents the number of users.

[0032] In formula (4), when k = 1, ..., K r When μ(k) = r, when k = K r +1, ..., K r +K t At that time, μ(k) = t, Let nk represent the channel coefficient vectors from the base station to the multilayer intelligent metasurface device and from the multilayer intelligent metasurface device to the k-th user, respectively. Let nk represent the Gaussian noise at the k-th user, with a mean of 0 and a variance of σ. 2 .

[0033] In formula (5), This represents the equivalent total noise and interference power at the k-th user.

[0034] In formula (6), the first constraint states that the total transmit power of the base station must not exceed P. b The second and third constraints represent the constant mode constraints of the scattering unit in the multilayer intelligent metasurface device. The scattering unit only changes the phase of the incident signal, but does not change the amplitude of the incident signal.

[0035] The beamforming method for a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device provided by this invention may also have the following feature: step 2 includes the following sub-steps:

[0036] Step 2-1: Using the relationship between mean square error and rate, the optimization problem of maximizing the sum of multiple users and rates is transformed into an equivalent problem of minimizing the weighted mean square error. The mathematical form of the problem of minimizing the weighted mean square error is as follows:

[0037]

[0038] subject to Tr(VV H )≤P b

[0039]

[0040]

[0041] In formula (7), Indicates the magnitude of the MSE value, w k Let ψ represent the receiver coefficients of the k-th user terminal. k An auxiliary weighting coefficient is used to establish the equivalence between the mean squared error (MSE) and the rate. The closed-form solutions are as follows: ψ k =e k -1 ,in,

[0042] Furthermore, the receiver coefficient matrix is ​​defined as Ψ = diag(ψ1, ..., ψ). K The auxiliary weight coefficient matrix W = diag(w1, ..., w) K E = diag(e1, ..., e) K Substituting ψ and W into the closed-form solution, the mathematical form of the problem of minimizing the weighted mean square error is transformed into:

[0043]

[0044] subject to Tr(VV H )≤P b

[0045]

[0046]

[0047]

[0048] Step 2-2: Optimize beamforming at the base station. By fixing Φ and Θ, optimize the digital precoding V. According to the KKT conditions, under power constraints, the optimal solution for V has the following closed-form solution:

[0049]

[0050] Steps 2-3: Optimize the first-layer smart metasurface. By fixing other variables, optimize the phase shift coefficient matrix Φ of the first-layer smart metasurface, defining... φ = diag(Φ), and apply the identity Tr(C) H ACB)=c H (A⊙B T c and Tr(AC) = a T c, the objective function simplifies to:

[0051] f = φ H Pφ-q H φ-φ H q (11)

[0052] To extract The objective function for the contribution of f is further transformed into:

[0053]

[0054] get The optimal solution: in By alternately optimizing each element using the coordinate iterative descent algorithm, the local optimal solution of Φ is obtained;

[0055] Steps 2-4: Optimize the smart metasurface of the second layer. By fixing other variables, optimize the phase shift coefficient matrix Θ of the smart metasurface of the second layer, and decompose H into two parts related to the reflecting user and the transmitting user: in, After ignoring constants that are independent of Θ, the objective function f simplifies to:

[0056] f = θ H P′θ-q′ H θ-θ H q′ (13)

[0057] In formula (13), θ = diag(Θ), The local optimal solution of Θ is obtained by the coordinate iterative descent algorithm;

[0058] Steps 2-5 alternately optimize the receiver coefficient matrix W, the auxiliary weight coefficient matrix Ψ, the base station's digital precoding matrix V, the phase shift coefficient matrix Φ of the first-layer smart metasurface, and the phase shift coefficient matrix Θ of the second-layer smart metasurface. If the termination requirement is met, output the base station's digital beamforming and the simulated beamforming of the smart metasurfaces at different layers. If the termination requirement is not met, repeat steps 2-2 to 2-4.

[0059] In formula (10),

[0060] The role and effect of invention

[0061] According to the present invention, a multilayer smart metasurface device modulates an incident wireless signal and radiates it simultaneously to both the reflecting and transmitting users through reflection and transmission. Each smart metasurface can adjust the phase of the wireless signal through a scattering unit, thereby enabling independent control of beamforming at both the reflecting and transmitting ends. Furthermore, according to the present invention, a beamforming method for a downlink multi-user communication system assisted by a multilayer smart metasurface device first transforms the complex maximization and spectral efficiency optimization problem into a problem of minimizing the weighted mean square error. Then, for the transformed problem, the receiver coefficient matrix and auxiliary weight coefficient matrix, the base station digital beamforming design, and the simulated beamforming of different layers of smart metasurfaces are alternately optimized. Finally, the base station digital beamforming and the simulated beamforming of different layers of smart metasurfaces are obtained. The present invention provides a low-hardware-cost, low-power beamforming method that can independently control the simulated beamforming on both sides, improving the spectral efficiency of users on both sides of the multilayer smart metasurface device. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the structure of the multilayer intelligent metasurface device in an embodiment of the present invention;

[0063] Figure 2 This is a flowchart of a beamforming method for a downlink multi-user communication system assisted by a multilayer intelligent metasurface device in an embodiment of the present invention;

[0064] Figure 3 This is a schematic diagram of beamforming of a single-layer smart metasurface in an embodiment of the present invention;

[0065] Figure 4 This is a schematic diagram of beamforming of a dual-layer smart metasurface in an embodiment of the present invention;

[0066] Figure 5 This is a schematic diagram of a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device in an embodiment of the present invention;

[0067] Figure 6 This is the relationship between transmission power and user rate when the number of scattering units, the number of transmitting users, and the number of reflecting users are fixed in the embodiments of the present invention.

[0068] Figure 7 This is an embodiment of the invention showing the relationship between the number of scattering units of the smart metasurface and the number of users and the rate, under the condition of fixed emission rate, number of transmitting users, and number of reflecting users;

[0069] Figure 8This is an embodiment of the invention where, under the conditions of fixed emission rate, total number of users, and number of scattering units of the smart metasurface, the number of transmitting users is related to the number of users and the emission rate. Detailed Implementation

[0070] To make the technical means and effects of the present invention easy to understand, the present invention will be specifically described below in conjunction with embodiments and accompanying drawings.

[0071] <Example>

[0072] Figure 1 This is a schematic diagram of the structure of the multilayer intelligent metasurface device in an embodiment of the present invention.

[0073] like Figure 1 As shown, this embodiment of a multilayer smart metasurface device is used to modulate incident wireless signals and radiate them to both the reflecting end user and the transmitting end user simultaneously through reflection and transmission, comprising: two or more layers of smart metasurface.

[0074] This embodiment uses a dual-layer multilayer smart metasurface device (BIOS), which has a dual-layer smart metasurface, including a first smart metasurface (IOS1) and a second smart metasurface (IOS2).

[0075] Each layer of the intelligent metasurface consists of multiple scattering units and a digital control module connected to the scattering units.

[0076] The wireless signal emitted by the base station communicates with the user at the reflecting end after being reflected by the first layer of smart metasurface. The wireless signal emitted by the base station communicates with the user at the transmitting end after being transmitted multiple times through all the smart metasurfaces.

[0077] The shaped beams of the reflecting and transmitting users are independently controlled by controlling the state of different scattering units.

[0078] In this embodiment, each scattering unit can phase-shift the incident wireless signal and transmit it simultaneously at both the transmission and reflection ends. The two smart metasurfaces are very close together and are fixed at the factory; therefore, the communication channel between the two smart metasurfaces only has a direct path and no scattering path. For the reflection end, the user signal will only be reflected by the first smart metasurface, while for the transmission end, the user signal will be transmitted through both smart metasurfaces. Therefore, by appropriately setting the on / off state of the PIN junctions of all scattering units on the two smart metasurfaces, independent beamforming designs for the transmission and reflection ends can be achieved.

[0079] Figure 2 This is a flowchart of a beamforming method for a downlink multi-user communication system assisted by a multilayer intelligent metasurface device, as described in an embodiment of the present invention.

[0080] like Figure 2 As shown in this embodiment, a beamforming method for a downlink multi-user communication system assisted by a multilayer intelligent metasurface device includes the following steps:

[0081] Step 1: Model the downlink multi-user communication system assisted by the multi-layer intelligent metasurface device, and model the optimization problem of maximizing the number of users and the maximum speed.

[0082] Step 1 includes the following sub-steps:

[0083] Step 1-1: For a single-layer smart metasurface, it can be represented as a diagonal matrix in a communication system. For a smart metasurface with M scattering units, the coefficient matrix of a single-layer smart metasurface is expressed as follows:

[0084]

[0085] Steps 1-2: For a two-layer multilayer smart metasurface device, since the signal received by the user at the reflecting end is only reflected by the first layer of the smart metasurface, while the signal received by the user at the transmitting end is modulated by the transmission of both layers of the smart metasurface, the equivalent coefficient matrix of the multilayer smart metasurface is expressed as follows based on the coefficient matrix of the single-layer smart metasurface:

[0086]

[0087] In this embodiment, the base station in the downlink multi-user communication system assisted by a multilayer intelligent metasurface device (BIOS) has N antennas deployed, and the users are all single-antenna nodes, totaling K, including K1 and K2. r One reflector user and K t One transmission end user. In this embodiment, the users are numbered 1, ..., K. r Assigned to the reflection user, and the number K. r +1, ..., K r +K t Assigned to transmission users.

[0088] Steps 1-3, from the perspective of equivalent baseband representation, the mathematical expression of the base station's transmitted signal is as follows:

[0089] x = Vs (3);

[0090] Steps 1-4, the mathematical expression of the user's received signal is as follows:

[0091]

[0092] In steps 1-5, due to severe congestion, the direct channel between the user and the base station is negligible. Therefore, the mathematical form of the receiving rate for the k-th user is as follows:

[0093]

[0094] Steps 1-6: Given the magnitude of the receiving rate and the coefficient matrices Φ and Θ of the multilayer intelligent metasurface device and the precoding matrix V at the base station, the mathematical form of the optimization problem for maximizing the multi-user and rate maximization is as follows:

[0095]

[0096] subject to Tr(VV H )≤P b

[0097]

[0098]

[0099] In formula (1), M represents the number of scattering units in a single-layer smart metasurface, ∈ μ This represents the intensity coefficients of the reflected and transmitted signals of a single-layer smart metasurface. When μ = r, it represents reflection; when μ = t, it represents transmission. Since the smart metasurface is a passive element, it must satisfy the constraint of energy conservation. The sum of the energies of the transmitted and reflected signals equals the incident signal (ignoring the energy loss of the scattering unit), i.e., ∈ t +∈ r =1, φ m This represents the phase shift of the incident signal produced by the m-th scattering unit. In this embodiment, it is assumed that the phase shift produced by the scattering unit for the transmitted and reflected signals is the same.

[0100] In formula (2), These are the diagonal coefficient matrices corresponding to the first and second layer smart metasurfaces, respectively. μ ·ξ μ These are the intensity coefficients of the reflected and transmitted signals of the first and second smart metasurfaces, respectively. The channel matrix between the two smart metasurfaces is different for the beamforming matrices on the transmission and reflection sides of the multilayer smart metasurface device, which can realize separate control of beamforming at the transmission and reflection ends.

[0101] In formula (3), Represents the precoding matrix, This represents the information symbols sent to each user, where N represents the number of base station antennas and K represents the number of users.

[0102] In formula (4), when k = 1, ..., K r When μ(k) = r, when k = K r +1, ..., K r +K tAt that time, μ(k) = t, Let n represent the channel coefficient vectors from the base station to the multilayer intelligent metasurface device and from the multilayer intelligent metasurface device to the k-th user, respectively. k This represents the Gaussian noise at the k-th user terminal, with a mean of 0 and a variance of σ. 2 .

[0103] In formula (5), This represents the equivalent total noise and interference power at the k-th user.

[0104] In formula (6), the first constraint states that the total transmit power of the base station must not exceed P. b The second and third constraints represent the constant mode constraints of the scattering unit in the multilayer intelligent metasurface device. The scattering unit only changes the phase of the incident signal, but does not change the amplitude of the incident signal.

[0105] Step 2: Since the optimization problem is highly non-convex and involves the coupling of multiple optimization variables, it is very difficult to solve the original problem directly. The optimization problem of multiple users and the highest rate is transformed into the problem of minimizing the weighted mean square error. For the problem of minimizing the weighted mean square error, the receiver coefficient matrix and the auxiliary weight coefficient matrix, the digital beamforming design of the base station, and the simulated beamforming of the intelligent metasurface of different layers are alternately optimized. Finally, the digital beamforming of the base station and the simulated beamforming of the intelligent metasurface of different layers are obtained.

[0106] The optimization of simulated beamforming is based on the coordinate iterative descent algorithm.

[0107] Step 2 includes the following sub-steps:

[0108] Step 2-1: Using the relationship between mean square error and rate, the optimization problem of maximizing the sum of multiple users and rates is transformed into an equivalent problem of minimizing the weighted mean square error. The mathematical form of the problem of minimizing the weighted mean square error is as follows:

[0109]

[0110] subject to Tr(VV H )≤P b

[0111]

[0112]

[0113] In formula (7), Indicates the magnitude of the MSE value, w k Let ψ represent the receiver coefficients of the k-th user terminal. kAn auxiliary weighting coefficient is used to establish the equivalence between the mean squared error (MSE) and the rate. The closed-form solutions are as follows: ψ k =e k -1 ,in,

[0114] Furthermore, the receiver coefficient matrix is ​​defined as Ψ = diag(ψ1, ..., ψ). K The auxiliary weight coefficient matrix W = diag(w1, ..., w) K E = diag(e1, ..., e) K Substituting Ψ and W into the closed-form solution, the mathematical form of the problem of minimizing the weighted mean square error is transformed into:

[0115]

[0116] subjectto Tr(VV H )≤P b

[0117]

[0118]

[0119]

[0120] Step 2-2: Optimize beamforming at the base station. By fixing Φ and Θ, optimize the digital precoding V. According to the KKT conditions, under power constraints, the optimal solution for V has the following closed-form solution:

[0121]

[0122] Steps 2-3 optimize the first-layer smart metasurface by fixing other variables and optimizing the phase shift coefficient matrix Φ of the first-layer smart metasurface. Since its elements satisfy the constant modulus constraint, the problem remains non-convex, making it difficult to obtain the global optimum. However, because this constraint is decoupled for different matrix elements, a coordinate iterative descent algorithm can be used to find the local optimum. Definition φ = diag(Φ), and apply the identity Tr(C) H ACB)=c H (A⊙B T c and Tr(AC) = a T c(A and B are arbitrary matrices, C is an arbitrary diagonal matrix, a = diag(A), c = diag(C)), the objective function simplifies to:

[0123] f = φ H Pφ-q H φ-φH q (11)

[0124] To extract The objective function for the contribution of f is further transformed into:

[0125]

[0126] get The optimal solution: in By alternately optimizing each element using the coordinate iterative descent algorithm, the local optimal solution of Φ is obtained;

[0127] Steps 2-4: Optimize the second-layer smart metasurface. By fixing other variables, optimize the phase shift coefficient matrix Θ of the second-layer smart metasurface. Since only the rate of the transmitting end user is related to the second-layer smart metasurface, H is split into two parts related to the reflecting user and the transmitting user: in, After ignoring constants that are independent of Θ, the objective function f simplifies to:

[0128] f = θ H P′θ-q′ H θ-θ H q′ (13)

[0129] In formula (13), θ = diag(Θ), The local optimal solution of Θ is obtained by the coordinate iterative descent algorithm;

[0130] Steps 2-5 alternately optimize the receiver coefficient matrix W, the auxiliary weight coefficient matrix ψ, the base station's digital precoding matrix V, the phase shift coefficient matrix Φ of the first-layer smart metasurface, and the phase shift coefficient matrix Θ of the second-layer smart metasurface. If the termination requirement is met, output the base station's digital beamforming and the simulated beamforming of the smart metasurfaces at different layers. If the termination requirement is not met, repeat steps 2-2 to 2-4.

[0131] In formula (10),

[0132] In this embodiment, the beamforming method of the downlink multi-user communication system assisted by multi-layer intelligent omnidirectional surfaces can be simply derived from the beamforming method of the downlink multi-user communication system assisted by dual-layer intelligent omnidirectional surfaces.

[0133] In this embodiment, beamforming on both sides of a single-layer intelligent metasurface (IOS) and a double-layer intelligent metasurface (BIOS) is analyzed. Figure 3 This is a schematic diagram of beamforming of a single-layer smart metasurface in an embodiment of the present invention.

[0134] like Figure 3 As shown, for a single-layer smart metasurface, the beamforming on both sides is highly correlated because the phase shifts of the transmitted and reflected signals are coupled for each scattering unit. Specifically, a typical smart metasurface model assumes that the phase shifts of the reflected and transmitted signals are consistent across the scattering units; under this model, the beams on both sides of the smart metasurface will be identical. When users are randomly distributed on both sides of the smart metasurface, this could potentially result in some beams not pointing at any user, causing significant energy loss.

[0135] Figure 4 This is a schematic diagram of beamforming of a dual-layer smart metasurface in an embodiment of the present invention.

[0136] like Figure 4 As shown, for a dual-layer smart metasurface, since the signal received by the user at the reflecting end is obtained by reflection from the first layer of the smart metasurface, while the signal received by the user at the transmitting end will be transmitted through both layers of the smart metasurface, the dual-layer smart metasurface can achieve different beamforming on both sides, reducing energy loss.

[0137] Figure 5 This is a schematic diagram of a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device in an embodiment of the present invention.

[0138] like Figure 5 As shown, both smart metasurfaces are uniform square planar arrays with M scattering units. The first smart metasurface (IOS1) transmits (∈ r ) and reflection intensity coefficient (∈ t The values ​​are all 0.5, and the second layer of intelligent metasurface (IOS2) is set to full transmittance (ξ). r =0, ξ t =1). The base station is equipped with a uniform linear antenna array of 8 antennas, while the user ends are all single-antenna nodes. The spacing between all antenna arrays and scattering element arrays is half a wavelength, with no gaps between elements. The height of both the base station and the two-layer multilayer intelligent metasurface device (BIOS) is 5m, with a spacing of 20m between them. The spacing between the two layers of the multilayer intelligent metasurface device is 0.015m. Users are randomly distributed within a range of 3m to 10m from the multilayer intelligent metasurface device at a height of 1.5m. The Gaussian noise variance σ at the user end is... 2 = -80dBm.

[0139] Channel (G) from base station to multilayer smart metasurface device and channel from multilayer smart metasurface device to user terminal. Ricean channel model is used in all cases. For the channel (F) between the two smart metasurfaces, since the spacing between the two smart metasurfaces is extremely small, the channel between each scattering unit adopts a near-field channel model dominated by the direct path. In addition, in the numerical simulation, it is assumed that the direct path from the base station to the user can be ignored due to strong congestion.

[0140] To compare the performance of different smart metasurfaces, this embodiment performs numerical simulations on a two-layer multilayer smart metasurface device (BIOS), a smart reflective surface (IRS), and a single-layer smart metasurface (IOS). The position and number of scattering units of the smart reflective surface (IRS) are the same as the first layer IOS in the BIOS. The IRS can only reflect signals and therefore cannot serve users at the transmission end. The transmission and reflection intensities of the single-layer IOS are both set to 0.5, and the specific process is as follows:

[0141] When the number of scattering units of the smart metasurface is fixed Transmission user number (K) t The number of reflected users (K) is 3. r When the value is 2, the relationship between the transmit power and the user and rate in the three systems is obtained through numerical simulation. Figure 6 This describes the relationship between transmission power and user rate when the number of scattering units, the number of transmitting users, and the number of reflecting users are fixed in the embodiments of the present invention.

[0142] like Figure 6 As shown, for all three systems, the number of users and the data rate increase with increasing transmit power. However, the BIOS and IOS systems exhibit a significant performance advantage over the IRS system. This is because the BIOS and IOS can serve users on both sides simultaneously, while the IRS can only serve users on the reflector side. Furthermore, since the BIOS can implement different beamforming on both sides, its performance is superior to that of the IOS.

[0143] When the fixed emission rate (P) b =30dBm), number of transmitted users (K) t The number of reflected users (K) is 3. r When the value is 2, the number of scattering units of the smart metasurface in the three systems is obtained through numerical simulation. The relationship between users and speed. Figure 7 This is an embodiment of the invention showing the relationship between the number of scattering units of the smart metasurface and the number of users and the emission rate, with a fixed emission rate, number of transmitting users, and number of reflecting users.

[0144] like Figure 7As shown, increasing the scattering units of the smart metasurface improves beamforming gain, leading to increased user capacity and speed across all three systems. Meanwhile, the BIOS exhibits a significant performance advantage over the IOS and IRS.

[0145] When the fixed emission rate (P) b =20dBm), total number of users (K=5), number of scattering units of the smart metasurface In this case, the number of transmission users (K) in the three systems is obtained through numerical simulation. t The relationship between users and speed. Figure 8 This is an embodiment of the invention where, under the conditions of fixed emission rate, total number of users, and number of scattering units of the smart metasurface, the number of transmitting users is related to the number of users and the emission rate.

[0146] like Figure 8 As shown, for the IRS system, since it can only serve users on the reflecting end, its performance will monotonically decrease to 0 when more users are distributed on the transmitting end. For the IOS system, since both the transmission and reflection intensity coefficients are 0.5, the IOS's ability to serve users on both sides is completely consistent, so the performance of the IOS system hardly changes with the number of transmitting users. For the BIOS system, due to the presence of the second-layer IOS, the BIOS can control the beamforming on both sides almost independently, thus having better performance than the other two systems. Since the transmission and reflection coefficients of the first-layer IOS are both 0.5, the power allocated to the reflecting and transmitting ends is almost identical. Therefore, when users are evenly distributed on both sides of the BIOS (K... t =2 or 3), the number of users and the speed of the BIOS system reach their maximum value. Additionally, when the number of users at the transmission end is large (K... t =4 or 5), the BIOS system still has a significant advantage in terms of users and speed compared to IOS. This is because for users at the transmission end, their signals can be transmitted through two layers of IOS, and their beamforming quality is better than that at the reflection end.

[0147] In summary, in this embodiment, by using the multi-layer intelligent metasurface device (BIOS) of the present invention to assist in a multi-user multiple-input single-output scenario, better performance is achieved compared to IOS and IRS systems, and the performance difference is maximized when users are evenly distributed on both sides of the BIOS.

[0148] The role and effect of the embodiments

[0149] According to a multilayer smart metasurface device involved in this embodiment, the incident wireless signal is modulated and radiated simultaneously to both the reflecting and transmitting users through reflection and transmission. Each smart metasurface can adjust the phase of the wireless signal through a scattering unit, thereby enabling independent control of beamforming at both the reflecting and transmitting ends. Furthermore, according to a beamforming method for a downlink multi-user communication system assisted by a multilayer smart metasurface device in this embodiment, firstly, the original complex maximization and spectral efficiency optimization problem is equivalently transformed into a problem of minimizing the weighted mean square error. Then, for the transformed problem, the receiver coefficient matrix and auxiliary weight coefficient matrix, the base station digital beamforming design, and the simulated beamforming of different layers of smart metasurfaces are alternately optimized. Finally, the base station digital beamforming and the simulated beamforming of different layers of smart metasurfaces are obtained. This embodiment provides a beamforming method with low hardware cost and low power consumption that can independently control the simulated beamforming on both sides, improving the spectral efficiency of users on both sides of the multilayer smart metasurface device.

[0150] The above embodiments are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention.

Claims

1. A beamforming method for a downlink multi-user communication system assisted by a multilayer intelligent metasurface device, characterized in that, The multilayer smart metasurface device is used to modulate incident wireless signals and radiate them simultaneously to both the reflecting and transmitting users. The multilayer smart metasurface device comprises two or more smart metasurfaces. Each layer of the intelligent metasurface consists of multiple scattering units and a digital control module connected to the scattering units. The wireless signal emitted by the base station communicates with the user at the reflecting end through reflection via the first layer of the smart metasurface. The wireless signal emitted by the base station communicates with the user at the transmission end after being transmitted multiple times through all the smart metasurfaces. The shaping beams of the reflecting end user and the transmitting end user are independently controlled by controlling the state of different scattering units. The method includes the following steps: Step 1: Model the downlink multi-user communication system assisted by the multi-layer intelligent metasurface device, and model the optimization problem of maximizing multi-user and speed. Step 2: The optimization problem of maximizing the number of users and the maximum rate is transformed into a problem of minimizing the weighted mean square error. For the problem of minimizing the weighted mean square error, the receiver coefficient matrix and auxiliary weight coefficient matrix, the digital beamforming design at the base station, and the simulated beamforming of the smart metasurface at different layers are alternately optimized to finally obtain the digital beamforming at the base station and the simulated beamforming of the smart metasurface at different layers. The downlink multi-user communication system assisted by the multi-layer intelligent metasurface device includes a base station with multiple antennas, the multi-layer intelligent metasurface device for receiving wireless signals from the base station and simultaneously reflecting and transmitting them, and multiple users, including reflecting end users and transmitting end users. The optimization of the simulated beamforming is based on the coordinate iterative descent algorithm.

2. The beamforming method for a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device according to claim 1, characterized in that: in, Step 1 includes the following sub-steps: Step 1-1, the coefficient matrix of the single-layer smart metasurface is expressed as follows: ， In the above formula, This represents the coefficient matrix of the single-layer intelligent metasurface. This represents the diagonal coefficient matrix of the intelligent metasurface in the first layer. Indicates the intensity coefficients of the reflected and transmitted signals of the single-layer smart metasurface, when Time indicates reflection, when Time indicates transmission. , Indicates the first The phase shift of the incident signal by each scattering unit This indicates the number of scattering units in the single-layer intelligent metasurface. , Represents the imaginary unit; Steps 1-2: Based on the coefficient matrix of the single-layer smart metasurface, the equivalent coefficient matrix expression of the multilayer smart metasurface is obtained as follows: ， ， In the above formula, This represents the equivalent coefficient matrix of the multilayer smart metasurface in reflection mode. This represents the equivalent coefficient matrix of the multilayer smart metasurface in transmission mode. This is the diagonal coefficient matrix corresponding to the second layer of intelligent metasurface. , These represent the intensity coefficients of the reflected and transmitted signals of the intelligent metasurface in the first layer, respectively. This represents the intensity coefficient of the transmitted signal of the intelligent metasurface in the second layer. The channel matrix between the two smart metasurface layers. Represents the set of complex numbers; Steps 1-3, the mathematical expression of the base station's transmitted signal is as follows: ， In the above formula, Represents a digital precoding matrix. Symbols representing information sent to each user. Indicates the number of base station antennas. Indicates the number of users; Steps 1-4, the mathematical expression of the user's received signal is as follows: ， In the above formula, Indicates the first The signal received by each user, when hour, ,when hour, , , These respectively represent the connection from the base station to the multilayer intelligent metasurface device and the connection from the multilayer intelligent metasurface device to the first... Channel coefficient vectors for each user Indicates the first The user terminals have Gaussian noise with a mean of 0 and a variance of . ; Steps 1-5, the mathematical form of the user-side speed is as follows: ， In the above formula, Indicates the first User terminal speed, Indicates the first Equivalent noise and total interference power at each user location Indicates the first The base station precoding vector corresponding to each user; Steps 1-6, the mathematical form of the optimization problem of maximizing the sum of multiple users and the rate is as follows: ， In the above formula, the first constraint states that the total transmit power of the base station must not exceed [a certain value]. The second and third constraints represent the constant mode constraints of the scattering unit in the multilayer intelligent metasurface device. The scattering unit only changes the phase of the incident signal, but does not change the amplitude of the incident signal.

3. The beamforming method for a downlink multi-user communication system assisted by a multi-layer intelligent metasurface device according to claim 2, characterized in that: in, Step 2 includes the following sub-steps: Step 2-1: Using the relationship between mean square error and rate, the optimization problem of maximizing the sum of multiple users and rates is transformed into the equivalent problem of minimizing the weighted mean square error. The mathematical form of the problem of minimizing the weighted mean square error is as follows: ， In the above formula, Indicates the magnitude of the MSE value. Indicates the first Receiver coefficients at each user terminal An auxiliary weighting coefficient is used to establish the equivalence between the mean squared error (MSE) and the rate. The closed-form solutions are as follows: , ,in, , Define the receiving end coefficient matrix Auxiliary weight coefficient matrix , and substitute , After obtaining the closed-form solution, the mathematical form of the problem of minimizing the weighted mean square error is transformed into: ， ； Step 2-2, optimize the beamforming at the base station end by fixing and Optimize digital precoding According to the KKT conditions, under power constraints, The optimal solution has the following closed-form solution: ， In the above formula, , , express An identity matrix of dimensionality; Steps 2-3: Optimize the smart metasurface of the first layer by fixing other variables and optimizing the diagonal coefficient matrix of the smart metasurface of the first layer. ,definition , and use identities and The objective function simplifies to: ， To extract right The objective function is further transformed to reflect the contribution of the objective function: ， get The optimal solution: ,in By alternately optimizing each element using a coordinate iterative descent algorithm, we obtain... The local optimal solution; Steps 2-4: Optimize the smart metasurface of the second layer by fixing other variables and optimizing the diagonal coefficient matrix of the smart metasurface of the second layer. ,Will It can be broken down into two parts related to reflection users and transmission users: ,in, , ignoring After irrelevant constants, the objective function Simplified to: ， In the above formula, , , , , , The coordinate iterative descent algorithm is used to obtain The local optimal solution; Steps 2-5: Alternately optimize the receiver coefficient matrix. The auxiliary weight coefficient matrix Digital precoding matrix of base station The diagonal coefficient matrix of the first layer of the intelligent metasurface and the phase shift coefficient matrix of the intelligent metasurface in the second layer. If the termination requirement is met, the digital beamforming of the base station and the simulated beamforming of the smart metasurface of different layers are output. If the termination requirement is not met, steps 2-2 to 2-4 are repeated.

Images