A multi-user transmission method assisted by multi-intelligent metasurfaces

Abstract

This invention discloses a multi-user transmission method assisted by multiple intelligent metasurfaces. This method is applicable to a single-cell downlink multi-user transmission system. First, the phase shift matrix of each RIS is fixed, and the base station's transmit precoding matrix for each user is designed. Then, the base station's transmit precoding matrix for each user is fixed, and the phase shift matrix of each RIS is designed. The transmit precoding matrix and the phase shift matrix are alternately optimized until the system's traversal and rate converge to optimal values. This method utilizes statistical channel state information between each intelligent metasurface, the base station, and the user to design the base station's transmit precoding and each intelligent metasurface's phase shift matrix. This method has the advantages of strong robustness and high system traversal and rate.

Description

Technical Field

[0001] The present invention relates to a multi-user precoding joint design method assisted by Reconfigurable Intelligent Surfaces (RIS), and belongs to the technical field of wireless communication. Background Art

[0002] In cellular mobile systems, cell-edge users are far from the base station, resulting in significant path loss during signal transmission and easy obstruction by buildings and other obstacles. This results in poor quality of service and low throughput for cell-edge users. Improving performance for cell-edge users typically requires deploying hundreds of antennas at the base station, which introduces challenges such as high cost and complexity. In recent years, smart reflective surfaces (RIS) have been recognized as one of the most promising 6G-enabling technologies. Unlike traditional active base stations, RIS consists of a large number of passive, reconfigurable reflective units. By adaptively adjusting the reflection coefficient of each reflective unit, RIS can enhance or weaken the signal in a target area, improving system energy efficiency, and other goals.

[0003] Most existing research results design the RIS phase shift by obtaining instantaneous channel state information. However, accurate instantaneous CSI is difficult to obtain in certain special scenarios, such as high-speed motion. Furthermore, the large number of reflectors on the RIS further increases the overhead of obtaining instantaneous CSI. Compared to instantaneous CSI, statistical CSI changes relatively slowly, requiring only limited channel state information, such as the beam arrival angle within direct line of sight, for transmission design. Furthermore, because statistical CSI changes slowly, the number of channel bits required for feedback is significantly reduced, thereby reducing channel feedback overhead.

[0004] In order to fully realize the potential of RIS-assisted communication system, multiple RIS blocks can be used, and the base station's transmit precoding matrix and RIS phase shift matrix can be jointly optimized based on statistical CSI to maximize the system energy efficiency. Summary of the Invention

[0005] Technical problem: The purpose of the present invention is to provide a multi-user transmission method assisted by a multi-intelligent metasurface in a RIS-assisted multi-user communication system, which can jointly design the base station transmit precoding matrix and the RIS phase shift matrix based on the system's statistical CSI, so as to maximize the traversal and rate of the system users; first, the phase shift matrix of each RIS is fixed, and the base station transmit precoding matrix for each user is designed; then, the base station transmit precoding matrix for each user is fixed, and the phase shift matrix of each RIS is designed; by alternately optimizing the transmit precoding matrix and the phase shift matrix, the traversal and rate of the system converge to the optimal value.

[0006] Technical solution: The present invention is a multi-user transmission method assisted by multiple intelligent metasurfaces. This method is applicable to a single-cell downlink multi-user transmission system. The cell is equipped with L intelligent metasurfaces to serve K single-antenna users. The cell base station adopts a uniform linear antenna array containing M antennas; each intelligent metasurface adopts a uniform planar antenna array containing N l =h l ×v l reflection units, where h l is the number of reflection units of the lth smart metasurface in the horizontal direction, v l is the number of reflection units in the vertical direction of the lth smart metasurface, l=1,…,L. The method specifically includes the following steps:

[0007] Step 1. Set the convergence threshold ε, initialize the number of iterations t = 1, and set the initial value of the base station's transmission precoding vector for the kth user k = 1, ..., K Among them 1 M×1 Represents an M×1 dimensional column vector with all elements set to 1, setting the initial value of the phase shift matrix of the first block of the intelligent metasurface In order to satisfy the diagonal matrix with diagonal element module value of 1, the initial values of the L-block smart metasurface phase shift matrix are composed of the auxiliary matrix The initial value of the iteration That is is a block diagonal matrix of diagonal matrices;

[0008] Step 2. Calculate the t-th iteration value of the precoding vector sent by the base station to the k-th user

[0009] Step 3. Calculate the t-th iteration value Ψ of the auxiliary matrix Ψ (t) ;

[0010] Step 4. Determine whether the following formula is true:

[0011]

[0012] in and Calculate using the following formulas

[0013]

[0014]

[0015] If the above inequality does not hold, set t = t + 1 and go to step 2; if the above inequality holds, the algorithm ends and get Output and Φ l (t)As the base station's transmission precoding vector for the kth user and the phase shift matrix of the lth smart metasurface.

[0016] in,

[0017] In step 2, the base station sends the t-th iteration value of the precoding vector to the k-th user The method comprises the following steps:

[0018] b1) Calculate the K×1 auxiliary vector q for the tth iteration (t) , whose kth element The calculation method is as follows:

[0019]

[0020] where Ψ (t-1) is the t-1th iteration value of the auxiliary matrix Ψ, is the t-1th iteration value of the precoding vector sent by the base station to the kth user, is the t-1th iteration value of the precoding vector sent by the base station to the jth user, and the auxiliary matrix Q in the expression 1k and Q 2k They are:

[0021]

[0022]

[0023] Auxiliary variable Q k for is the noise power received by the kth user;

[0024] b2) Calculate the auxiliary matrix of the tth iteration using the following formula

[0025]

[0026] b3) The matrix Perform Kolaski decomposition to obtain the auxiliary matrix of the tth iteration Satisfy

[0027] b4) Calculate the auxiliary variable for the tth iteration using the following formula Value:

[0028]

[0029] b5) Calculate the auxiliary vector for the tth iteration

[0030] b6) Calculate the auxiliary vector for the tth iteration

[0031] b7) for real number λ i Perform a binary search so that the base station sends the precoding vector calculated using the following formula: satisfy

[0032]

[0033] where γ k is the weight of the kth user, P max is the maximum transmission power of the base station; Update is the t-th iteration value of the precoding vector sent by the base station.

[0034] The step three is to calculate the t-th iteration value Ψ of the auxiliary matrix Ψ (t) , specifically including the following sub-steps:

[0035] c1) Set the convergence threshold ξ, initialize the number of iterations r = 0, and take the t-1th iteration value Ψ of the auxiliary matrix Ψ (t-1) The diagonal elements of the vector form a vector as an auxiliary vector The initial value of iteration η (0) ;

[0036] c2) Calculate the following function:

[0037]

[0038]

[0039] The auxiliary matrix and Calculate using the following formulas

[0040]

[0041]

[0042] diag(·) represents a diagonal matrix with the vectors in brackets as diagonal elements;

[0043] c3) Calculate the following function

[0044]

[0045] The symbol Re{·} represents the real part of a complex number, and the superscript (·) * represents conjugation, ⊙ represents Hadamard product;

[0046] c4) Calculate auxiliary vector Where τ is the step size determined using the Armijo criterion;

[0047] c5) Calculation Where unit(·) means normalizing the modulus of each element in the vector;

[0048] c6) Calculation

[0049] c7) judge |g(η (r+1) )-g(η (r) )|≤ξ is established, if so, proceed to step c8), if not, set r=r+1 and proceed to step c2);

[0050] c8) Update the auxiliary matrix Ψ to its t-th iteration value Ψ (t) =diag{η (r)}.

[0051] Beneficial effects: The present invention provides a multi-user transmission method assisted by multiple intelligent metasurfaces. Compared with the prior art, this method has the following advantages:

[0052] (1) The present invention only requires statistical CSI of the channel, which requires a small amount of channel information and has low feedback overhead, making it suitable for RIS-assisted wireless communication systems.

[0053] (2) The alternating optimization design scheme of the transmission precoding matrix and the phase shift matrix in the present invention can maximize the system traversal and rate, and is relatively easy to implement. DETAILED DESCRIPTION

[0054] The technical solutions provided by the present invention will be described in detail below in conjunction with specific implementation cases. It should be understood that the following specific implementation methods are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0055] The present invention relates to a multi-user transmission method assisted by multiple intelligent metasurfaces: the method is applicable to a single-cell downlink multi-user transmission system, in which there are K single-antenna users in the cell whose service quality is poor or cannot be covered by the base station signal due to factors such as building obstruction, and it is necessary to create a virtual line-of-sight path through L intelligent metasurfaces equipped in the cell to provide services; the cell base station adopts a uniform linear antenna array containing M antennas; each intelligent metasurface adopts a uniform planar antenna array containing N l =h l ×v l reflection units, where h l is the number of reflection units of the lth smart metasurface in the horizontal direction, v l is the number of reflection units in the vertical direction of the lth smart metasurface, l=1, ...L. The method specifically includes the following steps:

[0056] Step 1. Set the convergence threshold ε, initialize the number of iterations t = 1, and set the initial value of the base station's transmission precoding vector for the kth user k = 1, ..., K Among them 1 M×1 Represents an M×1 dimensional column vector with all elements set to 1, setting the initial value of the phase shift matrix of the first block of the intelligent metasurface In order to satisfy the diagonal matrix with diagonal element module value of 1, the initial values of the L-block smart metasurface phase shift matrix are composed of the auxiliary matrix The initial value of the iteration That is is a block diagonal matrix of diagonal matrices;

[0057] Step 2. Calculate the t-th iteration value of the precoding vector sent by the base station to the k-th user

[0058] In step 2, the base station sends the t-th iteration value of the precoding vector to the k-th user The calculation method includes the following steps:

[0059] b1) Calculate the K×1 auxiliary vector q for the tth iteration (t) , whose kth element The calculation method is as follows:

[0060]

[0061] where Ψ (t-1) is the t-1th iteration value of the auxiliary matrix Ψ, is the t-1th iteration value of the precoding vector sent by the base station to the kth user, is the t-1th iteration value of the precoding vector sent by the base station to the jth user, and the auxiliary matrix Q in the expression 1k and Q 2k They are:

[0062]

[0063]

[0064] Auxiliary variable Q k for is the noise power received by the kth user;

[0065] b2) Calculate the auxiliary matrix of the tth iteration using the following formula

[0066]

[0067] b3) The matrix Perform Kolaski decomposition to obtain the auxiliary matrix of the tth iteration Satisfy

[0068] b4) Calculate the auxiliary variable for the tth iteration using the following formula Value:

[0069]

[0070] b5) Calculate the auxiliary vector for the tth iteration

[0071] b6) Calculate the auxiliary vector for the tth iteration

[0072] b7) for real number λ i Perform a binary search so that the base station sends the precoding vector calculated using the following formula: satisfy

[0073]

[0074] where γ k is the weight of the kth user, P max is the maximum transmission power of the base station; Update is the t-th iteration value of the precoding vector sent by the base station.

[0075] Step 3. Calculate the t-th iteration value Ψ of the auxiliary matrix Ψ (t) , specifically including the following sub-steps:

[0076] c1) Set the convergence threshold ξ, initialize the number of iterations r = 0, and take the t-1th iteration value Ψ of the auxiliary matrix Ψ (t-1) The diagonal elements of the vector form a vector as an auxiliary vector The initial value of iteration η (0) ;

[0077] c2) Calculate the following function:

[0078]

[0079]

[0080] The auxiliary matrix and Calculate using the following formulas

[0081]

[0082]

[0083] diag(·) represents a diagonal matrix with the vectors in brackets as diagonal elements;

[0084] c3) Calculate the following function

[0085]

[0086] The symbol Re{·} represents the real part of a complex number, and the superscript (·) * represents conjugation, ⊙ represents Hadamard product;

[0087] c4) Calculate auxiliary vector Where τ is the step size determined using the Armijo criterion;

[0088] c5) Calculation Where unit(·) means normalizing the modulus of each element in the vector;

[0089] c6) Calculation

[0090] c7) judge |g(η (r+1) )-g(η (r) )|≤ξ is established, if so, proceed to step c8), if not, set r=r+1 and proceed to step c2);

[0091] c8) Update the auxiliary matrix Ψ to its t-th iteration value Ψ (t) =diag{η (r)}.

[0092] Step 4. Determine whether the following formula is true:

[0093]

[0094] in and Calculate using the following formulas

[0095]

[0096]

[0097] If the above inequality does not hold, set t = t + 1 and go to step 2; if the above inequality holds, the algorithm ends and Obtain Φ l (t) , l=1,…,L, output and Φl (t) As the base station's transmission precoding vector for the kth user and the phase shift matrix of the lth smart metasurface.

[0098] In summary, this invention surpasses traditional transmission design methods in both computational complexity and algorithmic performance. It also requires only a small amount of statistical channel information to complete the transmission method design, reducing both the accuracy requirements for channel estimation errors and the latency requirements in signal processing. The process of the multi-user transmission method assisted by multi-intelligent metasurfaces is shown in Table 1.

[0099] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-user transmission method assisted by multiple intelligent metasurfaces. First, the phase shift matrix of each RIS is fixed, and the base station's transmit precoding matrix for each user is designed. Then, the base station's transmit precoding matrix for each user is fixed, and the phase shift matrix of each RIS is designed. The transmit precoding matrix and phase shift matrix are alternately optimized until the system's traversal and rate converge to optimal values. The multi-user transmission method is applicable to a single-cell downlink multi-user transmission system. The cell is equipped with L intelligent metasurfaces to serve K single-antenna users. The cell base station adopts a uniform linear antenna array containing M antennas. Each intelligent metasurface adopts a uniform planar antenna array containing N l =h l ×v l reflection units, where h l is the number of reflection units of the lth smart metasurface in the horizontal direction, v l is the number of reflection units in the vertical direction of the lth smart metasurface, l = 1, ... L; it is characterized by: The method specifically comprises the following steps: Step 1. Set the convergence threshold ε, initialize the number of iterations t = 1, and set the initial value of the base station's transmission precoding vector for the kth user k = 1, ..., K Among them 1 M×1 Represents an M×1 dimensional column vector with all elements set to 1, setting the initial value of the phase shift matrix of the first block of the intelligent metasurface In order to satisfy the diagonal matrix with diagonal element module value of 1, the initial values of the L-block smart metasurface phase shift matrix are composed of the auxiliary matrix The initial value of the iteration That is is a block diagonal matrix of diagonal matrices; Step 2. Calculate the t-th iteration value of the precoding vector sent by the base station to the k-th user Step 3. Calculate the t-th iteration value Ψ of the auxiliary matrix Ψ (t) ; Step 4. Determine whether the following formula is true: in and Calculate using the following formulas Ψ (t-1) is the t-1th iteration value of the auxiliary matrix Ψ, is the t-1th iteration value of the precoding vector sent by the base station to the kth user, and are the t-th and t-1-th iteration values of the precoding vector sent by the base station to the j-th user, respectively. The auxiliary matrix Q in the expression 1k and Q 2k They are: Auxiliary variable Q k for is the noise power received by the kth user, is the channel matrix between the base station and the lth smart metasurface The sight distance component, auxiliary matrix α l and ρ l are the Ricean factor and large-scale fading factor of the channel between the base station and the lth smart metasurface, respectively; is the channel between the lth intelligent metasurface and the kth user The sight distance component, β kl and are the Ricean factor and large-scale fading factor of the channel between the lth intelligent metasurface and the kth user, respectively, and the auxiliary vector Superscript (·) H represents the conjugate transpose; if the above inequality does not hold, let t = t + 1 and go to step 2; if the above inequality holds, the algorithm ends and Obtain Φ l (t) ,l=1,…,L,output and Φ l (t) As the base station's transmission precoding vector for the kth user and the phase shift matrix of the lth smart metasurface.

2. The multi-user transmission method assisted by a multi-intelligent metasurface according to claim 1, characterized in that: In step 2, the base station sends the t-th iteration value of the precoding vector to the k-th user The calculation method includes the following steps: b1) Calculate the K×1 auxiliary vector q for the tth iteration (t) , whose kth element The calculation method is as follows: where Ψ (t-1) is the t-1th iteration value of the auxiliary matrix Ψ, is the t-1th iteration value of the precoding vector sent by the base station to the kth user, is the t-1th iteration value of the precoding vector sent by the base station to the jth user; b2) Calculate the auxiliary matrix of the tth iteration using the following formula b3) The matrix Perform Kolaski decomposition to obtain the auxiliary matrix of the tth iteration Satisfy b4) Calculate the auxiliary variable for the tth iteration using the following formula Value: b5) Calculate the auxiliary vector for the tth iteration b6) Calculate the auxiliary vector for the tth iteration b7) for real number λ i Perform a binary search so that the base station sends the precoding vector calculated using the following formula: satisfy where γ k is the weight of the kth user, P max is the maximum transmission power of the base station; Update is the t-th iteration value of the precoding vector sent by the base station.

3. The multi-user transmission method assisted by a multi-intelligent metasurface according to claim 1, characterized in that: In step 3, the t-th iteration value Ψ of the auxiliary matrix Ψ is calculated (t) , specifically including the following sub-steps: c1) Set the convergence threshold ξ, initialize the number of iterations r = 0, and take the t-1th iteration value Ψ of the auxiliary matrix Ψ (t-1) The diagonal elements of the vector form a vector as an auxiliary vector The initial value of iteration η (0) ; c2) Calculate the following function: The auxiliary matrix and Calculate using the following formulas diag(·) represents a diagonal matrix with the vectors in brackets as diagonal elements; c3) Calculate the following function The symbol Re{·} represents the real part of a complex number, and the superscript (·) * represents conjugation, ⊙ represents Hadamard product; c4) Calculate auxiliary vector Where τ is the step size determined using the Armijo criterion; c5) Calculation Where unit(·) means normalizing the modulus of each element in the vector; c6) Calculation c7) judge |g(η (r+1) )-g(η (r) )|≤ξ is established, if so, proceed to step c8), if not, set r=r+1 and proceed to step c2); c8) Update the auxiliary matrix Ψ to its t-th iteration value Ψ (t) =diag{η (r) }.

Images