Massive mimo satellite communication joint polarization shaping and precoding method and system
By combining polarimetric and fixed polarization antennas in satellite communication systems, a precoding and polarization shaping matrix was designed, and the overall matrix was optimized to decouple precoding and polarization shaping. This solved the problems of polarization mismatch and difficulty in estimating channel state information, thereby improving communication performance and speed.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SOUTHEAST UNIV
- Filing Date
- 2026-04-21
- Publication Date
- 2026-07-14
AI Technical Summary
Existing satellite communication systems experience a significant performance degradation when faced with atmospheric propagation phenomena such as polarization mismatch caused by rainfall and ice clouds. Furthermore, the long distance between satellites and users and their high mobility make it difficult to estimate channel state information, thus affecting communication performance.
By combining polarized adjustable antennas and fixed polarized antennas, a precoding matrix and a polarization shaping matrix are designed. The overall matrix is optimized using a convex-concave process algorithm, and the precoding and polarization shaping matrices are decoupled to maximize user rate and system performance.
By utilizing polarization degrees of freedom and statistical channel state information, the system's communication performance is improved, the complexity of the optimization problem is reduced, the computation speed is accelerated, and the communication rate is increased.
Smart Images
Figure CN122394610A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to satellite communications, and in particular to a method and system for joint polarization shaping and precoding in massive MIMO satellite communications. Background Technology
[0002] Low Earth Orbit (LEO) satellite constellations have attracted significant attention due to their low transmission latency and minimal path loss. The combination of LEO satellites and massive MIMO technology can leverage spatial degrees of freedom, resulting in a substantial improvement in spectral efficiency. However, existing satellite systems generally use fixed-polarization antennas. Atmospheric phenomena such as rainfall and ice clouds can alter the receiver's polarization, leading to polarization mismatch and a significant degradation in system performance. Therefore, it is necessary to investigate the use of polarization-tunable antennas on the satellite side and the adoption of polarization shaping techniques to effectively combat channel depolarization effects. Furthermore, the long distance between satellites and users and their high mobility result in long propagation delays and large Doppler shifts, making it difficult for the satellite side to estimate instantaneous channel state information. Summary of the Invention
[0003] Purpose of the invention: In view of the shortcomings of the prior art, the purpose of this invention is to provide a method and system for joint polarization shaping and precoding in large-scale MIMO satellite communication, which utilizes polarization degrees of freedom to resist the depolarization effect of the channel and achieve a higher achievable rate.
[0004] Technical Solution: To achieve the above-mentioned objectives, in a first aspect, this invention provides a joint polarization shaping and precoding method for large-scale MIMO satellite communication. The satellite side is equipped with a polarization-tunable antenna, and the user side is equipped with a fixed-polarization antenna. A precoding matrix and a polarization shaping matrix are jointly designed to maximize user capacity and data rate. The method includes the following steps:
[0005] Based on statistical channel state information that takes into account satellite-side polarization shaping, a closed-form expression for downlink ergodicity and rate upper bound is determined.
[0006] An optimization problem is established to maximize the traversal and rate upper bound. The precoding matrix and polarization shaping matrix are treated as a whole matrix. The convex-concave process algorithm is used to solve the optimization problem and obtain the optimal whole matrix.
[0007] The problem of minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized global matrix is established, thereby decoupling the precoding matrix and the polarization shaping matrix.
[0008] Furthermore, considering satellite-side polarization shaping, the channel is the sum of the direct path component and the scattered path component, where the direct path component is the direct path component of the channel without considering polarization shaping and the matrix used to describe the dual polarization effect. The Kronecker product, wherein the scattering path component is the scattering path component of the channel without considering polarization shaping, and... The phase shift matrix resulting from the scattering differences of the four polarization channels The Hadamard product and the Kronecker product; where This indicates the cross-polarization discrimination.
[0009] Furthermore, the satellite side The polarization shaping vector of each transmitting antenna is defined as follows: ;in represents an imaginary number, For the first Phase shift of each transmitting antenna, The number of transmitting antennas; the satellite-side polarization shaping matrix is defined as follows: .
[0010] Furthermore, users The statistical channel state information considering satellite-side polarization shaping is represented as follows: ;in It is statistical channel state information that does not consider polarization shaping. It represents the Kronecker product.
[0011] Furthermore, with the goal of maximizing user traversal and speed, the following optimization problem is established:
[0012]
[0013] in Indicates user The traversal speed, Indicates user and the encoding vector, For the number of users, Expressing expectations, The logarithm of the determinant Indicates noise power. The dimension is unit array, Indicates user The precoded vector, This represents the channel matrix after considering polarization shaping. This indicates the conjugate transpose. This indicates a downlink power constraint.
[0014] Furthermore, the precoding matrix and polarization shaping matrix After treating it as a single matrix, The closed upper limit is ;in , , , The dimension is unit array, This represents the channel matrix after considering polarization. This represents the fixed polarization matrix for user k.
[0015] Furthermore, the closed-form expression for the downlink ergodicy and rate upper bound based on statistical channel state information is as follows:
[0016]
[0017] in This represents the deterministic portion of the channel matrix after considering polarization. The Kronecker product of the eigenma and the identity matrix of the covariance matrix of the user's side-scattering path response vector is represented. The matrix function represents the Kronecker product of the conjugate transpose of the satellite-side array response vector and the identity matrix. The elements satisfy:
[0018]
[0019] in To take into account the statistical channel state information of satellite-side polarization shaping, Let the Kronecker product of the eigenvalues of a diagonal matrix consisting of the eigenvalues of a covariance matrix whose mean covariance is the user-side scattering path response vector and the scattering path polarization response matrix be expressed. The optimization problem is rewritten as:
[0020]
[0021] The convex-concave process algorithm is used to solve the rewritten problem.
[0022] Furthermore, the step of minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized overall matrix, and decoupling the precoding matrix and the polarization shaping matrix, includes:
[0023] The overall matrix is obtained using the convex-concave process algorithm. Then, dimensionality reduction is achieved using EVD decomposition or Gaussian randomization, and the following definition is given. ;in express The vector after dimensionality reduction;
[0024] Polarization shaping matrix Rewritten as ,in , Representing antennas The phase shift of the two ports; definition ,in , Indicates rounding up;
[0025] Establish the following questions:
[0026]
[0027] in Indicates the transmit power; for a fixed value... precoding matrix The optimization of this problem is a projection problem, and its solution is:
[0028]
[0029] By bringing back the objective function, the optimization problem is transformed into:
[0030]
[0031] definition , subscript This indicates that the rows and columns of the matrix are respectively taken as... arrive elements, , , The optimization problem is equivalent to
[0032]
[0033] Introducing auxiliary variables The problem is transformed into:
[0034]
[0035] in express The problem can be solved using an inexact maximize-minimize algorithm for finding the convex hull.
[0036] Secondly, this invention provides a joint polarization shaping and precoding system for large-scale MIMO satellite communication. The satellite side is equipped with a polarization-tunable antenna, and the user side is equipped with a fixed-polarization antenna. A precoding matrix and a polarization shaping matrix are jointly designed to maximize user access and data rate. The system is used to implement the method described in the first aspect, comprising: a problem construction module, used to determine closed-form expressions for downlink ergodicity and rate upper bounds based on statistical channel state information considering satellite-side polarization shaping, and to establish an optimization problem that maximizes the ergodicity and rate upper bounds; a global matrix solving module, used to treat the precoding matrix and polarization shaping matrix as a single global matrix, and to solve the optimization problem using a convex-concave process algorithm to obtain the optimal global matrix; and a decoupling module, used to establish a problem that minimizes the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized global matrix, thus decoupling the precoding matrix and the polarization shaping matrix.
[0037] Thirdly, the present invention provides a computer system including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, it implements the steps of the large-scale MIMO satellite communication joint polarization shaping and precoding method described in the first aspect.
[0038] Beneficial Effects: This invention utilizes a polarization-tunable antenna, leveraging polarization degrees of freedom, which further enhances the system's communication performance compared to traditional fixed-polarization antenna schemes. Simultaneously, it employs statistical channel state information in its design, avoiding the difficulty in estimating instantaneous channel state information. Furthermore, it derives a closed-form expression for the sum and rate upper bound using statistical channel state information, avoiding Monte Carlo operations in expectation computation. By treating the precoding matrix and polarization shaping matrix as a single matrix and employing a convex-concave process algorithm to solve the optimization problem, and obtaining the optimal overall matrix, it establishes a problem to minimize the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized overall matrix. This decouples the precoding matrix and the polarization shaping matrix, reducing the complexity of solving the optimization problem and implementing it at the physical layer, and accelerating the computation speed. Attached Figure Description
[0039] Figure 1 This is a schematic diagram of the overall method flow of an embodiment of the present invention.
[0040] Figure 2 This is a system model diagram of the present invention.
[0041] Figure 3 This is a simulation effect diagram of the present invention. Detailed Implementation
[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0043] like Figure 1As shown in the figure, this invention discloses a joint polarization shaping and precoding method for large-scale MIMO satellite communication. The satellite is equipped with multiple polarization-tunable antennas to provide services to multiple users equipped with fixed polarization antennas. The method jointly designs a precoding matrix and a polarization shaping matrix to maximize the number of users and the rate. Specifically, it includes: determining closed-form expressions for downlink ergodicity and rate upper bounds based on statistical channel state information considering satellite-side polarization shaping; establishing an optimization problem to maximize the ergodicity and rate upper bounds, treating the precoding matrix and the polarization shaping matrix as a single global matrix, and solving the problem using the Convex-Concave Process (CCCP) algorithm to obtain the optimal global matrix; and establishing a problem to minimize the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized global matrix, thereby decoupling the precoding matrix and the polarization shaping matrix.
[0044] In this embodiment, the channel considering satellite-side polarization shaping is the sum of the direct path component and the scattered path component. The direct path component is the direct path component of the channel without polarization shaping and the matrix used to describe the dual polarization effect. The Kronecker product, wherein the scattering path components are the scattering path components of the channel without considering polarization shaping and the matrix. The phase shift matrix resulting from the scattering differences of the four polarization channels The Hadamard product and the Kronecker product. The statistical channel state information considering satellite-side polarization shaping is the statistical channel state information without polarization shaping and the matrix. The Kronecker product.
[0045] The method of this embodiment will be explained in more detail below with a specific scenario.
[0046] Part 1: Constructing a Dual-Polarization Channel Model for Large-Scale MIMO Satellite Communication
[0047] The satellite side is equipped with a uniform planar array response, including One polarization-tunable antenna, among which and The x-axis and y-axis represent the number of antenna elements. Each polarization-tunable antenna is connected to an RF chain and contains two vertical antenna elements, H and V, representing horizontal and vertical polarization respectively. Each polarization-tunable antenna contains a phase shifter to adjust the phase difference between the two components, thus dynamically adjusting the antenna polarization. All users are equipped with... One fixed polarization antenna, of which and express shaft and The number of antenna elements on the axis. Figure 2 The system configuration and model are shown.
[0048] Define user The first between and satellite The departure angle and arrival angle of the strip are respectively and Then the array response vectors on the satellite side and the user side can be expressed as:
[0049]
[0050]
[0051]
[0052] Where n represents the number of antennas. It represents the Kronecker product.
[0053] user The frequency response of the channel can be expressed as
[0054]
[0055] in Represents the average channel energy. Represents Rice factor. This represents the direct path component. Since the distance between the satellite and the user is relatively large, and most of the scattered radiation occurs near the user, while the scattering object is far from the satellite, the departure angle on the satellite side is approximately the same for the same user on different paths. Therefore, it is omitted. . Represents the scattering path component. satisfy and .in Represents the covariance matrix. This indicates the trace operation.
[0056] Channels with polarization It can be represented as
[0057]
[0058] in It represents the Hadamah accumulation. Used to describe the bipolarization effect, denoted as ,in This indicates the cross-polarization discrimination. This represents the phase shift caused by the scattering differences of the four polarization channels, and each element satisfies .
[0059] definition For the first The phase shift of the first transmitting antenna, then the first... The polarization shaping vector of each transmitting antenna is defined as follows:
[0060]
[0061] The satellite side polarization shaping matrix is defined as
[0062]
[0063] in This represents a block-based diagonal matrix. This indicates the phase shift of the antenna on the satellite side; in addition, the user... The polarization matrix is defined as
[0064]
[0065] in Represents the user-side polarization vector. This represents the phase shift of user k antenna n. This represents the phase shift of the user's k-antenna.
[0066] After considering polarization shaping, the user The channel matrix between the satellite and the satellite is
[0067]
[0068] The signal transmitted by the satellite is
[0069]
[0070] in This represents the precoding matrix on the satellite side. Indicates the transmitted signal, satisfying Then the user The received signal is
[0071]
[0072] in This represents additive white Gaussian noise. Based on the above model, the user... The traversal rate is
[0073]
[0074] With the goal of maximizing user traversal and speed, the following optimization problem can be formulated.
[0075]
[0076] in This indicates a downlink power constraint.
[0077] Part Two: Derivation of the Closed-Form Expressions for System Downward Traversal and Rate Upper Bound
[0078] The closed upper limit is
[0079]
[0080] definition , , It can be organized into
[0081]
[0082] expect Closed-form expressions can be obtained using statistical channel information. The eigenvalue decomposition is denoted as ,make ,but ,but It can be represented as
[0083]
[0084] Right now The Kronecker product of the eigenma and the identity matrix of the covariance matrix of the user's side-scattering path response vector is represented. Let Kronecker product be the eigenvalues of a diagonal matrix consisting of the eigenvalues of a covariance matrix that satisfies zero mean and covariance equal to the user-side scattering path response vector, and the scattering path polarization response matrix. This represents the conjugate transpose of the satellite-side array response vector and the Kronecker product of the identity matrix.
[0085] Define the statistical channel information
[0086]
[0087] in .
[0088] Define matrix functions
[0089]
[0090] It can be proven It is a diagonal matrix, and its elements satisfy...
[0091]
[0092] but and It can be re-represented as
[0093]
[0094]
[0095] in This represents the deterministic component in the channel matrix considering polarization.
[0096] Part 3: Solving the optimal global matrix using the CCCP algorithm
[0097] The optimization problem can be rewritten as:
[0098]
[0099] Replace the second term in the objective function with a first-order Taylor expansion.
[0100]
[0101] superscript Indicates the first The next iteration.
[0102] definition
[0103]
[0104] Then there is
[0105]
[0106] in
[0107]
[0108]
[0109] The question has been changed to:
[0110]
[0111] This problem is a convex problem, and the CCCP algorithm is used to solve it.
[0112] Part Four: Decoupling the precoding matrix and polarization shaping matrix using the method of minimizing Euclidean distance.
[0113] Obtained using the CCCP algorithm Then, dimensionality reduction is achieved using EVD decomposition or Gaussian randomization, and the following definition is given. ,in express The vector after dimensionality reduction, .because It is only used to describe the phase difference between two elements of an antenna, therefore it can be used to describe the phase difference between two elements of an antenna. Rewritten as ,in This is equivalent to the phase shift of two elements of an antenna. For ease of representation, we define... ,in Let F represent the feasible region of the polarization shaping matrix, ensuring that each element has a constant modulus and maintaining the block diagonal structure of matrix F. Consider the following problem:
[0114]
[0115] Here This indicates the transmit power. For a fixed value... , The optimization of this problem is a projection problem, and its solution is:
[0116]
[0117] By bringing back the objective function, the optimization problem is transformed into:
[0118]
[0119] definition , , , , The optimization problem is equivalent to
[0120]
[0121] Introducing auxiliary variables The problem is transformed into:
[0122]
[0123] in express The problem can be solved using an inexact maximize-minimize algorithm for finding the convex hull.
[0124] The technical effects of this invention can be demonstrated through the following simulation experiments.
[0125] Simulation setup: Considering a scenario with three users, two equipped with circularly polarized antennas and one equipped with a linearly polarized antenna, satellite side... User side The system has a bandwidth of 20 MHz, antenna gains of 6 dBi on the satellite side and 0 dBi on the user side, an equivalent noise temperature of 290 K, and cross-polarization discrimination. Set to 10 dB. Figure 3 The effects of the method proposed in this invention are demonstrated, showing that the system downlink traversal and speed of the proposed solution are much higher than those of existing fixed polarization schemes.
[0126] This invention discloses a joint polarization shaping and precoding system for large-scale MIMO satellite communication, used to implement the aforementioned joint polarization shaping and precoding method. The system includes: a problem construction module, used to determine closed-form expressions for downlink ergodicity and rate upper bounds based on statistical channel state information considering satellite-side polarization shaping, and to establish an optimization problem maximizing the ergodicity and rate upper bounds; a global matrix solving module, used to treat the precoding matrix and polarization shaping matrix as a single global matrix, and to solve the optimization problem using a convex-concave process algorithm to obtain the optimal global matrix; and a decoupling module, used to establish a problem minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized global matrix, thus decoupling the precoding matrix and the polarization shaping matrix.
[0127] This invention also discloses a computer system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the large-scale MIMO satellite communication joint polarization shaping and precoding method.
[0128] Any aspects of this invention not described in detail are well-known to those skilled in the art.
[0129] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.
Claims
1. A joint polarization shaping and precoding method for large-scale MIMO satellite communication, characterized in that, The satellite side is equipped with a polarization-tunable antenna, and the user side is equipped with a fixed-polarization antenna. A precoding matrix and a polarization shaping matrix are jointly designed to maximize user capacity and data rate. The method includes the following steps: Based on statistical channel state information that takes into account satellite-side polarization shaping, a closed-form expression for downlink ergodicity and rate upper bound is determined. An optimization problem is established to maximize the traversal and rate upper bound. The precoding matrix and polarization shaping matrix are treated as a whole matrix. The convex-concave process algorithm is used to solve the optimization problem and obtain the optimal whole matrix. The problem of minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized global matrix is established, thereby decoupling the precoding matrix and the polarization shaping matrix.
2. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 1, characterized in that, The channel considering satellite-side polarization shaping is the sum of the direct path component and the scattered path component. The direct path component is the direct path component of the channel without polarization shaping and the matrix used to describe the dual polarization effect. The Kronecker product, wherein the scattering path component is the scattering path component of the channel without considering polarization shaping, and... The phase shift matrix resulting from the scattering differences of the four polarization channels The Hadamard product and the Kronecker product; where This indicates the cross-polarization discrimination.
3. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 1, characterized in that, Satellite side The polarization shaping vector of each transmitting antenna is defined as follows: ;in represents an imaginary number, For the first Phase shift of each transmitting antenna, This refers to the number of transmitting antennas; The satellite side polarization shaping matrix is defined as .
4. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 1, characterized in that, user The statistical channel state information considering satellite-side polarization shaping is represented as follows: ;in It is statistical channel state information that does not consider polarization shaping. Indicates the Kronecker product. This indicates the cross-polarization discrimination.
5. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 1, characterized in that, To maximize user traversal and speed, the following optimization problem is established. ； in Indicates user The traversal speed, Indicates user and the encoding vector, For the first Phase shift of each transmitting antenna, This refers to the number of transmitting antennas. For the number of users, Expressing expectations, The logarithm of the determinant Indicates noise power. The dimension is unit array, Indicates user The precoded vector, This represents the channel matrix after considering polarization shaping. This indicates the conjugate transpose. This indicates a downlink power constraint.
6. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 5, characterized in that, precoding matrix and polarization shaping matrix After treating it as a single matrix, The closed upper limit is ;in , , , The dimension is unit array, This represents the channel matrix after considering polarization. This represents the fixed polarization matrix for user k.
7. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 6, characterized in that, The closed-form expression for the downlink ergodicy and rate upper bound based on statistical channel state information is as follows: ； in This represents the deterministic portion of the channel matrix after considering polarization. The Kronecker product of the eigenma and the identity matrix of the covariance matrix of the user's side-scattering path response vector is represented. The matrix function represents the Kronecker product of the conjugate transpose of the satellite-side array response vector and the identity matrix. The elements satisfy ； in To take into account the statistical channel state information of satellite-side polarization shaping, Let the Kronecker product of the eigenvalues of a diagonal matrix consisting of the eigenvalues of a covariance matrix whose mean covariance is the user-side scattering path response vector and the scattering path polarization response matrix be expressed. The optimization problem is rewritten as: ； The convex-concave process algorithm is used to solve the rewritten problem.
8. The method for joint polarization shaping and precoding in large-scale MIMO satellite communication according to claim 3, characterized in that, The problem of minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized overall matrix, and decoupling the precoding matrix and the polarization shaping matrix, includes: The overall matrix is obtained using the convex-concave process algorithm. Then, dimensionality reduction is achieved using EVD decomposition or Gaussian randomization, and the following definition is given. ;in express The vector after dimensionality reduction Number of users; Polarization shaping matrix Rewritten as ,in , Representing antennas The phase shift of the two ports; definition ,in , Indicates rounding up; Establish the following questions: ； in Indicates the transmit power; for a fixed value... precoding matrix The optimization of this problem is a projection problem, and its solution is: ； By bringing back the objective function, the optimization problem is transformed into: ； definition , subscript This indicates that the rows and columns of the matrix are respectively taken as... arrive elements, , , The optimization problem is equivalent to ； Introducing auxiliary variables The problem is transformed into: ； in express The problem can be solved using an inexact maximize-minimize algorithm for finding the convex hull.
9. A joint polarization shaping and precoding system for massive MIMO satellite communication, characterized in that, The satellite side is equipped with a polarization-tunable antenna, and the user side is equipped with a fixed polarization antenna. The precoding matrix and polarization shaping matrix are jointly designed to maximize user and data rate. The system is used to implement the method according to any one of claims 1-8, comprising: The problem construction module is used to determine the closed-form expression of downlink ergodicity and rate upper bound based on statistical channel state information considering satellite-side polarization shaping, and to establish an optimization problem that maximizes the ergodicity and rate upper bound. The global matrix solving module is used to treat the precoding matrix and the polarization shaping matrix as a global matrix, and use the convex-concave process algorithm to solve the optimization problem to obtain the optimal global matrix; The decoupling module is used to solve the problem of minimizing the Euclidean distance between the product of the polarization shaping matrix and the precoding matrix and the optimized overall matrix, thereby decoupling the precoding matrix and the polarization shaping matrix.
10. A computer system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the large-scale MIMO satellite communication joint polarization shaping and precoding method according to any one of claims 1-8.