Holographic MIMO positioning method and system based on dynamic maximum likelihood
By adopting a dynamic maximum likelihood positioning method in the holographic MIMO system, the combination of dynamic sampling and maximum likelihood criterion is solved, and a high-precision radiation source position estimation is achieved.
Patent Information
- Application Number
- CN202411645443.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-18
- Publication Date
- 2025-02-25
AI Technical Summary
The prior art has huge observations in the holographic MIMO system, resulting in the computational complexity of the maximum likelihood method and the minimum mean square error method, and methods such as multiple signal classification and rotation invariant technology cannot be applied to the spherical wave propagation model of holographic MIMO.
The holographic MIMO positioning method with dynamic maximum likelihood is adopted to reduce the computational complexity through dynamic sampling, and the radiation source position is estimated in combination with the maximum likelihood criterion. The specific steps include the first sampling as a coarse estimate, reducing the number of sampling points, determining the area where the radiation source is located through the maximum likelihood criterion, reducing the sampling range for the second sampling, increasing the sampling density, and finally determining the location of the radiation source through the maximum likelihood criterion.
While ensuring good estimation accuracy, it effectively reduces the computational complexity, solves the problem of excessive computational complexity in holographic MIMO systems, and is suitable for spherical wave propagation models.
Smart Images
Figure CN119511194A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of wireless communication technology, and provides a technology for estimating the position of a radiation source, specifically a holographic MIMO positioning method and system based on dynamic maximum likelihood for a holographic multiple-input multiple-output (MIMO) system. Background Art
[0002] Traditional positioning methods include compressed sensing channel estimation methods based on the sparse characteristics of holographic MIMO far-field angular domain channels, two-stage channel estimation methods based on spherical wave channel models, and three-dimensional positioning methods proposed in reconfigurable intelligent surface systems. However, there is currently no specific positioning method in holographic MIMO systems. If traditional positioning methods are applied to holographic MIMO positioning, they will face the following two limitations: the maximum likelihood method and the minimum mean square error method have extremely high computational complexity in holographic MIMO due to the extremely large number of observations; methods such as multiple signal classification and rotation invariance technology are based on the traditional plane wave propagation model and are not applicable to the spherical wave propagation model of holographic MIMO. Based on this, the present invention proposes a positioning technology for holographic MIMO systems, which simplifies the calculation process while ensuring good estimation accuracy. Summary of the invention
[0003] In view of the above problems existing in the prior art, the present invention proposes a dynamic sampling method for sampling the surface electric field matrix and radiation source position of holographic MIMO, and a positioning method for estimating the radiation source position using the maximum likelihood criterion, specifically a holographic MIMO positioning method and system based on dynamic maximum likelihood. The present invention has high estimation accuracy and effectively reduces the computational complexity.
[0004] The application scenario of the present invention is: the transmitting end is a radiation source, and the receiving end adopts a holographic MIMO antenna array; the antenna array is a uniform planar array, including dozens or hundreds of antennas.
[0005] In the holographic MIMO system, its three-dimensional space coordinate system is represented by (x, y, z). There is a radiation source on the x-axis, which is a dipole antenna with a length of l s , the coordinates of the centroid are (x C ,y C ,z C ), the direction vector is t x ,t y ,t z They are The components on the x, y, and z axes, the current density is Where δ is the Dirac function, I inis the current, and s is the space vector at any point in the dipole.
[0006] There is a planar observation area on the yoz plane, which is located in the far field of the dipole. Take any point on the observation area, the space vector from the dipole to the point is r, and the electric field generated by the dipole at the point is e(r). Assume that the space has no boundaries, and the medium in the space is homogeneous and isotropic, that is, in the line of sight (LOS) scenario:
[0007]
[0008] Among them, G(r)=-jκZ0g(r), κ is the wave number, Z0 is the inherent impedance of the medium, is the coefficient Green's function, r=||r||.
[0009] Assume that the observation area is placed The dipoles constitute the holographic MIMO surface, where L is the length of the holographic MIMO plane, and λ is the spacing of the sampling points in the holographic MIMO plane. The dipoles are placed perpendicular to the y-axis on a square grid with a length of l r =λ / 10, the set of centroid positions is:
[0010] {(x,y,z)|x=0,y=mλ / 2,z=nλ / 2,m=-N,-N+1,L,N,n=-N,-N+1,L,N}
[0011] The received voltages are combined into a matrix The voltage received by the (m,n)th dipole is [V] m,n = l r [E] m,n +ν mn ,in, [E] m,n is e(r mn ) on the Z axis, expressed as:
[0012]
[0013] in, σ 2 is the external noise variance.
[0014] On the holographic MIMO surface, according to the received voltage matrix V, the maximum likelihood criterion is used to estimate the coordinates of the centroid position of the radiation source, and its estimated position coordinates are:
[0015]
[0016] in, r' mnis the coordinate (x' C ,y' C ,z' C ) to the origin (0,0,0), r' mnx ,r' mny ,r' mnz is the coordinate (x' C ,y' C ,z' C ) are the distances from their projection points on the x, y, and z axes to the origin.
[0017] Through the above definition and analysis of the system, the position coordinates of the center of mass of the radiation source can be estimated.
[0018] The present invention adopts the following technical solution:
[0019] The holographic MIMO positioning method based on dynamic maximum likelihood has the following specific steps:
[0020] Step 1: First sampling:
[0021] 1) The sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1.
[0022] 2) The radiation source plane is divided into P1 regions, and the radiation source sampling point is in the middle of each region.
[0023] 3) Determine the area where the radiation source is located using the maximum likelihood criterion.
[0024] Step 2: Second sampling:
[0025] 1) The sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2.
[0026] 2) The area of the radiation source plane where the first sampling is performed is further divided into P2 areas, and the radiation source sampling point is in the middle of each area.
[0027] 3) The location of the radiation source is finally determined by the maximum likelihood criterion.
[0028] Preferably, in step 1, equally spaced sampling is performed on the dipole array of the observation plane, and S is a set of sampling points (m, n) to obtain a low-dimensional sampled voltage signal. The number of sampling points corresponding to m and n is That is, the number of samples on the y-axis and z-axis are That is, the total sampling points are (2N+1) 2 .
[0029] Preferably, step 2 performs a second sampling of the radiation source position.
[0030] Sampling is performed in the radiation source plane, C is the sampling point (x' C ,y' C ,z' C ), construct the likelihood function for any sampling point, and its expression is Constructing the likelihood function tensor As the sampling result of each time.
[0031] Preferably, the maximum likelihood criterion is used p, q, r are the numbers of the sampling points on the x, y, and z axes. The corresponding sampling points (x' C0 ,y' C0 ,z' C0 ) is used as the estimated position coordinates of the center of mass of the radiation source each time.
[0032] In the present invention, two samplings are dynamically adjusted in each execution. The first sampling is a rough estimate, which reduces the number of sampling points of the holographic MIMO plane and the radiation source plane, and determines the area where the radiation source is located by the maximum likelihood criterion. The second sampling narrows the sampling range, increases the sampling density of the holographic MIMO plane and the radiation source plane, and finally determines the location of the radiation source by the maximum likelihood criterion.
[0033] The present invention also discloses a holographic MIMO positioning system based on dynamic maximum likelihood, which is used to execute the above method and includes the following modules:
[0034] The first sampling module: the sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1; the radiation source plane is divided into P1 regions, and the radiation source sampling point is in the middle of each region; the region where the radiation source is located is determined by the maximum likelihood criterion.
[0035] Second sampling module: the sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2; the area of the radiation source plane determined by the first sampling is divided into P2 areas, and the radiation source sampling point is in the middle of each area; the position of the radiation source is finally determined by the maximum likelihood criterion.
[0036] Compared with the prior art, the present invention has the following beneficial effects:
[0037] The present invention adopts a dynamic sampling method to solve the problem of high computational complexity of the maximum likelihood method in holographic MIMO due to the huge amount of observations. At the same time, the dynamic combination of sampling and the maximum likelihood method solves the problem that methods such as multiple signal classification and rotation invariance technology cannot be applied to the holographic MIMO spherical wave propagation model. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a scene diagram of the holographic MIMO surface positioning system.
[0039] Figure 2 It is a specific diagram of the dynamic adjustment sampling process of the radiation source plane.
[0040] Figure 3 is the error variance (m 2 )Simulation diagram of the change of signal-to-noise ratio (dB).
[0041] Figure 4 It is a block diagram of a holographic MIMO positioning system based on dynamic maximum likelihood according to a preferred embodiment of the present invention. DETAILED DESCRIPTION
[0042] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to specific examples.
[0043] like Figure 1 As shown, in the holographic MIMO system, its three-dimensional space coordinate system is represented by (x, y, z). There is a radiation source on the x-axis, which is a dipole antenna with a length of l s , the coordinates of the centroid are (x C ,y C ,z C ), the direction vector is The current density is Among them, I in is the uniform current level, and s is the space vector at any point within the dipole.
[0044] There is a planar observation area on the yoz plane, which is located in the far field of the dipole. Take any point on the observation area, the space vector from the dipole to the point is r, and the electric field generated by the dipole at the point is e(r). Assume that the space has no boundaries, and the medium in the space is homogeneous and isotropic, that is, in the line of sight (LOS) scenario:
[0045]
[0046] Among them, G(r)=-jκZ0g(r), κ is the wave number, Z0 is the inherent impedance of the medium, is the coefficient Green's function, r=||r||.
[0047] In this embodiment, the x-axis position of the radiation source of point C is randomly generated in the range of 0 to 10 m, and the y and z axis coordinates are fixed to 0. The length of the holographic MIMO surface is defined as L = 3 m, the first sampling spacing of the holographic MIMO surface is λ = 0.03 m, the second sampling spacing is λ = 0.01 m, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO surface is N = L / λ. For the radiation source plane, the number of first sampling area divisions P1 = 10 3 , the number of sampling area divisions for the second time is P2 = 400. The first sampling area is 0~10m, the first sampling estimation point is (x1,0,0), and the second sampling area is (x1-l0)~(x1+l0), where l0=0.1m.
[0048] The specific steps of a holographic MIMO positioning method based on dynamic maximum likelihood in this embodiment are as follows:
[0049] Step 1: First sampling:
[0050] 1) The sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1.
[0051] 2) The radiation source plane is divided into P1 regions, and the radiation source sampling point is in the middle of each region.
[0052] 3) Determine the area where the radiation source is located using the maximum likelihood criterion.
[0053] The dipole array on the observation plane is sampled at equal intervals, and S is the set of sampling points (m, n) to obtain a low-dimensional sampled voltage signal. The number of sampling points corresponding to m and n is That is, the number of samples on the y-axis and z-axis are That is, the total sampling points are (2N+1) 2 .
[0054] Step 2: Second sampling:
[0055] 1) The sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2.
[0056] 2) Taking the position determined by the first sampling as the center, construct an area of (x1-l0)~(x1+l0), which is further divided into P2 areas, and the radiation source sampling point is in the middle of each area.
[0057] 3) The location of the radiation source is finally determined by the maximum likelihood criterion.
[0058] Sampling is performed in the radiation source space, C is the sampling point (x' C ,y'C ,z' C ), construct the likelihood function for any sampling point, and its expression is Construct the likelihood function tensor, denoted by y.
[0059] Maximum Likelihood Criterion The corresponding sampling point (x' C0 ,y' C0 ,z' C0 ) is used as the estimated position coordinates of the center of mass of the radiation source each time.
[0060] In this embodiment, the first sampling is a rough estimate, which reduces the number of sampling points of the holographic MIMO plane and the radiation source plane, and determines the area where the radiation source is located by the maximum likelihood criterion. The second sampling narrows the sampling range, increases the sampling density of the holographic MIMO plane and the radiation source plane, and finally determines the location of the radiation source by the maximum likelihood criterion.
[0061] Figure 2 This is a diagram of the dynamic sampling of the present invention. The first sampling determines the approximate location of the radiation source, and the second sampling narrows down the estimated location of the radiation source. Dynamic sampling is implemented to reduce the number of sampling points and simplify calculations.
[0062] Figure 3 This is a simulation diagram of the error variance of the present invention changing with the signal-to-noise ratio, where the horizontal axis is the signal-to-noise ratio and the vertical axis is the error variance. Figure 3 It can be seen that the simulation results of the maximum likelihood estimation method of the present invention are basically close to the location of the radiation source estimated by large-scale sampling. Large-scale sampling estimation has too many sampling points and a large amount of calculation. The present invention solves this problem well by reducing the number of sampling points and effectively reducing the amount of calculation.
[0063] like Figure 4 As shown, this embodiment discloses a holographic MIMO positioning system based on dynamic maximum likelihood, which is used to execute the above method. The system includes the following modules:
[0064] The first sampling module: the sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1; the radiation source plane is divided into P1 regions, and the radiation source sampling point is in the middle of each region; the region where the radiation source is located is determined by the maximum likelihood criterion.
[0065] Second sampling module: the sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2; the area of the radiation source plane determined by the first sampling is divided into P2 areas, and the radiation source sampling point is in the middle of each area; the position of the radiation source is finally determined by the maximum likelihood criterion.
[0066] For other contents of this embodiment, please refer to the above method embodiment.
[0067] In summary, the present invention discloses a holographic MIMO positioning method and system based on dynamic sampling and maximum likelihood criterion. At present, there is no specific positioning method in the holographic MIMO system. The present invention dynamically samples the surface electric field matrix of the holographic MIMO and the likelihood function matrix of the radiation source position sampling, and uses the maximum likelihood criterion to estimate the positioning of the radiation source position.
[0068] The present invention solves the problem that the maximum likelihood method and the minimum mean square error method have high computational complexity in holographic MIMO due to the large number of observations; methods such as multiple signal classification and rotation invariance technology are based on the traditional plane wave propagation model and are not applicable to the spherical wave propagation model of holographic MIMO. The present invention simplifies the calculation while ensuring good estimation accuracy, and innovates the positioning technology.
[0069] The above contents are further detailed descriptions of the present invention in combination with specific implementation methods, but it cannot be determined that the specific implementation of the present invention is limited to these descriptions. Any operations within the above-mentioned design principles and implementation points of the present invention, after corresponding modifications, replacements, improvements, etc., should be included in the protection scope of the present invention.
Claims
1. A holographic MIMO positioning method based on dynamic maximum likelihood, the application scenario of this method is: the transmitting end is a radiation source, and the receiving end uses a holographic MIMO antenna array; the antenna array is a uniform planar array; its characteristics are: The specific steps of the method are as follows: Step 1: First sampling: 1) The sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1; 2) Divide the radiation source plane into P1 regions, and the radiation source sampling point is in the middle of each region; 3) Determine the area where the radiation source is located by using the maximum likelihood criterion; Step 2: Second sampling: 1) The sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2; 2) The area of the radiation source plane where the radiation source is sampled for the first time is further divided into P2 areas, and the radiation source sampling point is in the center of each area; 3) The location of the radiation source is finally determined by the maximum likelihood criterion.
2. The holographic MIMO positioning method based on dynamic maximum likelihood as claimed in claim 1, characterized in that: In step 1, equally spaced sampling is performed on the dipole array on the observation plane, where S is the set of sampling points (m, n) to obtain a low-dimensional sampled voltage signal; the number of sampling points corresponding to m and n is The number of samples on the y-axis and z-axis is The total number of sampling points is (2N+1) 2 .
3. The holographic MIMO positioning method based on dynamic maximum likelihood as claimed in claim 2, characterized in that: Sampling is performed in the radiation source plane, C is the sampling point (x' C ,y' C ,z' C ), construct the likelihood function for any sampling point, and the expression is Constructing the likelihood function tensor As the sampling result of each time; where [V] m,n = is the voltage observed at the output of the receiving dipole at the sampling point (m,n) of the dipole array on the observation plane, j is an imaginary unit, and the dipole is placed perpendicular to the y-axis on a square grid with a length of l r =λ / 10, e is the base of natural logarithm, κ is the wave number, t x ,t y ,t z are the components of the direction vector of the dipole antenna on the x, y, and z axes, respectively, r m ' n is the coordinate (x' C ,y' C ,z' C ) to the origin (0,0,0), r m ' nx ,r m ' ny ,r m ' nz is the coordinate (x' C ,y' C ,z' C ) are the distances from their projection points on the x, y, and z axes to the origin; Z0 is the inherent impedance of the medium, I in is the current, l s is the length of the dipole antenna, and λ is the spacing of the sampling points in the holographic MIMO plane.
4. The holographic MIMO positioning method based on dynamic maximum likelihood as claimed in claim 4, characterized in that: Maximum Likelihood Criterion p, q, r are the numbers of the sampling points on the x, y, and z axes, and the corresponding sampling points (x' C0 ,y' C0 ,z' C0 ) is used as the estimated position coordinate of the centroid of the radiation source in step 1 or step 2.
5. A holographic MIMO positioning system based on dynamic maximum likelihood, used to execute the method according to any one of claims 1 to 4, characterized in that: The system includes the following modules: First sampling module: the sampling spacing of the holographic MIMO plane is λ1, the length of the holographic MIMO plane is L, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N1=L / λ1; the radiation source plane is divided into P1 regions, and the radiation source sampling point is in the middle of each region; the region where the radiation source is located is determined by the maximum likelihood criterion; Second sampling module: the sampling spacing of the holographic MIMO plane is reduced to λ2, and the number of sampling points in a row on the positive semi-axis of the holographic MIMO plane x is N2=L / λ2; the area of the radiation source plane determined by the first sampling is divided into P2 areas, and the radiation source sampling point is in the middle of each area; the position of the radiation source is finally determined by the maximum likelihood criterion.