A channel parameter estimation method based on beam dispersion pattern detection

Abstract

The application discloses a kind of channel parameter estimation methods based on beam dispersion mode detection in the technical field of communication, comprising the following: step 1 constructs the sparse representation model of wideband millimeter wave MIMO channel, utilizes index set beam dispersion mode to capture the power of channel to detect physical channel direction, obtains physical channel direction index;Step 2 generates support detection window by extending beam dispersion mode, recovers the sparse channel support determined by physical channel direction index;Step 3 obtains channel parameter matrix based on sparse channel support degree.The application utilizes the sparse characteristics of wideband channel and beam dispersion effect, can correctly detect physical channel direction and corresponding sparse channel support degree, simulation results show that, in the line-of-sight channel wideband millimeter wave MIMO system scene in all signal-to-noise ratio regions are superior to existing scheme, with good performance.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a channel parameter estimation method based on beam dispersion mode detection. Background Technology

[0002] In wideband millimeter-wave MIMO systems, due to the large signal bandwidth, the spatial phase of subcarriers differs significantly, resulting in a large gap between them; this effect is called beam dispersion. Because of beam dispersion, the beam energy distribution in wideband millimeter-wave beamdomain systems varies significantly with the subcarrier frequency. Beam dispersion in the beam-frequency domain poses a serious challenge to wideband millimeter-wave beamdomain channel estimation. Beam dispersion produces different spatial channel directions on different subcarriers, i.e., different sparse support. Therefore, it is difficult to accurately detect sparse channels using a Common Support Detection Window (CSDW) on all subcarriers, and the ideal general sparse support assumption is unreasonable. Summary of the Invention

[0003] To address the aforementioned problems in channel parameter estimation and to perform parameter estimation more accurately and effectively, this invention provides a channel estimation algorithm based on beam dispersion mode detection (BSPD).

[0004] A channel parameter estimation method based on beam dispersion mode detection includes the following steps:

[0005] Step 1: Construct a sparse representation model of a broadband millimeter-wave MIMO channel, and use the index set beam dispersion mode to capture the power of the channel to detect the physical channel direction and obtain the physical channel direction index.

[0006] Step 2: Generate a support detection window by extending the beam dispersion mode to restore the sparse channel support determined by the physical channel direction index;

[0007] Step 3: Obtain the channel parameter matrix based on sparse channel support.

[0008] Furthermore, the specific process of step 1 is as follows:

[0009] The sparse characterization model of a broadband millimeter-wave MIMO channel is as follows:

[0010]

[0011] in, Represents the overall observation matrix. , , . This represents the pilot signal received on the m-th subcarrier. This represents the wideband beam domain channel on subcarrier m. Indicates effective noise. As can be seen from the sparse representation model, through the observation matrix... and received pilot Recover the channel parameter matrix H.

[0012] The residual matrix of the received signal Initialize to ,in This represents the residual of the m-th subcarrier;

[0013] Assume the physical channel direction of the l-th path component Located in Angular domain channel samples Above, beam dispersion mode (BSP) Defined as

[0014]

[0015] in, f is the frequency of the m-th subcarrier. c For the center frequency, Indicates the predefined spatial orientation of the antenna array, index set With physical channel direction There is a one-to-one correspondence, which contains the element index of the angular domain channel with the highest power on each subcarrier.

[0016] Energy of the correlation matrix C captured using BSP

[0017]

[0018] The exponent of the physical channel direction of the l-th path component for

[0019] .

[0020] Furthermore, the specific process of step 2 is as follows:

[0021] Extend BSP to generate support for detection window SDW:

[0022]

[0023] in , It is the size of SDW.

[0024] The first one is obtained through SDW Sparse channel support of each path component at different subcarriers ，

[0025]

[0026] On the m-th subcarrier Sparse channel support of each path for

[0027] .

[0028] Furthermore, the specific process of step 3 is as follows:

[0029] The least squares algorithm is used to calculate the position of the m-th subcarrier. Non-zero elements of each path component

[0030]

[0031] Eliminate the The impact of each path, and update the residual matrix to

[0032]

[0033] The sparse channel support of the m-th subcarrier is

[0034] .

[0035] The sparse angular domain channel of the m-th subcarrier is

[0036]

[0037] in, This represents all signals received by the m-th subcarrier, where m = 1, 2, ..., M.

[0038] The channel parameter matrix is

[0039] .

[0040] The beneficial effects of this invention include:

[0041] The channel parameter estimation method based on beam dispersion mode detection (BSPD) provided by this invention utilizes a defined BSP to capture the channel power in order to detect the direction of the physical channel. Due to the beam dispersion effect in wideband millimeter-wave MIMO channels, the ideal ordinary sparse support assumption is unreasonable. This invention leverages the frequency-dependent sparse channel support implied by the beam dispersion effect, which can accurately detect the physical channel direction and the corresponding sparse channel support, achieving better channel estimation accuracy. Attached Figure Description

[0042] Figure 1 This is a flowchart of an embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram comparing the normalized mean square error (NMSE) performance of the method provided by this invention with SOMP-based and OMP-based schemes.

[0044] Figure 3 The present invention provides a schematic diagram comparing the NMSE performance of the method, SOMP-based scheme, and OMP-based scheme for pilot length P. Detailed Implementation

[0045] The following will be combined with the appendix Figure 1-3 The embodiments are provided to clearly and completely describe the implementation of the present invention. It should be understood that the embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art fall within the scope defined by the appended claims.

[0046] A channel parameter estimation method based on beam dispersion mode detection (BSPD) includes the following steps:

[0047] Step 1: Construct a sparse representation model of a broadband millimeter-wave MIMO channel, and use the index set beam dispersion mode to capture the power of the channel to detect the physical channel direction and obtain the physical channel direction index.

[0048] Beam dispersion results in different spatial channel orientations on different subcarriers, meaning they have varying sparsity support. Channel parameters exhibit significant sparsity, which can be addressed by sparsifying them when constructing channel parameter estimation models, thus refining the broadband channel estimation problem as a joint sparse recovery problem.

[0049] 1) Construct the original transmission variation relationship in the broadband millimeter-wave MIMO channel between the transmitter and receiver;

[0050] For the transmitted pilot signal, define P as the pilot transmission length and assume... The signal received on the m-th subcarrier It can be represented as

[0051]

[0052] in Represents the overall observation matrix. Indicates effective noise. For noise, obey , This represents the broadband beam domain channel on subcarrier m.

[0053] For the transmission pilot at the m-th subcarrier in time slot p The received vector on the m-th subcarrier for

[0054]

[0055] in, It is an adaptive selection network, which is fixed for different subcarriers.

[0056] For a wideband beam domain channel on subcarrier m It can be obtained by converting a traditional spatial domain channel through a uniform antenna array.

[0057]

[0058] in, Represents the traditional spatial domain channel on subcarrier m; Represents the N-dimensional DFT transform of the received signal. Indicates the spatial orientation predefined by the antenna array; This represents the l-th path component on subcarrier m in the beam domain.

[0059]

[0060] in, This represents the Dirac sinc function. Let l be the spatial phase of the l-th path at the m-th subcarrier.

[0061] The m-th subcarrier Spatial domain channel It can be represented as

[0062]

[0063] In the formula, L, , These represent the number of paths, the path complex gain of the l-th path, and the time delay of the l-th path, respectively. The spatial phase of the l-th path at the m-th subcarrier

[0064]

[0065] in Let m be the frequency of the m-th subcarrier. B is the center frequency, B is the signal bandwidth, and c represents the speed of light. Antenna spacing, It is the angle of incidence for the l-th path. The response vector of a uniform linear array

[0066]

[0067] 2) Perform sparse representation:

[0068] Based on channel correlation, a sparse representation is obtained by jointly estimating the channel at different subcarrier frequencies.

[0069]

[0070] in , , Based on the observation matrix and received pilot Restore the broadband sparse angular domain channel H.

[0071] The sparse representation model obtained in step 1 is solved using a channel parameter estimation method based on beam dispersion mode detection (BSPD).

[0072] The specific process of step 2 is as follows:

[0073] 1) Define a beam dispersion mode (BSP) to capture the power of the channel in order to detect the physical channel direction of the path component.

[0074] The received residual matrix Initialize to ,in This represents the residual of the m-th subcarrier.

[0075] Assuming physical channel direction Located in angular domain samples Above, BSP Defined as

[0076]

[0077] Direction of physical channel There is a one-to-one correspondence, containing the element index of the angular domain channel with the highest power on each subcarrier, therefore BSP It can be viewed as a specific characteristic of the physical channel direction, BSP It can be used to estimate the physical channel .

[0078] Calculate the correlation matrix C

[0079]

[0080] The energy of the correlation matrix C is captured using BSP, and the exponent of the physical channel direction of the l-th path component is determined.

[0081]

[0082] Since the physical channel direction is in one-to-one correspondence with the BSP, the above formula can guarantee the physical channel direction. The accuracy of the estimation. This is crucial when obtaining the physical channel direction. Then, the sparse channel support at different subcarriers can be determined.

[0083] Step 2: Generate a support detection window by extending the beam dispersion mode to restore the sparse channel support determined by the physical channel direction index;

[0084] definition As a supporting detection window SDW, in which , Since the physical channel direction and BSP are in one-to-one correspondence, the above formula can guarantee the physical channel direction. The accuracy of the estimation. This is crucial when obtaining the physical channel direction. Then, the sparse channel support at different subcarriers can be determined.

[0085] The sparse channel support of the l-th path component at different subcarriers is obtained by expanding the detection window SDW generated by the BSP. ，

[0086]

[0087] Sparse channel support of the l-th path on the m-th subcarrier

[0088]

[0089] in,

[0090] Step 3: Obtain the channel parameter matrix based on sparse channel support.

[0091] The non-zero elements of the l-th path component at the m-th subcarrier are calculated using the least squares (LS) algorithm.

[0092]

[0093] in, This is the observation matrix.

[0094] Eliminate the influence of the l-th path and update the residual matrix as follows:

[0095]

[0096] The sparse channel support of the m-th subcarrier is

[0097] .

[0098] The m-th subcarrier sparse corner domain channel is

[0099]

[0100] in, This represents all signals received by the m-th subcarrier, where m = 1, 2, ..., M.

[0101] Channel parameter matrix

[0102]

[0103] Figure 2 This paper presents channel estimation techniques based on BSPD and existing channel estimation techniques, including SOMP-based and OMP-based schemes, and their Normalized Mean Square Error (NMSE) performance for signal-to-noise ratio. For the OMP-based scheme, we perform the OMP algorithm once every 16 subcarriers. Then, based on the general support assumption, we obtain the sparse channel support of these 16 subcarriers using the OMP algorithm. For all considered schemes, each user uses P=10 time slots for pilot transmission. For the BSPD-based scheme proposed in this invention, the SDW size is set to ∆=4. For a fair comparison, we assume that the sparsity level of the OMP-based and SOMP-based schemes is L(2∆+2)=27.

[0104] Figure 3 The NMSE performance for pilot length P is presented, with the signal-to-noise ratio set to 20 dB. Other parameters are... Figure 2 The parameters are the same. From Figure 3 As can be seen, the NMSE decreases with the length of the pilot sequence. Among all considered pilot lengths P, the proposed BSPD-based scheme achieves better NMSE performance than existing schemes. Especially when the pilot length is short (e.g., P=4 and P=8), the performance gap between the proposed BSPD-based scheme and existing schemes is quite large.

[0105] Figure 2 as well as Figure 3As can be seen, the proposed channel estimation scheme based on beam dispersion mode (BSPD) detection outperforms existing schemes in all signal-to-noise ratio regions, indicating that the BSPD-based scheme can effectively reduce pilot overhead for channel estimation. Since the proposed scheme utilizes the frequency-dependent sparse channel support implied by the beam dispersion effect, it can accurately detect the physical channel direction and the corresponding sparse channel support. Therefore, the proposed scheme can achieve better channel estimation accuracy.

Claims

1. A channel parameter estimation method based on beam dispersion mode detection, characterized in that, Includes the following steps: Step 1: Construct a sparse representation model of a broadband millimeter-wave MIMO channel, and use the index set beam dispersion mode to capture the power of the channel to detect the physical channel direction and obtain the physical channel direction index. Step 2: Generate a support detection window by extending the beam dispersion mode to restore the sparse channel support determined by the physical channel direction index; Step 3: Obtain the channel parameter matrix based on sparse channel support. The specific process of step 1 is as follows: The sparse characterization model of a broadband millimeter-wave MIMO channel is as follows: in, Represents the overall observation matrix. , , ; This represents the pilot signal received on the m-th subcarrier. This represents the wideband beam domain channel on subcarrier m. Indicates effective noise. From the sparse representation model, we know that through the observation matrix... and received pilot Recover the channel parameter matrix H; The residual matrix of the received signal Initialize to ,in This represents the residual of the m-th subcarrier; Assume the physical channel direction of the l-th path component Located in Angular domain channel samples Above, beam dispersion mode (BSP) Defined as: ； in, f is the frequency of the m-th subcarrier. c For the center frequency, Indicates the predefined spatial orientation of the antenna array; index set With physical channel direction There is a one-to-one correspondence, which contains the element index of the corner domain channel with the highest power on each subcarrier; Energy captured using BSP for correlation matrix C: The exponent of the physical channel direction of the l-th path component for 。 2. The channel parameter estimation method based on beam dispersion mode detection according to claim 1, characterized in that: The specific process of step 2 is as follows: Extend BSP to generate support for detection window SDW: in , It is the size of SDW. For index set; The first one is obtained through SDW Sparse channel support of each path component at different subcarriers ， On the m-th subcarrier Sparse channel support of each path for 。

Images