A Channel Estimation Filtering Method and System for High-Speed ​​Rail Scenarios

Abstract

This invention provides a channel estimation and filtering method and system for high-speed rail scenarios, belonging to the field of channel estimation technology. The method includes: receiving demodulated reference signals simultaneously transmitted from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals and demodulated reference signals have quasi-co-address correlation; calculating frequency offset and power values, wherein the frequency offset includes Doppler frequency shift and Doppler spread; obtaining the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculating the first frequency domain correlation matrix of the symbol containing the demodulated reference signal in combination with the Doppler spread value; correcting the frequency domain correlation matrix according to the Doppler frequency shift and power values; calculating time-domain filter coefficients; and performing time-domain filtering. Beneficial effects: This invention distinguishes between Doppler frequency shift and Doppler spread in the correlation-based time-domain filtering, improving the accuracy of time-domain filtering, thereby improving the accuracy of channel estimation and making the channel estimation filtering more robust.

Description

Technical Field

[0001] This invention relates to the field of channel estimation technology, and in particular to a channel estimation filtering method and system for high-speed rail scenarios. Background Technology

[0002] In high-speed rail networks, the network side typically employs a multi-transmitter and receiver point (TRP) co-cell networking approach to reduce frequent cell handovers caused by high-speed movement. When the train moves at high speed, the signal frequency at the receiver changes, which degrades the receiver's demodulation performance. The train moves at high speed relative to two TRPs, such as... Figure 1 As shown, when the receiver reaches the middle position between two TRPs, it will receive the same signal from both TRPs at the same time. However, the Doppler frequency shift of the two identical signals is opposite, causing time-selective fast fading and affecting the channel estimation accuracy of the receiver.

[0003] High-speed rail channels typically use a single-path model. The high-speed movement generates dual Doppler shift, which is a Doppler frequency shift. In this environment, multiple pilot symbols are transmitted intermittently in the downlink time slot. When performing time-domain channel estimation filtering and interpolation, the receiver needs to first estimate the Doppler frequency shift and power values ​​of the two transmitted TRP signals, then calculate the correlation between the pilot symbols, generate filter coefficients, and perform time-domain filtering and interpolation operations. Figure 2 The diagram shows the power of the signal received by the user equipment (UE) in the same time slot under the dual Doppler effect.

[0004] In current technology, traditional time-domain filters typically mix Doppler frequency shift and Doppler spread values ​​together, without distinguishing between them, and use them together to calculate the time-domain correlation matrix as the Doppler spread. When the Doppler frequency shift exceeds a certain range, the difference between it and the correlation introduced by the Doppler spread increases, leading to inaccurate correlation calculations and affecting the accuracy of time-domain filtering. Furthermore, current time-domain filtering processes generally do not consider the transmit power of the two transmitting and receiving points; when the power difference is significant, it also affects the accuracy of time-domain correlation calculations. Summary of the Invention

[0005] To address the above technical problems, this invention provides a channel estimation and filtering method and system for high-speed rail scenarios.

[0006] The technical problem solved by this invention can be achieved by the following technical solutions:

[0007] A channel estimation filtering method for high-speed rail scenarios includes:

[0008] The system receives demodulation reference signals transmitted simultaneously from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals have a quasi-co-address association with the demodulation reference signals.

[0009] Calculate the frequency offset and power value of each pilot signal, wherein the frequency offset includes the Doppler frequency shift and the Doppler spread;

[0010] Obtain the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol of the demodulated reference signal in combination with the Doppler spread value;

[0011] The first frequency domain correlation matrix is ​​corrected based on the Doppler frequency shift value and the power value to obtain the second frequency domain correlation matrix;

[0012] Calculate the time-domain filter coefficients based on the second frequency-domain correlation matrix;

[0013] The coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate.

[0014] Preferably, the calculation of the frequency offset of each pilot signal is specifically implemented using the following formula:

[0015]

[0016]

[0017] Where K represents the total number of subcarriers; k represents the subcarrier index; H(k, l0) represents the channel estimate of the k-th subcarrier of the l0-th symbol; H(k, l1) represents the channel estimate of the k-th subcarrier of the l1-th symbol; Corr represents frequency offset compensation; Δf represents the subcarrier spacing; Δl i Indicates the relationship with the lth i The symbol interval between symbols; f i This indicates the frequency offset corresponding to the symbol index of the pilot signal.

[0018] Preferably, the calculation of the power value of each pilot signal is specifically implemented using the following formula:

[0019]

[0020] Where S represents the total number of symbols; K represents the total number of subcarriers; H(k, l) s ) indicates the lth S The channel estimate of the k-th subcarrier of a symbol; P i This represents the power value of symbol i, where the pilot signal is located.

[0021] Preferably, the Doppler frequency shift value is specifically achieved using the following formula.

[0022]

[0023] Where M is the number of times the pilot signal is counted within a predetermined time period; f i (m) represents the frequency offset of the pilot signal in symbol i in the m-th statistical analysis; This represents the Doppler frequency shift value of the symbol i containing the pilot signal.

[0024] Preferably, the Doppler extension value is specifically implemented using the following formula:

[0025]

[0026] Where M is the number of times the pilot signal is counted within a predetermined time period; f i (m) represents the frequency offset of the pilot signal in symbol i in the m-th statistical analysis; This represents the power value of symbol i, where the pilot signal is located.

[0027] Preferably, the step of obtaining the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculating the first frequency domain correlation matrix of the symbol containing the demodulated reference signal in combination with the Doppler spread value, is specifically implemented using the following formula:

[0028] C = sinc(f) d *Δl*dt);

[0029] Among them, f d The value represents the Doppler spread; Δl represents the symbol spacing matrix where the demodulated reference signal is located; dt represents the symbol duration; and C represents the first frequency domain correlation matrix.

[0030] Preferably, the step of correcting the frequency domain correlation matrix based on the Doppler frequency shift value and the power value to obtain the second frequency domain correlation matrix is ​​specifically implemented using the following formula:

[0031]

[0032] Among them, P i The power value is represented by C; the first frequency domain correlation matrix is ​​represented by C; Δl is represented by the symbol spacing matrix where the demodulated reference signal is located; and dt is represented by the symbol duration. This represents the Doppler frequency shift value of the symbol i containing the pilot signal; This represents the second frequency domain correlation matrix.

[0033] Preferably, the step of performing time-domain filtering on the coarse channel estimate of the demodulated reference signal based on the time-domain filter coefficients to obtain the filtered channel estimate is specifically implemented using the following formula:

[0034]

[0035] Where, σ 2 Here, I represents the noise variance; I denotes the identity matrix. W represents the second frequency domain correlation matrix; mmse This represents the time-domain filter coefficients.

[0036] Preferably, the step of performing time-domain filtering on the coarse channel estimate of the demodulated reference signal based on the time-domain filter coefficients to obtain the filtered channel estimate is specifically implemented using the following formula:

[0037]

[0038] Wherein, H is the coarse channel estimate of the symbol containing the demodulated reference signal; W mmse Represents the time-domain filter coefficients; This represents the filtered channel estimate.

[0039] This invention also provides a channel estimation and filtering system for high-speed rail scenarios, used to implement the channel estimation and filtering method for high-speed rail scenarios as described above, including:

[0040] A receiving unit is configured to receive demodulation reference signals simultaneously transmitted from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals and the demodulation reference signals have a quasi-co-address association.

[0041] A calculation unit, connected to the receiving unit, is used to calculate the frequency offset and power value of each of the pilot signals, wherein the frequency offset includes Doppler frequency shift and Doppler spread.

[0042] A frequency domain correlation matrix calculation unit, connected to the calculation unit, is used to obtain the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol of the demodulated reference signal in combination with the Doppler spread value;

[0043] The correction unit, connected to the frequency domain correlation matrix calculation unit, is used to correct the first frequency domain correlation matrix according to the Doppler frequency shift value and the power value to obtain the second frequency domain correlation matrix;

[0044] A filter coefficient calculation unit, connected to the correction unit, is used to calculate the time-domain filter coefficients based on the second frequency-domain correlation matrix.

[0045] The time-domain filtering unit, connected to the filter coefficient calculation unit, is used to perform time-domain filtering on the coarse channel estimate of the demodulated reference signal based on the time-domain filter coefficients to obtain the filtered channel estimate.

[0046] The advantages or beneficial effects of the technical solution of this invention are as follows:

[0047] This invention estimates the Doppler frequency shift and Doppler spread of the system by using the demodulation reference signal and pilot signal associated with quasi-co-location, and distinguishes between Doppler frequency shift and Doppler spread in the correlation-based time-domain filtering, thereby improving the accuracy of time-domain filtering and thus improving the accuracy of channel estimation, making the channel estimation filtering more robust. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of a 5G high-speed rail scenario in the current technology.

[0049] Figure 2 This is a waveform diagram of the received signal power in the prior art.

[0050] Figure 3 This is a flowchart illustrating the channel estimation and filtering method for a high-speed rail scenario, as shown in a preferred embodiment of the present invention.

[0051] Figure 4 This is a waveform diagram showing the comparison of frequency domain and extended correlation in a preferred embodiment of the present invention;

[0052] Figures 5a-5h This is a schematic diagram of a simulation comparing filter compensation with and without compensation in a preferred embodiment of the present invention.

[0053] Figure 6 This is a structural block diagram of a channel estimation and filtering system for a high-speed rail scenario, as described in a preferred embodiment of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0057] See Figure 3In a preferred embodiment of the present invention, based on the above-mentioned problems existing in the prior art, a channel estimation filtering method for high-speed rail scenarios is provided, comprising:

[0058] A1 receives demodulation reference signals transmitted simultaneously from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals and demodulation reference signals have a quasi-co-address association.

[0059] This embodiment of the invention takes two transmitting and receiving points as an example, including TRP0 and TRP1. The two transmitting and receiving points transmit two sets of identical data based on the same port, and simultaneously send demodulation reference signals DMRS0 and DMRS1 to the user terminal UE. To achieve time-frequency synchronization, the two TRPs additionally transmit two pilot signals at different time-frequency positions, named TRS0 and TRS1 respectively. TRS0 and DMRS0 are in a quasi-co-address (QCL) relationship, and TRS1 and DMRS1 are also in a quasi-co-address (QCL) relationship. The specific implementation steps are as follows:

[0060] A2, calculate the frequency offset and power value of each pilot signal. The frequency offset includes the Doppler frequency shift and the Doppler spread.

[0061] In a preferred embodiment, calculating the frequency offset of each pilot signal includes:

[0062] The channel estimates of the two interval symbols TRS0 and TRS1 for each subcarrier on all subcarriers are conjugated and then summed.

[0063] Perform angle calculation on the summation result;

[0064] The frequency offset is calculated using the following formula:

[0065]

[0066]

[0067] Where K represents the total number of subcarriers; k represents the subcarrier index; H(k, l0) represents the channel estimate of the k-th subcarrier of the l0-th symbol; H(k, l1) represents the channel estimate of the k-th subcarrier of the l1-th symbol; Corr represents frequency offset compensation; Δf represents the subcarrier spacing; Δl i Indicates the relationship with the lth i The symbol interval between symbols; f i This indicates the frequency offset corresponding to the symbol index of the pilot signal; * indicates conjugate operation; angle indicates angle calculation operation.

[0068] Furthermore, the frequency offset f0 of TRS0 and the frequency offset f1 of TRS1 can be calculated using the above formula; where, when i = 0, Δl0 = l1 - l0; when i = 1, Δl1 = l0 - l1, where l0 and l1 are DMRS symbol indices.

[0069] Specifically, the frequency offsets f0 and f1 calculated above include Doppler frequency shift values. and Doppler extension value f d .

[0070] In a preferred embodiment, assuming the high-speed train's speed remains constant over a predetermined time period, the Doppler frequency shift value is calculated as follows:

[0071] The Doppler frequency shift value is obtained by averaging the frequency offsets f0 and f1 over a predetermined time period. The following formula is used to achieve this:

[0072]

[0073] Where M is the number of times the pilot signal is counted within a predetermined time period; f i (m) represents the frequency offset of the pilot signal in symbol i in the m-th statistical analysis; This represents the Doppler frequency shift value of the symbol i containing the pilot signal.

[0074] In a preferred embodiment, the Doppler extension value f d The calculation methods include:

[0075] For a predetermined time period, the frequency offset f0 and the Doppler frequency shift value The difference, frequency offset f1 and Doppler frequency shift value The average of the differences is the Doppler expansion value f. d Specifically, the following formula is used:

[0076]

[0077] Where M is the number of times the pilot signal is counted within a predetermined time period; f i (m) represents the frequency offset of the pilot signal in symbol i in the m-th statistical analysis; This represents the power value of symbol i, where the pilot signal is located.

[0078] In a preferred embodiment, calculating the power values ​​P0 and P1 of each pilot signal includes:

[0079] The power of symbols on all subcarriers of TRS0 and TRS1 within a predetermined time period is statistically analyzed, and the average value is calculated. This average value is the power value P0 and P1 obtained from the solution. The specific implementation is achieved using the following formula:

[0080]

[0081] Where S represents the total number of symbols; K represents the total number of subcarriers; H(k, l) s ) indicates the lth S The channel estimate of the k-th subcarrier of a symbol; P i This represents the power value of symbol i, where the pilot signal is located.

[0082] A3, obtain the symbol spacing matrix Δl and symbol duration dt of the demodulated reference signal, and combine them with the Doppler spread value f. d Calculate the first frequency domain correlation matrix C of the symbol containing the demodulated reference signal;

[0083] In a preferred embodiment, the Doppler power spectrum is further described as an average power spectrum. Therefore, in this embodiment, the correlation follows a sinc function distribution.

[0084] Obtain the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol containing the demodulated reference signal using the Doppler spread value, including:

[0085] The symbol spacing matrix Δl, symbol duration dt, and Doppler spread value f of the demodulated reference signal are considered. d The product is processed by the sinc function to obtain the first frequency domain correlation matrix C, which is implemented using the following formula:

[0086] C = sinc(f) d *Δl*dt);

[0087]

[0088] Among them, f d denoted by Δl; denoted by dt; denoted by C; denoted by the first frequency domain correlation matrix.

[0089] A4. The first frequency domain correlation matrix is ​​corrected based on the Doppler frequency shift and power values ​​to obtain the second frequency domain correlation matrix;

[0090] In a preferred embodiment, the frequency domain correlation matrix is ​​corrected based on the Doppler frequency shift and power values ​​to obtain a second frequency domain correlation matrix, specifically achieved using the following formula:

[0091]

[0092] Among them, P iThe power value is represented by C; the first frequency domain correlation matrix is ​​represented by Δl; the symbol spacing matrix of the demodulated reference signal is represented by dt; and the symbol duration is represented by dt. This represents the Doppler frequency shift value of the symbol i containing the pilot signal; This represents the second frequency domain correlation matrix.

[0093] Specifically, considering traditional time-domain filters, Doppler frequency shift and Doppler spread values ​​are generally mixed together and used as the Doppler spread correlation matrix. When the Doppler frequency shift exceeds a certain range, the correlation difference introduced by the Doppler spread increases, leading to inaccurate correlation calculations. Furthermore, existing time-domain filters do not consider the transmit power of the two TRPs; when the power difference is large, it also significantly affects the correlation. In this embodiment, due to the existence of Doppler frequency shift and power difference between DMRS symbols, the correlation needs to be corrected using the Doppler frequency shift and power difference through the above formula when calculating the inter-symbol correlation. Further, when the power between DMRS symbols is comparable, the frequency domain and spread correlation comparison diagram is as follows: Figure 6 As shown.

[0094] A5, calculate the time-domain filter coefficients based on the second frequency domain correlation matrix;

[0095] In a preferred embodiment, the coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate, specifically using the following formula:

[0096]

[0097] Where, σ 2 Here, I represents the noise variance; I denotes the identity matrix. W represents the second frequency domain correlation matrix; mmse This represents the time-domain filter coefficients.

[0098] A6. Based on the time-domain filter coefficients, perform time-domain filtering on the coarse channel estimate of the demodulated reference signal to obtain the filtered channel estimate.

[0099] In a preferred embodiment, the coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate, specifically using the following formula:

[0100]

[0101] Where H is the coarse channel estimate of the symbol containing the demodulated reference signal; W mmse Represents the time-domain filter coefficients; This represents the filtered channel estimate.

[0102] Specifically, existing technologies do not distinguish between Doppler frequency shift and spread when calculating the time-domain correlation matrix, treating them together as Doppler spread for correlation calculation. This leads to inaccurate correlation calculations, affecting the accuracy of time-domain filtering. Furthermore, existing time-domain filtering processes do not consider the impact of the transmit power of the two transmitting and receiving points on the accuracy of time-domain correlation calculations. In this embodiment, based on the 5G protocol, the Doppler frequency shift and Doppler spread values ​​of the system are estimated using the demodulation reference signal and pilot signal associated with quasi-co-location. In the time-domain filtering of correlation, Doppler frequency shift and Doppler spread are distinguished and compensated for using different methods in the correlation calculation. By considering the influence of the Doppler frequency shift and Doppler spread of the two TRPs, the optimal time-domain filtering coefficients are calculated, improving the accuracy of time-domain filtering. While preserving the signals with the two Doppler frequency shifts, noise and interference are suppressed to the greatest extent, improving the accuracy of channel estimation. In the high-speed rail scenario, the time-domain filter coefficients are adjusted in real time based on the estimated Doppler values, making the channel estimation filtering more robust.

[0103] like Figures 5a-5h The figure shows a simulation comparison of time-domain filter performance with and without dual Doppler compensation (comp) for the HST-SFN channel, with the high-speed train located at eight different positions between two TRPs. Block error rate (BLER) is the ratio of erroneous blocks to the total number of blocks received by the digital circuit, used to measure system performance. Signal-to-noise ratio (SNR or S / N) is the ratio of the power of the amplifier's output signal to the power of the simultaneously output noise. The simulation results show that at the same signal-to-noise ratio (SNR), the block error rate (BLER) is significantly reduced, improving system performance.

[0104] This invention also provides a channel estimation and filtering system for high-speed rail scenarios, used to implement the channel estimation and filtering method for high-speed rail scenarios described above. Figure 6 As shown, it includes:

[0105] The receiving unit 1 is used to receive demodulation reference signals transmitted simultaneously from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals and demodulation reference signals have a quasi-co-address association.

[0106] The calculation unit 2, connected to the receiving unit 1, is used to calculate the frequency offset and power value of each pilot signal. The frequency offset includes the Doppler frequency shift value and the Doppler spread value.

[0107] Frequency domain correlation matrix calculation unit 3, connected to calculation unit 2, is used to obtain the symbol interval matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol of the demodulated reference signal in combination with the Doppler spread value;

[0108] Correction unit 4 is connected to frequency domain correlation matrix calculation unit 3 and is used to correct the first frequency domain correlation matrix according to the Doppler frequency shift value and power value to obtain the second frequency domain correlation matrix;

[0109] The filter coefficient calculation unit 5 is connected to the correction unit 4 and is used to calculate the time-domain filter coefficients based on the second frequency domain correlation matrix.

[0110] The time-domain filtering unit 6 is connected to the filter coefficient calculation unit 5. It is used to perform time-domain filtering on the coarse channel estimate of the demodulated reference signal based on the time-domain filter coefficients to obtain the filtered channel estimate.

[0111] The above description is merely a preferred embodiment of the present invention and does not limit the implementation and protection scope of the present invention. Those skilled in the art should realize that any equivalent substitutions and obvious changes made using the content of this specification and illustrations should be included within the protection scope of the present invention.

Claims

1. A channel estimation filtering method for high-speed rail scenarios, characterized in that, include: The system receives demodulation reference signals transmitted simultaneously from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals have a quasi-co-address association with the demodulation reference signals. Calculate the frequency offset and power value of each pilot signal, wherein the frequency offset includes the Doppler frequency shift and the Doppler spread; Obtain the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol of the demodulated reference signal in combination with the Doppler spread value; The first frequency domain correlation matrix is ​​corrected based on the Doppler frequency shift value and the power value to obtain the second frequency domain correlation matrix; Calculate the time-domain filter coefficients based on the second frequency-domain correlation matrix; The coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate. The second frequency domain correlation matrix is ​​obtained by correcting the frequency domain correlation matrix based on the Doppler frequency shift value and the power value, specifically using the following formula: ； Among them, P i The power value of symbol i where the pilot signal is located is represented, i=0,1; C represents the first frequency domain correlation matrix; Δl represents the symbol spacing matrix of the demodulated reference signal; dt represents the symbol duration; This represents the Doppler frequency shift value of the symbol i containing the pilot signal; This represents the second frequency domain correlation matrix.

2. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The calculation of the frequency offset of each pilot signal is specifically achieved using the following formula: ； ； Where K represents the total number of subcarriers; k represents the subcarrier index; H(k, l0) represents the channel estimate of the k-th subcarrier of the l0-th symbol; H(k, l1) represents the channel estimate of the k-th subcarrier of the l1-th symbol; Corr represents frequency offset compensation; Δf represents the subcarrier spacing; Indicates the relationship with the lth i The symbol interval between symbols; f i This indicates the frequency offset corresponding to the symbol index of the pilot signal.

3. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The calculation of the power value of each pilot signal is specifically achieved using the following formula: ； Where S represents the total number of symbols; K represents the total number of subcarriers; Indicates the lth S The channel estimate of the k-th subcarrier of a symbol; P i This represents the power value of symbol i, where the pilot signal is located.

4. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The Doppler frequency shift value is specifically achieved using the following formula. ； Where M is the number of times the pilot signal is counted within a predetermined time period; This represents the frequency offset of the pilot signal in symbol i, as determined in the m-th statistical analysis. This represents the Doppler frequency shift value of the symbol i containing the pilot signal.

5. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The Doppler extension value is specifically achieved using the following formula: ； Where M is the number of times the pilot signal is counted within a predetermined time period; This represents the frequency offset of the pilot signal in symbol i, as determined in the m-th statistical analysis. This represents the power value of symbol i, where the pilot signal is located.

6. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The process of obtaining the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculating the first frequency domain correlation matrix of the symbol containing the demodulated reference signal in combination with the Doppler spread value, is specifically implemented using the following formula: ； Among them, f d The value represents the Doppler spread; Δl represents the symbol spacing matrix where the demodulated reference signal is located; dt represents the symbol duration; and C represents the first frequency domain correlation matrix.

7. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate, which is specifically achieved using the following formula: ； Where, σ 2 Here, I represents the noise variance; I denotes the identity matrix. W represents the second frequency domain correlation matrix; mmse This represents the time-domain filter coefficients.

8. The channel estimation and filtering method for high-speed rail scenarios according to claim 1, characterized in that, The coarse channel estimate of the demodulated reference signal is filtered in the time domain based on the time domain filter coefficients to obtain the filtered channel estimate, which is specifically achieved using the following formula: ； Wherein, H is the coarse channel estimate of the symbol containing the demodulated reference signal; W mmse Represents the time-domain filter coefficients; This represents the filtered channel estimate.

9. A channel estimation filtering system for high-speed rail scenarios, characterized in that, The channel estimation filtering method for high-speed rail scenarios as described in any one of claims 1-8 includes: A receiving unit is configured to receive demodulation reference signals simultaneously transmitted from at least two transmitting and receiving points in opposite directions, as well as additionally transmitted pilot signals, wherein the pilot signals and the demodulation reference signals have a quasi-co-address association. A calculation unit, connected to the receiving unit, is used to calculate the frequency offset and power value of each of the pilot signals, wherein the frequency offset includes Doppler frequency shift and Doppler spread. A frequency domain correlation matrix calculation unit, connected to the calculation unit, is used to obtain the symbol spacing matrix and symbol duration of the demodulated reference signal, and calculate the first frequency domain correlation matrix of the symbol of the demodulated reference signal in combination with the Doppler spread value; The correction unit, connected to the frequency domain correlation matrix calculation unit, is used to correct the first frequency domain correlation matrix according to the Doppler frequency shift value and the power value to obtain the second frequency domain correlation matrix; A filter coefficient calculation unit, connected to the correction unit, is used to calculate the time-domain filter coefficients based on the second frequency-domain correlation matrix. The time-domain filtering unit, connected to the filter coefficient calculation unit, is used to perform time-domain filtering on the coarse channel estimate of the demodulated reference signal based on the time-domain filter coefficients to obtain the filtered channel estimate.

Images