Channel estimation implementation method in high-speed mobile scenarios

By combining frequency offset and Doppler frequency estimation with a pre-stored MMSE-sinc interpolation matrix, the performance and computational complexity issues of channel estimation in high-speed mobile scenarios are solved, achieving a performance improvement and simplified channel estimation process.

CN117527486BActive Publication Date: 2026-11-17NANJING TICOM TECH +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311351653.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-18
Publication Date
2026-11-17
Estimated Expiration
2043-10-18

AI Technical Summary

Technical Problem

In high-speed mobile scenarios, the performance of existing channel estimation methods deteriorates due to time-domain fading and frequency offset. In particular, MMSE estimation interpolation is difficult to implement when Doppler information is lacking, and traditional linear estimation interpolation is insufficient in high-speed scenarios.

Method used

By using frequency offset estimation and compensation, Doppler frequency estimation, and the pre-stored MMSE-sinc estimation interpolation matrix, the system implementation process is simplified, storage overhead and table lookup times are reduced, and channel estimation performance is improved.

Benefits of technology

In high-speed mobile scenarios, it improves channel estimation performance, reduces the computational overhead and storage requirements of system implementation, and simplifies the channel estimation process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117527486B_ABST
    Figure CN117527486B_ABST
Patent Text Reader

Abstract

This invention discloses a channel estimation method for high-speed mobile scenarios. Based on the demodulation reference signal (DMRS) designed for transmission time slots in the 5G NR standard, this invention first estimates the frequency offset and maximum Doppler frequency of the received signal when performing time-domain channel estimation interpolation at the receiver of a 5G CP-OFDM system. Then, frequency offset compensation is performed on the received signal based on the obtained frequency offset estimate. The coefficient matrix is ​​obtained by looking up the obtained maximum Doppler frequency estimate in a lookup table and time-domain MMSE channel estimation interpolation is performed. The interpolation result is then used to obtain a complete channel response estimate through frequency offset inverse compensation. This method offers a performance gain compared to traditional linear estimation interpolation in high-speed mobile scenarios and is suitable for wireless communication in 5G high-speed mobile scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of wireless communication transmission technology, specifically a method for channel estimation in high-speed mobile scenarios. Background Technology

[0002] With the development of 5G NR mobile communication technology, wireless communication technology in high-speed mobile scenarios has gradually attracted attention. In mobile scenarios, the channel will experience small-scale time-domain fading, manifested as Doppler frequency. As the movement speed between the transmitter and receiver increases, the Doppler frequency of the channel will increase accordingly, exacerbating the time-domain fading and degrading the time-domain channel estimation interpolation performance at the receiver. Currently, most practical systems used for mobile wireless communication employ nearest-neighbor estimation interpolation or linear estimation interpolation schemes to complete the time-domain channel interpolation in channel estimation, performing interpolation only based on the positional relationship between the demodulation reference signal (DMRS) and data symbols, which shows good performance in low-speed scenarios. However, in high-speed scenarios, the channel Doppler frequency will have a significant impact on the performance of these two estimation interpolation schemes. To address the time-domain fading problem, minimum mean square error (MMSE) estimation uses the time-domain correlation function of the channel to generate a channel interpolation matrix. The matrix elements change with the channel Doppler frequency, making it more suitable for high-speed mobile scenarios. However, the implementation of MMSE estimation interpolation requires the use of the channel's Doppler information and time-domain correlation function, which are often unpredictable before actual transmission, thus posing a challenge to the system implementation of MMSE estimation interpolation.

[0003] Meanwhile, some studies indicate that in uplink transmission of wireless communication in high-speed mobile scenarios, the Doppler frequency of the signal received at the base station will contain an additional frequency offset component. In practical applications, the system often uses the frequency of the base station RRU as a reference frequency. Therefore, in downlink transmission, the effective carrier frequency of the signal obtained by the user-side equipment through Automatic Frequency Control (AFC) will include the Doppler frequency of the downlink channel. Consequently, during uplink transmission, the user-side transmitted signal will use the carrier frequency obtained by AFC, which is equivalent to introducing an additional frequency offset into the uplink received signal. This component will also interfere with the time-domain channel estimation interpolation at the receiver. Therefore, before performing time-domain channel estimation interpolation, it is necessary to estimate and compensate for the frequency offset of the received signal to reduce its impact on the time-domain channel estimation interpolation results. Summary of the Invention

[0004] To address the problems of existing technologies, this invention provides a channel estimation implementation method for high-speed mobile scenarios. Compared with the traditional linear estimation interpolation used in most practical systems, this method offers a performance improvement in high-speed mobile scenarios. It avoids the overhead caused by operations such as inversion in real-time calculation of the MMSE coefficient interpolation matrix. At the same time, while ensuring performance gains, it reduces the storage overhead of the interpolation coefficient matrix and the number of loops during table lookups, thus simplifying the system implementation process.

[0005] A method for channel estimation in high-speed mobile scenarios includes the following steps:

[0006] 1) Perform frequency offset estimation based on the channel interpolation results of the DMRS symbol, and then perform frequency offset compensation based on the results to obtain the channel estimate of the DMRS symbol after frequency offset compensation. Meanwhile, the system estimates the channel Doppler frequency using the channel time-domain correlation function corresponding to the uniform Doppler spectrum;

[0007] 2) The system will look up the pre-stored time-domain MMSE-sinc estimation interpolation matrix based on the channel Doppler frequency estimation results, and then use it to estimate the channel value of the DMRS symbol after frequency offset compensation. The channel estimate of the data symbols is obtained by weighting the data symbols. The principle of time-domain MMSE channel estimation interpolation is to generate an interpolation matrix based on the time-domain correlation function of the channel. The matrix form is as follows:

[0008]

[0009] Among them, R HH_time and Both are correlation matrices, σ x 2 With σ n 2 Let represent the variances of the transmitted signal and the noise variable, respectively, and I represent the identity matrix. Similar to Doppler frequency estimation, MMSE estimation interpolation is performed using the correlation function corresponding to the uniform Doppler spectrum. Then, the elements of the two correlation matrices are generated by the sinc function and satisfy the following relationship:

[0010] R HH_time [p,p']=sinc[2f d (l p -l p' )T symb ]

[0011]

[0012] Where p and p' represent the subscripts of the DMRS symbol, l p l p'This represents the OFDM symbol subscripts of DMRS symbols p and p' within the time slot, where l represents the OFDM symbol subscript. Furthermore, since the signal-to-noise ratio (SNR) effect is considered during frequency domain channel interpolation, it can be disregarded during time domain interpolation, i.e., (σ) is omitted. n 2 / σ x 2 I, where the coefficient interpolation matrix is ​​represented as

[0013]

[0014] 3) Inverse compensation for frequency offset: After completing the time-domain MMSE-sinc channel estimation interpolation, inverse compensation for frequency offset is required on the channel interpolation results of all resource elements, because the frequency offset FO affects the received signal, thus obtaining the complete channel estimate value on all resource elements:

[0015]

[0016] Step 1) describes the frequency offset estimation process based on the channel interpolation results of the DMRS symbols as follows: Assume the system has R receiving antennas and T transmitting antennas, the time-domain OFDM symbol length of one slot is L, the total number of frequency-domain subcarriers is K, and P demodulation reference signal DMRS symbols are set in one slot; after the receiving system completes the frequency-domain channel estimation interpolation, the estimated DMRS symbol channel response value is obtained, denoted as... Where r is the receive antenna index, t is the transmit antenna index, p is the DMRS symbol index, and k is the frequency domain subcarrier subscript.

[0017] Based on the estimated values The specific process for compensation is as follows:

[0018] The phase difference between adjacent DMRS symbol channel responses is entirely introduced by the frequency offset:

[0019]

[0020] Where a represents the amplitude attenuation factor, f o Frequency offset, l p l p+1 T represents the time-domain symbol subscripts of DMRS symbols p and p+1 in the time slot, respectively. symb Given the OFDM symbol interval; first, calculate the cross-correlation coefficient of the frequency domain channel response estimates for two adjacent DMRS symbols p and p+1:

[0021]

[0022] Among them, (g) *This indicates taking the conjugate. Accordingly, an OFDM symbol interval T symb The estimated phase difference within can be expressed as:

[0023]

[0024] Where angle(g) represents calculating the phase angle. When more than two DMRS symbols are configured in the time slot, the above algorithm is repeated to obtain the estimated phase difference value corresponding to each group of adjacent DMRS symbols. The final estimation result is obtained by averaging:

[0025]

[0026] For subsequent compensation operations, the frequency offset estimation result is retained in phase form. After obtaining the estimation result, the system will perform frequency offset compensation on the channel response estimate at the DMRS symbol position, thereby reducing its impact on the time-domain channel interpolation.

[0027]

[0028] The specific process of channel Doppler frequency estimation in step 1) is as follows:

[0029] Doppler estimation is performed by selecting the correlation function corresponding to the uniform Doppler spectrum. This correlation function is expressed as follows:

[0030] R(Δl·T symb ) = sinc(2f d Δl·T symb )

[0031] First, calculate the autocorrelation coefficient based on the channel response estimate of the DMRS symbol:

[0032]

[0033] Where, σ 2 It is the noise power estimated by the system;

[0034] Calculate the cross-correlation coefficient between adjacent DMRS symbols and take the modulo, the expression is:

[0035]

[0036] This parameter can be obtained from R in the frequency shift estimation. p,p+1 Obtained by taking the modulus;

[0037] According to the expression of the relevant function, R0, R1 and the Doppler frequency f d The relationship between them is satisfied:

[0038]

[0039] Meanwhile, the Taylor expansion of the function sinc(x) at x = 0 is:

[0040]

[0041] Therefore, the Doppler frequency estimate f obtained from DMRS symbols p and p+1 d,p Represented as

[0042]

[0043] When a time slot contains two or more DMRS symbols, the above algorithm is repeated to obtain the Doppler frequency estimate for each group of adjacent DMRS symbols. The final estimate is obtained by averaging.

[0044]

[0045] The beneficial effects of this invention are as follows:

[0046] (1) A system implementation method for time-domain MMSE channel estimation interpolation is provided, which has a certain performance improvement compared with the traditional linear estimation interpolation used by most practical systems in high-speed mobile scenarios;

[0047] (2) By using matrix pre-generation and interval partitioning, the overhead caused by operations such as inversion in real-time calculation of MMSE coefficient interpolation matrix is ​​avoided. At the same time, while ensuring performance gains, the storage overhead of interpolation coefficient matrix and the number of loops when looking up tables are reduced, simplifying the system implementation process. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 A schematic diagram illustrating the steps of the segmented implementation method for MMSE-sinc channel estimation interpolation;

[0050] Figure 2 This is a schematic diagram of the system's BLER performance. Detailed Implementation

[0051] 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.

[0052] Figure 1 The execution flow diagram of the segmented implementation of MMSE-sinc channel estimation interpolation in high-speed mobile scenarios of this patent is given, and in more detail, it includes the following steps:

[0053] Frequency offset estimation and compensation based on additional DMRS:

[0054] Assume the system has R receive antennas and T transmit antennas, the time-domain OFDM symbol length of one slot is L, the total number of frequency-domain subcarriers is K, and P demodulation reference signals (DMRS) are set in one slot. After the receiving system completes frequency-domain channel estimation interpolation, the estimated DMRS symbol channel response is obtained, denoted as . Where r is the receive antenna index, t is the transmit antenna index, p is the DMRS symbol index, and k is the frequency domain subcarrier subscript. To reduce the impact of frequency offset on time-domain channel estimation interpolation, it is necessary to estimate the frequency offset component in the received signal and then apply the estimated value to... Compensation is performed. In high-speed mobile scenarios, additional DMRS symbols are usually required within a time slot to provide more reference information, thereby adapting to the time-varying characteristics of the channel in high-speed mobile scenarios. Frequency offset estimation based on additional DMRS directly uses the channel response estimates of these DMRS symbols to estimate the frequency offset.

[0055] Ideally, the phase difference between the channel responses of adjacent DMRS symbols is entirely introduced by the frequency offset:

[0056]

[0057] Where a represents the amplitude attenuation factor, f o Frequency offset, l p l p+1 T represents the time-domain symbol subscripts of DMRS symbols p and p+1 in the time slot, respectively. symb Let p be the OFDM symbol interval. Therefore, to obtain an estimate of the frequency offset, we first calculate the cross-correlation coefficient of the frequency domain channel response estimates for two adjacent DMRS symbols p and p+1:

[0058]

[0059] Among them, (g)* This indicates taking the conjugate. Accordingly, an OFDM symbol interval T symb The estimated phase difference within can be expressed as:

[0060]

[0061] Where angle(g) represents calculating the phase angle. When more than two DMRS symbols are configured in the time slot, the above algorithm is repeated to obtain the estimated phase difference value corresponding to each group of adjacent DMRS symbols. The final estimation result is obtained by averaging:

[0062]

[0063] For subsequent compensation operations, the frequency offset estimation result can be retained in phase form without needing phase-to-frequency conversion. After obtaining the estimation result, the system will perform frequency offset compensation on the channel response estimate at the DMRS symbol position, thereby reducing its impact on time-domain channel interpolation.

[0064]

[0065] Doppler estimation based on uniform Doppler spectrum:

[0066] For time-domain channel interpolation, when using MMSE-sinc estimation interpolation, the Doppler frequency f of the channel within the current time slot needs to be known. d This generates the interpolation coefficient matrix. In practical implementations, the system often cannot obtain the Doppler frequency in advance, so it is necessary to estimate the Doppler frequency.

[0067] Similar to MMSE estimation interpolation algorithms, current Doppler estimation primarily relies on the channel's time-domain correlation function, which is also unavailable in advance during practical applications. Therefore, one approach is to pre-define the correlation function and derive the relationship between the correlation function value and the Doppler frequency based on a pre-defined expression. Considering adaptability to different channels, a correlation function corresponding to the uniform Doppler spectrum is selected for Doppler estimation. This correlation function can be expressed as:

[0068] R(Δl·T symb ) = sinc(2f d Δl·T symb )

[0069] Therefore, in order to obtain an estimate of the Doppler frequency, the autocorrelation coefficient is first calculated based on the channel response estimate of the DMRS symbol:

[0070]

[0071] Where, σ 2This is the noise power estimated by the system. Further, the cross-correlation coefficient between adjacent DMRS symbols is calculated and modulo taken; the expression is:

[0072]

[0073] This parameter can be obtained from R in the frequency shift estimation. p,p+1 The modulus is taken. According to the expression of the relevant function, R0, R1, and the Doppler frequency f... d The relationship between them is satisfied:

[0074]

[0075] Meanwhile, the Taylor expansion of the function sinc(x) at x = 0 is:

[0076]

[0077] Therefore, the Doppler frequency estimate f obtained from DMRS symbols p and p+1 d,p It can be represented as

[0078]

[0079] When a time slot contains two or more DMRS symbols, the above algorithm is repeated to obtain the Doppler frequency estimate for each group of adjacent DMRS symbols. The final estimate is obtained by averaging.

[0080]

[0081] The DMRS symbol channel estimates used in Doppler estimation have not yet undergone frequency offset compensation, so this step can be performed simultaneously with frequency offset estimation.

[0082] Time-domain segmented MMSE-sinc estimation interpolation:

[0083] After obtaining the channel Doppler frequency estimate, the system will look up the pre-stored time-domain MMSE-sinc estimation interpolation matrix based on the estimation result, and then use this matrix to estimate the channel value of the DMRS symbol after frequency offset compensation. Weighted summation yields the channel estimate for the data symbols. The principle of time-domain MMSE channel estimation interpolation is to generate an interpolation matrix based on the channel's time-domain correlation function. The matrix form is as follows:

[0084]

[0085] Among them, R HH_time and Both are correlation matrices, σ x 2 With σ n2 Let represent the variances of the transmitted signal and the noise variable, respectively, and I represent the identity matrix. Similar to Doppler frequency estimation, MMSE estimation interpolation is performed using the correlation function corresponding to the uniform Doppler spectrum. The elements of the two correlation matrices are then generated by the sinc function and satisfy the following relationship:

[0086] R HH_time [p,p']=sinc[2f d (l p -l p' )T symb ]

[0087]

[0088] Where p and p' represent the subscripts of the DMRS symbol, l p l p' This represents the OFDM symbol subscripts of DMRS symbols p and p' within the time slot, where l represents the OFDM symbol subscript. Furthermore, since the signal-to-noise ratio (SNR) effect is considered during frequency domain channel interpolation, it can be disregarded during time domain interpolation, i.e., (σ) is omitted. n 2 / σ x 2 In this case, the coefficient interpolation matrix can be expressed as:

[0089]

[0090] In practical applications, to reduce the storage overhead and table lookup complexity of the pre-generated coefficient matrix, the system only generates coefficient matrices corresponding to a finite number of discontinuous Doppler frequencies. After performing Doppler estimation, the system will use the estimated values... The Doppler frequency f at maximum speed d max The interpolation is performed by selecting the corresponding matrix based on the interval of the ratio. The specific segmentation is shown in the table below.

[0091]

[0092] Table 1. Interval Division of MMSE-sinc Interpolation Coefficient Matrix

[0093] After obtaining the coefficient matrix, the system applies the interpolation coefficients to the frequency offset compensation. After weighting, based on the corresponding element information, its matrix form is as follows:

[0094]

[0095] Among them, (g) T Indicates transpose. The dimension is K×P. The channel estimate is calculated for all resource units, with a dimension of K×L. Time-domain MMSE-sinc channel estimation interpolation is performed for all data streams by traversing the receive and transmit antennas.

[0096] Inverse compensation for frequency offset:

[0097] After completing the time-domain MMSE-sinc channel estimation interpolation, frequency offset inverse compensation needs to be performed on the channel interpolation results of all resource elements, because the frequency offset FO affects the received signal, thus obtaining the complete channel estimate value on all resource elements:

[0098]

[0099] Figure 2 A comparison of the system BLER performance of the time-domain MMSE-sinc channel estimation interpolation segmentation implementation scheme and the traditional linear estimation interpolation scheme at a mobile speed of 1000 km / h is presented. It can be seen that within the confidence interval, the system BLER performance using the time-domain channel estimation interpolation scheme of this patent is significantly improved compared to the traditional linear estimation interpolation. The linear estimation interpolation scheme exhibits a bit error plateau at a signal-to-noise ratio of 18 dB, while the system BLER performance of this scheme continues to improve with increasing signal-to-noise ratio, and there is an approximately 3 dB gain in signal-to-noise ratio when the BLER equals its bit error plateau. The corresponding system parameter configurations are shown in the table below.

[0100] Table 2 Simulation System Parameter Configuration

[0101] Simulation system parameters Parameter value Simulation system parameters Parameter value Simulated Channel PUSCH Subcarrier spacing 15kHz carrier frequency 700MHz FFT points 1024 bandwidth 10MHz Number of time slot symbols 14 Modulation method QPSK Channel coding LDPC Number of antennas 1T1R DMRS type Type A, type 1 Additional DMRS symbol number 3 DMRS start symbol subscript 2

[0102] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. In particular, for the device embodiments, the above descriptions are merely preferred embodiments of the present invention. Since they are fundamentally similar to the method embodiments, the descriptions are relatively simple, and relevant parts can be referred to the descriptions of the method embodiments. The above descriptions are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention, without departing from the principle of the present invention, should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for channel estimation in high-speed mobile scenarios, characterized in that, Includes the following steps: 1) Perform frequency offset estimation based on the channel interpolation results of the DMRS symbol, and then perform frequency offset compensation based on the results to obtain the channel estimate of the DMRS symbol after frequency offset compensation. Simultaneously, the system estimates the channel Doppler frequency using the channel time-domain correlation function corresponding to the uniform Doppler spectrum, where... For the receiving antenna index, For transmitting antenna index, For DMRS symbol index, For frequency domain subcarrier subscripts; 2) The system will look up the pre-stored time-domain MMSE-sinc estimation interpolation matrix based on the channel Doppler frequency estimation results, and then use this matrix to estimate the channel value of the DMRS symbol after frequency offset compensation. The channel estimate of the data symbols is obtained by weighting the data symbols. The principle of time-domain MMSE channel estimation interpolation is to generate an interpolation matrix based on the time-domain correlation function of the channel. The matrix form is as follows: ; in, and All are correlation matrices. and Let these represent the variances of the transmitted signal and the noise variable, respectively. Let represent the identity matrix; similar to Doppler frequency estimation, if the correlation function corresponding to the uniform Doppler spectrum is selected for MMSE estimation interpolation, then the elements of the two correlation matrices are generated by the sinc function and satisfy the following relationship: ; ; in, , The subscript indicates the DMRS symbol. , Indicates DMRS symbol , The subscript of the OFDM symbol occupied in the time slot, This indicates the OFDM symbol subscript. Also, since the signal-to-noise ratio (SNR) effect is considered during frequency domain channel interpolation, it can be disregarded during time domain interpolation, i.e., omitted. At this point, the coefficient interpolation matrix is ​​represented as ; 3) Inverse compensation for frequency offset: After completing the time-domain MMSE-sinc channel estimation interpolation, it is necessary to perform inverse compensation for frequency offset on the channel interpolation results of all resource elements, because the frequency offset FO affects the received signal, thus obtaining the complete channel estimate value on all resource elements: 。 2. The channel estimation method for high-speed mobile scenarios according to claim 1, characterized in that: Step 1) The frequency offset estimation process based on the channel interpolation results of the DMRS symbols is as follows: Assume the system has a total of Root receiving antenna and For a single transmit antenna, the time-domain OFDM symbol length of one slot is... The total number of frequency domain subcarriers is And set in a time slot One demodulated reference signal DMRS symbol; after the receiving system completes frequency domain channel estimation interpolation, the estimated channel response value of the DMRS symbol is obtained, denoted as . ,in For the receiving antenna index, For transmitting antenna index, For DMRS symbol index, This is the subcarrier index in the frequency domain.

3. The channel estimation method for high-speed mobile scenarios according to claim 2, characterized in that: Based on the estimated values The specific process for compensation is as follows: The phase difference between the channel responses of adjacent DMRS symbols is entirely introduced by the frequency offset: ; in, Indicates the amplitude attenuation factor. Indicates frequency offset. , These represent DMRS symbols. , The time-domain symbol subscript occupied in the time slot, The OFDM symbol interval is calculated first between two adjacent DMRS symbols. , The cross-correlation coefficient of the frequency domain channel response estimate: ; in, This indicates taking the conjugate, and based on this, an OFDM symbol interval The estimated phase difference within can be expressed as: ; in, To calculate the phase angle, when two or more DMRS symbols are configured in the time slot, the above algorithm is repeated to obtain the estimated phase difference value corresponding to each group of adjacent DMRS symbols. The final estimation result is obtained by averaging. ; For subsequent compensation operations, the frequency offset estimation result is retained in phase form. After obtaining the estimation result, the system will perform frequency offset compensation on the channel response estimate at the DMRS symbol position, thereby reducing its impact on the time-domain channel interpolation. 。 4. The channel estimation method for high-speed mobile scenarios according to claim 1, characterized in that: Step 1) The specific process of channel Doppler frequency estimation is as follows: Doppler estimation is performed by selecting the correlation function corresponding to the uniform Doppler spectrum. This correlation function is expressed as follows: ; First, calculate the autocorrelation coefficient based on the channel response estimate of the DMRS symbol: ; in, It is the noise power estimated by the system; Calculate the cross-correlation coefficient between adjacent DMRS symbols and take the modulo, the expression is: ; This parameter can be obtained from the frequency shift estimation. Obtained by taking the modulus; Based on the expression of the relevant function, , With Doppler frequency The relationship between them is satisfied: ; At the same time, the function exist Taylor expansion at is ; Therefore, according to DMRS symbols , The obtained Doppler frequency estimates Represented as ; When more than two DMRS symbols are configured in a time slot, the above algorithm is repeated to obtain the Doppler frequency estimate for each group of adjacent DMRS symbols. The final estimate is obtained by averaging: 。

Citation Information

Patent Citations

  • Self-adapting channel estimating method applied to 3GPP LTE system

    CN102571650A

  • Statistical characteristic-based channel estimation method for LTE system

    CN107743106A