WO2016058476A1

WO2016058476A1 - Method and device for estimating lte uplink system channel in interference condition

Info

Publication number: WO2016058476A1
Application number: PCT/CN2015/090735
Authority: WO
Inventors: 熊高才; 秦洪峰; 林灯生
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-10-17
Filing date: 2015-09-25
Publication date: 2016-04-21
Also published as: CN105577582A

Abstract

A method and device for estimating an LTE uplink system channel in an interference condition. The method comprises: receiving a guidance vector of a baseband signal, so as to obtain a guidance vector Y_m of a frequency-domain baseband signal via a fast Fourier transform; processing the Y_m in a frequency domain, and deleting some interference spectrum lines, so as to obtain a vector y_m after the frequency domain is suppressed; conducting least square estimation (LS) on the y_m, so as to obtain an LS estimation output vector of h_LS = X^-1y_m, X being an LTE uplink demodulation reference signal (DMRS); and weighting the LS estimation output vector, so as to obtain a channel estimation value. The technical solution can fully preserve the channel energy while filtering noise and suppressing interference, thereby improving the channel estimation accuracy and having the advantage of relatively low complexity at the same time.

Description

LTE uplink system channel estimation method and device under interference conditions

Technical field

This document relates to, but is not limited to, the field of communications, and in particular, to a method and apparatus for estimating LTE uplink system channels under interference conditions.

Background technique

The LTE (Long Term Evolution) system has adopted the SC-FDMA (Single-Carrier Frequency-Division Multiple Access) technology as the uplink multiple access technology. Since the LTE system uses global frequency reuse, neighbor cell interference. The phenomenon is more serious; in addition, in some frequency bands, LTE uplink may suffer from interference from a variety of different systems, such as:

(1) Cordless telephone (2.4 or 5.xGHz);

(2) Bluetooth personal area networking equipment (2.4GHz);

(3) Bluetooth wireless headset;

(4) Microwave oven (50% of the busyness in the 2.4 GHz band will generate pulse interference).

The presence of interference will seriously affect the normal operation of the LTE uplink receiver, thereby reducing the cell throughput. Therefore, certain measures must be taken to suppress it.

The channel estimation algorithms commonly used in the LTE uplink mainly include the following three types:

(1) LS (Least Squares) algorithm: In the "On channel estimation in OFDM system" article, the LS channel estimation application error squared the smallest sum of the criteria to estimate the impulse response of the channel, is the most For simple channel estimation, the LS channel estimation value is represented by h _LS =X ^-1 y _m , X is the designed steering vector, () ^-1 represents the inverse of the matrix, and y _m is the received steering vector;

(2) LMMSE (Linear Minimum Mean Square Error) algorithm: The LMMSE algorithm proposed in the article "OFDM channel estimation by singular value decomposition" is an improvement on the LS algorithm, which is to minimize the mean square The linear channel estimation with error is the criterion, which is to correct the LS channel estimation by channel autocorrelation matrix to suppress noise. It is a linear optimal channel estimation method, and its channel estimation value is h _LMMSE =R _h (R _h +S ^H R _j S+N ₀ I) ^-1 h _LS means that R _h =E{HH ^H } represents the channel autocorrelation matrix, R _j is the interference signal autocorrelation matrix, S is the diagonal matrix, and the diagonal elements are pilots Sequence, N ₀ represents the noise power, I is the unit matrix, () ^H represents the conjugate transpose of the matrix, and E{} represents the expected value;

(3) DFT-Based (Fourier Transform) algorithm: DFT-Based channel estimation proposed in the article "DFT-Based Channel Estimation and Noise Variance Estimation Techniques for Single-Carrier FDMA" is based on LS channel estimation technology. The characteristics of the domain channel energy concentration achieve noise reduction and suppression of interference. The DFT-Based channel estimation transforms the LS channel estimation value into the time domain through IDFT, and then performs time domain windowing to achieve noise reduction and interference suppression. After the windowing process is completed, the DFT transform is performed to the frequency domain. The channel estimate is represented by h _DFT-Based = FDF ^H h _LS , F represents the DFT transformation matrix, and () ^H represents the conjugate transpose of the matrix.

Represents a windowed matrix, and I _m represents an m-order identity matrix.

Analysis of the above three types, respectively have the following characteristics:

The LS algorithm has the advantages of simple implementation, but does not have the noise cancellation capability, and has poor performance at low SNR, and the performance will deteriorate further in the presence of interference;

The linear optimal LMMSE algorithm has the best noise cancellation effect. However, in the implementation of LMMSE channel estimation, there is a need for more a priori information and high computational complexity. In the scenario where interference exists, it is necessary to know the frequency domain statistical information of the interference signal to implement the LMMSE channel estimation, and the channel characteristics, the interference signal is generally unknown in the actual communication environment;

The traditional DFT-Based technology uses a fixed window length filter matrix, which causes insufficient noise and interference filtering, or a large signal energy loss under high SNR conditions. Due to the energy diffusion problem in the time domain channel response, The DFT-Based technology suffers from severe degradation in performance when the user occupies fewer subcarriers.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims. (The following is a brief summary of subject matter that is described in greater detail herein.This summary is not intended to be limiting as to The scope of the claims.)

Embodiments of the present invention provide a channel estimation method and apparatus for an LTE uplink system under interference conditions, to solve the technical problem of how to simply and sufficiently preserve channel energy and improve channel estimation accuracy while filtering noise and suppressing interference.

An embodiment of the present invention provides a channel estimation method for an LTE uplink system under interference conditions, including:

Receiving a steering vector of the baseband signal, and obtaining a steering vector Y _{m of the} frequency domain baseband signal through fast Fourier transform;

Y _m is processed in the frequency domain, and part of the interference spectrum is deleted, and the vector y _m after frequency domain suppression is obtained;

The least squares estimation LS is performed on y _m to obtain an LS estimated output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS; and the LS estimation output vector is weighted to obtain a channel estimation value.

Optionally, the step of processing Y _m in the frequency domain, deleting part of the interference spectrum, and obtaining the frequency domain suppressed vector y _m includes:

The pilot vector Y _{m of} the frequency domain baseband signal is sorted by the square size, and the half of the spectral value of the minimum modulus value is selected; the mean square value of the selected spectral line is obtained, and then a factor θ is multiplied as a gate Limit, θ is the ratio of noise to useful signal;

Comparing the modulus value of the pilot vector Y _m spectral line of the frequency domain baseband signal with a threshold value, deleting a spectral line larger than the threshold value, and maintaining a spectral line smaller than the threshold value , get the vector y _m after the line is deleted.

Optionally, the step of weighting the LS estimation output vector to obtain the channel estimation value includes:

S401, storing a U, U ^H matrix and a non-zero element number Q1 calculated in advance;

S402. Calculate a weight matrix according to the stored U, U ^H, and Q1 values, and weight the LS estimation output vector to obtain a channel estimation value.

Optionally, the calculation process of the U, U ^H, and Q1 is as follows:

The DFT-Based channel estimation result of the LS estimated output vector is:

make

will

Recorded as:

Wherein, T ₁₁ is a square matrix of N _SC order, and N _SC is a number of subcarriers occupied by the user.

For the estimated values that need to be obtained, there are:

among them,

0, EVD decomposition of T ₁₁ gives:

T ₁₁ =U Λ U ^H , U is a eigenvector obtained by eigenvalue decomposition of T ₁₁ ;

Where Λ is a diagonal matrix, the number of non-zero elements is Q1, and Q1<Q, Q is a rectangular window length.

Optionally, the weight matrix is calculated according to the stored U, U ^H, and Q1 values, and the LS estimation output vector is weighted, and the obtained channel estimation value includes:

The least squares estimate is windowed in the feature domain:

h _U is the channel estimation value obtained by performing the weighting. .

An embodiment of the present invention further provides an LTE uplink system channel estimation apparatus under interference conditions, including:

Fast Fourier transform module configured to receive steering vector base band signal, obtained through the fast Fourier transform steering vectors Y _m of frequency domain baseband signal;

The frequency domain processing module is configured to process Y _m in the frequency domain, delete part of the interference spectrum line, and obtain a vector y _m after frequency domain suppression;

The channel estimation module is configured to perform least squares estimation LS on y _m to obtain an LS estimation output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS; weighting the LS estimation output vector to obtain a channel estimated value.

Optionally, the frequency domain processing module includes:

a threshold generation submodule, configured to sort the pilot vector Y _{m of} the frequency domain baseband signal by a modulus square, select a half of the spectral value with the smallest modulus value; and obtain a norm average of the selected spectral line, and then Multiply a factor θ as the threshold value, and θ is the ratio of the noise to the useful signal;

a spectral line deletion submodule, configured to compare a modulus value of the pilot vector Y _m spectral line of the frequency domain baseband signal with a threshold value, and delete a spectral line greater than the threshold value, less than the gate The spectral line of the limit remains unchanged, and the vector y _m after the line deletion is obtained.

Optionally, the channel estimation module includes:

The least squares estimation submodule is set to perform least squares estimation LS on y _m to obtain an LS estimated output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS;

Weighted sub-modules, including:

a storage unit, configured to save a previously calculated U, U ^H matrix and a non-zero element number Q1;

The calculating unit is configured to calculate the U, U ^H matrix and the non-zero element number Q1, calculate a weight matrix according to the stored U, U ^H and Q1 values and weight the LS estimated output vector to obtain a channel estimation value.

Optionally, the calculation process of the U, U ^H, and Q1 is as follows:

The DFT-Based channel estimation result of the LS estimated output vector is:

make

will

Recorded as:

For the estimated values that need to be obtained, there are:

among them,

0, EVD decomposition of T ₁₁ gives:

Optionally, the calculating unit is configured to calculate a weight matrix according to the stored U, U ^H, and Q1 values and weight the LS estimated output vector, where the channel estimation value is obtained by:

The least squares estimate is windowed in the feature domain:

h _U is the channel estimation value obtained by performing the weighting.

An embodiment of the present invention provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the foregoing method.

The embodiment of the invention proposes a channel estimation scheme based on frequency domain deletion and EVD decomposition (eigenvalue decomposition), which solves the problem that the traditional DFT-Based channel estimation technology can not suppress strong interference due to energy diffusion and channel estimation accuracy. The problem of deterioration. Under the premise of preserving the channel energy as much as possible, the influence of noise and interference is fully eliminated, and the channel estimation accuracy is improved.

Advantageous effects of embodiments of the present invention include:

(1) In the frequency domain, the power spectrum of the narrowband interference is only distributed in a narrow frequency band, and the frequency domain steering vector is frequency-domain deleted, a part of the interference spectrum is deleted, the influence of the narrowband interference is reduced, and the LS channel estimation is improved. Accuracy

(2) Consider the DFT-Based method as a weighting process, and then perform EVD on the weight matrix. Decomposition, using the characteristics of the concentration of channel energy, the EVD-Based weighting value is obtained, and the weighting value is used for the channel estimation value to achieve better estimation accuracy.

Other aspects will be apparent upon reading and understanding the drawings and detailed description. (Other aspects will be appreciated upon reading and understanding the attached figures and detailed description)

BRIEF abstract

1 is a flowchart of a method for estimating an LTE uplink system channel under interference conditions according to an embodiment of the present invention;

2 is a structural diagram of a baseband receiver of a typical LTE uplink system in an embodiment of the present invention;

3 is a structural diagram of an LTE uplink system channel estimation apparatus under interference conditions according to an embodiment of the present invention;

4 is a schematic structural diagram of a weighting submodule according to an embodiment of the present invention;

FIG. 5 is a performance comparison diagram of a common channel estimation method according to an embodiment of the present invention in a case where a dry signal ratio is changed and a signal to noise ratio is fixed;

FIG. 6 is a performance comparison diagram of a common channel estimation method in a case where the dry signal ratio is fixed and the signal to noise ratio is changed according to an embodiment of the present invention.

Preferred embodiment of the invention

The technical solution of the present invention will be described in more detail below with reference to the accompanying drawings and embodiments.

It should be noted that, if not conflicting, the embodiments of the present invention and the various features of the embodiments may be combined with each other, and are all within the protection scope of the present invention. Additionally, although logical sequences are shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.

Embodiment 1 A method for estimating an LTE uplink system channel under interference conditions, including:

Sl, reception baseband signal steering vectors, and then through a Fast Fourier Transform FFT to obtain steering vectors Y _m of frequency domain baseband signal;

S2, processing Y _m in the frequency domain, deleting part of the interference spectrum, and obtaining the vector y _m after frequency domain suppression;

S3, performing least squares estimation on y _m to obtain an LS estimated output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS;

S4. Weight the LS estimation output vector to obtain a channel estimation value.

In an implementation manner of this embodiment, the S2 includes:

S201: Calculate a threshold value of the spectral line deletion according to the guiding vector Y _{m of} the frequency domain baseband signal, and calculate the threshold value as follows:

The steering vector Y _{m is} sorted by the square of the modulus, and the half of the spectrum with the smallest modulus value is selected. Since the narrowband signal is distributed in a narrow frequency band, it is considered that the partial line is not contaminated by narrowband interference, only noise and useful. signal. Then find the mean square of the partial line, and then multiply a factor θ as the threshold, θ is the ratio of the noise to the useful signal;

S202, delete the line:

Comparing the modulus value of the guiding vector Y _m spectral line with the threshold value, deleting the spectral line larger than the threshold value, and maintaining the spectral line smaller than the threshold value, and obtaining the spectral line deleted Vector y _m .

In an implementation manner of this embodiment, the step S4 includes:

S401. Store the U, U ^H matrix and the non-zero element number Q1 calculated in advance into the storage unit, where U, U ^H and Q1 are calculated as follows:

Let the expected user's least squares estimate over the entire frequency band be

Recorded as:

Where h _LS =X ^-1 y _m is the least squares estimate of the desired user at its occupied subcarriers,

Valuation for virtual LS, not available, right

Do IFFT get

For N-point IFFT,

It can be seen that the time domain tap of the channel does not exceed

The first M elements, where M is the length of the CP (cyclic prefix); therefore, the time domain window is selected to filter out most of the interference and noise while saving all channel energy, and then transform the result into the frequency domain to obtain DFT. The -Based channel estimation result is:

Diag() is a diagonal matrix,

When the value of Q exceeds the channel time domain tap length, all the energy of the channel is collected, and the DFT-Based method is regarded as a weighting process, namely:

make

Divide the weight matrix, you can

Recorded as:

Where T ₁₁ is an N _SC order square matrix (N _SC is the number of subcarriers occupied by the user),

For the estimated value that needs to be obtained,

A channel estimation value corresponding to a subcarrier other than the subcarrier occupied by the user, which is not an estimated value to be obtained for the desired user,

Among them, because

Unable to get, record it as 0, and perform EVD decomposition on T ₁₁ to get:

Where Λ is a diagonal matrix, the number of non-zero elements is Q1, and Q1<Q, Q is a rectangular window window length;

S402. Calculate a weight matrix according to the stored U, U ^H, and Q1 values and weight the LS estimation output vector to obtain a channel estimation value h _U .

The least squares estimate is windowed in the feature domain:

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the above method.

Embodiment 2: An LTE uplink system channel estimation apparatus under interference conditions, including:

The fast Fourier transform module 1 is configured to receive a pilot vector of the baseband signal, and obtain a pilot vector Y _{m of the} frequency domain baseband signal through fast Fourier transform;

The frequency domain processing module 2 is configured to process Y _m in the frequency domain, delete part of the interference spectrum line, and obtain a vector y _m after frequency domain suppression;

The channel estimation module 3 is configured to perform least squares estimation LS on y _m to obtain an LS estimation output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS; weighting the LS estimation output vector to obtain Channel estimate.

In an implementation manner of this embodiment, the frequency domain processing module 2 includes:

The threshold generation sub-module 21 is configured to sort the guidance vector Y _{m of} the frequency domain baseband signal by a square size, and select a half of the spectral line with the smallest modulus value; and obtain a modulus average of the selected spectral line. Then multiply a factor θ as the threshold value, θ is the ratio of the noise to the useful signal;

The spectral line deletion sub-module 22 is configured to compare a modulus value of the pilot vector Y _m spectral line of the frequency domain baseband signal with a threshold value, and delete a spectral line greater than the threshold value, less than the The spectral line of the threshold remains unchanged, and the vector y _m after the spectral line deletion is obtained.

In an implementation manner of this embodiment, the channel estimation module 3 includes:

The least squares estimation sub-module 31 is configured to perform least squares estimation LS on y _m to obtain an LS estimated output vector h _LS =X ^-1 y _m , where X is an LTE uplink demodulation reference signal DMRS;

The weighting sub-module 32 includes:

The storage unit 321 is configured to save the U, U ^H matrix and the non-zero number Q1 calculated in advance;

The calculating unit 322 is configured to calculate a U, U ^H matrix and a non-zero element number Q1, and calculate a weight matrix according to the stored U, U ^H and Q1 values and weight the LS estimated output vector to obtain a channel estimation value.

In this embodiment, optionally, the calculation process of the U, U ^H, and Q1 is as follows:

The DFT-Based channel estimation result of the LS estimated output vector is:

Diag() is a diagonal matrix,

make

will

Recorded as:

For the estimated value that needs to be obtained,

among them,

0, EVD decomposition of T ₁₁ gives:

In this embodiment, optionally, the calculating unit calculates the weight matrix according to the stored U, U ^H, and Q1 values and weights the LS estimation output vector to:

The computing unit performs windowing on the feature domain by least squares estimation:

As shown in FIG. 2, a typical LTE uplink system baseband receiver includes: an FFT module, a frequency domain processing module, a channel estimation module, a demodulation module, and a sink module, where:

The FFT module performs FFT transformation on a baseband received signal vector;

The demodulation module performs demodulation processing on the data to obtain bit level data.

As shown in FIG. 3, an LTE uplink system channel estimation apparatus under interference conditions includes a fast Fourier transform module 1, a frequency domain processing module 2, and a channel estimation module 3. The frequency domain processing module 2 includes a threshold. The generating sub-module 21 and the spectral line deleting sub-module 22:

Threshold generation sub-module 21: calculating a threshold value of the spectral line deletion according to the frequency domain receiving pilot signal vector;

The spectral line deletion sub-module 22 compares the modulus value of the received vector Y _m spectral line with the threshold value, and removes the spectral line larger than the threshold value, and the spectral line smaller than the threshold value remains unchanged, and the spectral line deletion is obtained. After the vector y _m .

As shown in FIG. 4, the channel estimation module includes a least squares estimation sub-module 31 and a weighting sub-module 32:

The least squares estimation sub-module 31: performs LS channel estimation according to the designed pilot symbol and the received pilot symbol to obtain an output LS channel estimation value as follows: h _LS =X ^-1 y _m ;

Weighting sub-module 32: Calculating the weight vector acts on the input vector to obtain a channel estimate.

The weighting sub-module 32 includes a storage unit 321 and a computing unit 322 configured to store a first calculated U, U ^H matrix and a non-zero number Q1; the computing unit 322 is configured to calculate a channel estimate.

FIG. 5 is a performance comparison diagram of a common channel estimation method in a case where a dry signal ratio (ISR) is changed and a signal to noise ratio (SNR) is fixed, wherein the MSE represents a mean square error of the channel estimation, and ISR=-10: 2:10dB, SNR=0dB. It can be seen from the figure that the performance of the dual-domain channel estimation is significantly better than that of the DFT-Based channel estimation in both the low dry signal ratio and the high dry signal ratio.

FIG. 6 is a performance comparison diagram of a common channel estimation method in a fixed-signal ratio (ISR) and a signal-to-noise ratio (SNR) according to a common channel estimation method, where SNR=0: 2:16 dB, ISR=10 dB, from the figure It can be seen that in the case of strong interference, the performance of the dual-domain channel estimation is significantly better than the DFT-Based channel estimation.

The present invention has been described in detail by way of specific embodiments thereof, and the description of the above embodiments is provided to enable those skilled in the art to make or use the present invention, and various modifications of these embodiments are readily apparent to those skilled in the art Understand. The invention is not limited to these examples, or some aspects thereof. The scope of the invention is to be described in detail by the appended claims.

The above description shows and describes a preferred embodiment of the present invention, but as described above, it should be understood that the present invention is not limited to the form disclosed herein, and should not be construed as Other combinations, modifications, and environments are possible and can be modified by the above teachings or related art or knowledge within the scope of the inventive concept described herein. All changes and modifications made by those skilled in the art are intended to be within the scope of the appended claims.

One of ordinary skill in the art will appreciate that all or a portion of the steps of the above-described embodiments can be implemented using a computer program flow, which can be stored in a computer readable storage medium, such as on a corresponding hardware platform (eg, The system, device, device, device, etc. are executed, and when executed, include one or a combination of the steps of the method embodiments.

Alternatively, all or part of the steps of the above embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve.

The devices/function modules/functional units in the above embodiments may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.

When each device/function module/functional unit in the above embodiment is implemented in the form of a software function module and sold or used as a stand-alone product, it can be stored in a computer readable storage medium. The above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Industrial applicability

The above technical solution reduces the influence of narrowband interference, improves the accuracy of LS channel estimation, and improves the accuracy of channel estimation.

Claims

A long-term evolution LTE uplink system channel estimation method under interference conditions, comprising:

Receiving a steering vector of the baseband signal, and obtaining a steering vector Y m of the frequency domain baseband signal through fast Fourier transform;

Y m is processed in the frequency domain, and part of the interference spectrum is deleted, and the vector y m after frequency domain suppression is obtained;

Performing a least squares estimation LS on y m to obtain an LS estimated output vector h LS =X -1 y m , where X is an LTE uplink demodulation reference signal DMRS;

The LS estimated output vector is weighted to obtain a channel estimate.
The method according to claim 1, wherein said processing Y m in the frequency domain to delete a part of the interference line, and obtaining the frequency domain suppressed vector y m comprises:

The pilot vector Y m of the frequency domain baseband signal is sorted by the square size, and the half of the spectral value of the minimum modulus value is selected; the mean square value of the selected spectral line is obtained, and then a factor θ is multiplied as a gate Limit, θ is the ratio of noise to useful signal;

Comparing the modulus value of the pilot vector Y m spectral line of the frequency domain baseband signal with a threshold value, deleting a spectral line larger than the threshold value, and maintaining a spectral line smaller than the threshold value , get the vector y m after the line is deleted.
The method of claim 1 wherein said step of weighting said LS estimate output vector to obtain a channel estimate comprises:

Preserving the previously calculated U, U H matrix and non-zero element number Q1;

Calculating a weight matrix according to the stored U, U H and Q1 values and weighting the LS estimation output vector to obtain a channel estimation value.
The method of claim 3 wherein said U, U H and Q1 calculations comprise:

The channel estimation result based on the Fourier transform of the LS estimated output vector is:

make

will
Recorded as:

Wherein, T 11 is a square matrix of N SC order, and N SC is a number of subcarriers occupied by the user.
For the estimated values that need to be obtained, there are:

among them,
0, EVD decomposition of T 11 gives:

T 11 =U Λ U H , U is a eigenvector obtained by eigenvalue decomposition of T 11 ;

Where Λ is a diagonal matrix, the number of non-zero elements is Q1, and Q1<Q, Q is a rectangular window length.
The method of claim 4, wherein said calculating a weight matrix from said stored U, U H and Q1 values and weighting said LS estimated output vector comprises obtaining:

The least squares estimate is windowed in the feature domain:

h U is the channel estimation value obtained by performing the weighting.
A Long Term Evolution (LTE) uplink system channel estimation apparatus under interference conditions, comprising:

Fast Fourier transform module configured to receive steering vector base band signal, obtained through the fast Fourier transform steering vectors Y m of frequency domain baseband signal;

The frequency domain processing module is configured to process Y m in the frequency domain, delete part of the interference spectrum line, and obtain a vector y m after frequency domain suppression;

The channel estimation module is configured to perform least squares estimation LS on y m to obtain an LS estimation output vector h LS =X -1 y m , where X is an LTE uplink demodulation reference signal DMRS; weighting the LS estimation output vector to obtain a channel estimated value.
The apparatus of claim 6 wherein said frequency domain processing module comprises:

a threshold generation submodule, configured to sort the pilot vector Y m of the frequency domain baseband signal by a modulus square, select a half of the spectral value with the smallest modulus value; and obtain a norm average of the selected spectral line, and then Multiply a factor θ as the threshold value, and θ is the ratio of the noise to the useful signal;

a spectral line deletion submodule, configured to compare a modulus value of the pilot vector Y m spectral line of the frequency domain baseband signal with a threshold value, and delete a spectral line greater than the threshold value, less than the gate The spectral line of the limit remains unchanged, and the vector y m after the line deletion is obtained.
The apparatus of claim 6 wherein said channel estimation module comprises:

The least squares estimation submodule is set to perform least squares estimation LS on y m to obtain an LS estimated output vector h LS =X -1 y m , where X is an LTE uplink demodulation reference signal DMRS;

Weighted sub-modules, including:

a storage unit, configured to save a previously calculated U, U H matrix and a non-zero element number Q1;

The calculating unit is configured to calculate a U, U H matrix and a non-zero element number Q1, calculate a weight matrix according to the stored U, U H and Q1 values and weight the LS estimated output vector to obtain a channel estimation value.
The apparatus of claim 8, wherein the calculation unit is configured to implement calculations of U, U H, and Q1 by:

The channel estimation result based on the Fourier transform of the LS estimated output vector is:

make

will
Recorded as:

Wherein, T 11 is a square matrix of N SC order, and N SC is a number of subcarriers occupied by the user.
For the estimated values that need to be obtained, there are:

among them,
0, EVD decomposition of T 11 gives:

T 11 =U Λ U H , U is a eigenvector obtained by eigenvalue decomposition of T 11 ;

Where Λ is a diagonal matrix, the number of non-zero elements is Q1, and Q1<Q, Q is a rectangular window length.
The apparatus according to claim 9, wherein said calculating unit is configured to calculate a weight matrix from the stored U, U H and Q1 values and weight the LS estimated output vector to obtain a channel estimate by: value:

The least squares estimate is windowed in the feature domain:

h U is the channel estimation value obtained by performing the weighting.
A computer storage medium having stored therein computer executable instructions for performing the method of any one of claims 1 to 5.