CN102075220B - Channel estimating device and method based on time domain noise reduction - Google Patents

Channel estimating device and method based on time domain noise reduction Download PDF

Info

Publication number
CN102075220B
CN102075220B CN 200910222635 CN200910222635A CN102075220B CN 102075220 B CN102075220 B CN 102075220B CN 200910222635 CN200910222635 CN 200910222635 CN 200910222635 A CN200910222635 A CN 200910222635A CN 102075220 B CN102075220 B CN 102075220B
Authority
CN
China
Prior art keywords
matrix
noise reduction
transform
channel estimating
channel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN 200910222635
Other languages
Chinese (zh)
Other versions
CN102075220A (en
Inventor
李家海
史凡
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN 200910222635 priority Critical patent/CN102075220B/en
Publication of CN102075220A publication Critical patent/CN102075220A/en
Application granted granted Critical
Publication of CN102075220B publication Critical patent/CN102075220B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention provides a channel estimating device and method based on time domain noise reduction, wherein the method comprises the steps of: obtaining an original channel estimation value of each array element of each uplink user, received by a base station array antenna, to form a channel estimation matrix, and decomposing the channel estimation matrix into a channel estimation amplitude matrix and a channel estimation phase matrix for carrying out signal conversion on each array element of the channel estimation amplitude matrix and then carrying out denoising treatment, and converting the denoised channel estimation amplitude matrix and then restoring phase information of the channel estimation amplitude matrix. By adopting the technical scheme provided by the invention, the noise influence on a channel estimation signal tap is reduced by using a time domain denoising method, thus the accuracy of the channel estimation of the system is further improved and the performance of a TD-SCDMA (Time Division-Synchronization Code Division Multiple Access) system is improved.

Description

A kind of channel estimating apparatus and method based on time domain noise reduction
Technical field
The present invention relates to a kind of wireless communication technology, more specifically, relate to a kind of channel estimating apparatus based on time domain noise reduction and method.
Background technology
Channel estimating in TD-SCDMA (Time Division-Synchronous Code Division Multiple Access, the TD SDMA access) system is the basis of the key technologies such as joint-detection, measurement, smart antenna.Fig. 1 is the structure chart of the burst of TD-SCDMA business time-slot, wherein, training sequence (Midamble) is as carrying out channel estimating, for same residential quarter, system determines a basic Midamble code as basic code, and different user adopts the different cyclic shift versions conducts training sequence separately of this basic code to be used for channel estimating.
In the existing TD-SCDMA system, adopt the Steiner estimator of broad sense to carry out the channel estimating of array antenna, the result who obtains is called original channel response: h=G -1e Mid, e wherein MidThe training sequence that expression receives, G represent the circular matrix that the basic code of this residential quarter forms.Estimate to compare with ideal communication channel, original channel estimates to have comprised the impact of interference and noise, therefore, is obtaining will further to carry out reprocessing after original channel is estimated to suppress interference and the noise in the channel estimation results.
The method of channel estimating reprocessing of the prior art is by fixing or adaptive power thresholding, removes the noise tap and the small-signal tap that did not have thresholding in the subscriber channel estimating window, thereby improves the precision that communication system channel is estimated.But the method is unable to estimate and is removed for the residual noise in the channel estimating signal tap, so precision of channel estimation is relatively relatively poor.
Summary of the invention
The technical problem to be solved in the present invention provides a kind of channel estimating apparatus based on time domain noise reduction and method, can realize reducing the noise effect of channel estimating signal tap, further improves the accuracy that system channel is estimated.
In order to address the above problem, the invention provides a kind of channel estimation methods based on time domain noise reduction, described method comprises:
The original channel estimated value of obtaining up each each array element of user that base station array antenna receives forms channel estimate matrix, and described channel estimate matrix is decomposed into the channel estimating magnitude matrix With the channel estimating phasing matrix
Figure G2009102226355D00022
To described channel estimating magnitude matrix
Figure G2009102226355D00023
Each array element carry out making noise reduction process after the signal conversion, and the channel estimating magnitude matrix behind the noise reduction is carried out recovering after the inverse transformation its phase information.
Further, described signal mapping mode comprises discrete Fourier transform, discrete cosine transform and wavelet transform.
Further, when adopting discrete cosine transform, to described channel estimating magnitude matrix
Figure G2009102226355D00024
Each array element carry out W point discrete cosine transform conversion, afterwards N data of the HFS after the conversion are carried out the transform domain noise reduction process and obtain array To each array
Figure G2009102226355D00026
Do to form new channel estimating magnitude matrix A after the inverse discrete cosine transform k, with described matrix A kWith the channel estimating phasing matrix Make point multiplication operation, obtain through the channel estimation value H after the time domain discrete cosine transform k
Described W is that user's channel estimation window is long.
Further, when adopting wavelet transform, to described channel estimating magnitude matrix
Figure G2009102226355D00028
Each array element carry out W point wavelet transform, afterwards N data of the HFS after the conversion are carried out the transform domain noise reduction process and obtain new channel estimating magnitude matrix
Figure G2009102226355D00029
Again to described
Figure G2009102226355D000210
Every row carry out respectively W point discrete wavelet inverse transformation and obtain matrix A k, again with matrix A kWith Obtain through the channel estimation value H behind the time domain discrete wavelet transformation as point multiplication operation k
Described W is that user's channel estimation window is long.
Further, described N data to the HFS after the conversion are carried out the transform domain noise reduction process and are referred to, N data of described HFS are set to 0.
The present invention also provides a kind of channel estimating apparatus based on time domain noise reduction, and described device comprises:
Generation module, the original channel estimated value that is used for obtaining up each each array element of user that base station array antenna receives generates channel estimate matrix;
Decomposing module is used for described channel estimate matrix is decomposed into the channel estimating magnitude matrix
Figure G2009102226355D000212
With the channel estimating phasing matrix
Figure G2009102226355D00031
Conversion module is used for described channel estimating magnitude matrix Each array element carry out the signal conversion process;
Noise reduction module, the transform domain data medium-high frequency data partly that are used for obtaining after the signal conversion process are carried out the transform domain noise reduction process;
Inverse transform module is used for the channel estimating magnitude matrix after the noise reduction process is carried out inverse transformation, recovers its phase information.
Further, the signal mapping mode of described conversion module employing comprises discrete Fourier transform, discrete cosine transform and wavelet transform.
Further, described conversion module is to described channel estimating magnitude matrix
Figure G2009102226355D00033
Each array element carry out the signal conversion and refer to, conversion module is to described channel estimating magnitude matrix
Figure G2009102226355D00034
Each array element carry out W point discrete cosine transform conversion;
N data of the transform domain data medium-high frequency part that described noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained array
The channel estimating magnitude matrix of described inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to each array Do to form new channel estimating magnitude matrix A after the inverse discrete cosine transform k, and with described matrix A kWith the channel estimating phasing matrix
Figure G2009102226355D00037
Make point multiplication operation, obtain through the channel estimation value H after the time domain discrete cosine transform k
Described W is that user's channel estimation window is long.
Further, described conversion module is to described channel estimating magnitude matrix
Figure G2009102226355D00038
Each array element carry out the signal conversion and refer to, conversion module is to described channel estimating magnitude matrix
Figure G2009102226355D00039
Each array element carry out W point wavelet transform;
N data of the transform domain data medium-high frequency part that described noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained new channel estimating magnitude matrix
Figure G2009102226355D000310
The channel estimating magnitude matrix of described inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to described
Figure G2009102226355D000311
Every row carry out respectively W point discrete wavelet inverse transformation and obtain matrix A k, again with matrix A kWith
Figure G2009102226355D000312
Obtain through the channel estimation value H behind the time domain discrete wavelet transformation as point multiplication operation k
Described W is that user's channel estimation window is long.
Further, N data of the HFS of described noise reduction module after to conversion are carried out the transform domain noise reduction process and are referred to, noise reduction module is set to 0 with N data of described HFS.
In sum, the invention provides a kind of channel estimating apparatus based on time domain noise reduction and method, the upward signal that the base station termination is received, obtaining its original channel estimates, use the noise effect of the method reduction channel estimating signal tap of time domain noise reduction, further improve the accuracy that system channel is estimated, promoted the performance of TD-SCDMA system.
Description of drawings
Fig. 1 is the structural representation of the burst of TD-SCDMA business time-slot;
Fig. 2 is apparatus of the present invention structural representations;
Fig. 3 is the inventive method schematic flow sheet.
Embodiment
In order further to reduce the noise effect of channel estimating signal tap, further improve the accuracy that system channel is estimated, the elevator system performance the invention provides a kind of channel estimating apparatus based on time domain noise reduction and method, and the method can reduce noise to the impact of channel estimating.
The present embodiment provides a kind of channel estimating apparatus based on time domain noise reduction, as shown in Figure 2, comprises generation module, decomposing module, conversion module, noise reduction module and inverse transform module; This channel estimating apparatus is positioned at base station side.
Generation module, the original channel estimated value that is used for obtaining up each each array element of user that base station array antenna receives generates channel estimate matrix;
Decomposing module, the channel estimate matrix that is used for generating is decomposed into the channel estimating magnitude matrix
Figure G2009102226355D00041
With the channel estimating phasing matrix
Figure G2009102226355D00042
Conversion module is used for the channel estimating magnitude matrix
Figure G2009102226355D00043
Each array element carry out the signal conversion process;
Noise reduction module, the transform domain data medium-high frequency data partly that are used for obtaining after the signal conversion process are carried out the transform domain noise reduction process;
Inverse transform module is used for the channel estimating magnitude matrix after the noise reduction process is carried out inverse transformation, recovers its phase information.
The signal mapping mode that conversion module adopts can but be not limited to comprise DFT (Discrete FourierTransformation, discrete Fourier transform), DCT (Discrete Cosine Transformation, discrete cosine transform), the mode such as DWT (Discrete Wavelet Transform, wavelet transform);
If user's channel estimation window length is W.
When adopting dct transform, conversion module is to described channel estimating magnitude matrix
Figure G2009102226355D00051
Each array element carry out W point discrete cosine transform conversion; N data of the transform domain data medium-high frequency part that noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained array The channel estimating magnitude matrix of inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to each array
Figure G2009102226355D00053
Do to form new channel estimating magnitude matrix A after the inverse discrete cosine transform k, and with described matrix A kWith the channel estimating phasing matrix
Figure G2009102226355D00054
Make point multiplication operation, obtain through the channel estimation value H after the time domain discrete cosine transform k
When adopting the DWT conversion, conversion module is to the channel estimating magnitude matrix
Figure G2009102226355D00055
Each array element carry out W point wavelet transform; N data of the transform domain data medium-high frequency part that noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained new channel estimating magnitude matrix
Figure G2009102226355D00056
The channel estimating magnitude matrix of inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to described
Figure G2009102226355D00057
Every row carry out respectively W point discrete wavelet inverse transformation and obtain matrix A k, again with matrix A kWith
Figure G2009102226355D00058
Obtain through the channel estimation value H behind the time domain discrete wavelet transformation as point multiplication operation k
Processing when adopting other mapping modes is similar.
The present embodiment provides a kind of channel estimation methods based on time domain noise reduction, as shown in Figure 3, may further comprise the steps:
Step 301: after obtaining the original channel estimated value of up each each array element of user that base station array antenna receives, generate channel estimate matrix;
Channel estimate matrix can be take the channel estimation value on each array element as matrix the row vector, the channel estimation value of each antenna is as the matrix column vector on each channel estimating tap position;
Step 302: the subscriber channel estimated matrix is decomposed into channel estimating magnitude matrix and channel estimating phasing matrix;
Step 303: the element of locational each signal tap of same array element in the channel estimating magnitude matrix (being same delegation) is formed one group, carry out the signal conversion process, to realize noise reduction process;
The signal conversion here refers to realize the conversion of transform domain noise reduction process, can but be not limited to comprise DFT (Discrete Fourier Transformation, discrete Fourier transform), DCT (Discrete CosineTransformation, discrete cosine transform), DWT (Discrete Wavelet Transform, wavelet transform) mode such as, accordingly, transform domain is then distinguished corresponding DFT transform domain, dct transform domain, DWT transform domain etc.;
Step 304: the transform domain data medium-high frequency data partly that obtain after the signal conversion process are carried out the transform domain noise reduction process;
In the enforcement, the data of transform domain data medium-high frequency part carry out the transform domain noise reduction process can for: with the data zero setting of transform domain data medium-high frequency part.
Step 305: the data after the noise reduction process are carried out inverse transformation, obtain the channel estimating amplitude data behind the transform domain noise reduction;
In the enforcement, with the data zero setting of the transform domain data medium-high frequency that obtains part, carry out the transform domain noise reduction process, and then the data after zero setting processed carry out inverse transformation, obtain the channel estimating amplitude data behind the transform domain noise reduction;
Step 306: the channel estimating magnitude matrix behind the noise reduction and corresponding channel estimating phasing matrix are carried out obtaining the channel estimation value after amplitude has partly been carried out the transform domain noise reduction after the channel estimating phase information restores;
Step 307: the tap power to the channel estimation value after the transform domain noise reduction process carries out the reprocessing of channel estimating power threshold.
Below further specify the present invention by several application examples.
The method of in the present embodiment TD-SCDMA time domain noise reduction employing is that discrete cosine transform (Discrete Cosine Transform, DCT) and wavelet transform (Discrete WaveletTransform, DWT) method may further comprise the steps:
Application example one:
In this application example, take TD-SCDMA unit 8 line array as example, then TD-SCDMA time domain DCT channel estimating can be processed as follows:
Step 401, receiver utilize the steiner estimator to obtain all users' original channel estimation on each bay on the conventional arrays antenna:
h ka = G - 1 e mid ka , Wherein, ka ∈ 1...Ka, Ka are the number of antennas of conventional arrays antenna; G represents the circular matrix that is made of the midamble code; e Mid KaRepresent the training sequence that receives on the ka antenna; h KaRepresent the original channel estimation on the ka antenna.
Step 402, the channel impulse response from the original channel estimated value in the channel estimation window of intercepting user k are established in the present embodiment, and the channel estimation window of user k is long to be W=16, and antenna number is 8, generate the multi-antenna channel estimated matrix of 8 * 16 user k Be expressed in matrix as:
Figure G2009102226355D00073
Wherein, the multi-antenna channel estimated matrix of user k
Figure G2009102226355D00074
In element be
Figure G2009102226355D00075
Ka=1,2 ..., Ka, w=1,2 ..., W, h ^ w ka = a ^ w ka · exp ( j θ ^ w ka ) , Will
Figure G2009102226355D00077
Be decomposed into the channel estimating magnitude matrix
Figure G2009102226355D00078
With the channel estimating phasing matrix
Figure G2009102226355D00079
Obtain:
Figure G2009102226355D000710
Figure G2009102226355D000711
Step 403, each row of channel estimating magnitude matrix is carried out W point dct transform, namely time tap (w tap) data of each bay of magnitude matrix are carried out dct transform, since be based on the DCT that carries out in the time tap, so be referred to as time domain DCT:
y i j = w ( i - 1 ) Σ m = 0 W - 1 a ^ i j cos π ( 2 m + 1 ) ( i - 1 ) 2 W , i = 1 , . . . , W
Wherein
w ( i ) = 1 / W i = 1 2 / W i ≠ 1
Figure G2009102226355D00083
Be the channel estimating range value of i tap on j the bay, wherein W is the length of each subscriber channel estimating window, and the present embodiment is got W=16.
Step 404, the data of the HFS in the array behind the dct transform being carried out the transform domain noise reduction process, can be N data zero setting with HFS, (N＜W), obtain new array
Figure G2009102226355D00084
For:
Figure G2009102226355D00085
Then the new array that obtains is carried out inverse discrete cosine transform (IDCT)
Figure G2009102226355D00086
Wherein
w ( m ) = 1 / W m = 1 2 / W m ≠ 1
Choosing of N can obtain by emulation.
The array later through inverse discrete cosine transform forms new magnitude matrix A k, a i jIt is matrix A kElement.
Step 405, to the matrix A after the DCT inverse transformation (IDCT) kRecover its original phase information, i.e. matrix A kWith Obtain through the channel estimation value H behind the time domain DCT as point multiplication operation k:
H k = A k · θ ^ k
Step 406, to the channel estimate matrix H behind the time domain DCT kCarry out channel post-processing, the same prior art of operation afterwards no longer describes in detail herein.
Application example two:
This application example is take TD-SCDMA unit 8 line array as example, and then TD-SCDMA time domain DWT channel estimation methods can be processed as follows:
Step 501, receiver utilize the steiner estimator to obtain all users' original channel estimation on each bay on the conventional arrays antenna:
h ka = G - 1 e mid ka , Wherein, ka ∈ 1...Ka, Ka are the number of antennas of conventional arrays antenna; G represents the circular matrix that is made of the midamble code; e Mid KaRepresent the training sequence that receives on the ka antenna; h KaRepresent the original channel estimation on the ka antenna.
Step 502, the channel impulse response from the original channel estimated value in the channel estimation window of intercepting user k are established in the present embodiment, and the channel estimation window of user k is long to be W=16, and antenna number is 8, generate the multi-antenna channel estimated matrix of 8 * 16 user k
Figure G2009102226355D00092
Be expressed in matrix as:
Figure G2009102226355D00093
Wherein, the multi-antenna channel estimated matrix of user k
Figure G2009102226355D00094
In element be
Figure G2009102226355D00095
Ka=1,2 ..., Ka, w=1,2 ..., W, h ^ w ka = a ^ w ka · exp ( j θ ^ w ka ) , Will
Figure G2009102226355D00097
Be decomposed into the channel estimating magnitude matrix
Figure G2009102226355D00098
With the channel estimating phasing matrix
Figure G2009102226355D00099
Obtain:
Figure G2009102226355D000910
Figure G2009102226355D000911
Step 503, each row of magnitude matrix is carried out W point DWT conversion, namely time tap (w tap) data of each bay of channel estimating magnitude matrix are carried out the DWT conversion, owing to being based on the DWT that carries out in the time tap, so be referred to as time domain DWT.Use haar wavelet transform (Haar wavelet transform) expression in the present embodiment
Figure G2009102226355D00101
Wherein, B HaarExpression W=16 point Harr wavelet transformation.
Figure G2009102226355D00102
Obtain through the channel estimating magnitude matrix behind the time domain DWT
Figure G2009102226355D00103
Step 504, to through the channel estimating magnitude matrix after the time domain DWT conversion
Figure G2009102226355D00104
In the HFS data carry out noise reduction process, can be the element zero setting (N＜W), obtain new channel estimating magnitude matrix with last N row
Figure G2009102226355D00105
Figure G2009102226355D00111
Wherein
y ^ w ′ ka = y ^ w ka w ≤ W - N 0 w > W - N
Choosing of N can obtain by emulation.
Figure G2009102226355D00113
It is new channel estimating magnitude matrix
Figure G2009102226355D00114
Element.
Step 505, the new channel estimating magnitude matrix to obtaining again
Figure G2009102226355D00115
Every row carry out respectively W point IDWT:
A k = Y ^ k ′ * B Haar
To the matrix A after the DWT inverse transformation (IDWT) kRecover its original phase information, i.e. matrix A kWith
Figure G2009102226355D00117
Obtain through the channel estimation value H behind the time domain DWT as point multiplication operation k:
H k = A k · θ ^ k
Step 506, the channel estimating behind the time domain DWT is carried out channel post-processing, the same prior art of operation afterwards no longer describes in detail herein.

Claims (8)

1. the channel estimation methods based on time domain noise reduction is characterized in that, described method comprises:
The original channel estimated value of obtaining up each each array element of user that base station array antenna receives forms channel estimate matrix, and described channel estimate matrix is decomposed into the channel estimating magnitude matrix
Figure FDA00003118088100011
With the channel estimating phasing matrix
Figure FDA00003118088100012
To described channel estimating magnitude matrix
Figure FDA00003118088100013
Each array element carry out making noise reduction process after the signal conversion, and the channel estimating magnitude matrix behind the noise reduction is carried out recovering after the inverse transformation its phase information;
Described signal mapping mode comprises discrete Fourier transform, discrete cosine transform and wavelet transform;
When adopting discrete cosine transform, to described channel estimating magnitude matrix
Figure FDA00003118088100014
Each array element carry out the discrete cosine transform of W point, afterwards N data of the HFS after the conversion are carried out the transform domain noise reduction process and obtain array
Figure FDA00003118088100015
To each array
Figure FDA00003118088100016
Do to form new channel estimating magnitude matrix A after the inverse discrete cosine transform k, with described matrix A kWith the channel estimating phasing matrix
Figure FDA00003118088100017
Make point multiplication operation, obtain through the channel estimation value H after the time domain discrete cosine transform k
Described W is that user's channel estimation window is long.
2. the method for claim 1, it is characterized in that: described N data to the HFS after the conversion are carried out the transform domain noise reduction process and are referred to, N data of described HFS are set to 0.
3. the channel estimation methods based on time domain noise reduction is characterized in that, described method comprises:
The original channel estimated value of obtaining up each each array element of user that base station array antenna receives forms channel estimate matrix, and described channel estimate matrix is decomposed into the channel estimating magnitude matrix
Figure FDA00003118088100018
With the channel estimating phasing matrix
Figure FDA00003118088100019
To described channel estimating magnitude matrix
Figure FDA000031180881000110
Each array element carry out making noise reduction process after the signal conversion, and the channel estimating magnitude matrix behind the noise reduction is carried out recovering after the inverse transformation its phase information;
Described signal mapping mode comprises discrete Fourier transform, discrete cosine transform and wavelet transform;
When adopting wavelet transform, to described channel estimating magnitude matrix
Figure FDA000031180881000111
Each array element carry out W point wavelet transform, afterwards N data of the HFS after the conversion are carried out the transform domain noise reduction process and obtain new channel estimating magnitude matrix Again to described
Figure FDA000031180881000113
Every row carry out respectively W point discrete wavelet inverse transformation and obtain matrix A k, again with matrix A kWith
Figure FDA000031180881000114
Obtain through the channel estimation value H behind the time domain discrete wavelet transformation as point multiplication operation k
Described W is that user's channel estimation window is long.
4. method as claimed in claim 3, it is characterized in that: described N data to the HFS after the conversion are carried out the transform domain noise reduction process and are referred to, N data of described HFS are set to 0.
5. the channel estimating apparatus based on time domain noise reduction is characterized in that, described device comprises:
Generation module, the original channel estimated value that is used for obtaining up each each array element of user that base station array antenna receives generates channel estimate matrix;
Decomposing module is used for described channel estimate matrix is decomposed into the channel estimating magnitude matrix With the channel estimating phasing matrix
Conversion module is used for described channel estimating magnitude matrix
Figure FDA00003118088100023
Each array element carry out the signal conversion process; The signal mapping mode that described conversion module adopts comprises discrete Fourier transform, discrete cosine transform and wavelet transform;
Noise reduction module, the transform domain data medium-high frequency data partly that are used for obtaining after the signal conversion process are carried out the transform domain noise reduction process;
Inverse transform module is used for the channel estimating magnitude matrix after the noise reduction process is carried out inverse transformation, recovers its phase information;
Described conversion module is to described channel estimating magnitude matrix
Figure FDA00003118088100024
Each array element carry out the signal conversion and refer to, conversion module is to described channel estimating magnitude matrix
Figure FDA00003118088100025
Each array element carry out the discrete cosine transform of W point;
N data of the transform domain data medium-high frequency part that described noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained array
Figure FDA00003118088100026
The channel estimating magnitude matrix of described inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to each array
Figure FDA00003118088100027
Do to form new channel estimating magnitude matrix A after the inverse discrete cosine transform k, and with described matrix A kWith the channel estimating phasing matrix
Figure FDA00003118088100028
Make point multiplication operation, obtain through the channel estimation value H after the time domain discrete cosine transform k
Described W is that user's channel estimation window is long.
6. device as claimed in claim 5 is characterized in that: N data of the HFS of described noise reduction module after to conversion are carried out the transform domain noise reduction process and are referred to, noise reduction module is set to 0 with N data of described HFS.
7. the channel estimating apparatus based on time domain noise reduction is characterized in that, described device comprises:
Generation module, the original channel estimated value that is used for obtaining up each each array element of user that base station array antenna receives generates channel estimate matrix;
Decomposing module is used for described channel estimate matrix is decomposed into the channel estimating magnitude matrix
Figure FDA00003118088100031
With the channel estimating phasing matrix
Figure FDA00003118088100032
Conversion module is used for described channel estimating magnitude matrix
Figure FDA00003118088100033
Each array element carry out the signal conversion process; The signal mapping mode that described conversion module adopts comprises discrete Fourier transform, discrete cosine transform and wavelet transform;
Noise reduction module, the transform domain data medium-high frequency data partly that are used for obtaining after the signal conversion process are carried out the transform domain noise reduction process;
Inverse transform module is used for the channel estimating magnitude matrix after the noise reduction process is carried out inverse transformation, recovers its phase information;
Described conversion module is to described channel estimating magnitude matrix
Figure FDA00003118088100034
Each array element carry out the signal conversion and refer to, conversion module is to described channel estimating magnitude matrix
Figure FDA00003118088100035
Each array element carry out W point wavelet transform;
N data of the transform domain data medium-high frequency part that described noise reduction module obtains after with the signal conversion process are carried out the transform domain noise reduction process and are obtained new channel estimating magnitude matrix
The channel estimating magnitude matrix of described inverse transform module after to noise reduction process carried out inverse transformation, recovers its phase information and refers to, inverse transform module is to described
Figure FDA00003118088100037
Every row carry out respectively W point discrete wavelet inverse transformation and obtain matrix A k, again with matrix A kWith
Figure FDA00003118088100038
Obtain through the channel estimation value H behind the time domain discrete wavelet transformation as point multiplication operation k
Described W is that user's channel estimation window is long.
8. device as claimed in claim 7 is characterized in that: N data of the HFS of described noise reduction module after to conversion are carried out the transform domain noise reduction process and are referred to, noise reduction module is set to 0 with N data of described HFS.
CN 200910222635 2009-11-23 2009-11-23 Channel estimating device and method based on time domain noise reduction Expired - Fee Related CN102075220B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200910222635 CN102075220B (en) 2009-11-23 2009-11-23 Channel estimating device and method based on time domain noise reduction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200910222635 CN102075220B (en) 2009-11-23 2009-11-23 Channel estimating device and method based on time domain noise reduction

Publications (2)

Publication Number Publication Date
CN102075220A CN102075220A (en) 2011-05-25
CN102075220B true CN102075220B (en) 2013-10-16

Family

ID=44033583

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200910222635 Expired - Fee Related CN102075220B (en) 2009-11-23 2009-11-23 Channel estimating device and method based on time domain noise reduction

Country Status (1)

Country Link
CN (1) CN102075220B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11843481B2 (en) 2019-10-10 2023-12-12 Huawei Technologies Co., Ltd. Channel estimation method and apparatus

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108845975B (en) * 2018-06-22 2021-04-13 深圳大学 Phase recovery method based on array antenna
CN114095319B (en) * 2021-11-05 2023-08-18 成都中科微信息技术研究院有限公司 Time-frequency smoothing method for 5G channel estimation by adopting amplitude-phase separation technology
CN115189993B (en) * 2022-06-28 2024-01-19 延安大学 TDCS basic modulation waveform design method, device and information transmitting and receiving method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1555782A2 (en) * 2004-01-15 2005-07-20 Samsung Electronics Co., Ltd. A method for designing an uplink pilot signal and a method and a system for estimating a channel in a multicarrier code division multiple access system
CN1829206A (en) * 2005-12-19 2006-09-06 北京邮电大学 A channel estimation method for TD-SCDMA system
CN102055692A (en) * 2009-10-28 2011-05-11 中兴通讯股份有限公司 Channel estimation method and device in multiaerial system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1555782A2 (en) * 2004-01-15 2005-07-20 Samsung Electronics Co., Ltd. A method for designing an uplink pilot signal and a method and a system for estimating a channel in a multicarrier code division multiple access system
CN1829206A (en) * 2005-12-19 2006-09-06 北京邮电大学 A channel estimation method for TD-SCDMA system
CN102055692A (en) * 2009-10-28 2011-05-11 中兴通讯股份有限公司 Channel estimation method and device in multiaerial system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11843481B2 (en) 2019-10-10 2023-12-12 Huawei Technologies Co., Ltd. Channel estimation method and apparatus

Also Published As

Publication number Publication date
CN102075220A (en) 2011-05-25

Similar Documents

Publication Publication Date Title
CN102739573B (en) Channel estimation methods and channel estimator
CN103701730B (en) Channel estimation method and device based on channel time-domain correlation and low-complexity compressed sensing
CN104410590A (en) Short-wave OFDM (Orthogonal Frequency Division Multiplexing) interference suppression joint channel estimation method based on compressed sensing
CN102387115B (en) OFDM pilot scheme design and channel estimation method
Huang et al. Comparison of sparse recovery algorithms for channel estimation in underwater acoustic OFDM with data-driven sparsity learning
CN101702696A (en) Implement method and device of channel estimation
CN101616104A (en) The channel estimation methods of ofdm system and device
CN1688143A (en) Method for channel estimation in multi-antenna system
CN101707582A (en) Method for estimating MIMO channel on basis of multi-phase decomposition
CN106302298A (en) A kind of method eliminating OFDM underwater sound communication system clipped noise
CN103685096A (en) Optimal pilot frequency based MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) system channel estimation method
CN102075220B (en) Channel estimating device and method based on time domain noise reduction
CN103944852B (en) A kind of impulse noise based on compressed sensing is estimated and removing method and device
CN105871440A (en) Method for receiving signals based on mixed carrier multi-antenna component transmission
CN102055692B (en) Channel estimation method and device in multiaerial system
CN109802901B (en) 3D MIMO channel estimation method and system based on angle of arrival measurement
CN100479444C (en) Block mode channel estimating method in electric line communication system
CN101409574B (en) Channel estimation method, apparatus and receiver for TD-SCDMA system
CN102377699B (en) Channel estimation method and device for multi-user multi-input multi-output (MU-MIMO) system
CN102487364B (en) Channel estimation method and apparatus thereof
CN114244675B (en) MIMO-OFDM system channel estimation method based on deep learning
CN103179058B (en) The method of estimation of channel impulse response length and device
CN113726376B (en) 1bit compression superposition CSI feedback method based on feature extraction and mutual-difference fusion
CN102832986B (en) A kind of multi-antenna diversity merges method of reseptance and equipment
CN113315551B (en) Signal detection method and device of layered space-time code system and computer equipment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20131016

Termination date: 20201123

CF01 Termination of patent right due to non-payment of annual fee