CN100502371C - A Channel Estimation Method in MIMO-OFDM System - Google Patents
A Channel Estimation Method in MIMO-OFDM System Download PDFInfo
- Publication number
- CN100502371C CN100502371C CNB2006101129941A CN200610112994A CN100502371C CN 100502371 C CN100502371 C CN 100502371C CN B2006101129941 A CNB2006101129941 A CN B2006101129941A CN 200610112994 A CN200610112994 A CN 200610112994A CN 100502371 C CN100502371 C CN 100502371C
- Authority
- CN
- China
- Prior art keywords
- channel
- antenna
- estimation
- initial
- frequency
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000001228 spectrum Methods 0.000 claims abstract description 16
- 238000001914 filtration Methods 0.000 claims description 20
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 abstract description 3
- 125000004122 cyclic group Chemical group 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 239000004579 marble Substances 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
Landscapes
- Radio Transmission System (AREA)
Abstract
This invention relates to input cross frequency multiple system signal channel estimation method in wireless communication technique field, which comprises the following steps: according to the system to receive cross training sequence and to isolate multiple antenna into several parallel single antenna signal channel to get initial signal estimation; getting frequency OLS estimation on initial channel estimation for square limitation according to system signal zone length L. This invention method uses system characteristics to isolate multiple channel to get single antenna initial channel and then uses rough channel zone responding length to process non spectrum filter section to get rough bar LS frequency signal channel estimation with high accuracy.
Description
Technical field
The present invention relates to the channel estimation methods in a kind of multiple-input and multiple-output-ofdm system, relate in particular to the processing method of robust least square (hereinafter to be referred as LS) channel estimating when having unloaded ripple in multiple-input and multiple-output-OFDM (hereinafter to be referred as MIMO-OFDM) system, belong to wireless communication technology field.
Background technology
Adopt MIMO-OFDM technology under the prerequisite that does not increase channel width and transmitting power, to significantly improve power system capacity.No matter be carry out data stream separation at receiving terminal, decoding when empty, data coherency detects or carry out precoding at transmitting terminal, it all is necessary that precise channels is estimated.Channel estimating in MIMO-ofdm system will be estimated the channel parameter between every pair of transmit receive antenna.Not only wait that the number of estimating parameter obviously increases a lot than a single aerial system this moment, and because received signal is the stack of each transmission antennas transmit signal, exist between all the other each antennas and disturb, can have a strong impact on performance for estimating channel and the further decoding performance that influences system, so channel estimating is a problem very crucial in MIMO-ofdm system.
In MIMO-ofdm system can by the design quadrature training sequence separate belong to different antennae to channel, the channel estimating in MIMO-ofdm system has just changed into the channel estimation problems in a single aerial system like this.The performance of mimo system is more more responsive to precision of channel estimation than a single aerial system, channel estimation errors can cause many antenna interference in MIMO-ofdm system in demodulating process, traditional a single aerial system channel estimation methods can not satisfy the performance requirement of mimo system, therefore needs the higher channel estimation methods of research precision.
Traditional single antenna ofdm system utilizes training sequence to carry out the LS channel estimating at frequency domain usually.Generally training sequence is designed to frequency domain and is ± 1 sequence in order to reduce computation complexity, when carrying out frequency domain LS channel estimating, can avoid matrix inversion operation like this, thereby can just can obtain frequency domain channel on corresponding each subcarrier by simple multiplying.Training sequence by the design quadrature in MIMO-ofdm system can be converted into the problem that mimo channel is estimated the problem that single antenna channel is estimated, still can use said method to obtain frequency domain channel and estimate.But such frequency domain LS precision of channel estimation generally is not enough for MIMO-ofdm system, thereby will cause the remarkable decline of systematic function.Frequency domain LS estimates not utilize the prior information of any channel, generally can know some prior informations of channel in the reality, such as rough channel time domain response length L, utilize this information to estimate that further carrying out filtering improves channel estimating performance to frequency domain LS.All subcarriers all are used for transmission signals when system, are called entire spectrum filtering in conjunction with the method for above-mentioned filtering.General ofdm system can vacate the direct current carrier wave usually and several high frequency subcarriers reduces problems such as direct current biasing and adjacent-channel interference, has same problem in MIMO-ofdm system, therefore has unloaded ripple equally.Call effective subcarrier removing unloaded ripple all subcarriers in addition.When there was unloaded ripple in system, the method that the length information that utilizes channel time domain to respond carries out filtering was called non-entire spectrum filtering.The estimation error variance difference of frequency domain channel estimated value on different subcarriers that adopts non-entire spectrum filtering to obtain, wherein the channel estimation error variance on the subcarrier of contiguous unloaded ripple is bigger.For the ofdm communication system based on the bag transmission, any one subcarrier causes error code because of evaluated error is excessive, all can seriously reduce systematic function.Therefore if want to improve the actual ofdm system performance that free carrier wave is arranged, must address the above problem.The performance of cyclic shift sign indicating number quadrature training sequence is best when frequency deviation and phase noise, so cyclic shift sign indicating number quadrature training sequence has better comprehensive performance.
Summary of the invention
The objective of the invention is to propose the channel estimation methods in a kind of multiple-input and multiple-output-ofdm system, at practical MIMO-ofdm system that free carrier wave is arranged, solve channel estimating big problem of estimation error variance on the subcarrier of contiguous unloaded ripple when having unloaded ripple.
Channel estimation methods in multiple-input and multiple-output-ofdm system that the present invention proposes comprises following each step:
(1) according to the quadrature training sequence that receives in multiple-input and multiple-output-ofdm system, multi-antenna channel is separated into a plurality of parallel single antenna channel, the initial channel estimation of the single antenna channel after obtaining separately is designated as j transmitting antenna to the initial channel estimation between i the reception antenna
(2) according to channel time domain response length information L in the system, respectively above-mentioned initial channel estimation is carried out the non-entire spectrum filtering of square restriction, obtain frequency domain least square channel estimating, be designated as H
Fine
Above-mentioned channel estimation methods wherein is separated into multi-antenna channel the method for a plurality of parallel single antenna channel, may further comprise the steps:
(1) receive the radiofrequency signal that sends in multiple-input and multiple-output-ofdm system, carry out down-frequency conversion, and periodic sampling, the base-band digital received signal obtained;
(2) from the received signal of each reception antenna of receiver, intercepting is used for the data of the quadrature training sequence of channel estimating, adopts the frequency domain least square method to obtain unsegregated domain channel response H
All
(3) according to above-mentioned quadrature training sequence, with above-mentioned unsegregated domain channel response H
AllBe separated into a plurality of parallel single antenna channel, the initial channel estimation of each single antenna channel after obtaining separating
Above-mentioned channel estimation methods, wherein according to channel time domain response length information L, the method to initial channel estimation is carried out the non-entire spectrum filtering of square restriction may further comprise the steps:
(1) estimates from frequency domain initial
In the corresponding element of all effective subcarriers in intercepting and the system, constitute the frequency domain initial estimate vector of effective subcarrier, be designated as H
Cut
(2) from standard fast fourier transform matrix F, intercept and the corresponding row of above-mentioned all effective subcarriers, and be listed as, constitute a part fast fourier transform matrix F with the corresponding preceding L of channel time domain response length L
p
(3) utilize square non-entire spectrum filtering method of restriction, to the frequency domain initial estimate vector H of above-mentioned effective subcarrier
CutCarry out filtering, the frequency domain channel that obtains on effective subcarrier is estimated
(4) above-mentioned
In with the corresponding position of unloaded ripple on insert 0, obtain frequency domain least square method channel estimating H
Fine
Channel estimation methods in multiple-input and multiple-output-ofdm system that the present invention proposes, for the MIMO-ofdm system that utilizes quadrature training sequence, utilize the orthogonal property of training sequence that multi-antenna channel is separated into a plurality of parallel single antenna channel, the initial channel estimation of the single antenna channel that obtains separating earlier.Utilize rough channel time domain response length information L that the initial channel estimation of the single antenna channel separated is carried out the non-entire spectrum filtering of square restriction again, obtain the higher robust LS frequency domain channel of precision and estimate.If length of field was L when channel was true
Real, require channel time domain length L 〉=L among the present invention
Real, and L approaches L more
RealPerformance is good more.The inventive method has the following advantages:
(1) the present invention only needs rough channel time domain response length information, only requires L 〉=L
RealGet final product, be convenient in the reality realize;
(2) the present invention has reduced the dimension of matrix by non-entire spectrum filtering, thereby has reduced computation complexity;
(3) the present invention has solved the big problem of estimation error variance on the contiguous unloaded marble carrier wave of non-entire spectrum filtering effectively by adopting the diagonal square restriction that loaded simple realization;
(4) diagonal loading coefficient α can obtain by prior search among the present invention, can search out the diagonal loading coefficient corresponding to different channels time-domain response length L in practice in advance, reduces amount of calculation.
Description of drawings
Fig. 1 is the FB(flow block) of an embodiment of the inventive method.
Fig. 2 is an embodiment of the inventive method, receives in MIMO-ofdm system at 44, adopts the structure chart of the training sequence of cyclic shift sign indicating number quadrature.
Embodiment
Channel estimation methods in multiple-input and multiple-output-ofdm system that the present invention proposes, at first according to the quadrature training sequence that receives in multiple-input and multiple-output-ofdm system, multi-antenna channel is separated into a plurality of parallel single antenna channel, the initial channel estimation of the single antenna channel after obtaining separately is designated as j transmitting antenna to the initial channel estimation between i the reception antenna
According to channel time domain response length information L, wherein L 〉=L in the system
Real, respectively above-mentioned initial channel estimation is carried out the non-entire spectrum filtering of square restriction, obtain frequency domain least square channel estimating, be designated as H
Fine
In the said method, the method that multi-antenna channel is separated into a plurality of parallel single antenna channel is: at first receive the radiofrequency signal that sends in multiple-input and multiple-output-ofdm system, carry out down-frequency conversion, and periodic sampling, the base-band digital received signal obtained; From the received signal of each reception antenna of receiver, intercepting is used for the data of the quadrature training sequence of channel estimating, adopts the frequency domain least square method to obtain unsegregated domain channel response H
AllAccording to above-mentioned quadrature training sequence, with above-mentioned unsegregated domain channel response H
AllBe separated into a plurality of parallel single antenna channel, the initial channel estimation of each single antenna channel after obtaining separating
In the said method, according to channel time domain response length information L, the method for initial channel estimation being carried out the non-entire spectrum filtering of square restriction is: estimate from frequency domain initial
In the corresponding element of all effective subcarriers in intercepting and the system, constitute the frequency domain initial estimate vector of effective subcarrier, be designated as H
CutFrom standard fast fourier transform matrix F, intercept and the corresponding row of above-mentioned all effective subcarriers, and be listed as, constitute a part fast fourier transform matrix F with the corresponding preceding L of channel time domain response length L
pUtilize square non-entire spectrum filtering method of restriction, to the frequency domain initial estimate vector H of above-mentioned effective subcarrier
CutCarry out filtering, the frequency domain channel that obtains on the higher effective subcarrier of precision is estimated
Can utilize following diagonal loading technique to realize,
In the following formula, α is the diagonal loading coefficient, and L is relevant with channel time domain response length, and its optimal value can be according to the known rough prior information of channel circumstance is obtained by search in advance; Above-mentioned
In with the corresponding position of unloaded ripple on insert 0, obtain frequency domain least square method channel estimating H
Fine
Introduce the embodiment of the inventive method below in conjunction with Fig. 1, cyclic shift is counted between the Q representative group among Fig. 1, and cyclic shift is counted in the P representative group:
MIMO of the present invention-ofdm system robust LS channel estimation methods at first will separate and belong to the right channel of different transmit receive antennas.At 4 the 4 cyclic shift sign indicating number quadrature training sequences of receiving shown in Figure 1, at first from the training sequence that each reception antenna receives, take out the long code that is used for channel estimating, adopt frequency domain LS to estimate to obtain all transmitting antennas superpose and channel response.
If s
tThe elementary cell of expression long code, H
bBe the channel frequency domain response corresponding to a long code elementary cell, N is a noise, receives data so and can be expressed as Y=H
bs
t+ N.
Can be expressed as corresponding to a long code elementary cell of all transmitting antenna stacks with channel response
In real system, because the value of each element of long code only may be ± 1 (s
tIn the value of each element can only be ± 1, and it is diagonal matrix, therefore
), following formula can be write as:
Utilize the sign indicating number orthogonal property and the cyclic shift characteristic of training sequence that multi-antenna channel is separated into a plurality of parallel single antenna channel below respectively.The training sequence of sign indicating number orthohormbic structure can be expressed as:
Wherein, C indication code orthogonal matrix, and CC
H=I,
Expression Kronecker is long-pending, S
tRepresent whole sign indicating number quadrature training sequence.By S
tOn the estimated whole long code training sequence that goes out with channel response H
AllFor:
The training sequence of sign indicating number quadrature separately not during on the same group channel response method as follows:
Wherein, H
DecouplingThe expression component from after channel, be time quadrature, the channel that belongs to has so not on the same group just separated.
The characteristic of cyclic shift is separated the channel response on the different antennae in same group in time domain in utilizing on the same group again.Only consider a group below.Expression formula according to received signal:
y
1,2(n)=s(n)⊙h
1(n)+s((n-N
cs)
N)⊙h
2(n)+v(n)
=s(n)⊙(h
1(n)+h
2((n-N
cs)
N))+v(n),
Wherein, s ((n-N
Cs)
N) represent s (n) cyclic shift N
CsLater sequence, N be ofdm system FFT/IFFT count h
1(n) and h
2(n) represent in this group time domain channel response between first and second transmitting antenna and the reception antenna respectively, ⊙ represents circular convolution.
Work as N
CsDuring 〉=L, h
1(n) and h
2(n) be quadrature in time domain, can separate.With H
DecouplingTransform to time domain, separate h in time domain
1(n) and h
2(n), more respectively to h
1(n) and h
2(n) do the FFT conversion and can obtain channel frequency domain response on each antenna
Thereby MIMO-OFDM channel just is separated into a plurality of single antenna channel, and the initial channel estimation of each single antenna channel is
Each carries out following processing respectively to the channel after separating.Getting frequency domain initial estimates
In form a vector corresponding to the element of all effective subcarriers, be designated as H
CutGet in the standard FFT matrix F corresponding to the row of effective subcarrier and preceding L row and obtain part FFT matrix, be designated as F corresponding to channel time domain response length L
pTo H
CutCarry out non-entire spectrum filtering, and introduce square restriction and obtain the higher robust LS frequency domain channel of precision and estimate, can realize a square restriction, reduce estimation error variance on the contiguous unloaded marble carrier wave, thereby obtain by the diagonal loading technique:
Wherein, α is the diagonal loading coefficient, can obtain optimum α value, channel time domain response length L difference, then F by search
pThereby the value of difference α is also different.
Claims (2)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2006101129941A CN100502371C (en) | 2006-09-14 | 2006-09-14 | A Channel Estimation Method in MIMO-OFDM System |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2006101129941A CN100502371C (en) | 2006-09-14 | 2006-09-14 | A Channel Estimation Method in MIMO-OFDM System |
Publications (2)
Publication Number | Publication Date |
---|---|
CN1925472A CN1925472A (en) | 2007-03-07 |
CN100502371C true CN100502371C (en) | 2009-06-17 |
Family
ID=37817940
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNB2006101129941A Expired - Fee Related CN100502371C (en) | 2006-09-14 | 2006-09-14 | A Channel Estimation Method in MIMO-OFDM System |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN100502371C (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101459645B (en) * | 2009-01-15 | 2012-11-14 | 清华大学 | Detection method based on sub-band in multi-antenna OFDM system |
CN102014401B (en) * | 2009-09-07 | 2013-05-08 | 鼎桥通信技术有限公司 | Channel estimate method and system in multi-antenna descending system, and base station as well as terminal thereof |
CN101945068B (en) * | 2010-09-28 | 2013-05-08 | 电子科技大学 | Detection method of receiver of single-carrier frequency domain equalizing system with low transmission power |
CN102546487B (en) * | 2011-12-13 | 2014-11-05 | 华为技术有限公司 | Channel response and power offset jointed estimation method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003034644A1 (en) * | 2001-10-17 | 2003-04-24 | Nortel Networks Limited | Scattered pilot pattern and channel estimation method for mimo-ofdm systems |
CN1579063A (en) * | 2002-09-26 | 2005-02-09 | 株式会社东芝 | Transmission signals, method and apparatus |
CN1809039A (en) * | 2005-01-19 | 2006-07-26 | 北京邮电大学 | Method of implementing estimated performance of TD-LS channels in MIMO-OFDM system |
-
2006
- 2006-09-14 CN CNB2006101129941A patent/CN100502371C/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003034644A1 (en) * | 2001-10-17 | 2003-04-24 | Nortel Networks Limited | Scattered pilot pattern and channel estimation method for mimo-ofdm systems |
CN1579063A (en) * | 2002-09-26 | 2005-02-09 | 株式会社东芝 | Transmission signals, method and apparatus |
CN1809039A (en) * | 2005-01-19 | 2006-07-26 | 北京邮电大学 | Method of implementing estimated performance of TD-LS channels in MIMO-OFDM system |
Non-Patent Citations (2)
Title |
---|
MIMO2OFDM 无线通信系统鲁棒的二维导频辅助信道估计. 李汀,杨绿溪.扬州大学学报(自然科学版),第7卷第1期. 2004 |
MIMO2OFDM 无线通信系统鲁棒的二维导频辅助信道估计. 李汀,杨绿溪.扬州大学学报(自然科学版),第7卷第1期. 2004 * |
Also Published As
Publication number | Publication date |
---|---|
CN1925472A (en) | 2007-03-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7852906B2 (en) | Wireless communication device and channel estimation and separation method in multi input system | |
US7885360B2 (en) | Wireless communication apparatus and receiving method | |
US8654879B2 (en) | Multi-antenna channel estimation method based on polyphase decomposition | |
US8290462B2 (en) | Receiver and method for estimating a plurality of estimated transfer functions corresponding to wireless channels in a multiple-input system | |
CN102208972B (en) | Noise estimation for wireless communication | |
CN102739573B (en) | Channel estimation methods and channel estimator | |
CN102204197B (en) | OFDM channel estimation method and apparatus | |
CN101588335B (en) | Multiple input multiple output (MIMO) detecting method and MIMO detecting system utilizing correlation of signal channels | |
CN104168241B (en) | Multiple input multiple output orthogonal frequency division multiplexing communication system and method for compensating signal | |
EP1643660A1 (en) | Multiple input multiple output orthogonal frequency division multiplexing mobile communication system and channel estimation method | |
CN101286760B (en) | Channel estimating device and method in orthogonal frequency division multiplexing system | |
CN106534030B (en) | A kind of channel estimation methods based on joint training sequence and pilot tone in 802.11n multiple antennas ofdm system | |
CN110034804B (en) | Method and apparatus for estimating angle information for wireless communication system | |
CN103873422A (en) | Method for eliminating multipath interference in system symbol in underwater sound orthogonal frequency-division multiplexing system | |
CN101512996B (en) | Equalizing structure and equalizing method | |
US7557752B2 (en) | Apparatus and method for communication | |
CN100502371C (en) | A Channel Estimation Method in MIMO-OFDM System | |
CN102685060A (en) | Multi-user multiple input multiple output (MIMO) receiving method and device for orthogonal frequency division multiplexing system | |
CN101039305B (en) | Balancing technique-based receiver and receiving method | |
CN101507213B (en) | Equalizing structure and equalizing method | |
CN112995078A (en) | Frequency offset compensation algorithm of OFDMA uplink | |
CN102075482B (en) | Method and device for estimating channels in MIMO-OFDM (Multiple Input Multiple Output- Orthogonal Frequency Division Multiplexing) system | |
CN105187351B (en) | A kind of OFDM Timing Synchronization detection methods under multipath channel | |
CN102223336B (en) | Wireless communication method and equipment | |
KR20100049909A (en) | Method for inter-carrier interference cancellation in high mobility mimo-ofdm system |
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 | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090617 Termination date: 20120914 |