KR20100070478A - A method for channel and interference estimation in a wireless communication system and an apparatus thereof - Google Patents

Abstract

PURPOSE: A method for estimating a channel and interference of a wireless communication system and a device thereof are provided to perform channel estimation and interference estimation by using a reference symbol of a received signal and a temporarily detected data symbol. CONSTITUTION: A detector detects a temporary data symbol in a received signal which the channel compensation and interference cancellation are performed. A channel estimator(380) presumes an initial channel by using a reference symbol of a received signal including a reference symbol and a data symbol. The channel estimator presumes a final channel by using the temporary data symbol. An interference estimator presumes an initial interference by using the reference symbol of a received signal including the reference symbol and the data symbol. The interference estimator presumes a final interference by using the presumed final channel and the temporary data symbol.

Description

A method for channel and interference estimation in a wireless communication system and an apparatus

The present invention relates to a method and apparatus for estimating channel and interference in a wireless communication system, and more particularly, to a method and apparatus for estimating channel and interference using a data symbol detected in a received signal. .

In recent mobile communication systems, orthogonal frequency division multiple access (OFDM), or a similar method, is useful for high-speed data transmission in a wireless channel. Frequency division multiple access (hereinafter referred to as SC-FDMA) has been actively studied.

In addition, a multiple input multiple output using multiple transmit antennas and multiple receive antennas in a manner to maximize diversity gain and throughput using multiple antennas. In the following, "MIMO") is being actively studied.

Currently, the 3rd Gerneration Partnership Project (3GPP), an asynchronous cellular mobile communication standard organization, is studying the next generation mobile communication system, Long Term Evolution (LTE), based on OFDM and MIMO.

In order to recover a transmission signal in the mobile communication system as described above, a channel estimation technique should be used to compensate for a distorted channel during transmission, and interference should be removed by interference estimation.

Existing channel estimation algorithms focus on selecting effective CIR components from the estimated CIR to reduce the effects of noise from the estimated initial channel estimates using pilot signals. However, depending on the environment of the channel used, all of the selected CIR components may not be valid CIRs. In this case, there is a disadvantage that does not sufficiently eliminate the influence of noise.

In addition, since only the channel estimation value of the pilot subcarrier position is used, the noise attenuation effect is dependent on the degree of noise attenuation of the pilot signal. In addition, when the channel estimate value of the data subcarrier through interpolation is included, the performance degradation occurs at high SNR due to the error due to the inaccuracy of the interpolation method rather than the noise attenuation effect of the data subcarrier. Therefore, there is a need for a method for increasing the accuracy of channel response estimation by maximizing the noise attenuation effect of the existing DFT-based channel estimation algorithm.

In addition, the existing interference estimation technique was analyzed under the assumption that the receiver fully knows the information of the multipath channel for the transmitted signal and that the interference and noise components can be completely separated from the received signal. However, when considering the realistic channel estimation, the error due to the error of the channel estimation causes the error of the interference estimation in the reference symbol, and the performance is degraded. In addition, since demodulation is performed using information of the estimated channel estimate value during the demodulation process, stable system performance cannot be maintained because an error in channel estimation causes performance degradation. The conventional channel estimation and interference estimation method using the reference symbol has a disadvantage in that the noise attenuation of the channel and interference estimated at the reference symbol position is limited.

Accordingly, an object of the present invention is to provide a channel and interference estimation method and apparatus for a wireless communication system that does not depend on reference symbols in channel estimation and interference estimation.

In accordance with an aspect of the present invention, there is provided a method for estimating a channel of a wireless communication system, including: estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol; Compensating for the channel of the received signal using the estimated initial channel, demodulating and modulating a data symbol according to the compensated channel, and generating a temporary data symbol; and using the generated temporary data symbol Estimating the final channel.

The process of estimating the initial and final channels may include estimating a channel in a frequency domain according to a Least Square (LS) estimation technique, and estimating a channel in a time domain according to a Discrete Fourier Transform (DFT) based channel estimation technique. It includes.

According to an aspect of the present invention, there is provided a channel estimation apparatus for a wireless communication system, including: an equalizer configured to compensate a channel of a received signal according to an estimated channel; A demodulator and modulator for demodulating and modulating data symbols according to the compensated channel to generate temporary data symbols; And a channel estimator for estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol, and estimating a final channel using the temporary data symbol.

The channel estimator includes an LS channel estimator for estimating a channel in a frequency domain according to a least square (LS) estimation technique; And a DFT-based channel estimator for estimating a channel in the time domain according to a Discrete Fourier Transform (DFT) -based channel estimation technique.

In the wireless communication system and the interference estimation method according to an embodiment of the present invention for achieving the above object, the initial channel using the reference symbol of the received signal including the reference symbol and the data symbol in the beauty system Estimating, estimating initial interference using the estimated initial channel and the reference symbol, compensating a channel of the received signal according to the estimated initial channel, and receiving the received signal according to the estimated initial interference Detecting a temporary data symbol by removing interference of a signal, estimating a final channel using the detected temporary data symbol, and finally using the detected temporary data symbol and the estimated final channel Estimating the process.

The process of estimating the initial and final channels may include estimating a channel in a frequency domain according to a Least Square (LS) estimation technique, and estimating a channel in a time domain according to a Discrete Fourier Transform (DFT) based channel estimation technique. It includes.

Estimating the initial and final interference includes estimating the instantaneous interference (SCM) and estimating the interference according to a low-pass smoothing (LPS) technique.

In order to achieve the above object, a channel and interference estimation apparatus of a wireless communication system according to an embodiment of the present invention compensates for a channel of a received signal according to an initial estimated channel and performs reception of a received signal according to an initial estimated interference. An equalizer for removing interference; A detector for detecting temporary data symbols in the received signal from which the channel compensation and the interference cancellation are performed; A channel estimator estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol, and estimating a final channel using the temporary data symbol; And an interference estimator for estimating initial interference using a reference symbol of a received signal including a reference symbol and a data symbol, and further estimating the final interference using the estimated final channel and the temporary data symbol.

The channel estimator includes an LS channel estimator for estimating a channel in a frequency domain according to a least square (LS) estimation technique; And a DFT-based channel estimator for estimating a channel in the time domain according to a Discrete Fourier Transform (DFT) -based channel estimation technique.

The interference estimator includes a temporal interference estimator for estimating temporary interference; And an LPS interference estimator for estimating interference according to the LPS technique.

In the channel and interference estimation method according to the embodiment of the present invention as described above, by performing the channel estimation and interference estimation using not only the reference symbols of the received signal but also temporarily detected data symbols, the noise reduction effect is more effective than the conventional techniques. Excellent, more accurate channel estimation and interference estimation are possible.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, only parts necessary for understanding the operation according to the embodiment of the present invention will be described, and the description of other parts will be omitted so as not to disturb the gist of the present invention.

The terms or words used in the specification and claims described below should not be construed as being limited to the ordinary or dictionary meanings, and the inventors are appropriate to the concept of terms in order to explain their invention in the best way. It should be interpreted as meanings and concepts in accordance with the technical spirit of the present invention based on the principle that it can be defined. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only one of the most preferred embodiments of the present invention, and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

The present invention estimates a channel or estimates interference using data symbols included in a received signal transmitted by a transmitter in an OFDM system. First, channel estimation will be described in the first embodiment in consideration of an OFDM system, and channel estimation and interference estimation will be described in the second embodiment in consideration of a MIMO system.

First embodiment

1 is a diagram illustrating a baseband transmitter structure of an OFDM communication system according to an exemplary embodiment of the present invention.

Referring to FIG. 1, a baseband transmitter includes a binary data generator 110, a data symbol mapper 120, a pilot symbol generator 130, and a symbol inserter. 140), an Inverse Fast Fourier Transform (IFFT) machine 150, and a Guard Interval (GI) inserter 160.

Binary data to be transmitted is generated from the binary data generator 110, and is modulated (mapping / modulated) by the data symbol mapper 120, and the pilot symbol generator 130 generates pilot symbols that are identically known between the transmitter and the receiver. do. The mapped data symbols and pilot symbols are assigned to the subcarriers in the symbol inserter 130. This signal is converted into a signal in the time domain via IFFT 150, which produces a baseband signal that is finally transmitted via GI inserter 160.

2 is a diagram illustrating a frame structure of an OFDM communication system according to an exemplary embodiment of the present invention.

개의 부반송파를 갖는다. 주파수 동기 및 타이밍 동기 그리고 채널 추정을 위해 이용되는 프리앰블(210)은 송신기 및 수신기가 미리 알고 있는 신호를 할당한다. 도면부호 230을 참조하면, 데이터 전송을 위한 데이터 심벌(220)은 먼저 파일럿 심벌을 동일 간격

의 파일럿 용 부반송파(240)에 할당한 후 나머지 모든 가용 부반송파들을 데이터 용 부반송파(250)로 이용한다. 도면부호 260을 참조하면, 데이터 심벌(220)에 할당되는 파일럿 심벌은 분산 파일럿이 적용된다. 2, the OFDM frame structure is composed of a preamble symbol 210 and a data symbol 220, each OFDM symbol is

Subcarriers The preamble 210 used for frequency synchronization and timing synchronization and channel estimation allocates signals known to the transmitter and the receiver in advance. Referring to 230, the data symbols 220 for data transmission firstly space pilot symbols at equal intervals.

After allocating to the pilot subcarrier 240, all remaining available subcarriers are used as the data subcarrier 250. Referring to 260, a pilot is allocated to a data symbol 220 and distributed pilot is applied.

3 is a diagram illustrating a channel estimation apparatus of a receiver according to an exemplary embodiment of the present invention.

Referring to FIG. 3, a receiver according to an embodiment of the present invention may include a guard interval removal (310) device, a fast fourier transform (FFT) device 320, a pilot symbol extraction device (330), A data extractor 340, an equalizer 350, a data symbol demapper 360, a data symbol mapper 370, and a channel estimator 380 are included.

The channel estimator 380 includes an LS channel estimation (LS) estimator 410, an interpolation 420, and a DFT-based CFR estimation 430.

The GI remover 310 removes and outputs a guard interval of the received OFDM symbol. The fast fourier transform (FFT) performs a fast Fourier transform on the input time domain signal and outputs the signal in the frequency domain.

The pilot extractor 330 extracts a pilot symbol from the signal converted into the frequency domain and inputs it to the channel estimator 380.

The data extractor 340 extracts the data symbols from the signal converted into the frequency domain and inputs them to the equalizer 350.

The equalizer 350 equalizes input data symbols according to channel estimation of the channel estimator to compensate for the influence of the multipath channel.

The demodulator 360 outputs binary data by demodulating and demodulating the input data symbols according to a modulation scheme. The modulator 370 generates data symbols by mapping and modulating binary data output from the demodulator according to a modulation scheme, and inputs the data symbols to the channel estimator 380. Such demodulation and modulation scheme may be QPSK, QAM, 16-QAM, or the like.

The channel estimator 380 estimates a channel according to an embodiment of the present invention, but the channel estimator of the channel estimator has two steps. Step 1 uses only CFR information of the pilot subcarrier position as in the conventional method, and step 2 additionally includes a decision-directed method using a temporary demodulated received signal.

That is, in step 2, data symbols are again used for channel estimation through the symbol demapper and the symbol mapper.

The channel estimator includes an LS channel estimator 410 for estimating a channel according to an LS channel estimation technique in a frequency domain, an interpolator 420 for interpolating sample signal values for channel estimation, and a DFT-based estimating channel in a time domain. Channel estimator 430 is included. In this case, the interpolator 420 may be selectively configured.

Next, a channel estimation method according to an embodiment of the present invention will be described in detail.

The baseband signal as described in FIG. 1 is received through an antenna through a multipath fading channel. As such, when the receiver receives the baseband signal, the receiver removes the GI of the baseband signal and obtains a frequency domain signal through the FFT. The signal in this frequency domain is represented by Equation 1 below.

인 AWGN(Additive White Gaussian Noise)을 의미한다. In Equation 1, m and k represent OFDM symbol numbers and subcarrier numbers, respectively. Where Y (m, k) represents the signal received through the multipath fading channel. Where X (m, k) is the OFDM symbol of the m-th frequency domain generated by the transmitter, H (m, k) is the frequency response characteristic of the multipath fading channel, and W (m, k) has an average of zero. And dispersion

AWGN (Additive White Gaussian Noise).

Here, the frequency response H (m, k) of the channel and the response h (m, k) of the time domain are expressed by Equation 2 below.

는 채널의 지연 경로 수,

는

번째 채널 경로의 지연 시간,

는 복소 채널 이득이다. In Equation 2,

Is the number of delay paths for the channel,

Is

Delay time for the first channel path,

Is the complex channel gain.

The received signal as described above is affected by the frequency selective fading effect due to the multipath delay of the channel and the Doppler frequency due to the increase of the moving body speed, and the channel is estimated to compensate for the influence of the channel.

According to an embodiment of the present invention, channel estimation is performed twice. The first step estimates a channel using a reference symbol (pilot symbol), and the second step temporarily modulates the channel by using a reference symbol. The reconstructed signal is used again for channel estimation.

First, a channel is estimated using a reference symbol (pilot symbol). Channel estimation using a reference symbol (pilot symbol) is performed as follows. The LS channel estimator estimates an initial channel (CFR) by applying an LS estimation technique at a subcarrier location using pilot symbols separated from the pilot extractor 330. The LS estimation technique estimates the received symbol by one-to-one dividing the received symbol into a known symbol, that is, a pilot symbol, without considering the statistical characteristics of the noise. This LS estimation technique is represented by Equation 3 below.

는 주파수 영역에서 분산 파일럿 심벌이 할당된 부반송파 위치를 나타내는 집합이다. In <Equation 3>

Is a set of subcarrier positions to which distributed pilot symbols are allocated in the frequency domain.

As described above, after estimating the channel through the LS estimation method, the channel is performed according to the channel estimation based on the DFT. As an input of the DFT-based channel estimation block, the initial CFR may optionally include not only the channel value of the subcarrier at the pilot symbol position but also the channel estimation value of the data subcarrier through the interpolator 420.

4 is a diagram illustrating a DFT-based channel estimation technique according to an embodiment of the present invention.

4 shows a functional block for channel estimation based on DFT. As shown, the DFT-based channel estimator includes an IDFT unit 411, an effective CIR estimator 413, and a DFT unit 411.

DFT-based channel estimation according to Equation (3) estimates the CIR through an operation of transforming the signal into a time domain signal through the IDFT 411, and the CIR estimator 420 selects only a valid component in the estimated CIR. For those components other than the effective CIR value, it is considered noise and is forcibly replaced by zero. This is to remove the influence of noise. Viewing this in the frequency domain shows an effect of smoothing the estimated channel value. After selecting the useful CIR, the DFT operation 413 calculates the final CFR to compensate for the influence of the channel through the DFT operation.

을 IDFT 시켜서 구한 CIR은 <수학식 4>와 같다. When the m-th OFDM symbol is a primamble signal, which is a signal previously known at the transmitting and receiving end, the LS estimation technique is estimated.

The CIR obtained by IDFT is shown in Equation 4.

은

을 나타내고, 채널의 최대 지연 시간을 L이라 할 때, <수학식 2>와 같이 h(m,n)은 <수학식 5>와 같이 나타낼 수 있다. here

silver

When the maximum delay time of the channel is L, h (m, n) can be expressed as Equation 5 as shown in Equation 2.

과 순수 잡음 구간인

으로 나눌 수 있고 <수학식 6>과 같이 표현된다. Therefore, the CIR estimated by Equations 4 and 5 is an effective channel estimation interval.

And the pure noise interval

It can be divided into Equation (6).

Assuming the ideal situation of knowing the statistical delay time of the channel, the selection of the effective CIR can be represented by Equations 7 and 8 as follows.

의 설정에 따라서 채널 추정 성능이 상이하게 나타나는데 유효 채널 구간에서 MSE(Mean-Square Error)를 최소화하기 위한 것으로 잡음 전력의 두 배인

을 설정할 수 있다.In addition, the effective CIR selection method according to the threshold setting considering the path-to noise ratio (PNR) by calculating the variance of the noise of a certain period may be represented by Equation 9 as follows. Threshold

The channel estimation performance is different depending on the setting of. It is to minimize Mean-Square Error (MSE) in the effective channel section.

Can be set.

5 is a diagram for describing a method of selecting an effective CIR from an estimated CIR. This is an example of <Equation 7> to <Equation 9>.

As described above, the attenuation effect on the noise can be obtained by taking only the effective CIR component, that is, the multipath component of the channel, and forcing the remaining values to be zero for the remaining values.

However, this method assumes an ideal situation in which the position of the delay path of the channel is accurately known or the maximum delay position of the channel. Therefore, a separate estimator is needed to estimate the delay characteristics of the channel for practical application.

가 채널의 최대 지연 시간 L보다 길게 설정되는 OFDM 시스템의 특성을 고려하여 <수학식 10>과 같이 GI길이만큼의 CIR을 유효 성분으로 선택하게 되며, DFT 연산후의 최종적인 CFR은 <수학식 11>과 같이 나타낸다.So in general case, the length of the GI

In consideration of the characteristics of the OFDM system in which L is set to be longer than the maximum delay time L of the channel, the CIR as long as the GI length is selected as an active component as shown in Equation 10, and the final CFR after the DFT operation It is represented as

The equalizer 350 compensates for the influence of the multipath channel on the received data signal in the frequency domain separated from the data extractor 340. At this time, the equalizer 350 equalizes using the estimated CFR through the channel estimation process described above to compensate for the influence of the multipath channel. The signal whose channel influence is compensated for is historically demodulated and demodulated in demodulator 360 and restored to binary data.

)를 재생성하고, 데이터 부반송파의 주파수 채널 응답을 추정하기 위한 채널 추정기의 입력으로 사용된다. As described above, the recovered binary data is mapped and modulated again in the modulator 370 to generate a temporary received signal (

) Is used as an input to the channel estimator for regenerating and estimating the frequency channel response of the data subcarrier.

) 및 파일럿 심벌을 이용하여 다시 채널을 추정한다. 즉, 채널 추정기의 입력(

)은 다음의 <수학식 12>와 같이 파일럿 위치의 부반송파뿐만 아니라 데이터 위치의 부반송파 위치에서의 CFR을 포함한다. The channel estimator 380 then receives a temporary received signal (

) And the pilot symbol is used to estimate the channel again. That is, the input of the channel estimator (

) Includes the CFR at the subcarrier location of the data location as well as the subcarrier at the pilot location as shown in Equation 12 below.

Unlike Equation 3, Equation 12 includes a CFR for a data subcarrier differently from Equation 3, and uses reliable temporary received data, which is more effective than estimating a CFR for a data subcarrier by interpolation. Estimation errors can be significantly reduced, and when the DFT-based channel estimation algorithm is used again, the noise attenuation effect can be obtained not only at the pilot subcarrier position but also at the data subcarrier position. As described above, through input, the channel estimator estimates the channel again, equalizes the data symbols through the estimated channel value, and then demodulates to detect the binary data.

6 is a diagram illustrating a channel estimation method in an OFDM system according to an embodiment of the present invention.

를 추정한다. 이때, 채널 추정은 시간 영역에서의 추정 및 주파수 영역에서의 추정이 이루어진다. 주파수 영역에서의 채널 추정은 LS 추정(Estimation) 기법에 따라 이루어지며, 시간 영역에서의 추정은 DFT 기반의 채널 추정 기법에 따라 이루어진다. 이때, 선택적으로 주파수 영역에서의 채널 추정 후, 보간을 수행할 수 있다. Referring to FIG. 6, when the receiver receives a signal in step S601, the receiver estimates a channel of the received signal through a pilot symbol in step S603. That is, as shown in Equation 3, only the channel estimation value of the frequency-domain pilot subcarrier position is used.

Estimate In this case, channel estimation is performed in the time domain and in the frequency domain. Channel estimation in the frequency domain is performed according to LS estimation, and estimation in the time domain is performed according to DFT-based channel estimation. In this case, after channel estimation in a frequency domain, interpolation may be selectively performed.

Then, the receiver compensates for the data symbols distorted by the channel by the previously estimated channel value in step S605.

를 재생성한다. Subsequently, the receiver demodulates the data symbols in step S607, and modulates the data symbols demodulated in step S609 again to generate a temporary received signal (

Regenerate

및 파일럿 심벌을 이용 하여 최종적으로 채널을 추정한다. 이때, 채널 추정의 입력(

)은 <수학식 12>와 같이 파일럿 위치의 부반송파뿐만 아니라 데이터 위치의 부반송파 위치에서의 채널 추정 값(CFR)을 포함한다. The receiver receives a temporary reception signal (

Finally, the channel is estimated using the pilot symbols. At this time, input of channel estimation (

) Includes the channel estimation value (CFR) at the subcarrier location of the data location as well as the subcarrier at the pilot location as shown in Equation 12.

The receiver then compensates for the channel by the final channel estimate in step S613 and demodulates the data symbols according to the channel compensated in step S615.

As shown in Equation 3, Equation 12 includes a channel estimation value (CFR) for a data subcarrier differently from Equation 3, and by using reliable temporary received data, this is a channel for a data subcarrier by interpolation. It is possible to significantly reduce the channel estimation error rather than estimating, and when using the DFT-based channel estimation algorithm again, the noise attenuation effect can be obtained not only at the pilot subcarrier position but also at the data subcarrier position.

Next, the performance of the channel estimation method according to an embodiment of the present invention will be described. 7 to 9 are diagrams for explaining the performance of the channel estimation method according to an embodiment of the present invention.

The simulation for verifying the performance of the channel estimation method presented in the present invention was performed by recording statistical performance values through a sufficient number of iterations in a randomly varying multipath fading channel environment, and the results are shown in FIGS. 7 to 9. It was.

=8로 할당된 데이터 심벌만을 사용하여 모의실험을 하였다. 실험에 적용된 채널 환경과 각각의 파라미터는 도에 포함하여 나타냈으며, 성능 평가를 위해 비교 대상으로 설정한 채널 추정 기법은 다음과 같다.The preamble used for frequency synchronization and timing synchronization is not used and the interval between pilots is

We simulated using only data symbols assigned as = 8. The channel environment and each parameter applied to the experiment are shown in the figure, and the channel estimation method set as the comparison target for the performance evaluation is as follows.

① 2D (Dimensional) Linear Interpolation (1st linear interpolation among the conventional Lagrange interpolation techniques, interpolation to time axis precedes interpolation to frequency axis)

② Conventional Method (1) (Conventional DFT-based Channel Estimation Method for Selecting CIR as Long as Guard Period GI as Active Component)

③ Conventional Method (2) (a conventional DFT-based channel estimation technique that applies a 2D interpolation technique and selects a CIR equal to the guard interval GI length as an active component)

Conventional Method (3) A conventional DFT-based channel estimation technique that assumes an ideal case of knowing the delay component of each channel and selects only that component as the effective CIR.

⑤ Proposed Method (1) (DFT-based decision directed channel estimation technique of the present invention)

⑥ Proposed Method (2) (In the DFT-based decision directed channel estimation scheme, only the interpolation method is applied to the time axis when estimating the channel in the 1st step DFT)

7 is a graph showing MSE performance for each data subcarrier of the scheme proposed by the present invention in the frequency domain.

As shown in FIG. 7, "2D linear interpolation" shows relatively poor performance.

The conventional DFT-based channel estimation algorithm, which selects the CIR by the guard interval GI length as an effective component, shows better performance than the "2D linear interpolation".

In addition, the input of the conventional DFT-based channel estimation block of the "conventional method (2)" that selectively includes the channel value of the sub-carrier of the pilot symbol position, as well as the channel estimation value of the data sub-carrier through "2D linear interpolation". In this case, we can see the tendency of performance deterioration due to the error of interpolation.

It can be seen that the proposed method improves the accuracy of the CFR estimation for the entire data subcarriers by effectively improving the limitation of the channel estimation performance of the conventional DFT-based channel estimation algorithm using only the CFR of the pilot subcarriers.

In spite of applying realistic effective CIR selection technique, it is assumed that the ideal case of knowing the delay component of each channel, and the conventional DFT-based channel estimation technique that selects only the component as the effective CIR "conventional method (3)" It can be seen that the transient performance is the most similar.

8 is a graph comparing the MSE performance of the scheme proposed by the present invention according to the change of Eb / No.

The conventional DFT-based channel estimation algorithm shows better performance than "2D linear interpolation".

On the other hand, since the degree of noise attenuation is dependent on the degree of noise attenuation of the pilot subcarriers, it can be seen that the difference between the performance and the "conventional method (3)" assuming an ideal case is large.

In addition, the "conventional method (2)" shows better performance at the low SNR due to the effect of interpolation noise reduction, but as the SNR increases, the performance is due to the error due to interpolation rather than the noise attenuation effect. This tends to deteriorate.

의 MSE 성능으로 이상적인 경우를 가정하는 "conventional method(3)"과 근접한 성능을 나타낸다.As the Eb / No increases, the channel estimation method according to an embodiment of the present invention is much better than the conventional DFT-based channel estimation algorithm, and is weak at Eb / No = 30dB.

The performance is close to the "conventional method (3)", which assumes the ideal case of MSE performance.

9 is a graph comparing Uncoded BER performance of the method proposed by the present invention according to change of Eb / No.

에서

사이의 성능의 향상을 목표한다. 각각의 Uncoded BER 성능은 Eb/No의 변화에 따른 MSE 성능과 유사한 경향을 나타내며, 특히 "2D linear interpolation"과 "conventional method (2)"의 경우 60km/h의 이동체 속도와 Eb/No=30dB 이하의 실험 환경에서 Uncoded BER 성능이

이하로 내려가지 않고 에러 풀로어(error floor) 현상이 나타남을 확인할 수 있다. The IDEAL BER performance shown in FIG. 9 shows the Uncoded BER performance when the effect of the channel is ideally compensated. This performance represents the best performance in the multipath fading channel used. In addition, with Uncoded BER performance

in

Aims to improve performance between. Each Uncoded BER performance tends to be similar to the MSE performance according to the change of Eb / No. In particular, in the case of "2D linear interpolation" and "conventional method (2)", the moving speed of 60km / h and the Eb / No = 30dB or less Performance of Uncoded BER in

It can be seen that an error floor phenomenon appears without going below.

When the channel estimation algorithm according to the embodiment of the present invention is applied, it can be seen that the performance is superior to other channel estimation techniques, and the performance is very close to the IDEAL BER performance.

7 to 9, the channel estimation algorithm according to an embodiment of the present invention is more effective than the conventional technique even in an environment where severe Doppler shift is caused by the frequency selective fading effect of the multipath fading channel and the increase of the moving body speed. It can be seen that excellent performance can be exhibited. In particular, it is possible to effectively increase the accuracy of the channel response estimation by maximizing the noise attenuation effect of the existing DFT-based channel estimation algorithm without changing the method of determining the effective impulse response length.

Second embodiment

10 is a diagram for describing a structure of a baseband transmitter of a MIMO-OFDMA communication system according to an exemplary embodiment of the present invention.

The present invention considers a spatial multiplexing (SM) scheme that can obtain a capacity gain as a method of simultaneously transmitting multiple data streams through multiple antennas in a MIMO system.

Referring to FIG. 10, a transmitter according to another embodiment of the present invention includes a binary data generator 1110, a data symbol mapper 1120, and M transmitters 1130 corresponding to the number of antennas. do. Here, each transmitter 1130 may include a reference symbol generator 1131, a data symbol and reference symbol insertion 1132, an inverse fast fourier transform (IFFT) 1333, and a guard interval (GI) inserter. 1134.

Binary data generated by the binary data generator 1110 is mapped and modulated by the data symbol mapper 1120, and M transmitters 130 use different reference symbols for each transmit antenna. This reference symbol generator 1131 is generated.

Each data symbol and reference symbol are assigned to a subcarrier promised for each antenna in a symbol inserter 1132. This signal passes through the IFFT 1133 and is converted into a signal in the time domain, which generates a baseband signal through which the GI is inserted and finally transmitted through the GI inserter 1134.

11A and 11B illustrate an example of a frame structure of an OFDMA communication system according to another embodiment of the present invention.

11 shows an example of a downlink frame in which no preamble or a training sequence or a traning symbol is assigned. Here, reference numeral 210 denotes a case of a single-input multiple-output (SIMO) system of 1Tx-2Rx, and reference numeral 220 illustrates a case of a downlink frame according to the MIMO system of 2Tx-2Rx. In the downlink frame structure, the reference symbols 212 and 222 in the OFDM symbol having K subcarriers are staggered in the frequency domain in accordance with each transmit antenna in units of six subcarriers. It is allocated to different subcarriers and is located in the first OFDM symbol of each slot and the third OFDM symbol at the end of each slot.

In particular, as shown by reference numeral 220, when considering a plurality of transmit antennas, a null symbol 221 is allocated so that they are located in the same OFDM symbol on the time axis, but not alternately on the frequency axis. After the reference symbol and the null symbol are allocated, all remaining available subcarriers are used as the data subcarriers 211.

Next, a structure of a baseband receiver according to another embodiment of the present invention will be described. 12 to 14 are diagrams for describing a channel and interference estimation apparatus of a receiver according to an embodiment of the present invention.

Referring to FIG. 12, a receiver according to an exemplary embodiment of the present invention may include a guard interval removal (GI) device (not shown), a fast fourier transform (FFT) device (not shown), and an RS symbol extraction device (810). ), Data symbol extractor 820, Asynchronous ICI Mitigation 830, detector 840, LLR calculation (850), decoder (Turbo Decoder) 860, CFR Estimator 870 and an SCM estimator 880.

Here, the channel estimator 870 includes an LS channel estimator 910 and a DFT-based CFR estimator 885.

The GI eliminator (not shown) removes and outputs a guard interval of the received OFDM symbol. A Fast Fourier Transform (FFT) machine performs fast Fourier transform on the input time domain signal and outputs the signal in the frequency domain.

The RS extraction unit 810 extracts a reference symbol from the signal converted into the frequency domain and inputs the reference symbol to the channel estimator 870 and the interference estimator 880, respectively. The data symbol extractor 820 extracts a data symbol from the signal converted into the frequency domain and inputs the data symbol to the interference canceller 830.

The interference canceller 830 compensates the channel of the data symbol according to the channel estimation of the channel estimator 870, and removes the asynchronous ICI mitigation of the data symbol according to the estimated interference SCM of the interference estimator 860. Output

Detector 840 detects data symbols from which channel compensation and interference have been removed. In this case, the detection of data symbols is performed according to a maximum likelihood (ML) decision (ML) detection technique. The ML detection technique selects an input having a minimum Euclidean distance by substituting all transmittable symbols from all transmit antennas. Meanwhile, although an embodiment of the present invention is described as detecting data symbols according to a technique, the present invention is not limited thereto, and data symbols may be detected according to another detection technique such as MMSE.

The CFR Estimator 870 first estimates a channel using a reference symbol. In addition, the channel is estimated again using the data symbols and the reference symbols detected by the detector. The channel estimator 870 includes an LS channel estimator 911 for estimating a channel according to an LS channel estimation technique in a frequency domain, an interpolator 913 for interpolating values of a sample signal for channel estimation, and a channel in a time domain. A DFT-based channel estimator 920 is estimated. In this case, the interpolator 913 may be selectively configured.

The SCM estimator 880 first estimates interference using a reference symbol. Then, the interference is estimated again using the data symbols and the reference symbols detected by the detector.

The interference estimator 880 includes a temporal interference estimator 930 for estimating instantaneous interference and an LPS interference estimator 940 for estimating interference according to the LPS technique.

An LLR calculation 850 calculates the LLR of the data symbols.

The decoder 860 decodes the data symbols according to the calculated LLR and outputs binary data.

Next, a channel estimator and an interference estimator in a baseband receiver structure according to another embodiment of the present invention will be described in detail.

Considering M transmit antennas and N receive antennas, M _ICI neighbor cell interferences are received overlapping in an asynchronous ICI environment, and the received signal is represented by Equation 13 for the k th subcarrier of the j th OFDM symbol in the frequency domain. It is expressed as

인 잡음 벡터를 각각 나타낸다. 여기서, I_k,j는 asynchronous ICI와 잡음의 합이다. In Equation 13, Y _{k, j} is a reception signal vector, X _{k, j} is a transmission signal vector, H _{k, j} is a channel coefficient matrix for the transmission signal, and U _{k, j} is an asynchronous ICI signal vector, G _{k , j} is the channel coefficient matrix for the asynchronous ICI signal vector, and W _{k, j} is the mean "0" and the variance

Each represents a noise noise vector. Where I _{k, j} is the sum of asynchronous ICI and noise.

Equation 13 may be expressed in a matrix form as shown in Equation 14 in consideration of a transmission / reception antenna.

From the received signal vector (Y _{k, j)} as described above, the channel estimator () is H _{k, j} are, and estimates the channel to compensate for the channel distort for the channel coefficient matrix for the transmission signal, the interference estimator described above In order to estimate the remaining interference components except for the transmission signal vector (X _{k, j} ) and the channel coefficient matrix (H _{k, j} ) for the transmission signal in one received signal vector.

13 is a view for explaining the structure of a channel estimator according to an embodiment of the present invention.

Considering that a null symbol is allocated in a slot to avoid overlapping of reference symbols for each transmit antenna, the same channel estimation algorithm as that of a single input single output (SISO) system is applied in a MIMO system. Therefore, the channel estimation method described in FIG. 13 is based on the SISO system.

If the adjacent cell (Asynchronous IC) is not taken into account, i.e., the channel coefficient matrix G _k, for the asynchronous ICI signal vector U _{k, j} and the neighbor cell interference signal vector in Equation 13 _{, When} not considering _j ), in the SISO system, the received signal for the k-th subcarrier of the j-th OFDM symbol in the frequency domain may be expressed as Equation 15 below.

The received data signal as shown in Equation 15 is influenced by the frequency selective fading effect due to the multipath delay of the channel and the Doppler frequency due to the increase of the moving speed, and the channel is estimated to compensate for the influence of the channel.

The channel estimator 870 according to an embodiment of the present invention estimates a channel in two steps. In a first step, the channel estimator 870 estimates the channel by the reference signal. Accordingly, the receiver performs channel compensation according to the estimated channel, removes the interference component according to the interference estimated by the interference estimator 880, and then detects the data symbol.

Then, in the second step, the channel estimator 870 performs channel compensation according to the estimated channel, removes the interference component according to the interference estimated by the interference estimator 880, and then again uses the detected data symbol. Estimate

The channel estimator 870 includes an LS channel estimator 911 for estimating a channel according to the LS scheme, an interpolator 913 for interpolating values of a sample signal for channel estimation, and a DFT-based channel estimating a channel in a time domain. An estimator 920 is included. As described above, the interpolator 913 may be selectively configured.

The LS channel estimator 911 estimates a channel according to the LS technique of performing channel estimation in the frequency domain. In this case, the channel estimation in the first step described above uses only reference symbols, and the channel estimation in the second step further uses data symbols. Then, interpolation is performed through the interpolator 913 for the data symbol interval between the reference symbols.

)을 이용하여 다음의 <수학식 16>과 같이 IDFT기(921)를 통해 시간 영역의 신호로 변환하여 CIR을 추정한다. Next, the channel estimator 870 estimates the channel in the time domain through the DFT-based channel estimator 920. DFT-based channel estimator 920 includes an IDFT 921, a CIR estimator 923, and a DFT 925. The operation of the DFT-based channel estimator 920 may first include an LS channel estimate value for a reference symbol.

Equation (16) is used to estimate the CIR by converting the signal into a time domain signal through the IDFT unit 921 as shown in Equation 16 below.

The CIR estimator 923 selects only the components that are valid in the estimated CIR and considers them as noise for components other than the effective CIR value and forces them to zero. This is to remove the influence of noise. Viewing this in the frequency domain shows an effect of smoothing the estimated channel value.

At this time, to select the active ingredient, it is preferable to select the CIR as long as the guard interval (GI) length as the active ingredient, as shown in Equation 17 below.

After selecting the useful CIR, the DFT operation 925 calculates the final channel estimation value (CFR) to compensate for the influence of the channel through the DFT operation. As such, the estimated channel estimate values are input to the interference estimator 880 and the equalizer 830.

14 is a diagram for describing an interference estimator according to an exemplary embodiment of the present invention.

An interference estimator 880 according to an embodiment of the present invention estimates interference in two steps. In a first step, a channel coefficient for a transmission signal vector (X _{k, j} ) and a transmission signal in the above-described received signal vector (Y _{k, j} ) using the channel estimate estimated through the reference symbol of the channel estimator and the reference symbol. Exclude the matrix H _{k, j} and estimate the remaining interference components. This estimated interference value is input to the equalizer 830 to remove the interference component. In a second step, when a data symbol is detected through the detector 840 in a received signal in which channel compensation and interference cancellation are performed, the channel estimator 870 further includes a data symbol in a reference symbol to estimate a channel. Then, the interference estimator 880 further includes data symbols to estimate interference using the estimated channel estimate value and the detected data symbols. This estimated interference value is input back to the equalizer 830, through which the equalizer 830 finally removes the interference component.

The interference estimator 880 for this includes a temporal interference estimator 930 and an LPS interference estimator 940. In addition, the LPS interference estimator 940 includes a decomposer 941, an LPS processor 941, a decompressor 945, and a linear interpolator 947.

Assuming that ideal channel estimation is possible for a received signal such as Equation (13), the spatial covariance matrix estimation to remove and mitigate asynchronous ICI is performed with the asynchronous ICI. With respect to I _{k, j} , which is the sum of noise, the spatial covariance matrix (SCM) for the k-th subcarrier of the j-th OFDM symbol is obtained by averaging P received OFDM symbols as shown in Equation 18 below.

In the case of the multipath channel considering mobility, the characteristics of the SCM varying with respect to each OFDM symbol in the process of applying the average as shown in Equation 18, thereby deteriorating the cancellation and mitigation performance of adjacent interference. .

In particular, considering the structure of the assignment of reference symbols to which preambles, training sequences, etc. are not assigned, it is impossible to estimate the SCM as shown in Equation 18 within all valid subcarrier ranges. Therefore, it is necessary to estimate the SCMs for all valid subcarriers using the instantaneous SCM calculated as in Equation 19 or Equation 20.

The temporary time estimator 930 estimates the interference as shown in Equation 19 or Equation 20.

Equation 19 is applied to the interference estimation using the reference symbol and the initial channel estimation value of the first step, and Equation 20 uses the channel estimation value, the reference symbol and the detected data symbol of the second step. Used for interference estimation.

The LPS processor 943 removes the influence of the transform of the time and frequency domains and the noise of the uncorrelated components on the temporarily estimated interference value such as Equation 19 or Equation 20 through the LPS technique. As a result, interpolation and noise attenuation effects may be simultaneously obtained when the interference is estimated for the effective subcarriers. The interference estimation method according to the LPS method is as shown in Equation 21 below.

, m=n=1,2,···N이며, F는 K×K FFT 행렬, L은 추정된 임시 간섭(instantaneous SCM)의 시간 영역 성분에 대한 유효 성분을 검출하기 위해 설정된 것으로 보호 구간(GI, Guard Interval)의 길이와 동일하다. 이는 채널의 다중 경로 지연의 영향으로 인한 OFDM 심벌간 간섭을 방지하기 위해 삽입되는 보호 구간이 채널의 최대 지연 길이보다 길게 설정되어 있는 OFDM 시스템을 고려한 것이다. here,

, m = n = 1,2, ... N, where F is a K × K FFT matrix, and L is set to detect an effective component for the time domain component of the estimated instantaneous SCM. GI, Guard Interval). This takes into account an OFDM system in which a guard interval to be inserted is set longer than a maximum delay length of a channel to prevent interference between OFDM symbols due to the multipath delay of the channel.

)의 각각의 원소에 대하여 독립적으로 LPS 방식을 적용하게 되면, SCM의 허미션(Hermitian)과 positive definite 특성이 파괴되기 때문에 간섭 제거(interference mitigation) 성능의 열화를 가져온다. 이러한 성능의 열화를 방지하기 위하여 LPS 기법을 적용하기 이전에 instantaneous SCM에 대한 분해(decomposition) 과정이 필요하며, 따라서 분해기(941)는 "Cholesky Decomposition"기법에 따라 분해한다. 이는 2Tx-2Rx MIMO 시스템의

에 대하여 다음의 <수학식 22>와 같이 분해한다. However, the estimated temporary interference values (i.e., instantaneous SCM,

Independently applying the LPS method to each element of) results in the degradation of interference mitigation performance since the Hermitian and positive definite properties of SCM are destroyed. In order to prevent such deterioration of performance, a decomposition process for instantaneous SCM is required before the LPS technique is applied. Therefore, the decomposer 941 decomposes according to the "Cholesky Decomposition" technique. This is because the 2Tx-2Rx MIMO system

Decompose as shown in Equation 22 below.

)에 대하여 각 원소 별 LPS 기법을 적용한다. The LPS processor 943 then decomposes the decomposed U matrix (square-root of matix).

LPS method for each element is applied.

)를 복원하여 간섭 추정 값(SCM)을 재생성한다. The decompressor 945 is a signal to which the LPS technique is applied.

) Is regenerated to estimate the interference estimate (SCM).

The regenerated interference estimate is input to a linear interpolator 947, which interpolates the symbols in the time domain of the interference estimate.

Next, a channel and interference estimation method according to an embodiment of the present invention will be described.

15 illustrates a channel and interference estimation method according to another embodiment of the present invention.

)을 도출한다. Referring to FIG. 15, the receiver initially estimates a channel using a reference symbol in step S1501. In this case, the initial channel estimation is performed by estimating a channel by applying LS estimation and DFT-based channel estimation in order, thereby initial channel estimation value (CFR,

).

Next, the receiver initially estimates the interference using the reference symbol in step S1503.

)을 이용하여, 임시 간섭 추정(instantaneous SCM)을 수행한다. 이어서, 수신기는 다음의 <수학식 11>과 같이 LPC 기법에 따라 간섭을 추정하여 초기 추정 간섭 값(SCM,

)을 도출한다. In this case, the initial interference estimation is a channel estimation value (CFR, according to the initial channel estimation described above).

), To perform an instantaneous interference estimation (Instantaneous SCM). Subsequently, the receiver estimates the interference according to the LPC method as shown in Equation 11 below to determine the initial estimated interference value (SCM,

).

) 및 초기 추정 간섭 값(SCM,

)을 이용하여 수신 신호로부터 임시의 데이터 심벌을 검출한다. 이러한 검출이 ML 기법에 따라 이루어지는 경우, 검출된 데이터 심벌은 다음의 <수학식 23>과 같다. Subsequently, in step S1505, the receiver estimates an initial channel estimate value (CFR,

) And initial estimated interference values (SCM,

) Detects a temporary data symbol from the received signal. When the detection is performed according to the ML technique, the detected data symbol is expressed by Equation 23 below.

)을 이용하여 최종적으로 채널을 추정한다. 이러한 최종의 채널 추정을 위하여, 먼저, 수신기는 ML 기법을 사용하여 다음의 <수학식 24>에 따라 채널을 추정한다. Subsequently, the receiver detects the temporary data symbol detected in step S1507.

) To finally estimate the channel. For this final channel estimation, first, the receiver estimates the channel according to the following Equation 24 using the ML technique.

)에 대하여 DFT기반의 채널 추정 기법을 적용하여 채널을 추정한다. 즉, 최종 채널 추정 값(CFR,

)을 도출한다. The receiver then estimates the estimated channel value (

We estimate the channel by applying DFT-based channel estimation technique. That is, the final channel estimate (CFR,

).

)과 임시 데이터 심벌(

)을 이용하여 최종 간섭 추정을 수행한다. 최종 간섭 추정 값을 산출하기 위하여 먼저, 수신기는 임시 간섭 추정(instantaneous SCM)을 수행한다. 이러한 임시 간섭 추정은 다음의 <수학식 25>에 따라 이루어진다. Next, the receiver determines the S1509 final channel estimate (CFR,

) And temporary data symbols (

) Performs the final interference estimation. In order to calculate the final interference estimate, the receiver first performs an instantaneous interference estimate (Instantaneous SCM). Such temporary interference estimation is performed according to Equation 25 below.

)에 대하여 LPS 기반의 간섭 추정 기법을 재차 적용하여 최종적으로 간섭을 추정한다. 즉, 최종 간섭 추정 값(SCM

)을 도출한다. The receiver then determines the temporary interference estimated temporary interference estimate (

LPS-based interference estimation method is applied again to finally estimate the interference. That is, the final interference estimate value (SCM

).

)과 최종 채널 추정 값(CFR,

)에 따라 채널 보상 및 간섭 제거를 수행하여, S1513 단계에서 데이터 심벌을 복조하여 이진 데이터를 검출한다. Then, the receiver determines the final interference estimate value (SCM) in step S1511.

) And final channel estimates (CFR,

Channel compensation and interference cancellation are performed to detect binary data by demodulating the data symbols in step S1513.

As described above, when channel estimation and interference estimation are performed again using the detected temporary data symbols, the channel estimation and the interference estimation may be performed on the effective subcarriers. Noise reduction effect can be obtained. In addition, the interference estimation performance can be improved by estimating the power of "interference plus noise" that increases momentarily at the data symbol position more closely than the conventional interference estimation method.

Next, the performance according to the second embodiment of the present invention will be described. 16 to 21 are diagrams for explaining the performance of the channel and interference estimation method of the present invention.

The simulation for verifying the performance of the algorithms presented in the present invention is based on the 3GPP LTE MIMO-OFDMA standard. This was done by recording. In the 16QAM modulation scheme and the Cost 207 TU (Typical Urban) channel environment, 60km / h and 120km / h vehicle speeds were considered, and each parameter applied to the experiment was shown and included in each drawing (FIGS. For performance evaluation, the coded rate 1/3 and the internal and external interleaver used turbo code consisting of "mother interleaver" and "1 ^st block interleaver", respectively, and used eight iterations. The "Max-log Map turbo decoder" was applied.

Each technique set for comparison is as follows.

1D (Dimensional) Interpolation (Applying the same CFR and SCM estimated for the data OFDM symbol before the next reference symbol after the frequency axis interpolation with the first linear interpolation method in the conventional Lagrange interpolation method)

② 2D Interpolation (Freq.-Time) (First-order linear interpolation among conventional Lagrange interpolation techniques, frequency axis interpolation precedes interpolation to time axis)

③ 2D Interpolation (Time-Freq.) (First-order linear interpolation among the conventional Lagrange interpolation techniques, time axis interpolation precedes interpolation to frequency axis)

④ DFT-based (conventional DFT-based CFR estimation technique that selects CIR by the GI length as an active ingredient)

⑤ LPS (a conventional LPS-based SCM estimation technique that selects a time domain SCM as an active component as long as the GI length)

⑥ ML-based iterative (a technique using temporarily estimated received signal after ML decision of the present invention)

FIG. 16 is a graph showing BER performance for the CFR estimation scheme proposed in the present invention, assuming that the SCM for interference and noise is ideally known.

In FIG. 16, it is assumed that the SCM can be perfectly estimated at the receiver in order to evaluate the performance of various channel estimation schemes in the 1Tx-2Rx SIMO-OFDMA system.

In the case of CFR estimation based on linear interpolation, the change of channel is relatively insignificant with time in mobile environment of 120km / h or less, so that other linear interpolation is preceded by time axis interpolation. The performance is better than that of the However, since error due to neighboring cell interference increases, it is significantly degraded than ideal performance assuming ideal CFR estimation, and stable system performance cannot be maintained.

The general DFT-based channel estimation algorithm, which is more robust to errors due to the effects of adjacent cell interference (Asynchronous ICI), outperforms the linear interpolation-based channel estimation scheme, but is still ideal at BER = 10 ^-5 . (ideal) The difference in performance is about 3dB or more.

On the contrary, it can be seen that the scheme according to the embodiment of the present invention is almost similar to the ideal performance as the power of the interference signal increases. This is due to the improvement of the accuracy of the CFR estimation by effectively improving the limit of the channel estimation performance of the conventional DFT-based channel estimation algorithm using only the initial CFR estimated at the reference symbol subcarrier location. It is considered as an optimal CFR estimation method that can approach ideal performance in ICI) environment.

FIG. 17 is a graph showing BER performance for the SCM estimation scheme proposed in the present invention, assuming that the CFR for the received signal is ideally known. In FIG. 17, similarly to FIG. 16, it is assumed that the CFR of a received signal can be perfectly estimated at a receiver for performance evaluation between various SCM estimation schemes in a 1Tx-2Rx SIMO-OFDMA system, and according to a change in power of an interference signal. The results of the comparative analysis are shown.

Linear interpolation-based SCM estimation and LPS-based SCM estimation technique can accurately estimate the power of asynchronous ICI and noise that are instantaneously increased due to adjacent cell interference at the data subcarrier location between reference symbols. As a result, overall performance is poor. In addition, even though the SIR is increased, it can be seen that an error floor occurs in BER = 10 ⁻⁵ or more. This phenomenon is because, in the case of linear interpolation, an estimation error increases due to an instantaneous increase in SCM when estimating an SCM having different values according to subcarrier positions within an effective subcarrier range.

In addition, the LPS technique exhibits a limitation in performance due to the limitation of using only the instantaneous SCM estimated at the reference symbol position.

In contrast, the method according to the embodiment of the present invention uses the SCM information for all valid subcarriers using the detected temporary received symbols, thereby overcoming the limitations of the existing LPS scheme, thereby obtaining a more effective noise attenuation effect. In addition, it can be seen that the method according to the embodiment of the present invention shows better performance than the conventional SCM estimation method. In addition, the method according to an embodiment of the present invention shows a tendency to become closer to ideal performance without an error floor phenomenon as the SIR increases.

18 to 21 are graphs comparing the BER performance of a receiver structure in which the CFR and SCM estimation algorithms proposed by the present invention are linked.

18 and 19 consider a SIMO-OFDMA system of 1Tx-2Rx, and FIGS. 20 and 21 consider a MIMO-OFDMA system of 2Tx-2Rx.

The lower-bound disclosed in FIGS. 18 to 21 assumes ideal channel estimation (ideal CFR estimation), and illustrates the case in which the interference estimation technique proposed by the present invention is applied. This can be used as a measure of the degree of performance degradation of the proposed estimation algorithm due to channel estimation error, compared to the case of performing interference estimation in a non-ideal environment.

As can be seen from the performance evaluation results, the method of using CFR and SCM information for data symbols according to an embodiment of the present invention is more interference and compared to the conventional method using DFT-based CFR estimation and LPS-based SCM estimation. Excellent performance is achieved by improving the accuracy of CFR and SCM estimation due to the noise attenuation effect.

In addition, according to an embodiment of the present invention, even when considering realistic CFR estimation, it is confirmed that a stable performance of approx. 1 dB to a lower bound performance can be obtained in all environments without an error floor phenomenon. Can be.

In summary, the method according to the embodiment of the present invention is existing even in an environment of severe Doppler shift due to the frequency selective fading effect of the adjacent cell interference (asynchronous ICI) environment and the multipath fading channel and the increase of the moving speed. It can be seen that better performance than the technique. In addition, by performing channel estimation and interference estimation using data symbols, it is possible to ensure stable receiver performance by increasing accuracy of channel and interference estimation.

While the present invention has been described with reference to several preferred embodiments, these embodiments are illustrative and not restrictive. As such, those of ordinary skill in the art will appreciate that various changes and modifications can be made according to equivalents without departing from the spirit of the present invention and the scope of rights set forth in the appended claims.

1 is a diagram illustrating a baseband transmitter structure of an OFDM communication system according to an embodiment of the present invention.

2 is a diagram illustrating a frame structure of an OFDM communication system according to an exemplary embodiment of the present invention.

3 is a view for explaining a channel estimation apparatus of a receiver according to an embodiment of the present invention.

4 illustrates a DFT-based channel estimation technique according to an embodiment of the present invention.

5 is a diagram for explaining a method of selecting an effective CIR from an estimated CIR.

6 is a view for explaining a channel estimation method in an OFDM system according to an embodiment of the present invention.

7 to 9 are diagrams for explaining the performance of the channel estimation method according to an embodiment of the present invention.

10 is a view for explaining the structure of the baseband transmitter of the MIMO-OFDMA communication system according to an embodiment of the present invention.

11A and 11B are diagrams for explaining an example of a frame structure of an OFDMA communication system according to another embodiment of the present invention.

12 to 14 are diagrams for explaining a channel and interference estimation apparatus of a receiver according to an embodiment of the present invention.

15 illustrates a channel and interference estimation method according to another embodiment of the present invention.

16 to 21 are diagrams for explaining the performance of the channel and interference estimation method of the present invention.

Claims (10)

In the channel estimation method of a wireless communication system, Estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol; Compensating for the channel of the received signal by using the estimated initial channel; Generating a temporary data symbol by demodulating and modulating the data symbol according to the compensated channel; And estimating a final channel using the generated temporary data symbols. The method of claim 1, The process of estimating the initial and final channel is Estimating a channel in the frequency domain according to a Least Square (LS) estimation technique; And estimating a channel in the time domain according to a Discrete Fourier Transform (DFT) based channel estimation technique. In the channel estimation apparatus of a wireless communication system, An equalizer for compensating for the channel of the received signal according to the estimated channel; A demodulator and modulator for demodulating and modulating data symbols according to the compensated channel to generate temporary data symbols; And A channel estimator for estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol and estimating a final channel using the temporary data symbol Device. The method of claim 3, The channel estimator An LS channel estimator estimating a channel in a frequency domain according to a least square (LS) estimation technique; And And a DFT-based channel estimator for estimating a channel in the time domain according to a Discrete Fourier Transform (DFT) -based channel estimation technique. In the channel and interference estimation method of a wireless communication system, Estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol in a beauty system; Estimating initial interference using the estimated initial channel and the reference symbol; Compensating for the channel of the received signal according to the estimated initial channel and detecting the temporary data symbol by removing the interference of the received signal according to the estimated initial interference; Estimating a final channel using the detected temporary data symbols; Estimating a final interference using the detected temporary data symbol and the estimated final channel. The method of claim 5, The process of estimating the initial and final channel is Estimating a channel in the frequency domain according to a Least Square (LS) estimation technique; And estimating a channel in the time domain according to a Discrete Fourier Transform (DFT) based channel estimation technique. The method of claim 5, The process of estimating the initial and final interference Estimating the instantaneous interference (SCM), A method for estimating channel and interference in a wireless communication system, comprising estimating interference according to a low-pass smoothing (LPS) technique. In the channel and interference estimation apparatus of a wireless communication system, An equalizer for compensating for the channel of the received signal according to the initially estimated channel and canceling the interference of the received signal according to the initially estimated interference; A detector for detecting temporary data symbols in the received signal from which the channel compensation and the interference cancellation are performed; A channel estimator estimating an initial channel using a reference symbol of a received signal including a reference symbol and a data symbol, and estimating a final channel using the temporary data symbol; And And an interference estimator for estimating an initial interference using a reference symbol of a received signal including a reference symbol and a data symbol, and further estimating a final interference using the estimated final channel and the temporary data symbol. Channel and interference estimation apparatus of a wireless communication system. The method of claim 8, The channel estimator An LS channel estimator estimating a channel in a frequency domain according to a least square (LS) estimation technique; And And a DFT-based channel estimator for estimating a channel in a time domain according to a Discrete Fourier Transform (DFT) -based channel estimation technique. The method of claim 8, The interference estimator A temporary time estimator for estimating temporary interference; And LPS interference estimator for estimating interference in accordance with the LPS technique; Channel and interference estimation apparatus of a wireless communication system comprising a.

Images