KR101294283B1 - Method for estimating channel based on cross correlation of ofdm system and device thereof - Google Patents

Abstract

In the channel estimation scheme of the present invention, unlike the conventional method of cross-correlation channel estimation using a pilot symbol, the power of the pilot signal is reduced due to the decrease of the number of pilot symbols and the channel estimation performance is reduced. Channel estimation performance is improved.

Description

Cross-correlation-based channel estimation method and apparatus thereof in OPEM system TECHNICAL FIELD

The present invention relates to a method and apparatus for cross-correlation-based channel estimation of an OFDM system, and more particularly, to cross-correlation-based OFDM system for reducing the performance and complexity of a cross-correlation channel estimation technique using a pilot used in an OFDM system. A channel estimation method and apparatus are provided.

1 is a block diagram showing the configuration of a conventional OFDM system.

Referring to FIG. 1, when data bits are input to a modem, data symbols are output through a modulator. The output data is allocated to each subcarrier through S / P (Serial to Parallel) block.

The symbols assigned to each subcarrier are modulated into time-domain signals by an Inverse Fast Fourier Transform (IFFT), and inter-symbol interference (ISI) and inter-carrier interference by adjacent multi-channel channels. A Cyclic Prefix (CP) is inserted to prevent ICI and transmitted through a wireless channel.

The OFDM symbol at the nth sampling time of the transmitted data signal can be represented as follows.

은 전체 부반송파 수를 나타낸다.

는

번째 부반송파로 전송되는 주파수 영역 OFDM 심볼을 나타내며 다음과 같이 나타낼 수 있다.
here

Represents the total number of subcarriers.

The

A frequency domain OFDM symbol transmitted on the first subcarrier may be represented as follows.

와

는 각각 데이터 부반송파 인덱스와 파일럿 부반송파 인덱스의 집합을 나타낸다. here

Wow

Denotes a set of data subcarrier indexes and pilot subcarrier indexes, respectively.

개의 부반송파가

개의 데이터와

개의 파일럿 부반송파로 구성된다고 가정한다. 따라서

이며, 데이터 심볼과 파일럿 심볼은 중첩되지 않는 서로 다른 부반송파로 전송되며, 송신 심볼

은

로 나타낼 수 있다. 여기서,

와

는 각각 데이터 심볼과 파일럿 심볼을 나타낸다. here,

Subcarriers

Data and

Assume that the pilot subcarriers are configured. therefore

The data symbol and the pilot symbol are transmitted on different subcarriers that do not overlap, and the transmission symbol

silver

. here,

Wow

Denote data symbols and pilot symbols, respectively.

는 다음과 같이 나타낼 수 있다.
Signal transmitted during OFDM symbol period is received with noise after passing through multipath channel.

Can be expressed as follows.

은 기저대역에서 샘플링된

번째 다중경로 채널의 이산시간 임펄스 응답을 나타내며,

은 채널의 전체 샘플 수를 나타낸다. here

Is sampled at baseband

Discrete time impulse response of the first multipath channel,

Represents the total number of samples in the channel.

은 평균이 0이며 분산이

인 가산성 백색 가우스 잡음(Additive White Gaussian Noise; AWGN)을 나타낸다. 수신 신호를 FFT(Fast Fourier Transform)를 통해 복조하면 다음과 같이 나타낼 수 있다.
Also,

Is 0 and the variance is

Phosphorus Additive White Gaussian Noise (AWGN). Demodulating the received signal through the fast fourier transform (FFT) can be expressed as follows.

와

는 각각

번째 부반송파에서의 채널 주파수 응답(Channel Frequency Response; CFR)과 AWGN의 주파수 응답을 나타내며, CFR을 이용하여 데이터 심볼을 다음과 같이 검출할 수 있다.
here

Wow

Respectively

The channel frequency response (CFR) and the frequency response of the AWGN in the first subcarrier are shown, and data symbols can be detected as follows using the CFR.

를 다음과 같이 정의할 수 있다.
Cross-correlation signal between received signal including pilot and sample delayed time domain pilot signal

Can be defined as:

은 파일럿 심볼의 시간 영역 신호를 나타내며 다음과 같다.
here,

Denotes a time-domain signal of a pilot symbol and is as follows.

는 신호

와

의 상관함수를 나타내며 다음과 같이 정의된다.
Also,

The signal

Wow

It is the correlation function of and is defined as

는 송신된 데이터 신호와 파일럿 신호 간의 상호상관을 나타내며 다음과 같다.
here,

Denotes the cross correlation between the transmitted data signal and the pilot signal.

는 0이 된다. 또한,

는 파일럿 심볼 간의 상호상관을 나타내며 다음과 같다.
Since the data symbol and the pilot symbol are allocated to different subcarriers by Equation 2, Equation (9)

Becomes zero. Also,

Denotes cross-correlation between pilot symbols.

는 다음 수식과 같이 나타낼 수 있다.
Pilot symbols are generally transmitted via modulation such as BPSK or QPSK, where all symbols have the same power,

Can be expressed as

은 파일럿 신호의 전체 전력을 의미하며, 송신 신호의 전체 전력을 1이라고 가정하면 파일럿 신호의 전력은

이 된다. 전술한 수학식 9와 수학식 11에 의해 수학식 6은 다음과 같이 나타낼 수 있다.
here

Is the total power of the pilot signal. Assuming that the total power of the transmission signal is 1, the power of the pilot signal is

. Equation 6 may be expressed as follows by Equations 9 and 11 described above.

번재 다중경로 채널은 다음과 같이 추정할 수 있다.
therefore,

The multipath channel can be estimated as follows.

은 여전히 존재하게 되며, 그 값은 파일럿 심볼 수를 증가시켜

의 값이 증가하거나 평균을 구하는 데이터의 수가 증가할수록 감소한다. In Equation 13, interference with data is removed by orthogonality, but interference with AWGN

Is still present, and the value increases by increasing the number of pilot symbols.

The value of increases or decreases as the number of averaged data increases.

으로 고정되므로 채널 추정 성능 향상을 위해서는 파일럿 심볼 수를 증가시켜야 한다. 파일럿 심볼 수의 증가는 데이터 심볼 수의 감소를 초래하게 되어 데이터 전송 효율이 저하된다.
But the number of data

Since the number of pilot symbols must be increased for improving channel estimation performance. An increase in the number of pilot symbols results in a decrease in the number of data symbols, thereby degrading data transmission efficiency.

Accordingly, an object of the present invention is to improve the channel estimation performance by using a small number of pilot symbols, and at the same time occurs when using quadrature amplitude modulation (QAM) modulation instead of constant shift keying (PSK) modulation. The present invention provides a method of cross-correlation based channel estimation of an OFDM system that can reduce complexity by simply removing interference.

Another object of the present invention is to provide a channel estimating apparatus using the channel estimation method described above.

According to an aspect of the present invention, there is provided a cross-correlation-based channel estimation method of an OFDM system, by using a received signal in a time domain received from a transmitter and a pilot signal in a time domain known in advance. Calculating a channel estimated primary channel estimate obtained by calculating cross-correlation between a received signal and a first channel estimated signal delaying the pilot signal, and respectively receiving the received signal and the primary channel estimated value in the time domain Converting the received signal into the primary channel estimate value in the frequency domain, and demodulating the received signal in the frequency domain based on the primary channel estimate value in the frequency domain to detect a data signal included in the received signal. And normalizing the detected data signal, and normalizing the detected data signal to a data in a time domain. Converting the signal, adding the detected data signal and the pilot signal in the time domain to obtain a second channel estimation signal, and calculating a cross-correlation between the second channel estimation signal and the received signal in the time domain Calculating a final channel estimate, and performing channel estimation.

In accordance with another aspect of the present invention, a cross-correlation-based channel estimation apparatus of an OFDM system receives a received signal in a time domain and a pilot signal in a time domain received from a transmitter, and delays the received signal and the pilot signal. A cross-correlation calculator for outputting a channel estimated primary channel estimate obtained by calculating cross-correlation between the received signal in the time domain and the primary channel estimated value in the frequency domain And a fast Fourier transform calculator for converting the signal to the first signal and a demodulator for detecting the data signal included in the received signal by demodulating the received signal in the frequency domain based on the primary channel estimate value of the frequency domain. The apparatus may further include a normalization operator for multiplying the detected data signal by a normalization coefficient, and an inverse fast Fourier transformer for inverse fast Fourier transform of the normalized data signal. A feedback value obtained by summing the normalized data signal converted by the inverse fast Fourier transformer and the received signal in the time domain, and calculating a cross-correlation between the feedbacked summed result value and the received signal in the time domain to estimate a final channel. It characterized in that to perform.

According to the present invention, by estimating a channel using only pilot symbols, unlike the conventional case in which the channel estimation performance decreases when the number of pilot symbols decreases, the detected data and pilot symbols are used even when a small number of pilot symbols are used. By estimating the channel, interference with data is eliminated, which improves channel estimation performance.

In addition, since the interference generated when the QAM modulation method is used rather than the PSK modulation method can be easily removed using the detected data symbol, the complexity can be greatly reduced.

1 is a block diagram showing the configuration of a conventional OFDM system.
2 is a block diagram showing the overall configuration of a channel estimating apparatus according to an embodiment of the present invention.
FIG. 3 is a constellation diagram for describing a concept of a normalization process performed by the normalization operator illustrated in FIG. 2.
4 to 6 are diagrams showing simulation results according to a channel estimation technique proposed in an embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings, and terms such as a signal, a symbol, and a value described in the following embodiments appear. Unless otherwise defined, used interchangeably. Thus, it will be understood by those skilled in the art that the terminology of the signal may be replaced by terminology such as symbols or values.

2 is a block diagram showing the overall configuration of a channel estimating apparatus according to an embodiment of the present invention.

Referring to FIG. 2, the channel estimating apparatus 100 according to an exemplary embodiment of the present invention has a cross-correlation channel using a conventional pilot symbol in which power of a pilot signal is reduced due to a decrease in the number of pilot symbols, thereby degrading channel estimation performance. Unlike the estimation method, even when a small number of pilot symbols are used, pilot-based channel estimation is performed using cross-correlation, which improves channel estimation performance.

 To this end, the channel estimating apparatus 100 according to the embodiment of the present invention is largely divided into a delay unit 102, a mux 104, a first fast Fourier transform (FFT) operator 106, and a mutual shunt. It includes an operator 108, a second FFT operator 110, and a demodulator 112, and further includes a normalization operator 114 and an IFFT operator 116.

)를 d 샘플 지연된 제1 채널 추정 신호(

)를 출력한다.The delay unit 102 is a pilot signal of the time domain known to the receiver in advance (

) Is the first sample estimate signal (

).

) 및 상기 IFFT 연산기(116)의 출력값(

)과 상기 파일롯 신호(

) 간의 합한 최종 채널 추정 신호(

) 중 어느 하나를 선택적으로 출력한다. The mux 104 has a first channel estimate signal from the delay 102.

) And the output value of the IFFT operator 116 (

) And the pilot signal (

Sum of final channel estimates between

Outputs any one of

)와 함께 송신단으로부터 수신되는 수신 신호(

)를 입력받아서 이를 고속 푸리에 변환하여 주파수 영역의 수신 신호(

)로 변환한다. The first FFT operator 106 performs a pilot signal of the time domain known to the receiver in advance.

Along with the received signal from the transmitter (

) And then fast Fourier transform the received signal in the frequency domain (

).

)와 상기 송신단으로부터 수신된 상기 시간 영역의 수신 신호(

) 간의 상호 상관을 계산하는 과정이다. 최종 채널 추정값을 계산하는 과정은 상기 1차 채널 추정값을 계산하는 과정에 따라 계산된 결과치(

)를 이용하여 상기 복조기(112)에서 검출된 데이터 신호(

)를 상기 정규화 연산기(114)를 거쳐 IFFT 연산기(116)를 통해 출력되는 출력 값과 상기 시간 영역의 파일롯 신호(

)를 합산하는 과정과, 이 합산 과정에 의한 제2 채널 추정 신호(

)와 상기 시간 영역의 수신 신호(

) 간의 상호 상관을 다시 계산하여 최종 채널 추정 값(

)을 계산하는 과정으로 이루어진다. 이에 대한 보다 구체적인 상호상관 연산과정은 아래에서 제시되는 수학식을 통해 상세히 설명된다.Cross correlator operator 108 calculates the cross-correlation between the two signals. Computing cross-correlation by the cross-correlation operator 108 includes calculating a primary channel estimate and calculating a final channel estimate. The process of calculating the primary channel estimate includes the first channel estimate signal delayed by the delay unit 102.

) And the received signal of the time domain received from the transmitting end (

) Is the process of calculating the cross-correlation between The process of calculating the final channel estimate may include a result calculated according to the process of calculating the primary channel estimate.

Data signal detected by the demodulator 112 using

) Is outputted through the normalization operator 114 through the IFFT operator 116 and the pilot signal of the time domain.

), And the second channel estimation signal ()

) And the received signal in the time domain (

Recalculate the cross-correlation between

) Is calculated. More specific cross-correlation operation process is described in detail through the equations presented below.

) 또는 최종 채널 추정값(

)을 고속 푸리에 변환 연산을 통해 주파수 영역의 채널 추정값으로 변환하고, 이를 복조기(112)로 전달한다.The second FFT operator 110 calculates the first channel estimate calculated by the cross-correlation operator 108.

) Or final channel estimate (

) Is converted into a channel estimate of the frequency domain through a fast Fourier transform operation, and is transmitted to the demodulator 112.

)를 복조하여 데이터 신호(

)를 검출한다. The demodulator 112 receives the received signal in the frequency domain transformed by the first FFT operator 106 using the first and last channel estimates converted by the second FFT operator 110.

) Demodulates the data signal (

).

)를 정규화한다.The normalization operator 114 detects the detected data signal according to the primary channel estimate.

) Normalize

)를 역 고속 푸리에 변환한 결과치(

)를 출력한다. IFFT operator 116 is the result value normalized by the normalization operator 114 (

Inverse fast Fourier transform

).

)는 시간 영역의 수신 신호(

)와 합산되어 최종 채널값을 생성하기 위한 최종 채널 추정 신호(

)로서 먹스(104)에 전달되고, 먹스(104)는 최종 채널 추정 신호(

)를 상기 상호상관 연산기(108)로 전달한다. 이후, 상술한 바와 같이, 상호상관 연산기(108)는 상기 최종 채널 추정 신호와 상기 시간 영역의 수신 신호(

) 간의 상호상관을 계산하여 최종 채널 추정값을 출력한다.The result value calculated by the IFFT operator 116 (

) Is the received signal in the time domain (

) And the final channel estimate signal for summation to produce the final channel value (

) Is passed to the mux 104, and the mux 104 receives the final channel estimation signal (

) Is passed to the cross-correlation operator 108. Then, as described above, the cross-correlation operator 108 may receive the final channel estimate signal and the received signal in the time domain (

) And output the final channel estimate.

)을 생성하고, 이 변형 파일럿(

)을 이용하여 상호상관 기반 채널 추정을 수행한다. As described above, in the present invention, after estimating a channel based on cross-correlation between a received signal and a pilot signal, data is detected. Thereafter, the detected data signal is normalized and the final channel estimation is performed by recalculating the correlation between the result of summing the normalized data signal and the pilot signal and the received signal. That is, in the present invention, in order to improve channel estimation performance using a small number of pilot symbols, a modified pilot (

), This variant pilot (

) Is used to perform cross-correlation based channel estimation.

Meanwhile, each of the building blocks shown in FIG. 2 may be implemented with various logic circuit modules for performing respective calculation processes. In addition, each component block of FIG. 2 may be implemented as a chip-type logic circuit module that is separated from each other in an independent form and performs computational logic, respectively, and an operation process performed in each configuration block is implemented as one software logic. It is obvious that the software logic can be implemented as an integrated logic circuit module in the form of a single chip.

Hereinafter, an operation process of each component block of the channel estimating apparatus shown in FIG. 2 will be described in detail using the following equations. In addition, in FIG. 2, values (or signals) representing the inputs and outputs of the respective building blocks are shown in the following equations. Therefore, it will be apparent to those skilled in the art that the operation functions (or calculation processes) of the respective building blocks can be clearly understood through the following equations.

In the OFDM system, the cross-correlation channel estimation technique using pilot degrades the channel estimation performance when using a small number of pilot symbols. Accordingly, the present invention improves channel estimation performance by using a small number of pilot symbols. In addition, the complexity may be reduced by simply removing interference generated when using a Quadrature Amplitude Modulation (QAM) modulation method instead of a phase shift keying (PSK) modulation having a constant signal size.

In the conventional technique of cross-correlation channel estimation using pilot symbols, the power of the pilot signal is reduced by reducing the number of pilot symbols, thereby degrading channel estimation performance. In the present invention, a data symbol is detected as shown in Equation 5 using the channel estimated in Equation 13 described above.

A channel estimation signal may be generated using the detected data symbols and the pilot symbols as follows.

)는 다음과 같이 나타낼 수 있다.
Here, assuming that there is no error in the detected data symbol, the cross-correlation signal between the received signal and the d sample delayed channel estimation signal (

) Can be expressed as follows.

는 송신 신호와 채널 추정 신호와의 상호상관을 나타내며 다음과 같다.
here,

Denotes the correlation between the transmission signal and the channel estimation signal.

이 되어

이 된다. 수학식 15로부터

번째 경로의 채널을 추정하면, 다음과 같이 나타낼 수 있다.
Assuming there are no errors in the detected data symbols,

This

. From equation (15)

If the channel of the first path is estimated, it can be expressed as follows.

는 송신신호의 전력을 나타내며, 파일럿 심볼과 데이터 심볼의 전력을 포함하고 있으므로, 전술한 수학식 13에 비해 잡음에 의한 영향을 감소시킬 수 있다. here,

Denotes the power of the transmission signal, and includes the power of the pilot symbol and the data symbol, so that the influence of noise can be reduced as compared with Equation (13).

는 항상 0이 되지는 않으므로, 데이터 간의 상호상관 간섭인 CCI(Cross-Correlation Interference)가 발생하여 채널추정 성능에 영향을 미치게 되므로 CCI를 보상하여야 한다. However, unlike Equation 11 described above

Since is not always 0, CCI (Cross-Correlation Interference), which is a cross-correlation interference between data, occurs and affects channel estimation performance.

Equation (15) described above for the CCI compensation is represented by a matrix.

,

,

이며,

이다.
here,

,

,

Is,

to be.

을 추정할 수 있다.
A channel compensated for CCI by using Equation 18 as shown in Equation 19 below

Can be estimated.

의 값이

와 관계없이 동일한 값이 되므로,

가 0이 되어 CCI가 발생하지 않는다. As can be seen from Equation 16, when the PSK modulation scheme uses a constant power of a transmission symbol,

The value of

Is the same value regardless of,

Becomes 0 and no CCI occurs.

을 1로 정규화할 경우 제거 될수 있으므로, 이를 위해 검출된 데이터 심볼을 아래의 수학식 20과 같이 변환한다.
However, if the power of a transmission symbol is not constant, such as a QAM modulation scheme, CCI occurs and complexity increases. This CCI is shown in Figure 3

Since it can be removed when normalizing to 1, the detected data symbol is converted as shown in Equation 20 below.

는 정규화 계수를 나타내며, 아래의 수학식 21과 같이 구할 수 있다.
here,

Denotes a normalization coefficient and can be obtained as shown in Equation 21 below.

A channel estimation signal is generated as follows by using the temporary data symbol converted in Equation 20 above.

와

을 각각

와

으로 치환하면, 다음과 같이 상관신호

를 얻을 수 있다.
Of Equation 15 described above

Wow

Each

Wow

If replaced with the correlation signal as follows:

Can be obtained.

The channel is estimated using Equation 23 as follows.

행렬의 구성요소는 다음과 같다.
here,

The components of the matrix are as follows.

은 아래의 수학식 26과 같이 나타낼 수 있다.
Here, assuming that there is no error in the detected data symbol and substituting Equation 20 above,

May be represented by Equation 26 below.

는 다음과 같이 된다.
By Equation 26

Becomes

Therefore, when Equation 27 is substituted into Equation 24, it is as follows.

만큼 감소하는 것을 알 수 있다.
Equation 28 can be seen that the complexity is greatly reduced since the additional operation for channel estimation is not necessary as compared with Equation 19. In addition, the channel estimated by Equation 28 is the power of the noise component compared to the conventional technique

It can be seen that the decrease.

4 to 6 are diagrams showing simulation results according to a channel estimation technique proposed in an embodiment of the present invention.

The channel used in the simulation used a 2-path Rayleigh fading model. The total bandwidth 20MHz was divided into 1024 subcarriers, and the pilot used BPSK modulation. In addition, it is assumed that the carrier frequency offset (CFO) and symbol timing offset (STO) are perfectly compensated, and the multipath position of the channel is correctly estimated.

4 shows the performance of the conventional cross-correlation based channel estimation scheme according to the number of pilots. The pilot symbols are arranged at equal intervals in the frequency domain, and M represents the interval between pilot symbols. When M is 8, the total number of pilot symbols is 128 (= 1024/8), and when M is 64, a total of 16 pilots is used. In FIG. 4, it can be seen that, as the number of pilot symbols increases, the influence of AWGN decreases, thereby improving MSE performance.

In FIG. 5, when M is 8, the performance of the conventional scheme and the proposed cross-correlation based channel estimation scheme are shown. For the proposed scheme, data symbols use BPSK modulation. When the number of pilots is 128 symbols, it can be seen that the SNR performance of about 10 dB is improved in the MSE 10-4 compared to the conventional scheme. When the number of pilots is 16 symbols, the SNR performance is improved by about 22 dB.

6 shows the performance when 16 pilot symbols are used and data symbols are modulated with 16-QAM.

In the conventional scheme, since only pilot symbols are used, the same performance is obtained regardless of data symbols. However, in the channel estimation scheme according to the present invention, the probability of error is increased at low SNR due to the use of a high modulation scheme. It can be seen that occurs.

However, as the SNR increases, the data error decreases, and the SNR performance of about 16 dB is improved in the MSE 10-4 compared to the conventional technique.

In addition, as shown in FIG. 6, there is a difference between a method of removing adjacent correlation interference using Equation (24) and a method of removing adjacent correlation interference by zero-forcing using Equation 18 (Detected data). It can be seen that there is little. Both methods show similar performances, and it can be seen that there is almost no performance degradation even when the complexity of the modified pilot is used.

As described above, the present invention proposes a technique for improving the performance of cross-correlation based channel estimation in an OFDM system. In the conventional scheme, the channel estimation performance deteriorates as the number of pilot symbols decreases. However, the proposed cross-correlation-based channel estimation scheme of the present invention shows similar performance regardless of the number of pilots.

In addition, when the QAM modulation scheme is used instead of the constant PSK modulation scheme, the power of the data symbol has a disadvantage in that the complexity of the matrix operation is increased due to adjacent correlation interference, but a modified pilot symbol is used. You can simply remove the complexity.

The performance of the channel estimation scheme proposed in the present invention is as confirmed through the simulations illustrated in FIGS. 4 to 6, and it can be seen that the performance improvement is much higher than that of the conventional pilot technique. In addition, it can be seen that the technique using the modified (pilot) symbol shows a similar performance to the existing technique while greatly reducing the complexity.

Claims (6)

Calculating a primary channel estimate by cross-correlating a delayed pilot signal delaying a pilot signal in a time domain previously known at a receiver and a received signal in a time domain received from a transmitter;
Converting the received signal in the time domain and the primary channel estimate into a received signal in the frequency domain and a primary channel estimate in the frequency domain, respectively;
Detecting a data signal included in the received signal by demodulating the received signal in the frequency domain based on a primary channel estimate value of the frequency domain;
Normalizing the detected data signal and converting the normalized data signal into a data signal in a time domain;
Adding a data signal of the time domain and a pilot signal of the time domain to obtain a second channel estimation signal; And
Calculating a final channel estimate by cross-correlating the second channel estimate signal and the received signal in the time domain;
A cross-correlation based channel estimation method of an OFDM receiving system comprising a.
The method of claim 1, wherein the normalization of the detected data signal comprises:
The detected data signal

Denote the normalized detected data signal.

When we write,

Is performed through the equation of,
Here,

Is the normalization factor,

Correlation-based channel estimation method of the OFDM system, characterized in that it is calculated through the equation.
The method of claim 2, wherein the obtaining of the second channel estimation signal comprises:
The channel estimation signal

In other words,

Obtained through this equation,
Here,

Quot;

Inverse fast Fourier transform result of

The cross-correlation-based channel estimation method of the OFDM system, characterized in that for representing the pilot signal in the time domain.
The method of claim 1, wherein performing the channel estimation comprises:
And performing channel estimation of the received signal in the time domain and the pilot signal in the time domain modulated by a Quadrature Amplitude Modulation (QAM) by the transmitting end.
A cross-correlation calculator for correlating a delayed pilot signal delaying a pilot signal in a time domain known in advance by a receiver and a received signal in a time domain received from a transmitter, and outputting a first channel estimate value;
A fast Fourier transform calculator for converting the received signal in the time domain and the primary channel estimate into a received signal in the frequency domain and a primary channel estimated in the frequency domain, respectively;
A demodulator for demodulating a received signal of the frequency domain based on a first channel estimate value of the frequency domain to detect a data signal included in the received signal;
A normalization operator for multiplying the detected data signal by a normalization coefficient to normalize the detected data signal; And
An inverse fast Fourier transformer for inverse fast Fourier transforming the normalized data signal to a data signal in a time domain,
The cross-correlation calculator,
And a result of summing the result value of the data signal of the time domain and the pilot signal of the time domain and the received signal of the time domain to estimate a final channel.
The method of claim 5, wherein the received signal in the time domain and the pilot signal in the time domain received from the transmitter,
A cross correlation based channel estimation apparatus of an OFDM system, which is modulated by a phase shift keying (PSK) or quadrature amplitude modulation (QAM).

Images