KR20060117544A - Apparatus and method for channel estimation of ofdm system - Google Patents

Abstract

An apparatus and a method for estimating a channel of an OFDM(Orthogonal Frequency Division Multiplex) system are provided to obtain a channel estimation value strong to channel environment, by correcting a channel estimation value of a desired sub carrier with an average value of channel estimation values of adjacent frequencies according to frequency correlation during estimation. In an apparatus for estimating a channel of an OFDM(Orthogonal Frequency Division Multiplex) system, a first channel estimation part(25b) obtains a new channel estimation value by considering a channel estimation value of a sub carrier adjacent to a channel estimation value estimated by a preamble as to each sub carrier and a weight value of each sub carrier. A second channel estimation part(25c) adds the channel estimation values of each sub carrier obtained in the first channel estimation part, and obtains an average of the channel estimation values of each sub carrier, and then compensates a channel by multiplying the average by a received OFDM information symbol.

Description

Apparatus and method for channel estimation of OPM system {APPARATUS AND METHOD FOR CHANNEL ESTIMATION OF OFDM SYSTEM}

1A and 1B are block diagrams of a transmitter and a receiver of a typical OFDM system.

2 illustrates a general channel estimator using training symbols.

3 illustrates channel changes of an OFDM symbol occurring under a frequency selective fading channel.

4 is a block diagram of a channel estimating apparatus of an OFDM system according to the present invention.

5 and 6 are graphs showing the performance of the DDCE and the present invention according to the bandwidth.

7 and 8 are graphs showing the performance of the DDCE and the present invention according to the number of multipaths.

9 and 10 are graphs showing the performance of the DDCE and the present invention according to the transmission rate.

11 is a table summarizing the gains of the present invention and DDCE in BER 10 ^-4 .

<Explanation of symbols for the main parts of the drawings>

25a: Initial channel estimator 25b: First channel estimator

25c: second channel estimator

The present invention relates to an apparatus and method for channel estimation in an ultra-wideband wireless communication system. In particular, the present invention relates to a channel estimation value at a target frequency using a correlation between subcarriers of an orthogonal frequency division multiplex (OFDM) in a frequency selective fading channel. An apparatus and method for channel estimation in an OFDM system for estimating a channel by adapting the average of channel estimates.

Telematics is a compound word of communication and information technology that provides a new additional service that sends and receives information to a vehicle by using voice and data communication through a wireless network.

In order to provide telematics services, various wireless communication technologies, such as dedicated short range communication (DSRC), wireless LAN, and ultra wideband (UWB), have been proposed, and the dual UWB system is a connection between a terminal device in home networking and an in-vehicle telematics terminal device. Much research has been conducted to meet the needs of services and the need for high-speed data transmission in vehicles.

The UWB system is suitable for supplying a wireless PAN having a high data rate and defining a channel frequency bandwidth of 500 MHz or more in a spectrum mask of the 3.1-10.6 GHz band specified by the FCC.

Since the UWB system uses low power, the UWB system can provide a maximum transmission rate of about 1Gbps without affecting automobiles and other electronic products in the vehicle, and thus has various advantages as an in-vehicle wireless network application technology.

The OFDM method is a wireless access technology used in both a wireless LAN system and a UWB system and has a strong advantage against multipath interference causing performance degradation in a mobile reception environment.

In addition, in the mobile reception environment, effective channel estimation is needed to reduce the Doppler effect due to the multipath interference and mobility and to improve performance.

1 is a block diagram illustrating a general OFDM system, in which FIG. 1A illustrates a transmitter of an OFDM system and FIG. 1B illustrates a receiver of an OFDM system.

As shown in FIG. 1A, the transmitting side performs channel coding on the input data to be transmitted by the convolutional encoder 11, and then, through the interleaving unit 12, the modulator 13 through the quadrature phase shift keying (QPSK) scheme. Modulate.

After the modulation, a training symbol (Preamble) for channel estimation is inserted in the training symbol inserting unit (14), serial data is converted into parallel data through the S / P converter (15), and then Inverse Fast Fourie Transform. IFFT through the guard interval insertion unit 16 and a guard time is inserted to prevent inter symbol interference (ISI) due to multipath delay.

Thereafter, the digital signal is converted into an analog signal through the D / A converter 17 and transmitted to the receiving side through the channel through the antenna 18.

As shown in Fig. 1B, the receiving side processes data in the reverse order of the transmitting side.

That is, the antenna 21 receives the signal transmitted from the transmitting side, converts the received signal into a digital signal through the A / D converter 22, FFTs the FFT and the guard interval removing unit 23, and transmits the signal. Remove the protective gap inserted on the side.

Thereafter, the P / S converter 24 converts the parallel data into serial data, and the channel estimator 25 estimates the channel to recover the signal. After the training symbol is removed from the training symbol removing unit 25 and demodulated by the demodulation unit 26, the training symbol is decoded and output by the Viterbi decoder 28 via the deinterleaving unit 27.

2 shows a general channel estimator using training symbols in one subcarrier. It inserts the data that the receiver knows before the data information on the transmitting side (in the OFDM system, the packet system has a pilot symbol defined in the preamble structure, which is transmitted before the data symbol) so that they can change when passing through the channel. The initial channel estimator 25a applies the other information data Payload.

The equation for this is as shown in Equation 1 below.

(Equation 1)

R _k = hX _k + n _k

h _pilot = P /

는 수신된 파일럿 값,

는 파일럿에 의해 추정이 이루어진 심볼신호를 나타낸다. 이와 같이 파일럿에 의해 각 부반송파는 초기 추정값을 얻을 수 있게 된다. Here, R _k is a symbol signal received on a k-th subcarrier, h is a channel response, X _k is a transmitted k-th symbol signal, and n _k is a noise signal on a channel. P is the transmitted pilot value

Is the received pilot value,

Denotes a symbol signal estimated by a pilot. In this way, each subcarrier can obtain an initial estimate by the pilot.

In more detail, in general, OFDM consists of a symbol on a time side and a subcarrier on a frequency side, where a preamble refers to all subcarriers of an initial symbol in FIG. 3 to be described later, and each subcarrier of an initial preamble is of the same form as data. It consists of a pilot.

는 채널을 통과하여 수신된 프리앰블 심볼을 나타내며, 알고 있는 프리앰블의 각 부반송파의 파일럿값을 수신된 프리앰블의 각 부반송파의 파일럿값으로 나누고 이를 전체 부반송파의 수로 평균하여 채널 추정값 h_pilot을 얻게 되는 것이다.Therefore, p is a symbol of the transmitted preamble known to the receiving side,

Represents a preamble symbol received through the channel, and the pilot value of each subcarrier of the known preamble is divided by the pilot value of each subcarrier of the received preamble and averaged by the total number of subcarriers to obtain a channel estimate h _pilot .

On the other hand, a system using a wide bandwidth for high-speed transmission has a high correlation in time and low correlation between frequencies, resulting in severe frequency selective fading.

 FIG. 3 is a diagram illustrating a channel change of an OFDM symbol occurring under a frequency selective fading channel, and a portion represented by lines on the left and upper sides shows a channel change in frequency and time in a system having a wide bandwidth.

This is because T _o (Coherent Time) is much larger than T _s (Sampling Time) and becomes a highly correlated channel in terms of time, whereas B _o (Coherent Bandwidth) is much smaller than B _s (System Bandwidth). The correlation between the two subcarriers of the preamble, which is a training symbol, has a significant channel estimate.

That is, as described above, channel estimation in a conventional OFDM system is a channel estimation mainly used in a channel that changes slowly in time, and is a decision-directed channel estimation using inter-symbol data having high correlation. In the non-frequency environment, high gain is obtained, but as the system bandwidth increases, it becomes a serious frequency-selective fading environment, and the error flow phenomenon occurs when the reliability of the data decreases.

The present invention has been made to solve the above problems, and an object of the present invention is to estimate a channel using training symbols in an OFDM system, and to estimate a desired subcarrier as an average value of adjacent channel-side channel estimates according to frequency correlation. The present invention provides a channel estimation apparatus and method for an OFDM system so that a channel estimation value robust to a channel environment can be obtained by correcting a channel estimation value.

The channel estimation apparatus of the OFDM system according to the present invention for achieving the object of the present invention, the new channel estimation value in consideration of the channel estimation value of the subcarriers adjacent to each other and the weight of each subcarrier to the channel estimation value estimated by the preamble for each subcarrier A first channel estimator for obtaining the first channel estimation unit And a second channel estimator for adding a channel estimate of each subcarrier obtained by the first channel estimator, dividing it by the number of subcarriers, obtaining an average of channel estimates for each subcarrier, and multiplying the multiplication by the received OFDM information symbol. It features.

The first channel estimator has a channel estimate estimated by the preamble, h _k , k is an index of subcarriers, channel estimates of adjacent subcarriers are h _k-1 , h _{k + 1} , and a weight of each subcarrier is α, βh _k-1 when β A new channel estimate is obtained by + α _{h k} + β _{h k + 1} , and the weight is a relationship of β = (1-α) / 2.

In addition, the channel estimation method of the OFDM system of the present invention is characterized in that a new channel estimation value is obtained by considering channel estimation values of subcarriers adjacent to each other and the weight of each subcarrier with respect to the channel estimation value estimated by the preamble for each subcarrier.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the following examples are merely to illustrate the present invention is not limited to the contents of the present invention.

First, although the DDCE channel estimation technique shows a high gain in a non-frequency selective environment for the conventional channel estimator 25 as shown in FIG. 2, the system becomes a serious frequency selective paging environment as the system bandwidth increases. In order to compensate for the error flow when dropped, an OFDM system estimates a channel by applying a FE (Frequency Estimation) channel estimation technique, which has a strong characteristic even in a severe frequency selective paging environment.

4 illustrates a hardware implementation of such an algorithm, where k is an index of subcarriers and N is a total number of subcarriers.

)을 얻는 제1 채널 추정(Re Channel Estimation)부(25b), 상기 제1 채널 추정부(25b)에서 얻어진 각 부반송파들의 채널 추정값(

)을 가산하고 이를 부반송파의 수로 나누어 각 부반송파에 대한 채널 추정값들의 평균을 얻어 이를 수신한 OFDM 정보심볼에 곱하여 채널을 보상하는 제2 채널 추정부(25c)로 구성된다.As shown in FIG. 4, the present invention provides channel estimates h _{k-1 and} h of subcarriers adjacent to each other with respect to channel estimates h _k of the preamble estimated by the initial channel estimator 25a of FIG. 2 for each subcarrier. _{k + 1} ), and also considering the weighting factor (α, β, where β = (1-α) / 2) of each subcarrier, :

The first channel estimation unit 25b obtaining the channel estimation value of each of the subcarriers obtained by the first channel estimation unit 25b

), And dividing this by the number of subcarriers to obtain an average of channel estimates for each subcarrier and multiplying the received OFDM information symbol by a second channel estimator 25c.

)을 얻게 된다.According to the present invention configured as described above, the channel estimation value h _k-1 of the adjacent subcarriers is estimated by the first channel estimator 25b and the channel estimate h _k estimated by the initial channel estimator 25a of FIG. Considering the influence of, (h _{k + 1} ), we apply two weights (α, β)

)

)은

로 계산하여 얻게 된다.New channel estimate (

)silver

It is calculated by.

Here, the weights α and β vary depending on the relationship between adjacent subcarriers, that is, the correlation between frequencies (when the correlation between frequencies is high, the number of adjacent subcarriers is increased, the value of β is increased, and the frequency correlation is low. The number of adjacent subcarriers is reduced and the value of β is also reduced to increase the weight (α) of the channel estimate value to the original subcarrier. If this relationship is unknown, the average of the adjacent subcarriers is used.

Here, the use of the average of adjacent subcarriers means that if the correlation between frequencies is not known, the weight value of each subcarrier is the inverse of the subcarrier (normalization), and all the same weights, if all subcarriers, target subcarriers, and after subcarriers Using three of the neighboring channel estimates, etc., means that the weight is 0.33, and five means that the estimated value of adjacent subcarriers is added, such as 0.20, and the average value divided by the adjacent subcarriers (weighted value) is used.

)을 얻게 되면 제2 채널 추정부(25c)에서 각 부반송파들의 채널 추정값들을 모두 가산한 후, 부반송파의 수로 나누어 평균값을 얻으며, 이를 수신한 OFDM 정보심볼에 곱하여 해당 채널을 보상하게 된다.This new channel estimate (

), The second channel estimator 25c adds all channel estimates of each subcarrier, divides the number of subcarriers by the number of subcarriers, obtains an average value, and multiplies the received OFDM information symbol to compensate for the corresponding channel.

The following looks at the experimental example of the present invention.

First, the performance of the bandwidth is compared.

Referring to the change in the performance of the DDCE with respect to the bandwidth change of the system in Figure 5, the gain of the DDCE decreases as the bandwidth increases, resulting in the error flow of the DDCE due to fading of data at a bandwidth of 1056 MHz.

As the bandwidth of the system increases, the frequency selective tends to degrade the gain of the DDCE using the data due to fading of the data. At 264 MHz, the gain of DDCE is greater than at 528 MHz, but overall performance is degraded.

The reason for this is that T _s F _d that changes over time as so, the value of _{(Normalized Fading Factor, T s:} : Sampling Time, F d Maximum Doppler Frequency) larger by two times the increase in the amount of bandwidth to the channel of the temporally fast payout This degrades the overall performance.

However, as shown in FIG. 6, according to the FE according to the present invention, the performance according to the bandwidth of the system is not deteriorated in gain even when the bandwidth is increased and becomes frequency selective, whereas a slight gain is increased.

Therefore, at 1056 MHz, the performance of the DDCE is degraded while the performance of the FE is maintained. Therefore, the FE method is more useful in a channel environment with high frequency selective fading.

Next, compare the performance according to the number of multipaths.

7 illustrates the results of DDCE according to multipath in the case of 1056 MHz. Fading in the frequency domain is associated with delayed multipath. Accordingly, it can be seen that the gain of DDCE is further degraded as the number of multipaths is increased in FIG. 11 comparing the channel estimation technique gains in BER 10 ^-4 .

8 shows the performance of the FE in the environment as shown in FIG. In FIG. 11, unlike DDCE, even if the multipath is increased, the gain of the FE is not deteriorated but is robust. This is because the channel estimation value of the frequency side is obtained using only the preamble without using data.

Next, we compare the performance according to the transmission speed.

FIG. 9 shows performance according to transmission speed in the environment as shown in FIG. 7. As a result, an error flow phenomenon due to frequency selective fading appears. Transmission speeds of 220 Mbps, 110 Mbps, and 55 Mbps were implemented using time spread and frequency spread. In FIG. 11, the gain of DDCE deteriorates as the transmission speed increases. This is because the gain is degraded because spreading is not reliable.

Since the FE technique using the preamble is not influenced by data, it has a robust characteristic as shown in FIG. 10. Therefore, it can exhibit performance independent of spreading technique.

11 is a table summarizing the gains of FE and DDCE in BER 10 ^-4 . Since the DDCE uses the reliability of the data, an error flow phenomenon occurs due to fading of the data when the bandwidth is increased, and the gain deterioration can be seen by reducing the reliability of the data according to the multipath and the data rate. However, since the FE technique of the present invention is estimated with the correlation between frequencies using only the preamble, it can be seen that the FE technique operates independently in the above environment and has robust performance.

As described above, although described with reference to a preferred embodiment of the present invention, those skilled in the art various modifications of the present invention without departing from the spirit and scope of the invention described in the claims below Or it may be modified.

As described above, the present invention obtains a gain by adapting a channel estimate at a target frequency to an average of channel estimates of adjacent subcarriers by using correlation between OFDM subcarriers in a frequency selective fading channel, and using data as in DDCE. Because it uses the preamble instead of the data rate, it is independent of the transmission rate and the number of multipaths related to the reliability of the data, and thus can be gained even in a wide bandwidth system in which the error flow of the DDCE occurs, which is very useful in a severe frequency selective fading environment. There is this.

Claims (9)

In the channel estimation apparatus of an OFDM system, A first channel estimator configured to obtain a new channel estimate by considering channel estimates of adjacent subcarriers and weights of subcarriers adjacent to channel estimates estimated by the preamble for each subcarrier; And A second channel estimator for adding a channel estimate value of each subcarrier obtained by the first channel estimator, dividing it by the number of subcarriers, obtaining an average of channel estimates for each subcarrier, and multiplying the received channel information by the received OFDM information symbol to compensate a channel; Channel estimation apparatus of the OFDM system, characterized in that consisting of. The method of claim 1, wherein the first channel estimator When the channel estimate estimated by the preamble is h _k , k is an index of subcarriers, channel estimates of adjacent subcarriers are h _k-1 , h _{k + 1} , and the weight of each subcarrier is α, β, βh _{k -One} and a new channel estimate is obtained by + h _k + beta _{h k + 1} . The method of claim 2, wherein the weight is A channel estimating apparatus for an OFDM system, wherein β = (1-α) / 2. 4. The apparatus of claim 3, wherein the weight is applied differently according to the frequency correlation of adjacent subcarriers. In the channel estimation method of an OFDM system, And a new channel estimate is obtained by considering channel estimates of adjacent subcarriers and weights of each subcarrier with respect to channel estimates estimated by the preamble for each subcarrier. 6. The method of claim 5, wherein the channel estimate estimated by the preamble is h _k , k is an index of subcarriers, channel estimates of adjacent subcarriers are h _k-1 , h _{k + 1} , and the weight of each subcarrier is α. βh _k-1 A channel estimation method of an OFDM system, characterized by obtaining a new channel estimate by + αh _k + βh _{k + 1} . The method of claim 6, wherein the weight is A channel estimation method of an OFDM system, wherein β = (1-α) / 2. 8. The method of claim 7, wherein the weight is applied differently according to the frequency correlation of adjacent subcarriers. 6. The OFDM system as claimed in claim 5, wherein the new channel estimates of the subcarriers are added, divided by the number of subcarriers, the average of the channel estimates for each subcarrier is obtained, and multiplied by the received OFDM information symbols to compensate the channel. Channel estimation method.

Images