KR20110127362A - Apparatus and method for estimating doppler frequency of channel in a broadband wireless communication system - Google Patents

Abstract

PURPOSE: A device for estimating the doppler frequency of a channel in a broadband wireless communication system and a method thereof are provided to estimate spread delay or a Doppler frequency using an OFDM symbol. CONSTITUTION: An RF receiving unit(310) down-converts an RF band signal which is received through an antenna into a baseband signal. An OFDM symbol extracting unit(320) samples a baseband signal provided from the RF receiving unit to partition the baseband signal by an OFDM symbol unit. A PDP determining unit(330) determines delay spread of a channel using the sampled OFDM symbols. A Doppler estimating unit(340) estimates the Doppler frequency of a channel using the delay spread of the channel. The Doppler estimating unit includes an auto correlator(342), a power meter(344), a power compensator(346), and a reverse Bessel function operator(348).

Description

Apparatus and method for estimating the Doppler frequency of a channel in a broadband wireless communication system {APPARATUS AND METHOD FOR ESTIMATING DOPPLER FREQUENCY OF CHANNEL IN A BROADBAND WIRELESS COMMUNICATION SYSTEM}

The present invention relates to a broadband wireless communication system, and more particularly, to an apparatus and a method for estimating a Doppler frequency of a channel in a broadband wireless communication system.

In a wireless communication system, a base station and a terminal transmit and receive voice and data through a wireless channel. However, the terminal can move without communicating in a fixed position. Accordingly, the channel environment changes over time and the transmission path also changes. If the estimation of the channel considering the mobility of the terminal is not made correctly, demodulation of the received signal becomes difficult. In addition, since wireless communication systems use limited frequency and channel resources, it is important to allocate system resources through accurate channel estimation.

For accurate channel estimation, terminal speed information is required. In a time-varying channel such as a mobile communication system, speed information of a terminal works as an important factor. The speed information of the mobile terminal may be used for an adaptive receiver that adjusts coefficients such as channel tracker length, code rate, interleaver size, and the like. It is also important information for power control and handover processing.

In general, the speed estimation of the mobile terminal uses the characteristics of the Doppler shift. As the terminal moves, a Doppler shift of the received signal occurs. Doppler shift introduces frequency error, resulting in degradation of reception performance. This Doppler error occurs in proportion to the speed of the terminal. Therefore, an alternative for accurately estimating the Doppler frequency for the terminal should be presented.

Accordingly, an object of the present invention is to provide an apparatus and method for estimating Doppler frequency in a broadband wireless communication system.

Another object of the present invention is to provide an apparatus and method for estimating a spread delay in a broadband wireless communication system.

According to the first aspect of the present invention for achieving the above object, the Doppler frequency estimation method in a broadband wireless communication system, the autocorrelation value of the CP section and the data section corresponding to the original of the CP section is not affected by delay spread And calculating a power value, compensating power for the autocorrelation value, and determining a maximum Doppler frequency using the power-compensated autocorrelation value.

According to a second aspect of the present invention for achieving the above object, a spread delay estimation method in a broadband wireless communication system, the step of calculating the average correlation value for each sample between the CP interval and the original data of the CP interval, and for each sample Calculating difference values for average correlation values, calculating gain values of channel taps using the difference values, and extracting a valid gain value of at least one of the gain values. It features.

According to a third aspect of the present invention for achieving the above object, a wireless node device for estimating the Doppler frequency in a broadband wireless communication system, the CP interval and the data interval corresponding to the original of the CP interval is not affected by delay spread A correlator for calculating an autocorrelation value of a detector, a meter for calculating a power value of a CP section that is not affected by the delay spread, and a data section corresponding to an original of the CP section; Compensator and a calculator for determining the maximum Doppler frequency using the power-compensated autocorrelation value.

According to a fourth aspect of the present invention for achieving the above object, a wireless node device for estimating a spreading delay in a broadband wireless communication system, the first node for calculating the average correlation value for each sample between the CP interval and the original data of the CP interval A calculator, a difference calculator for calculating difference values for the average correlation values for each sample, a second calculator for calculating gain values of channel taps using the difference values, and an effective gain of at least one of the gain values And a comparator for extracting the value.

By estimating spreading delay and Doppler frequency using OFDM symbols in a broadband wireless communication system, the Doppler frequency can be effectively estimated without additional signals such as a preamble or a pilot.

1 is a diagram illustrating a signal region used for calculating an autocorrelation value and a power value in a broadband wireless communication system according to an embodiment of the present invention;
2 is a diagram illustrating an example of average correlation values for each sample between CP and original data in a wireless communication system according to an embodiment of the present invention;
3 is a block diagram of a wireless node in a wireless communication system according to an embodiment of the present invention;
4 is a block diagram of a PDP determination unit in a wireless communication system according to an embodiment of the present invention;
5 is a diagram illustrating a procedure of estimating Doppler frequency in a wireless communication system according to an embodiment of the present invention;
6 is a diagram illustrating a determination procedure of delay spread in a wireless communication system according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

Hereinafter, the present invention will be described a technique for estimating the Doppler frequency in a broadband wireless communication system. Hereinafter, the present invention will be described using an Orthogonal Frequency Division Multiplexing (OFDM) / Orthogonal Frequency Division Multiple Access (OFDMA) scheme as an example. The same applies to other wireless communication systems.

When scattering does not correlate with each other in a communication environment, and an angle of arrival (AOA) of a signal received at a terminal is equal, the channel has a characteristic as shown in Equation 1 below.

는 채널 계수의 자기상관값, 상기

는 채널 계수의 전력, 상기

는 0차 베셀 함수(Bessel function), 상기

는 자기상관길이, 상기

는 최대 도플러 주파수를 의미한다. In Equation 1,

Is the autocorrelation of the channel coefficient,

Is the power of the channel coefficient,

Is the 0th order Bessel function,

Is the autocorrelation length,

Is the maximum Doppler frequency.

As shown in Equation 1, the maximum Doppler frequency can be estimated by using the autocorrelation value of the channel coefficient and the zero-order Bessel function. Hereinafter, a process of estimating the Doppler frequency using the relationship as shown in Equation 1 will be described.

For convenience of description below, the present invention refers to a subject that estimates the Doppler frequency as a 'Doppler estimator'.

The Doppler estimator accepts an OFDM symbol consisting of CP and Cyclic Prefix. The Doppler estimator calculates autocorrelation values and power values of a CP section not affected by delay spread and a data section corresponding to the original of the CP section. Accordingly, the Doppler estimator needs delay spread information of the channel, and the determination of the delay spread information will be described later. 1 shows a signal region used for calculating the autocorrelation value and the power value. As shown in FIG. 1, the original data of the remainder 104 of the entire CP section 100 except for the front part 102 affected by delay spread and the rest 104 of the entire data section 150 are included. An interval 154 is used to calculate the autocorrelation value and the power value. For example, the autocorrelation value may be calculated as in Equation 2 below, and the power value may be calculated as in Equation 3 below.

는 i번째 수신한 OFDM 심벌 실수 성분의 자기상관값, 상기

는 i번째 수신한 OFDM 심벌 허수 성분의 자기상관값, 상기

는 수신 OFDM 심벌의 실수 성분, 상기

는 수신 OFDM 심벌의 허수 성분, 상기

는 샘플 시간, 상기

은 OFDM 심벌 내에서 데이터 구간의 길이, 상기

은 OFDM 심벌 내에서 CP 구간의 길이, 상기

는 채널에 의한 지연 확산 길이를 의미한다.In Equation 2,

Is an autocorrelation value of the i-th received OFDM symbol real component,

Is an autocorrelation value of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.

Referring to Equation 2, the autocorrelation value is calculated separately for the real component and the imaginary component, and is determined by multiplying and averaging received values of samples at positions corresponding to each other in the CP section and the original data section. do. Accordingly, one autocorrelation value of an imaginary component and one autocorrelation value of a real component are calculated per one OFDM symbol.

는 i번째 수신한 OFDM 심벌 실수 성분의 전력, 상기

는 i번째 수신한 OFDM 심벌 허수 성분의 전력, 상기

는 수신 OFDM 심벌의 실수 성분, 상기

는 수신 OFDM 심벌의 허수 성분, 상기

는 샘플 시간, 상기

은 OFDM 심벌 내에서 데이터 구간의 길이, 상기

은 OFDM 심벌 내에서 CP 구간의 길이, 상기

는 채널에 의한 지연 확산 길이를 의미한다.In Equation 3,

Is the power of the i-th received OFDM symbol real component,

Is the power of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.

Referring to Equation 3, the power value is calculated separately for the real component and the imaginary component, and after calculating the average of the squares of the received values of the samples in the CP interval and the original data interval, respectively, And multiplying averages of squares calculated from samples corresponding to each other in the original data interval and calculating a square root. Accordingly, one power value of an imaginary component and one power value of a real component are calculated per one OFDM symbol.

The autocorrelation values and power values of the real and imaginary components calculated using the received OFDM symbols are used for power compensation. That is, the Doppler estimator divides the autocorrelation value by the power value to compensate for the power so that the maximum power is 1, and calculates an average value of autocorrelation values of real components and autocorrelation values of imaginary components. For example, the power compensated autocorrelation value is calculated as in Equation 4 below.

는 i번째 수신한 OFDM 심벌의 전력 보상된 자기상관값, 상기

는 i번째 수신한 OFDM 심벌 실수 성분의 자기상관값, 상기

는 i번째 수신한 OFDM 심벌 허수 성분의 자기상관값, 상기

는 i번째 수신한 OFDM 심벌 실수 성분의 전력, 상기

는 i번째 수신한 OFDM 심벌 허수 성분의 전력을 의미한다.In Equation 4,

Is the power-compensated autocorrelation value of the i-th received OFDM symbol,

Is an autocorrelation value of the i-th received OFDM symbol real component,

Is an autocorrelation value of the i-th received OFDM symbol imaginary component,

Is the power of the i-th received OFDM symbol real component,

Denotes the power of the i-th received OFDM symbol imaginary component.

Referring to Equation 4, the power-compensated autocorrelation value is determined by dividing the autocorrelation value of each of the real component and the imaginary component by the power value and averaging the division result. In this case, in order to increase the reliability of the power compensated autocorrelation value measured in OFDM symbol units, the Doppler estimator uses an average autocorrelation value of a plurality of OFDM symbols. For example, an average autocorrelation value of the plurality of OFDM symbols is calculated as in Equation 5 below.

는 다수의 OFDM 심벌들에 대한 평균 자기상관값, 상기

는 평균화되는 다수의 OFDM 심벌들의 개수, 상기

는 i번째 수신한 OFDM 심벌의 전력 보상된 실수 성분의 자기상관값 및 전력 보상된 허수 성분의 자기상관값을 의미한다.In Equation 5,

Is an average autocorrelation value for a plurality of OFDM symbols,

Is the number of OFDM symbols to be averaged, where

Denotes the autocorrelation value of the power compensated real component and the power compensated imaginary component of the i-th received OFDM symbol.

As shown in Equation 1, the autocorrelation value of the received signal is expressed as a zero-order Bessel function with the maximum Doppler frequency as an input. Therefore, by taking the inverse Bessel function on the power compensated autocorrelation value, the maximum Doppler frequency is determined. For example, the maximum Doppler frequency is determined as in Equation 6 below.

는 최대 도플러 주파수, 상기

는 역베셀 함수, 상기

는 다수의 OFDM 심벌들에 대한 자기상관값들의 평균값, 상기

는 샘플 시간, 상기

은 OFDM 심벌 내에서 데이터 구간의 길이를 의미한다.In Equation 6,

Is the maximum Doppler frequency, said

Is the inverse Bessel function,

Is an average of autocorrelation values for a plurality of OFDM symbols,

Is the sample time,

Denotes the length of the data interval in the OFDM symbol.

Referring to Equation 6, the maximum Doppler frequency is determined by dividing the value obtained by the inverse Bessel function with respect to the power compensated autocorrelation value by the product of FFT size, sample time and 2π.

When the inverse Bessel function is implemented for the operation of Equation 6, a high amount of calculation is required. Therefore, in addition to the method of directly implementing the inverse Bessel function, a reverse Bessel look-up table may be used. In this case, a range of the maximum Doppler frequency that can be estimated is determined according to the size or resolution of the inverse Bessel lookup table. Therefore, the size or resolution of the inverse vessel lookup table is preferably adjusted according to the characteristics of the system in which the present invention is implemented.

As described above, the Doppler frequency can be measured. At this time, delay spread information is required. The delay spread information may be determined through various methods. According to one embodiment of the present invention, the delay spread information may be determined as follows.

For convenience of description below, the present invention refers to a subject that determines the delay spread information as a 'power delay profile (PDP) determiner'.

The PDP determiner calculates an average correlation value between a CP section and the original data of the CP in a normalized form. In this case, the average correlation value is calculated using a plurality of OFDM symbols. For example, the normalized mean correlation value is calculated as in Equation 7 below.

는 평균 연산자, 상기

는 k번째 샘플의 평균 상관값, 상기

는 k번째 샘플의 수신값,

은 OFDM 심벌 내에서 데이터 구간의 길이, 즉, FFT사이즈, 상기

은 OFDM 심벌 내에서 CP 구간의 길이, 상기

은 OFDM 심벌의 인덱스를 의미한다.In Equation 7,

Is the mean operator,

Is the mean correlation value of the k th sample,

Is the received value of the k th sample,

Is the length of the data interval in the OFDM symbol, i.e., the FFT size,

Is the length of the CP interval in the OFDM symbol,

Denotes an index of an OFDM symbol.

Referring to Equation 7, the average correlation value is a first value obtained by accumulating average correlation values of received values of samples corresponding to each other in a CP interval and a data interval, and samples of the CP interval. Calculates a second value accumulated over squares of a received value over a plurality of OFDM symbols and a third value accumulated over squares of a received value of a sample of a data interval over a plurality of OFDM symbols, and then calculates a third value; Is divided by the product of the square root of the second value and the square of the third value. Accordingly, average correlation values of each of the samples included in the CP are determined through the plurality of OFDM symbols.

In the absence of the effect of symbol interference and noise by the channel, the normalized mean correlation between the CP interval and the original data is 1 for all samples. However, in a general channel, since an ISI (Inter Symbol Interference) occurs due to a time delay of a channel, an average correlation value between the CP period and the original data decreases in proportion to the size of the ISI.

2 illustrates average correlation values for each sample between CP and original data when ISI occurs. In the graph of FIG. 2, the right section is a free section from the ISI, and it is confirmed that the channel tap causing ISI gradually increases toward the left side of the graph, and the average correlation value monotonously decreases. When the average correlation value calculated as described above is expressed by the gain of the channel tap, Equation 8 is obtained.

는 평균 연산자, 상기

는 평균 상관값, 상기

은 잡음의 분산, 상기

는 i번째 채널 탭의 시간 지연, 상기

는 i번째 채널 탭의 이득, 상기

는 k번째 샘플의 송신값, 상기

은 채널 탭의 총 개수를 의미한다. 여기서, 1번째 내지 N-n번째의 채널 탭들은 ISI를 발생시키지 않는 채널 탭들, N-n+1번째 내지 N번째의 채널 탭들은 ISI를 발생시키는 채널 탭들이다.In Equation 8,

Is the mean operator,

Is the mean correlation value,

Is the variance of noise,

Is the time delay of the i th channel tap,

Is the gain of the i th channel tap,

Is the transmission value of the k th sample,

Means the total number of channel taps. Here, the first through Nn channel taps are channel taps that do not generate ISI, and the N-n + 1 through Nth channel taps are channel taps that generate ISI.

The PDP determiner allows the maximum value of the average correlation value to be 1 to eliminate power reduction of the channel tap due to noise. When each sample of the OFDM symbol has a mutually independent characteristic, each mean correlation value may be approximated as shown in Equation 9 below.

는 평균 연산자, 상기

는 평균 상관값, 상기

은 잡음 전력, 상기

는 i번째 채널 탭의 시간 지연, 상기

는 i번째 채널 탭의 이득, 상기

는 k번째 샘플의 송신값, 상기

은 채널 탭의 총 개수를 의미한다. In Equation 9,

Is the average operator,

Is the mean correlation value,

Is noise power, said

Is the time delay of the i th channel tap,

Is the gain of the i th channel tap,

Is the transmission value of the k th sample,

Means the total number of channel taps.

According to the relationship between the average correlation value and the channel tap gain shown in Equation 9, gain values of each channel tap causing ISI can be expressed by using the average correlation change amount of adjacent samples as shown in Equation 10 below. Can be.

는 평균 연산자, 상기

는 k번째 샘플의 평균 상관값, 상기

는 i번째 채널 탭의 이득, 상기

은 채널 탭의 총 개수, 상기

은 모든 채널 탭들의 전력의 합을 의미한다.In Equation 10,

Is the mean operator,

Is the mean correlation value of the k th sample,

Is the gain of the i th channel tap,

Is the total number of channel taps,

Is the sum of the powers of all the channel taps.

According to Equation 10, the gain of the channel tap may be calculated using the average of the difference between the average correlation values of adjacent samples and the power of the channel. Accordingly, the PDP determiner calculates each average correlation value, calculates difference values between adjacent average correlation values, and calculates a gain of each channel tap by using the sum of the difference values and the power of the channel taps. do. In this case, the average correlation value may be calculated as in Equation 7 or may be calculated as in Equation 11 to reduce the amount of computation.

는 k번째 샘플의 n번째 순시 상관값까지 반영된 평균 상관값, 상기

는 평균화 상수,

는 k번째 샘플의 n번째 순시 상관값을 의미한다.In Equation 11,

Is an average correlation value reflected up to the n th instantaneous correlation value of the k th sample,

Is the averaging constant,

Denotes the n th instantaneous correlation value of the k th sample.

The value obtained through the above process has a form similar to that of a PDP of a channel. However, large and small errors are added to the PDP of the channel by the correlation characteristics and noise of the OFDM symbol. In order to minimize the error, the PDP determiner extracts only values greater than a certain level among the gain values of the obtained channel taps. The large value is extracted by comparison with a preset threshold value, and the threshold value is set using a maximum value among the calculated gain values of the channel taps. The error on the PDP is related to the number of iteration operations to calculate the noise and average correlation values at the receiver. Therefore, the threshold value may be set as in Equation 12 below.

는 문턱값, 상기

는 스케일링 상수, 상기

는 채널 탭들의 이득값들 중 최대값, 상기

은 신호대 잡음비, 상기

는 상관값 평균화를 위한 반복 연산의 횟수, 즉, 평균 상관값에 이용된 OFDM 심벌의 개수를 의미한다. 여기서, 상기

를 크게 설정할수록 문턱값이 커져 잡음에 의한 추정 오차가 적어지나, 동시에, 원래의 채널 탭 중 문턱값에 미치지 못하는 채널 탭들이 무시될 수 있다. 따라서, 상기

는 시스템 특성에 따라 적절한 값으로 사용됨이 바람직하다.
In Equation 12,

Is the threshold, said

Is the scaling constant,

Is the maximum of the gain values of the channel taps,

Is the signal-to-noise ratio,

Denotes the number of repetition operations for averaging correlation values, that is, the number of OFDM symbols used for the average correlation value. Where

The larger the value is set, the larger the threshold value is, thereby reducing the estimation error due to noise. At the same time, channel taps which do not reach the threshold value among the original channel taps may be ignored. Thus, the above

Is preferably used according to system characteristics.

Hereinafter, the configuration and operation of a wireless node estimating the Doppler frequency as described above will be described with reference to the drawings.

3 is a block diagram of a wireless node in a wireless communication system according to an exemplary embodiment of the present invention.

As shown in FIG. 3, the wireless node includes a radio frequency (RF) receiver 310, an OFDM symbol extractor 320, a PDP determiner 330, and a Doppler estimate 340.

The RF receiver 310 downconverts and amplifies an RF band signal received through an antenna into a baseband signal. The OFDM symbol extractor 320 samples the baseband signal provided from the RF receiver 310 and divides the baseband signal in units of ODFM symbols to provide the PDP decision unit 330 and the Doppler estimation unit 340. The PDP determiner 330 determines the delay spread of the channel using the sampled OFDM symbols. An example configuration of the PDP determination unit 330 will be described below with reference to FIG. 4. The Doppler estimation unit 340 estimates the Doppler frequency of the channel using the sampled OFDM symbols and the delay spread of the channel determined by the PDP determiner 330. In this case, the Doppler estimation unit 340 estimates the Doppler frequency using the relationship between the autocorrelation value of the channel coefficient, the zero-order Bessel function, and the maximum Doppler frequency, as shown in Equation (1).

In detail, the Doppler estimation unit 340 includes an autocorrelator 342, a power meter 344, a power compensator 346, and an inverse Bessel function operator 348.

The autocorrelator 342 identifies the CP section unaffected by the delay spread using the delay spread information, and determines the CP section and the data section corresponding to the original of the CP section unaffected by the delay spread. Calculate the autocorrelation value. Here, the autocorrelation value is calculated separately for the real component and the imaginary component, and is determined by multiplying and averaging received values of samples at positions corresponding to each other in the CP section and the original data section. Accordingly, one autocorrelation value of an imaginary component and one autocorrelation value of a real component are calculated per one OFDM symbol. For example, the autocorrelation value is determined as in Equation 2 above.

The power meter 344 identifies the CP section that is not affected by the delay spread by using the delay spread information, and determines the CP section that is not affected by the delay spread and the data section corresponding to the original of the CP section. Calculate the power value. Herein, the power value is calculated separately for the real component and the imaginary component, and after calculating an average of the squares of the received values of the samples in the CP interval and the original data interval, respectively, and corresponding to each other in the CP interval and the original data interval It is determined by multiplying the mean of squares calculated from the samples and calculating the square root. Accordingly, one power value of an imaginary component and one power value of a real component are calculated per one OFDM symbol. For example, the power value is determined as in Equation 3 above.

The power compensator 346 compensates the power so that the maximum power becomes 1 by dividing the autocorrelation value determined by the autocorrelator 342 by the power value determined by the power meter 344. The power compensator 346 calculates an average value of the autocorrelation value of the real component and the autocorrelation value of the imaginary component. That is, the power compensator 346 divides the autocorrelation value of each of the real component and the imaginary component by the power value, and determines the power-compensated autocorrelation value by averaging the division result. For example, the power compensated autocorrelation value is determined as in Equation 4 above. In addition, the power compensator 346 provides an average value of the autocorrelation value to the inverse Bessel function operator 348. In this case, in order to increase the reliability of the average value of the autocorrelation value measured in OFDM symbol units, an average autocorrelation value of a plurality of OFDM symbols may be used. In this case, the power compensator 346 determines the average autocorrelation value as shown in Equation 5, and provides the average autocorrelation value to the inverse Bessel function operator 348.

The inverse Bessel function operator 348 estimates the maximum Doppler frequency using the autocorrelation value of the channel provided from the power compensator 346. That is, the inverse Bessel function operator 348 determines the maximum Doppler frequency by taking an inverse Bessel function on the autocorrelation value of the channel. Specifically, the maximum Doppler frequency is determined by dividing the value of the inverse Bessel function with respect to the power compensated autocorrelation value by the product of FFT size, sample time and 2π. For example, the inverse Bessel function operator 348 estimates the maximum Doppler frequency as shown in Equation 6. In this case, the inverse Bessel function operator 348 uses a calculation algorithm that directly implements the inverse Bessel function or uses an inverse Bessel lookup table to perform the operation of the inverse Bessel function.

4 is a block diagram of the PDP determination unit 330 in a wireless communication system according to an embodiment of the present invention.

As shown in FIG. 4, the PDP determination unit 330 includes an average correlation value calculator 432, a difference 434, a gain value calculator 436, and a comparator 438. The PDP determiner 330 determines a delay spread of a channel through a correlation operation on a time axis using a CP included in all OFDM symbols. A detailed process is as follows.

The average correlation value calculator 432 calculates an average correlation value for each sample between the CP section and the original data of the CP section using the OFDM symbols provided from the OFDM symbol extractor 320. For example, the mean correlation value may be determined as in Equation 7 or Equation 11. According to Equation (7), the average correlation value is a first value obtained by accumulating a correlation value of received values of samples corresponding to each other in a CP interval and a data interval over a plurality of OFDM symbols, and a sample of a CP interval. A second value is obtained by accumulating the square of the received value over the plurality of OFDM symbols and a third value obtained by accumulating the square of the received value of the sample of the data interval over the plurality of OFDM symbols, and then calculates the first value. It is determined by dividing by the product of the square root of the second value and the square of the third value. According to Equation 11, the average correlation value is determined by summing the product of the recently calculated instantaneous correlation value and the weight with the previously calculated average correlation value. Accordingly, average correlation values of each of the samples included in the CP are determined through the plurality of OFDM symbols.

The difference unit 434 calculates difference values for the sample-specific mean correlation values determined by the mean correlation value calculator 432. In other words, the difference 434 calculates difference values between mean correlation values of adjacent samples.

The gain calculator 436 calculates gain values of the channel taps using the difference values of the average correlation values calculated by the difference unit 434. To this end, the gain calculator 436 measures the sum of the powers of all the channel taps, multiplies the sum of the powers of all the channel taps by the difference value of a particular sample pair, and the power of all the channel taps. The gain value of the tap corresponding to the position of one sample of the sample pair is calculated by calculating the square root of the sum of and the product of the difference values. For example, the gain value calculator 436 calculates gain values of each tap by using the relationship of Equation 10.

The comparator 438 removes a noise component among the gain values of the channel taps calculated by the gain value calculator 436 and extracts at least one valid gain value. To this end, the comparator 438 sets a threshold for removing the noise component and excludes gain values smaller than the threshold. For example, the comparator 438 determines the threshold value using a number indicating the maximum gain value and the channel quality and the number of OFDM symbols used for the average correlation value. For example, the threshold value may be determined as in Equation 12.

5 illustrates a procedure of estimating Doppler frequency in a wireless communication system according to an embodiment of the present invention. The operation of FIG. 5 is performed by a wireless node that transmits and receives a wireless signal. For example, the wireless node may be a base station or a terminal.

Referring to FIG. 5, in step 501, the wireless node identifies a CP section not affected by the delay spread using the delay spread information, and determines the CP section and the CP section not affected by the delay spread. The autocorrelation value of the data section corresponding to the original is calculated. Here, the autocorrelation value is calculated separately for the real component and the imaginary component, and is determined by multiplying and averaging received values of samples at positions corresponding to each other in the CP section and the original data section. Accordingly, one autocorrelation value of an imaginary component and one autocorrelation value of a real component are calculated per one OFDM symbol. For example, the autocorrelation value is determined as in Equation 2 above. Although not shown in FIG. 5, prior to calculating the autocorrelation value, the wireless node may perform an operation of converting an RF signal received through an antenna into a baseband signal and sampling the baseband signal. .

In step 503, the wireless node identifies a CP section not affected by the delay spread using the delay spread information, and corresponds to the CP section and the original of the CP section not affected by the delay spread. The power value of the data section is calculated. Herein, the power value is calculated separately for the real component and the imaginary component, and after calculating an average of the squares of the received values of the samples in the CP interval and the original data interval, respectively, and corresponding to each other in the CP interval and the original data interval It is determined by multiplying the mean of squares calculated from the samples and calculating the square root. Accordingly, one power value of an imaginary component and one power value of a real component are calculated per one OFDM symbol. For example, the power value is determined as in Equation 3 above.

After calculating the power value, the wireless node proceeds to step 505 to compensate the power so that the maximum power becomes 1 by dividing the autocorrelation value determined in step 501 by the power value determined in step 503. The wireless node averages the autocorrelation values of the real component and the imaginary component. That is, the wireless node determines the power-compensated autocorrelation value by dividing the autocorrelation value of each of the real component and the imaginary component by the power value, and averaging the division result. For example, the power compensated autocorrelation value is determined as in Equation 4 above. In this case, in order to increase the reliability of the autocorrelation value measured in OFDM symbol units, an average autocorrelation value of a plurality of OFDM symbols may be used. In this case, the wireless node determines the average autocorrelation value as shown in Equation (5).

In step 507, the wireless node estimates the maximum Doppler frequency using the autocorrelation value of the channel determined in step 505. That is, the wireless node determines the maximum Doppler frequency by taking an inverse Bessel function on the autocorrelation of the channel. Specifically, the maximum Doppler frequency is determined by dividing the value of the inverse Bessel function with respect to the power compensated autocorrelation value by the product of FFT size, sample time and 2π. For example, the wireless node estimates the maximum Doppler frequency as shown in Equation 6. In this case, the inverse Bessel function operator 348 uses a calculation algorithm that directly implements the inverse Bessel function or uses an inverse Bessel lookup table to perform the operation of the inverse Bessel function.

6 is a flowchart illustrating a determination of delay spread in a wireless communication system according to an exemplary embodiment of the present invention. 6 is performed by a wireless node that transmits and receives a wireless signal. For example, the wireless node may be a base station or a terminal.

Referring to FIG. 6, in step 601, the wireless node calculates a correlation value for each sample between a CP section and original data of the CP section using sampled OFDM symbols.

In step 603, the wireless node calculates an average correlation value obtained by averaging the correlation values for a plurality of OFDM symbols. For example, the mean correlation value may be determined as in Equation 7 or Equation 11. According to Equation (7), the average correlation value is a first value obtained by accumulating a correlation value of received values of samples corresponding to each other in a CP interval and a data interval over a plurality of OFDM symbols, and a sample of a CP interval. A second value is obtained by accumulating the square of the received value over the plurality of OFDM symbols and a third value obtained by accumulating the square of the received value of the sample of the data interval over the plurality of OFDM symbols, and then calculates the first value. It is determined by dividing by the product of the square root of the second value and the square of the third value. According to Equation 11, the average correlation value is determined by summing the product of the recently calculated instantaneous correlation value and the weight with the previously calculated average correlation value. Accordingly, average correlation values of each of the samples included in the CP are determined through the plurality of OFDM symbols.

After calculating the average correlation value, the wireless node proceeds to step 605 to calculate gain values of channel taps causing ISI. That is, the wireless node calculates difference values between average correlation values of adjacent samples, and calculates gain values of channel taps using the difference values. In detail, the wireless node measures the sum of the powers of all the channel taps, multiplies the sum of the powers of all the channel taps by a difference value of a specific sample pair, sums the power of all the channel taps and the By calculating the square root of the product of the difference values, the gain value of the tap corresponding to the position of one of the sample pairs is calculated. For example, the gain value calculator 436 calculates gain values of each channel tap by using the relationship of Equation 10.

After calculating the gain values, the wireless node proceeds to step 607 and extracts at least one valid gain value by comparing with a threshold value. In other words, the wireless node removes a noise component among the gain values of the channel taps and extracts at least one valid gain value. To this end, the wireless node sets a threshold for removing the noise component and excludes gain values smaller than the threshold. For example, the wireless node determines the threshold value using a number indicating the maximum gain value and the channel quality and the number of OFDM symbols used for the average correlation value. For example, the threshold value may be determined as in Equation 12.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the scope of the following claims, but also by the equivalents of the claims.

Claims (36)

In the Doppler frequency estimation method in a broadband wireless communication system,
Calculating an autocorrelation value and a power value of a CP section that is not affected by delay spread and a data section corresponding to the original of the CP section;
Compensating power for the autocorrelation value;
Determining a maximum Doppler frequency using the power compensated autocorrelation value.
The method of claim 1,
The autocorrelation value is calculated separately for a real component and an imaginary component, and is determined by multiplying and averaging received values of samples of positions corresponding to each other in the CP section and the original data section.
The method of claim 2,
The autocorrelation value is characterized by the following formula,

Where

Is an autocorrelation value of the i-th received OFDM symbol real component,

Is an autocorrelation value of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.
The method of claim 1,
The power value is calculated separately for the real component and the imaginary component, and after calculating an average of the squares of the received values of the samples in the CP section and the original data section, respectively, and corresponding to each other in the CP section and the original data section. And multiplying the mean of squares calculated from the samples to be calculated and calculating the square root.
The method of claim 4, wherein
The power value is characterized in that determined by the following formula,

Where

Is the power of the i-th received OFDM symbol real component,

Is the power of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.
The method of claim 1,
Compensating the power for the autocorrelation value,
Dividing the autocorrelation values of the real and imaginary components by power;
Determining a power compensated autocorrelation value by averaging the result of the division.
The method of claim 6,
Wherein the power compensated autocorrelation value is an average autocorrelation value for a plurality of OFDM symbols.
The method of claim 1,
Determining the maximum Doppler frequency,
And taking an inverse Bessel function on the power compensated autocorrelation value.
The method of claim 8,
The operation of the inverse Bessel function, characterized in that performed via a calculation algorithm or a reverse Bessel lookup table that implements the inverse Bessel function directly.
10. The method of claim 9,
The maximum Doppler frequency is characterized by the following formula,

Where

Is the maximum Doppler frequency, said

Is the inverse Bessel function,

Is an average of autocorrelation values for a plurality of OFDM symbols,

Is the sample time,

Is the length of the data interval in the OFDM symbol.
The method of claim 1,
Calculating an average correlation value for each sample between the CP section and the original data of the CP section;
Calculating difference values of the mean correlation values for each sample;
Calculating gains of channel taps using the difference values;
And extracting at least one valid gain value of the gain values.
A spread delay estimation method in a broadband wireless communication system,
Calculating a mean correlation value for each sample between the CP section and the original data of the CP section;
Calculating difference values of the mean correlation values for each sample;
Calculating gains of channel taps using the difference values;
Extracting at least one valid gain value of the gain values.
The method of claim 12,
The average correlation value is a first value obtained by accumulating a correlation value of received values of samples corresponding to each other in a CP interval and a data interval over a plurality of OFDM symbols, and a square of the received value of a sample of the CP interval in a plurality of OFDM symbols. Calculating a third value accumulated over a plurality of OFDM symbols by squaring a square of a second value accumulated over a field and a sample of a data interval, and then calculating the first value by the square root of the second value and the first value. Characterized by dividing by the product of the square of three values.
The method of claim 12,
Wherein said average correlation value is determined by summing a product of a recently calculated instantaneous correlation value and a weight to a previous average correlation value.
The method of claim 12,
The process of calculating the gain values,
Measuring the sum of power of all channel taps,
Multiplying the sum of the powers of all the channel taps by a difference value of a specific sample pair;
Calculating a gain value of the tap corresponding to the position of one sample of the pair of samples by calculating a square root of the sum of the powers of all the channel taps and the product of the difference values.
The method of claim 12,
Extracting the at least one valid gain value,
Setting the threshold,
And excluding at least one gain value less than the threshold value.
The method of claim 16,
The threshold value is determined using a maximum gain value, a value indicating channel quality, and the number of OFDM symbols used for the average correlation value.
The method of claim 17,
The threshold value is characterized in that the method is determined by the following formula,

Where

Is the threshold, said

Is the scaling constant,

Is the maximum of the gain values of the channel taps,

Is the signal-to-noise ratio,

Denotes the number of repetitive operations for averaging correlation values, that is, the number of OFDM symbols used for the average correlation value.
A wireless node apparatus for estimating Doppler frequency in a broadband wireless communication system,
A correlator for calculating an autocorrelation value of a CP section not affected by delay spread and a data section corresponding to the original of the CP section;
A measuring unit for calculating a power value of a CP section not affected by the delay spread and a data section corresponding to the original of the CP section;
A compensator for compensating power for the autocorrelation value;
And an operator for determining the maximum Doppler frequency using the power compensated autocorrelation value.
20. The method of claim 19,
The autocorrelation value is calculated separately for a real component and an imaginary component, and is determined by multiplying and averaging received values of samples of positions corresponding to each other in the CP section and the original data section.
The method of claim 20,
The autocorrelation value is characterized in that the device is characterized by the following formula,

Where

Is an autocorrelation value of the i-th received OFDM symbol real component,

Is an autocorrelation value of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.
20. The method of claim 19,
The power value is calculated separately for the real component and the imaginary component, and after calculating an average of the squares of the received values of the samples in the CP section and the original data section, respectively, and corresponding to each other in the CP section and the original data section. And multiply the averages of the squares calculated from the samples to be calculated and calculate the square root.
The method of claim 22,
The device is characterized in that the power value is determined as in the following formula,

Where

Is the power of the i-th received OFDM symbol real component,

Is the power of the i-th received OFDM symbol imaginary component,

Is a real component of the received OFDM symbol,

Is an imaginary component of the received OFDM symbol,

Is the sample time,

Is the length of the data interval within the OFDM symbol,

Is the length of the CP interval in the OFDM symbol,

Denotes the delay spread length by the channel.
20. The method of claim 19,
And the compensator divides the autocorrelation value of each of the real component and the imaginary component by the power value, and determines the power compensated autocorrelation value by averaging the result of the division.
25. The method of claim 24,
And the power compensated autocorrelation value is an average autocorrelation value for a plurality of OFDM symbols.
20. The method of claim 19,
And the operator takes an inverse Bessel function on the power compensated autocorrelation value.
The method of claim 26,
The operation of the inverse Bessel function, characterized in that performed via a calculation algorithm or a reverse Bessel lookup table that implements the inverse Bessel function directly.
The method of claim 27,
The maximum Doppler frequency is characterized in that the device is characterized by the following equation,

Where

Is the maximum Doppler frequency, said

Is the inverse Bessel function,

Is an average of autocorrelation values for a plurality of OFDM symbols,

Is the sample time,

Is the length of the data interval in the OFDM symbol.
20. The method of claim 19,
A first calculator configured to calculate an average correlation value for each sample between the CP section and the original data of the CP section;
A difference calculator for calculating difference values of the mean correlation values for each sample;
A second calculator for calculating gains of channel taps using the difference values;
And a comparator for extracting at least one valid gain value of said gain values.
A wireless node apparatus for estimating spreading delay in a broadband wireless communication system,
A first calculator for calculating an average correlation value for each sample between the CP section and the original data of the CP section;
A difference calculator for calculating difference values of the mean correlation values for each sample;
A second calculator for calculating gains of channel taps using the difference values;
And a comparator for extracting at least one valid gain value of the gain values.
The method of claim 30,
The average correlation value is a first value obtained by accumulating a correlation value of received values of samples corresponding to each other in a CP interval and a data interval over a plurality of OFDM symbols, and a square of the received value of a sample of the CP interval in a plurality of OFDM symbols. Calculating a third value accumulated over a plurality of OFDM symbols by squaring a square of a second value accumulated over a field and a sample of a data interval, and then calculating the first value by the square root of the second value and the first value. Device determined by dividing by the product of the square of three values.
The method of claim 30,
And said average correlation value is determined by summing a product of a weighted correlation value and weight recently calculated.
The method of claim 30,
The second calculator measures the sum of the powers of all the channel taps, multiplies the sum of the powers of all the channel taps by a difference value of a specific sample pair, and then sums the powers of the all channel taps and the difference. And calculating a gain value of the tap corresponding to the position of one of the sample pairs by calculating the square root of the product of the values.
The method of claim 30,
Wherein the comparator sets a threshold and excludes at least one gain less than the threshold.
The method of claim 34, wherein
The threshold value is determined by using the maximum gain value, the channel quality value and the average number of OFDM symbols used in the correlation value.
36. The method of claim 35,
The threshold value is characterized in that the device is determined by the following formula,

Where

Is the threshold, said

Is the scaling constant,

Is the maximum of the gain values of the channel taps,

Is the signal-to-noise ratio,

Denotes the number of repetitive operations for averaging correlation values, that is, the number of OFDM symbols used for the average correlation value.

Images