KR100733596B1 - System and method for estimating channel state information used in adaptive wireless transceiver - Google Patents

Abstract

 The present invention relates to an apparatus for estimating a moment in a channel spectrum representing statistical characteristics of a channel in a wireless communication system. The moment of the channel spectrum may provide information about the Doppler spectrum of the channel indicating the degree of channel change in the time domain and the distribution of the power delay spectrum of the channel indicating the degree of change of the channel in the frequency domain. We propose a technique that can be estimated using normalized autocorrelation characteristics.

An apparatus for estimating channel characteristics according to the present invention is an apparatus for estimating characteristics of a channel for adaptive transmission and reception, and includes: an autocorrelation unit processing a received signal to obtain a plurality of normalized autocorrelation values; A plurality of autocorrelation values are values obtained according to a relationship defined by an autocorrelation function, and assuming that the autocorrelation function is approximated by a polynomial, a coefficient estimate that approximates the coefficients of each term of the polynomial using a plurality of autocorrelation values government; And a channel spectrum information calculator for calculating distribution information of the channel spectrum from the coefficients of each term of the polynomial.

Description

Channel Characterization Apparatus and Method for Implementation of Adaptive Transmit and Receive Techniques {SYSTEM AND METHOD FOR ESTIMATING CHANNEL STATE INFORMATION USED IN ADAPTIVE WIRELESS TRANSCEIVER}

1 schematically illustrates a configuration of an apparatus for estimating channel characteristics according to an embodiment of the present invention.

2 illustrates one configuration of the autocorrelation unit

3 illustrates the relationship between autocorrelation values, normalized autocorrelation function and Taylor approximation polynomial.

4 shows an example of the moment estimation result of the channel spectrum obtained by applying the present invention.

5 is a flowchart schematically illustrating a channel characteristic estimation method according to another embodiment of the present invention.

 The present invention relates to an apparatus and method for estimating channel characteristics. More particularly, the present invention relates to an adaptive transceiver for changing a transmission / reception scheme according to channel characteristics, thereby more accurately displaying channel state information representing statistical characteristics of a channel. It is to provide an apparatus and method for identifying.

In wireless and mobile communications, the characteristics of the radio channel greatly influence the performance of the system. Therefore, by using the adaptive transmission / reception scheme for identifying the characteristics of the wireless channel and changing the transmission / reception scheme or parameters accordingly, it is possible to improve user capacity and reception performance compared to the fixed transmission / reception scheme operating regardless of the channel environment. . In order to apply the adaptive transmission / reception scheme, it is necessary to understand the characteristics of time-varying channels. As the channel state information that can represent the time-varying channel characteristics, the maximum Doppler frequency, the maximum path delay, and the Lysian factor have been considered, but these information are insufficient to represent the characteristics of the channel.

In general, statistical characteristics of a channel can be obtained from the power spectrum of the channel obtained by Fourier transforming the autocorrelation or autocorrelation characteristics of the channel shock response. If the autocorrelation curves in the time domain show large differences over time, or if the Doppler spectrum of the channel contains high frequency components, then these channels can be classified as fast fading channels, otherwise they are classified as slow fading channels. can do. As in the time domain, the autocorrelation characteristic curve in the frequency domain is classified as a frequency selective fading channel when the difference in frequency varies greatly or the power delay spectrum includes a long time delay component. If not, the frequency close to the non-selective fading channel.

To obtain these channel characteristics, it is necessary to know the Doppler spectrum in the time domain or the power delay spectrum in the frequency domain. For example, in the classic spectrum with typical Rayleigh dispersion characteristics, the shape of the spectrum is already known, so knowing the maximum Doppler frequency only gives an accurate picture of the Doppler spectrum. Therefore, most of the existing channel characteristic estimators use a method of estimating the maximum Doppler frequency.

However, in Rayleigh fading channels, where the direct path (line-of-sight, LOS) component is present and has a dispersive lysian dispersion, or a shape different from the classical spectrum, the spectral characteristics of the channel, even if the channel's maximum Doppler frequency is perfectly known You will not be able to figure out exactly. Accordingly, there is a problem that the estimation error is increased in the above-described channel characteristic estimators of the prior art that apply the same technique as that of the classical spectrum even in the case of a channel having no classical spectrum form.

Techniques for estimating the maximum delay value or root mean square (RMS) delay value have been proposed to extract the characteristics of the power delay spectrum in the frequency domain. However, even if these values are the same, the power delay spectrum of the channel may be greatly different in shape, and thus there is a problem in that the above values are incomplete to use as channel characteristic information in the frequency domain.

In addition to the spectral characteristics in the time and frequency domains of the channels mentioned above, much consideration has been given to the received signal power to noise power ratio (SIR), which can represent the magnitude of the received signal as a measure of the communication environment. Several methods have been proposed to estimate this. However, these methods also have the problem of requiring additional complexity to estimate the SIR.

The present invention is to solve the above problems of the prior art, unlike the prior art using only the maximum Doppler frequency or the maximum / RMS delay value, the channel spectrum in the time and frequency domain in order to increase the accuracy of the channel characteristic information more To provide an apparatus and method for estimating channel characteristics of a method of obtaining.

In addition, the present invention is based on the inventors' studies and studies that the channel spectrum can be completely obtained by knowing the derivative values of all orders or all orders of magnitude in the autocorrelation curve. An apparatus and method for estimating channel characteristics are proposed to obtain arbitrary order moments of a channel spectrum using normalized autocorrelation characteristics.

Furthermore, an object of the present invention is to provide an apparatus and method for estimating channel characteristics in which values of the Doppler spectral moment in the time domain and the power delay spectral moment in the frequency domain can be obtained for a desired order.

In addition, the present invention provides an apparatus and method for estimating channel characteristics based on observation that a zeroth order moment obtained from a power delay spectrum represents information on an average power of a channel. It is for.

In addition, the present invention can estimate the moment and SIR information in the time and frequency domain of the channel at the same time, so that when the present invention is applied to the adaptive radio transceiver, it is possible to more smoothly adapt to the time-varying radio channel situation An object and method for estimating channel characteristics is provided.

An apparatus for estimating channel characteristics according to a first aspect of the present invention for achieving the above object is an apparatus for estimating characteristics of a channel for adaptive transmission and reception, and has a plurality of normalized time or frequency intervals by processing a received signal. An autocorrelation unit for obtaining autocorrelation values for the samples; The plurality of autocorrelation values are values respectively obtained according to a relationship defined by an autocorrelation function, and the autocorrelation function is assumed to be approximated by a predetermined polynomial. A coefficient estimator for obtaining an approximation value of the two; And a channel spectrum information calculator for calculating distribution information of a channel spectrum including time and frequency spectrum information from coefficients of each polynomial term.

Preferably, in the apparatus for estimating channel characteristics of the present invention, the autocorrelation unit may be configured to obtain an autocorrelation value for samples having a plurality of time or frequency intervals among the received signals in a time or frequency domain. Alternatively, the autocorrelation values in the frequency domain may be suitably utilized together or selectively according to the algorithm of the adaptive transceiver.

Further, in the apparatus for estimating channel characteristics of the present invention, the coefficient estimator is a curve for the autocorrelation values to obtain an approximation of the coefficients of each term of the polynomial from the plurality of autocorrelation values obtained in a time or frequency domain. It may be to perform a curve fitting calculation. In this case, the coefficient estimator may use a curve approximation method such as a least square method to calculate the curve approximation.

From the coefficients of each term of the polynomial, channel spectrum information (hereinafter, referred to as 'CSI'), such as a k-th order moment (where k is an integer of 0 or more) in the time or frequency domain, can be obtained. See below for details.

Here, the distribution information of the channel spectrum may be a moment of the Doppler spectrum in the time domain or a moment of the power delay spectrum in the frequency domain.

In addition, as will be described in detail below, it is possible to obtain a received signal power-to-noise power ratio (SIR) by using the zero-order moment of the channel spectrum, and further including an SIR estimator for performing such an algorithm to estimate the channel characteristics of the present invention. The device may be implemented.

The channel characteristic estimating apparatus of the present invention may be used in an adaptive transmission / reception apparatus to detect a momentary change in channel characteristics.

A channel characteristic estimation method according to the second aspect of the present invention is a method for estimating a channel characteristic for adaptive transmission and reception, and processes autonomous correlation values for samples having a plurality of normalized time or frequency intervals by processing a received signal. Obtaining; The plurality of autocorrelation values are values respectively obtained according to a relationship defined by an autocorrelation function, and the autocorrelation function is assumed to be approximated by a predetermined polynomial. Obtaining an approximation of these; And calculating distribution information of a channel spectrum including time and frequency spectrum information from coefficients of each polynomial term.

Hereinafter, with reference to the accompanying drawings, it will be described in detail with respect to preferred embodiments of the present invention.

1 shows an apparatus for estimating channel characteristics according to a preferred embodiment of the present invention.

As shown, the channel characteristic estimating apparatus of the present embodiment includes an autocorrelation unit 100 that receives a received signal and obtains a normalized autocorrelation value of the received signal, and sets autocorrelation values to a least square LS curve. Coefficient estimator 101, 103, 105, 107 that approximates each term coefficient of the polynomial that approximates the least-squares curve, and calculates channel state information that calculates channel state information, such as channel moment and SIR, based on coefficient estimation results. It is divided into parts 102, 104, 106, and 108.

The embodiment illustrated in FIG. 1 is an example of an orthogonal frequency division multiplexing (OFDM) system in which transmission symbols are distributed in the time and frequency domains, and the moment of the Doppler spectrum in the time domain and the power of the frequency domain. This is an example configured to estimate the moment of the delay spectrum at the same time. For other communication systems that cover only one of the time and frequency domains, the moment of the desired channel spectrum can be determined by selectively applying a block for the corresponding region.

If the time delay and the shock response of the l- th path signal of the L channel having the number of multipaths are τ ₁ and h ₁ (t), respectively, the shock response characteristic h (t, τ) of the wireless channel is expressed by the following equation. It can be represented as 1.

는 크로네커 델타 (Kronecker delta) 함수이다. h ₁ (t)이 평균이 0인 복소 가우시안 분포를 따르며 각 경로가 독립이고 동일한 정규화된 시간 상관 함수

를 따른다고 가정하면, 시간 영역에서의 자기 상관 특성은 아래의 수학식 2와 같이 나타낼 수 있다. here

Is the Kronecker delta function. h ₁ (t) follows a complex Gaussian distribution with an average of 0, each path is independent, and the same normalized time correlation function

Assuming that follows, the autocorrelation property in the time domain can be expressed by Equation 2 below.

는 l번째 경로의 평균 전력값을 나타내며 정규화되었다고 가정한다(즉,

).Where E {X} is the expected value of X and the peak * is the complex conjugate. Also,

Denotes the average power value of the l- th path and assumes normalization (i.e.

).

 The channel shock response characteristic in the frequency domain at time t is expressed by Equation 3 below.

, 주파수 영역으로

만큼 떨어진 심벌간의 채널 충격 응답의 상관 특성은 아래 수학식 4와 같이 표현된다. At this time, the time domain

Into the frequency domain

The correlation characteristic of the channel shock response between symbols separated by is expressed by Equation 4 below.

는 주파수 영역의 자기 상관 특성을 나타내며

으로 구해진다. 심벌 시간 T _s , 부채널 간격

동안 채널 충격 응답 특성이 변화하지 않는다고 가정하면, 자기 상관 특성은 아래 수학식 5와 같이 심벌 단위로 표현할 수 있다. here

Represents the autocorrelation characteristics of the frequency domain

Obtained by Symbol time T _s , subchannel interval

Assuming that the channel shock response characteristic does not change during this time, the autocorrelation characteristic may be expressed in symbol units as shown in Equation 5 below.

이고

이다. 자기 상관 함수를 2차원 푸리에 변환한 전력 스펙트럼은 아래의 수학식 6과 같이 표현 가능하다. here

ego

to be. The power spectrum of the two-dimensional Fourier transform of the autocorrelation function can be expressed as in Equation 6 below.

와

는 각각 시간 영역의 상관 함수

와 주파수 영역의 상관함수

의 푸리에 변환값을 나타낸다. here

Wow

Is a correlation function of each time domain

Function in the frequency domain

Represents the Fourier transform of.

 If the k th subcarrier symbol of the n th symbol time transmitted from the transmitter is X [n, k], the symbol value Y [n, k] at the receiver is expressed by Equation 7 below.

인 백색 가우시안 분포를 따른다고 가정한다. Where H [n, k] represents the frequency domain shock response of the channel corresponding to the kth subcarrier of the transmitted nth symbol time, and Z [n, k] is the background noise and interference signal with an average of 0 and a variance

Suppose that follows a white Gaussian distribution.

는 n 번째 심벌 시간의 k번째 부반송파에 대한 채널 충격 응답의 순시 추정치를 나타내며, 아래 수학식 8과 같이 구할 수 있다. 2 is an exemplary view showing the configuration of the autocorrelation unit 100 of FIG. 1 in detail.

Denotes an instantaneous estimate of the channel shock response for the kth subcarrier of the nth symbol time, which can be obtained as shown in Equation 8 below.

Here, X [n, k] means a predetermined pilot symbol or demodulated data symbol.

 In order to estimate the moment of the Doppler spectrum in the time domain, a normalized autocorrelation value in the time domain is obtained as shown in Equation (9).

는 전체 수신 신호의 평균 전력을 의미하며 앞에서 1로 하였다.

는 수신 심벌의 평균 신호 전력대 잡음 전력비 (SIR)를 나타낸다. 실제적으로 수학식 9에서의 통계적 평균은 구할 수 없으므로 이를 수학식 10과 같이 시간 및 주파수 평균으로 근사화할 수 있다. here

Denotes the average power of all received signals.

Denotes the average signal power to noise power ratio (SIR) of the received symbol. In practice, since the statistical mean in Equation 9 cannot be obtained, it can be approximated to the time and frequency mean as in Equation 10.

Where n _s is the number of samples in each subchannel used for correlation, n _i is the index of the start symbol, and K is the number of subchannels.

는 테일러 급수를 사용하여 수학식 11과 같이 나타낼 수 있다. Autocorrelation Function in Time Domain

Can be expressed as in Equation 11 using the Taylor series.

Using Equations 9 and 11, the normalized autocorrelation function can be expressed as in Equation 12.

이다. 이 때, 수학식 12는 수학식 13과 같이 원점 부근에서 (2W-1)차의 다항식으로 근사 가능하다. here

to be. At this time, Equation 12 can be approximated as a polynomial of (2W-1) difference near the origin as in Equation 13.

및 이의 테일러 근사다항식을 나타낸 것으로

의 영역에서 4차의 근사식 사용이 매우 유효함을 알 수 있다.Figure 3 shows the autocorrelation function of the normalized time domain for a channel with classical spectrum at a maximum Doppler frequency of 100 Hz and symbol period (T _s ) of 25 Hz.

And Taylor approximation polynomials thereof

It can be seen that the fourth order approximation is very effective in the domain of.

의 테일러 근사다항식의 k차 계수 C_k는 [수학식 14]와 같이 도플러 스펙트럼의 k차 모멘트로 표현 가능하다.At this time,

The kth order C _k of the Taylor approximation polynomial of can be expressed as the kth moment of the Doppler spectrum as shown in [Equation 14].

는 수학식 15와 같이 도플러 스펙트럼의 n차 모멘트를 의미하며, 스펙트럼의 성분 분포에 대한 정보를 제공한다. here

Denotes the n-th order moment of the Doppler spectrum as shown in Equation 15, and provides information on the component distribution of the spectrum.

 Therefore, the normalized autocorrelation function can be approximated as in Equation 16 below.

As can be seen from Equation 16, the real part of the normalized autocorrelation function is related to the moment of even-order channel spectrum including zero, and the imaginary part is related to the moment of odd-order aberration. Therefore, the moment estimation is performed by dividing the real and imaginary parts as shown in FIG.

를 구하기 위해 도 3에서와 같이 W' 개의 시간 간격 샘플들에 대해

자기 상관값을 구하고(도 3의 경우, W'= 4), 이를 계수 추정부(101, 103)에 대입하면, 2W개

의 테일러 다항식 계수

를 구할 수 있다. 여기서 정규화의 정의 상 c₀은 1이므로, 0차 계수

는

이 되어 SIR를 구할 수 있다. 최종적으로 추정된

및

을 수학식 16에 대입함으로써 CSI 계산부(102, 104)에서는 원하는 채널 스펙트럼의 모멘트

및

등의 채널 상태 정보를 구할 수 있다.Coefficient of each term in (16)

For W 'time interval samples as in FIG.

If the autocorrelation value is obtained (W '= 4 in FIG. 3) and substituted into the coefficient estimating units 101 and 103, 2W

Taylor polynomial coefficients of

Can be obtained. Where c ₀ is 1 in the definition of normalization, so the 0th order coefficient

Is

The SIR can be obtained. Finally estimated

And

Is substituted into Equation (16), the CSI calculation unit (102, 104), the moment of the desired channel spectrum

And

Channel state information such as can be obtained.

In the frequency domain, the moment and SIR values can be estimated in the same manner as in the time domain. In this case, the only difference is that the autocorrelation value shown in FIG. 2 is obtained in the frequency domain rather than the time domain. Once the normalized autocorrelation value in the frequency domain is obtained, channel state information such as channel spectral moment in the desired frequency domain can be obtained through the coefficient estimator 105 and 107 and the CSI calculator 106 and 108.

FIG. 4 illustrates moment estimation results in an OFDM system when the experimental conditions given in Table 1 below are used. When three correlators are used in the time and frequency domains, the zero-order moments in the time and frequency domains (SIR) are illustrated. ) And the ratio of the estimated second moment to the actual exact moment value.

TABLE 1

As shown in FIG. 4, when the present invention is applied, the moment value can be estimated so that an error with the actual moment value can be within 5%.

5 shows a flow of a channel state estimation method according to another embodiment of the present invention. The channel state estimation method of the present invention is a method for estimating characteristics of a channel for adaptive transmission and reception, sampling a received signal (S10), and processing a received signal to have a plurality of normalized time or frequency intervals. Obtaining autocorrelation values for samples (S20), since the plurality of autocorrelation values are each obtained according to a relationship defined by an autocorrelation function, the autocorrelation function is approximated by a polynomial and the autocorrelation values are calculated. Estimating the coefficients of each term by a method such as curve fitting used (S30) and calculating channel spectrum distribution information including time and frequency spectrum information from the coefficients of the polynomial term (S40). It is characterized by including.

 The foregoing has focused on the preferred embodiments of the present invention, but the present invention is not limited to the above-described embodiments and drawings, and the general knowledge in the technical field to which the present invention belongs without departing from the technical spirit of the present invention. Since various changes and modifications may be made by those having the scope of the present invention, the scope of the present invention should be determined by the appended claims and their equivalents.

Using the present invention, distribution information of the channel spectrum in the time and frequency domain can be provided by estimating the moment of the time and frequency domain spectrum in a wireless communication system.

In addition, the zeroth order moment obtained by applying the present invention can be applied to calculating a signal-to-noise power ratio indicating the quality of a received signal.

When channel state information such as the moment of time and frequency spectrum obtained by the present invention is applied to an adaptive transmission / reception system that is variable in a channel environment, it is possible to obtain a significantly improved user capacity and reception performance compared to the prior art.

Claims (10)

An apparatus for estimating characteristics of a channel for adaptive transmission and reception, An autocorrelation unit processing the received signal to obtain autocorrelation values for samples having a plurality of normalized time or frequency intervals; The plurality of autocorrelation values are values respectively obtained according to a relationship defined by an autocorrelation function, and the autocorrelation function is assumed to be approximated by a predetermined polynomial. A coefficient estimator for obtaining an approximation value of the two; And And a channel spectrum information calculator for calculating distribution information of the channel spectrum including time and frequency spectrum information from the coefficients of the polynomial terms. delete The method of claim 1, The coefficient estimator performs a curve fitting calculation on the autocorrelation values to obtain an approximation of the coefficients of each term of the polynomial from the plurality of autocorrelation values obtained in a time or frequency domain. Channel characteristic estimation apparatus. The method of claim 3, And the coefficient estimating unit uses a least square method in the curve approximation calculation. The method of claim 1, The distribution information of the channel spectrum is a channel characteristic estimation apparatus, characterized in that the k-order moment (where k is an integer of 0 or more) of the channel spectrum in the time or frequency domain. The method of claim 1, And the distribution information of the channel spectrum is a moment of the Doppler spectrum in the time domain. The method of claim 1, And the distribution information of the channel spectrum is a moment of a power delay spectrum in a frequency domain. The method of claim 5, And an SIR estimator for obtaining a received signal power to noise power ratio (SIR) using the zeroth order moment of the channel spectrum. An adaptive transmission / reception device as set forth in any one of claims 1 to 8, wherein an instantaneous change in channel characteristics is detected using the channel characteristic estimating apparatus. In the method for estimating the characteristics of the channel for adaptive transmission and reception, Processing the received signal to obtain autocorrelation values for a sample having a plurality of normalized time or frequency intervals; The plurality of autocorrelation values are values respectively obtained according to a relationship defined by an autocorrelation function, and the autocorrelation function is assumed to be approximated by a predetermined polynomial. Obtaining an approximation of these; And And calculating distribution information of the channel spectrum including time and frequency spectrum information from the coefficients of each polynomial term.

Images