JP2010268220A - Channel response value estimating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a channel response value estimating device for reducing an error in channel response estimation in a notch etc. <P>SOLUTION: This channel response value estimating device is a device for estimating an estimated channel response value as an estimating value of an estimated channel response value of a third sub-carrier having a data symbol disposed therein between a first and second sub-carriers where a pilot symbol is disposed in a sub-carrier direction by using a channel response value of a pilot symbol disposed in the sub-carrier previously set by an OFDM symbol. The estimating device includes an equalizing section for equalizing the data symbol with a weighting coefficient of an inverse number of the estimated channel response value, and an adaptive signal processing section for obtaining an error between the equalized data symbol and a result of hard decision of the data symbol and obtaining a weighting coefficient minimizing the error. The adaptive signal processing section obtains an estimated channel response value of the third sub-carrier by using a channel response value of either the first or second sub-carrier. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a channel response value estimation apparatus in an Orthogonal Frequency Division Multiplexing (OFDM) system.

In an OFDM system, a channel response of a propagation path between a transmitting antenna of a transmitter and a receiving antenna of a receiving apparatus is used when receiving an OFDM symbol using a transfer function (hereinafter referred to as a channel response value) estimated by pilot symbols. Estimating the value is done.
That is, in wireless communication such as OFDM, as shown in FIG. 4, a transmission signal S _{(i, k)} , that is, an OFDM symbol is subjected to IFFT (Inverse Fast Fourier Transform ₎ and transmitted from the transmission side. When the transmitted signal S _{(i, k)} passes through the propagation path, the channel response value h _{(i, n)} is multiplied by the propagation path characteristics (characteristics between the transmitter and the radio). On the other hand, noise noise z _{(i, n)} is further added to change the phase and amplitude of the transmission signal, resulting in a reception signal r _{(i, n)} .
For this reason, on the receiver side, the reception signal r _{(i, n} ) is subjected to FFT (Fast Fourier Transform ₎ to obtain the reception signal R _{(i, k)} . At this time, the channel response value indicating the propagation path characteristic is estimated, and the inverse of the channel response value is multiplied by the modulation symbol arranged on the subcarrier in the received OFDM symbol to perform compensation (equalization). The correct transmitted OFDM symbol can be estimated from the received OFDM symbol.

Usually, pilot symbols are used to estimate and obtain the channel response value of the transmission path of each data symbol (see, for example, Non-Patent Document 1).
In an OFDM signal, pilot symbols with clear data are transmitted regularly for each of a plurality of symbols and a plurality of subcarriers as in the transmission frame shown in FIG. 2, and the channel response value of the data symbol between the pilot symbols is estimated. Is going. FIG. 2 is a conceptual diagram showing an overall configuration of an OFDM symbol for explaining the arrangement of pilot symbols and data symbols in subcarriers constituting the transmitted OFDM symbol. In FIG. 2, the row direction (horizontal direction) indicates the time direction in which the OFDM symbol is input, and the column direction (vertical direction) indicates the arrangement direction of subcarriers constituting the OFDM symbol (subcarrier frequency direction). Yes. The OFDM symbols are arranged in the order of transmission (time series), and k indicates the order of arrangement of subcarriers in each OFDM symbol (order of arrangement of frequency bands).

In the symbol arrangement for the elements in FIG. 2, the channel response value is estimated by first calculating the channel response value of each pilot subcarrier (corresponding to the element in which the pilot symbol is arranged) in the symbol direction as shown in FIG. Ask sequentially.
That is,
H ^ _ZF (i, k) = R (i, k) / D (i, k)
The channel response value of the subcarrier in which the pilot symbol is arranged is calculated by the following equation.
Next, in a data symbol sequence in which pilot symbols exist, channel response estimation values H ^ (i, k) of data symbols sandwiched between pilot symbols in the subcarrier direction are linearly calculated from the channel response values of pilot symbols on both sides. Estimate by interpolation.
And each received signal is represented by S ^ (i, k) = R (i, k) / H ^ (i, k)
The equalized received signal S ^ (i, k) is calculated by equalizing the received signal R (i, k) using the following equation.

Easy-to-understand OFDM technology, Makoto Itami, p. 66-p. 82, Ohmsha, May 10, 2003 (first edition)

However, linear interpolation is effective when the characteristics of the propagation path are static characteristics and a delay with a short delay time, but a large number of delayed waves arrive and there are many signal strength drops (notches) in the band due to frequency selective distortion. In some cases, effective interpolation cannot be performed.
That is, if the error of the channel response value obtained from the pilot symbol due to the notch is large in the estimation of the channel response estimation value of the data symbol included between the pilot symbols, the channel of the data symbol linearly interpolated from this channel response value The response estimation value also has a large error, and the accuracy of the modulation symbol after equalization is low.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a channel response value estimation device that reduces an error in channel response estimation in a notch or the like.

  The channel response value estimation apparatus according to the present invention uses a channel response value of a pilot symbol arranged in a subcarrier at a preset position in an OFDM symbol, and uses the first subcarrier in which the pilot symbol is arranged in the subcarrier direction. A channel response value estimation device that estimates a channel response estimation value that is an estimation value of a channel response value of a third subcarrier in which data symbols between a carrier and a second subcarrier are arranged, An error that is a difference between the equalization unit that performs equalization of the data symbol, the equalized data symbol, and the hard decision result of the data symbol is obtained by the inverse weighting coefficient, and the error is minimized. An adaptive signal processing unit for obtaining the weighting factor, wherein the adaptive signal processing unit is a first subcarrier or Using the channel response value of one of two subcarriers, and obtains a channel response estimate for the third sub-carrier.

  In the channel response value estimation apparatus of the present invention, the adaptive signal processing unit estimates the weighting factor by each of the first subcarrier and the second subcarrier sandwiching the data symbol in the subcarrier direction, The smallest error is compared, and the smaller one of the errors is used as the channel response estimation value of the third subcarrier.

  The channel response value estimation apparatus of the present invention is characterized in that the adaptive signal processing unit performs error minimization by LMS (Least Mean Square) or RLS (Recursive Least Squares) and outputs the channel response estimation value. To do.

  According to the present invention, in order to estimate the channel response value for each subcarrier of the OFDM signal adaptively according to the change in characteristics in the propagation path between the transmitter and the receiver, The error can be reduced.

It is a block diagram which shows the structural example of the OFDM receiver using the channel response value estimation apparatus by the 1st Embodiment and 2nd Embodiment of this invention. It is a conceptual diagram explaining the arrangement | positioning of the pilot symbol in the subcarrier which comprises an OFDM signal. It is a conceptual diagram of the two-dimensional plane which consists of Q value and I value explaining the determination of the hard decision value D ^ (i, k) from the estimated transmission signals S to (i, k) in the hard decision unit 11. It is a conceptual diagram explaining the change of the transmission signal by the characteristic of the propagation path between a transmitter and a receiver. It is a conceptual diagram explaining the estimation of the channel response value by ZF method.

<First Embodiment>
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration example of an OFDM receiver using a channel response value estimation apparatus according to an embodiment of the present invention. The OFDM receiver of this embodiment includes a frequency filter unit 1, an orthogonal detector 2, an A / D converter 3, an FFT unit 4, a coefficient multiplier 16, a demodulator 17, and a channel response value estimation device 100. Yes.
The channel response value estimation apparatus 100 includes a pilot subcarrier detection unit 5, a data subcarrier detection unit 6, a pilot subcarrier channel estimation unit 7, an equalization unit 8, an error calculation unit 9, a weight calculation unit 10, a hard decision. Unit 11 and adaptive signal processing unit 12.

The operation of the OFDM receiving apparatus using the channel response value estimating apparatus 100 according to the present embodiment will be described below with reference to FIG.
The frequency filter unit 1 extracts only the frequency components of a preset frequency band from the received signal received from the antenna element, that is, extracts the OFDM symbol transmission band and sends the OFDM symbol to the quadrature detection unit 2 at the next stage. Output as.
The quadrature detection unit 2 converts the input OFDM symbol from a real time signal to a complex time signal composed of an I signal (Ich) and a Q signal (Qch), and outputs the complex time signal to the A / D conversion unit in the next stage.
The A / D converter 3 performs A / D conversion on the I signal and the Q signal in the complex time signal, respectively, and converts the A / D converted I signal and Q signal (hereinafter referred to as a modulation symbol) into an FFT in the next stage. Output to unit 4.

The FFT unit 4 performs a Fourier transform on the OFDM signal, that is, performs a transform from the time domain to the frequency domain, and outputs a subcarrier modulation symbol in each subcarrier unit of the OFDM symbol as shown in FIG. 2 (described later). To do.
Channel response value estimation apparatus 100 obtains the channel response estimation value of the data symbol in each subcarrier from the channel response value of the pilot symbol in the OFDM symbol.
The coefficient multiplication unit 16 multiplies the reciprocal of the input channel response estimation value by the modulation symbol corresponding to the corresponding data symbol, and outputs the equalization modulation symbol that is the multiplication result to the demodulation unit 17.
The demodulator 17 performs demodulation processing on the equalization modulation symbol and outputs a demodulated signal.

Next, the calculation process of the channel response estimated value for the data symbol in the channel response value estimating apparatus 100 will be described in detail with reference to FIGS.
FIG. 2 shows a configuration of data subcarrier arrangement (subcarrier arrangement unit in each OFDM symbol) in each OFDM symbol in the OFDM signal for explaining estimation of channel response values (weighting factors) in the present embodiment. It is a conceptual diagram. FIG. 2 shows an example of the arrangement of OFDM symbols in the horizontal direction (time direction) and the arrangement of subcarriers in each OFDM symbol in the vertical direction. Further, a unit surrounded by a broken line is a block (a pilot symbol is arranged on a subcarrier at the apex), and in this embodiment, a channel response estimation value is estimated using this block as one unit.

The pilot subcarrier detection unit 5 starts from the modulation symbols sequentially input from the FFT unit 4 by the position of the subcarrier where the pilot symbol set in advance in the transmission frame is located (the position of the pilot subcarrier is known). , The modulation symbol R (i, k _p ) of the subcarrier in which the pilot symbol is arranged is extracted from the OFDM symbol, and is output to the pilot subcarrier channel estimation unit 7 in the next stage. In the figure, k−1 = k + 2 = k _p as pilot subcarriers, and k = k _d and k + 1 = k _{d + 1} as data subcarriers.

The pilot subcarrier channel estimation unit 7 divides the modulation symbol R (i, k _p ) by a set symbol D (i, k _p ) set in advance as a pilot symbol by the following equation (1), seek response value H ^ a (i.k _p).

Here, in the case of modulation symbol R (i, k _p ), pilot subcarrier channel estimation unit 7 outputs this channel response value H ^ (i.k _p ) to coefficient multiplication unit 16.
Then, the coefficient multiplication unit 16 obtains a weight coefficient W ^* (i, k _p ) that is an inverse number of the channel response value H ^ (i.k _p ), and multiplies the input modulation symbol R (i, k _p ). Then, the equalized reception signal S ^ (i, k _p ) is output. Here, the weight coefficient W ^* (i, k _p ) is a complex conjugate of the weight coefficient W (i, k _p ).

Further, when the channel response value H ^ (i.k _p ) is input to the weight calculation unit 10, the weight coefficient W ^* (i, k _p ) that is the reciprocal of this channel response value is represented by the following (2). Obtained by the equation and output to the equalization unit 8.

Further, the data subcarrier detection unit 6 uses the modulation symbols R (from the OFDM symbol to the modulation symbol R () of the data subcarrier by using the subcarriers in which the data symbols set in advance in the transmission frame are arranged. i, k _d ) is extracted and output to the equalization unit 8 and the adaptive signal processing unit 12 in the next stage.
Then, the equalization unit 8 multiplies the input modulation symbol R (i, k _d ) and the weighting coefficient W ^* (i, k _p ) as shown in the following equation (3), and estimates transmission signals S˜ (I, k _d ) is calculated and output to the hard decision unit 11 and the error calculation unit 9.

Here, the hard decision unit 11 performs a hard decision on the input estimated transmission signals S to (i, k _d ), and outputs a hard decision value D ^ (i, k _d ) according to the following equation (4).

Here, when the coding is performed by QPSK (quadrature phase shift keying), the hard decision unit 11 uses the estimated transmission signals S to (i, k _d ) in the QPSK coding as shown in FIG. Compared with the reference modulation symbol, the reference modulation symbol (signal point mapped to the IQ plane by modulation) at the closest position is output as a hard decision value D ^ (i, k _d ). That is, the function _{harddecision (S~ (i, k d} )) , in the two-dimensional coordinates of the IQ plane is the estimated transmission signal S~ (i, _{k d)} function for selecting the reference modulation symbol to the closest .

Next, the error calculation unit 9, the hard decision value D ^ (i, _{k d)} and the estimated transmission signal s to (i, _{k d)} the difference between the following (5) As shown in equation error e (i , K _d ) and output to the adaptive processing unit 12 in the next stage.

When the error e (i, _{k d)} is input, the adaptive signal processing section 12, the error e (i, _{k d)} error ^e * (i, _{k d)} is the complex conjugate of the weighting factor W ^* (i, _{k p)} weight coefficient W (i, _{k p)} is the complex conjugate of the modulation symbols R (i, _{k d)} by a, as shown in the following equation (6), the weighting factor W (i , K _d ). The coefficient μ (step size parameter) in the equation (6) is set by simulation or experiment.

Here, when the error e (i, k _d ) is input, the adaptive signal processing unit 12 uses a weighting factor W (i) by the LMS algorithm so that the power of the error e (i, k _d ) is minimized. , K _d ). This calculation of the error e (i, _{k d)} in the adaptive signal processing unit 12 to become the weighting factor W (i, _{k d)} are known technique using the method of least squares (LMS algorithm), detailed description It is omitted here.
The adaptive signal processor 12 for the data carrier _{k d} (i.e., modulation symbols R (i, _{k d)} for) the weighting coefficients of the time the weighting factor ^W * (i, _{k d)} and.

Next, the data subcarrier detection unit 6 outputs the detected modulation symbol R (i, k _{d + 1} ) to the equalization unit 8 and the adaptive signal processing unit 12.
Then, with the weighting factor W ^* (i, k _p ) as an initial value, the processing from Equation (3) to Equation (6) is performed, and the weighting factor W (i, k _{d + 1} ) for the modulation symbol R (i, k _{d + 1} ) is performed. k _{d + 1} ) is determined.
The estimation process described above uses the weight coefficient (reciprocal of the channel response value) of the subcarriers (data symbols) sandwiched between the subcarriers k-1 and k + 2 in which the pilot symbols of the block in the OFDM signal in FIG. Channel response values of subcarriers k−1 of the subcarriers k−1 are estimated as initial values, that is, channel response values W (i, k _d ) and W (i, k _{d + 1} ) of subcarriers k and k + 1 are estimated in the A direction.
Next, with the channel response value of subcarrier k + 2 of the other pilot symbol sandwiching the data symbol as an initial value, the weight coefficients W (i, k _{d + 1} ) and W ( i, k _d ) is estimated in the B direction.

Then, the adaptive signal processing unit 12 calculates a weighting coefficient with a small error calculated in the A direction or the B direction, that is, estimated with any one of the pilot symbols sandwiching the subcarrier of the data symbol in the subcarrier direction as an initial value. The subcarrier weight coefficients W (i, k _d ) and W (i, k _{d + 1} ) are sequentially output to the weight calculation unit 10. Here, when the adaptive signal processing unit 12 calculates the magnitude of the error, for example, the error e (i, k _d ) and the complex conjugate e ^* (i, k _d ) of the error e (i, k _d ) The values are compared based on the result of multiplication.
Then, the weight calculator 10 calculates complex conjugates of the weight coefficients W (i, k _d ) and W (i, k _{d + 1} ), respectively, and modulates the modulation symbols R (i, k _d ) and R (i, k), respectively. _{d + 1} ) are output as weight coefficients W ^* (i, k _d ) and W ^* (i, k _{d + 1} ) corresponding to each.
As a result, the coefficient multiplication unit 16 adaptively determines the weighting coefficient W ^* () for each of the modulation symbols R (i, k _d ) and R (i, k _{d + 1} ) input corresponding to the subcarriers. i, k _d ) and W ^* (i, k _{d + 1} ) are multiplied, and the equalized modulation symbol after equalization is output to the demodulator 17 as shown in the following equation (7).

As described above, in this embodiment, in the A direction and the B direction, the channel response values of different pilot symbols are used as initial values, and the weight coefficients of the subcarriers in which the data symbols sandwiched between these pilot symbols are arranged. Calculated.
For this reason, according to the present embodiment, even if the subcarrier in which any one of the pilot symbols is arranged is affected by a notch or the like, it is possible to use the weighting coefficient obtained as the initial value of the one not receiving it. Therefore, the accuracy of equalization of the received signal can be improved.

Further, subcarriers k and k + 1 in OFDM symbol S1 and subcarriers k and k + 1 in OFDM symbol S3 are estimated. The overall operation control described above is performed by a control unit (not shown).
Further, the channel response value of each subcarrier in the OFDM symbol S2 in which no pilot symbol is arranged in any subcarrier is estimated by the control of the control unit as follows.

Similar to the estimation processing of the channel response estimation value of the subcarrier of the data symbol in the subcarrier direction already described, in the OFDM symbol direction (C direction), the weighting factor W ^* (i) of the subcarrier k-1 of the OFDM symbol S2 , K _k−1 ), using the weighting factor W ^* (i, k _p ) of the subcarrier k−1 of the OFDM symbol S1 as an initial value.
Further, in the OFDM symbol direction (D direction), the weight coefficient W ^* (i, k _p ) of the subcarrier k-1 of the OFDM symbol S2 is set to the channel response value H ^ ( i.k _p ) is determined as an initial value.

The adaptive signal processing section 12, the weighting factor of the subcarrier k-1 in the OFDM symbol S1 or _{^{S3 W * (i, k p}} ) were determined as an initial value, the weighting factor of the OFDM symbol S2 ^W * (i, k _k−1 ), whichever is smaller in error e (i, k _k−1 ) is used as a coefficient of weight W ^* (i, k _k−1 ) of OFDM symbol S2 through coefficient calculation unit 10. To the unit 16.
Further, the control unit estimates the weighting factor W ^* (i, k _{k + 2} ) of the OFDM symbol S2 and calculates the OFDM symbol S1 similarly to the calculation of the weighting factor W ^* (i, k _k-1 ) of the OFDM symbol S2. Alternatively, the weighting coefficient W ^* (i, k _p ) of the subcarrier k + 2 of S3 is used.

Next, since the weighting factor W ^* (i, k _p ) of each of the subcarriers k−1 and k + 2 sandwiching the subcarriers k and k + 1 is obtained in the OFDM symbol S2 in the block, the control unit already has the OFDM symbol S1. as noted in the calculation process of the weighting factor, determined subcarrier k, k + 1 each weighting factor _{^{W * (i, k d)}} , W * a _{(i, k d + 1)} .
As described above, in the present embodiment, the weight coefficient of the subcarrier in which the data symbol is arranged in units of blocks in FIG. 2 can be obtained from the weight coefficient of the subcarrier in which the pilot symbol is arranged.

  In addition, by setting a subcarrier in which pilot symbols are always arranged for each OFDM symbol in FIG. 2, channel response estimation values can be estimated in units of OFDM symbols. Until the coefficient is determined, the memory capacity for storing the weighting coefficient corresponding to the OFDM symbol or the modulation symbol in the block can be reduced, and the processing speed can be increased.

<Second Embodiment>
The second embodiment of the present invention will be described below with reference to the drawings. The configuration is the same as that of the channel response value estimation apparatus 100 shown in FIG.
The difference between the second embodiment and the first embodiment is only the calculation of an adaptive weighting coefficient performed in the adaptive signal processing unit 12, and this point will be described below.
The error calculation unit 9 calculates the prior error ξ (i, k) in the following equation (8), similarly to the error e (i, k) in equation (5).

The adaptive signal processing section 12, the pre-error xi] (i, k) that is input, the weighting factor ^W * (i, _{k d)} and, Kalman gain K (i, k) and by the following weighting coefficient updating The weighting factor W (i, k) is calculated by the equation (9).

  In this equation (9), the Kalman gain K (i, k) is obtained by the following equation (10).

The above equation (10) can be obtained from the P matrix of the following equation (11). This P matrix is an inverse matrix of the correlation matrix and is called Riccati's equation. Λ (= 1−μ) is a forgetting factor, and is set by experiment or simulation. Here, P (i, k−1) = δ ⁻¹ is set as an initial value of the P matrix. This δ is a positive number smaller than 1.

The other calculation is the same, and the adaptive signal processing unit 12 calculates the weighting factor W (i, k _d ) by the RLS algorithm so that the prior error becomes the minimum value, and performs the A direction or the B direction. The smaller one of the weighting factors W (i, k _d ) in the above calculation is output as the weighting factor of the subcarrier of the corresponding data symbol.
Then, the weight calculation unit 10 calculates a weight coefficient W ^* (i, k _d ) that is a complex conjugate of the input weight coefficient W (i, k _d ), and outputs it to the coefficient multiplication unit 16.
That is, as described above, the adaptive signal processing unit 12 performs Kalman gain calculation, prior estimation error calculation, weight coefficient update, and correlation matrix update. Since these are known techniques using the recursive least square method (RLS algorithm) in the Kalman filter, detailed description thereof is omitted here.

  Further, a program for realizing the function of the channel response value estimation apparatus 100 of FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed. Interference wave removal processing may be performed. The “computer system” here includes an OS and hardware such as peripheral devices.

Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in the computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design and the like within a scope not departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 1 ... Frequency filter part 2 ... Quadrature detection part 3 ... A / D conversion part 4 ... FFT part 5 ... Pilot subcarrier detection part 6 ... Data subcarrier detection part 7 ... Pilot subcarrier channel estimation part 8 ... Equalization part 9 ... Error calculation unit 10 ... weight calculation unit 11 ... hard decision unit 12 ... adaptive signal processing unit 16 ... coefficient multiplication unit 17 ... demodulation unit

Claims (3)

Data between the first subcarrier and the second subcarrier in which the pilot symbol is arranged in the subcarrier direction using the channel response value of the pilot symbol arranged in a subcarrier at a preset position in the OFDM symbol A channel response value estimation device that estimates a channel response estimation value that is an estimation value of a channel response value of a third subcarrier in which symbols are arranged;
An equalization unit that performs equalization of the data symbols by a weighting factor that is a reciprocal of the channel response estimation value;
An adaptive signal processing unit that obtains an error that is a difference between the data symbol after equalization and the hard decision result of the data symbol, and obtains the weighting coefficient that minimizes the error;
The adaptive signal processing unit obtains a channel response estimated value of the third subcarrier by using the channel response value of either the first subcarrier or the second subcarrier. Estimating device.   The adaptive signal processing unit estimates the weighting factor by each of the first subcarrier and the second subcarrier sandwiching the data symbol in the subcarrier direction, and compares each of the minimum errors. 2. The channel response value estimation apparatus according to claim 1, wherein the smaller error is set as a channel response estimation value of the third subcarrier.   The channel response value estimation apparatus according to claim 1 or 2, wherein the adaptive signal processing unit performs error minimization by LMS or RLS and outputs the channel response estimation value.

Images