JP2006314088A - Ofdm receiving method and ofdm receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an OFDM (orthogonal frequency division multiplexing) receiving method and OFDM receiver that follows an synchronization error and transmission path fluctuations between transmission and reception. <P>SOLUTION: OFDM signals including a plurality of subcarriers, including data carriers and pilot carriers, orthogonal to each other are received by a plurality of receiver antennas. The received OFDM signals are each OFDM-demodulated into a plurality of receive subcarriers. Based on the plurality of receive subcarriers, the propagation coefficients of a plurality of space paths are estimated. An inverse matrix of a propagation matrix whose elements are the estimated propagation coefficients is calculated. Interference cancellation is performed for the plurality of receive subcarriers by using the inverse matrix, and transmit subcarriers multiplexed in space are estimated. A reliability of the inverse matrix is calculated. A pilot carrier is extracted from the estimated transmit subcarriers, and the extracted pilot carrier is weighted according to the reliability. An error included in the plurality of receive subcarriers is corrected based on the weighted pilot carrier. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an OFDM receiving method and an OFDM receiving apparatus that perform multiplex communication between transmission and reception via a plurality of paths by space division.

  In recent years, in wireless LAN and the like, an OFDM (Orthogonal Frequency Division Multiplexing) modulation method, which is a kind of multicarrier transmission, has been adopted as a modulation method strong against frequency selective fading that occurs in a multipath environment in mobile communication. In order to further improve the frequency utilization efficiency, a MIMO (Multi Input Multi Output) channel is configured using a plurality of transmission antennas and a plurality of reception antennas, and multiplexed between transmission and reception via a plurality of paths by space division. A method for performing communication has been proposed. On the receiving side, the inverse function of the propagation coefficient of multiple paths is estimated from the signals of each receiving antenna and equalized to separate the transmitted signals from each transmitting antenna and increase the number of channels by the number of transmitting antennas. Can do.

  Thus, for example, Patent Document 1 has conventionally proposed a MIMO-OFDM modulation scheme that combines multipath-resistant OFDM and MIMO that improves frequency utilization efficiency. FIG. 18 and FIG. 19 are diagrams showing configurations of an OFDM transmission apparatus 200 and an OFDM reception apparatus 220 using the conventional MIMO disclosed in Patent Document 1. In FIG. FIG. 18 and FIG. 19 illustrate the case of 2 × 2 MIMO-OFDM with two transmission antennas and two reception antennas.

  The data modulated by the data modulation unit 201 is distributed to the transmission antennas 206 and 207, and OFDM modulated by the OFDM modulation units 202 and 203, respectively. At this time, signals necessary for reception processing such as preambles 601 and 602 necessary for synchronization and training symbols 603 and 604 necessary for propagation coefficient estimation are added to constitute transmission frames 1 and 2 (FIG. 20). The transmission frames 1 and 2 are converted into radio frequencies by the frequency converters 204 and 205 and transmitted from the transmission antennas 206 and 207.

Signals transmitted from the plurality of transmission antennas 206 and 207 reach the plurality of reception antennas 208 and 209 via different paths. Here, the propagation coefficient between each transmitting antenna and each receiving antenna is h _{j, i} . i is a transmitting antenna number and j is a receiving antenna number. In the case of 2 × 2 MIMO, there are four transmission paths, h _1,1 , h _1,2 , h _2,1 and h _2,2 . In this case, the relationship between the transmission signal Si and the reception signal Rj is expressed by the following equations (1) and (2).
R1 = _h1,1 * S1 + _h1,2 * S2 (1)
R2 = _h2,1 * S1 + _h2,2 * S2 (2)

Here, if h _{j, i} are uncorrelated with each other and an inverse function of h _{j, i} can be obtained, the multiplexed transmission signal can be separated from the reception signal. This can be achieved, for example, by obtaining an inverse matrix of the propagation matrix H having h _{j, i} as elements and multiplying the inverse matrix by the matrix R composed of the received signal Rj. That is, S = [S1,..., SN] ^T , a transmission signal matrix having elements Si as signals transmitted from N transmission antennas, and a reception signal matrix having elements Rj received by M reception antennas as elements. When R = [R1,..., RM] ^T and a propagation matrix having M × N propagation coefficients h _{j, i} between transmitting and receiving antennas as elements, H = h _{j, i} , the received signal R is expressed by the following equation: It is represented by (3).
R = HS (3)
Here, when the inverse matrix of the propagation matrix H is set to W = H ⁻¹ and multiplied on both sides of the equation (3), WR = WHS = H ⁻¹ HS = S, and the transmission signal S can be separated.

On the receiving side, the radio signals received by the receiving antennas 208 and 209 are converted into frequency bands suitable for signal processing by the frequency converters 210 and 211, respectively. The converted received signals are OFDM demodulated by OFDM demodulation sections 212 and 213, respectively, and separated into a plurality of subcarrier signals as shown in FIG. The transmission path estimation unit 214 estimates a propagation coefficient h _{j, i} for each path using a training symbol added for propagation coefficient estimation. The inverse matrix calculation unit 215 obtains an inverse matrix of the propagation matrix H having h _{j, i} as elements. The interference cancellation unit 216 performs interference cancellation from the received subcarrier signal using the inverse matrix of the propagation matrix H, and separates (channel separates) the multiplexed transmission signals. The separated transmission signal is demodulated by the data demodulator 217.

  The OFDM demodulating units 212 and 213 perform carrier frequency synchronization, clock synchronization, and symbol synchronization using the synchronization preamble, and correct the frequency error and timing error. Thereafter, the time axis signal is converted into a frequency axis signal and separated into subcarrier signals.

  Here, when a synchronization estimation error occurs, a phase error occurs in each subcarrier signal. In the conventional OFDM transmission, a known phase is assigned to a specific subcarrier for transmission (pilot carrier), and the phase error is estimated and corrected using the received pilot carrier. It was.

  FIG. 22 shows an example of a conventional OFDM receiver 230 using MIMO for performing phase error correction. In addition, the same code | symbol is attached | subjected to the same component as FIG. 19, and description is abbreviate | omitted. Pilot extraction section 501 extracts only the pilot carrier from each subcarrier channel-separated by interference cancellation section 216. Phase error estimation section 502 compares the extracted pilot carrier phase with a known phase at the time of transmission, and estimates a phase error. The correction unit 503 corrects the data carrier so as to correct the phase error obtained by the estimation, and the data demodulation unit 217 performs data demodulation.

When such demodulation processing is repeated several times to increase the accuracy of interference cancellation, the following can be further performed. The data demodulated by the data demodulator 217 is modulated again by the data modulator 504 to generate a transmission signal. The replica generation unit 505 generates a replica signal by multiplying the remodulated transmission signal by the estimated propagation coefficient. The replica signal is regarded as an interference signal and is subtracted from the received signal, and the remaining signals are sequentially separated. Such demodulating stages 506 and 507 may be connected in a required number of stages.
Japanese Patent No. 3590008

  However, as in the OFDM receiving apparatus 220 using MIMO shown in Patent Document 1 above, the following problems remain if the OFDM modulation scheme is simply combined with the transmission path estimation method based on the MIMO configuration. That is, when the inverse matrix of the propagation coefficient matrix is multiplied in channel separation, the amplitude of the separated signal is normalized regardless of the original reception level. Therefore, the noise level of a signal having a low original reception level is enhanced by normalizing the amplitude. For this reason, if the phase error is obtained from the separated pilot carrier, the error of the estimation result may become very large due to noise enhancement. As a result, the phase correction of the separated data carrier cannot be performed correctly, causing a demodulation error.

  In addition, in the configuration in which a replica signal is generated and repeatedly demodulated while being subtracted from the received signal, as in the OFDM receiver 230 using MIMO for performing phase error correction, a pilot carrier is extracted for each demodulation stage, and the phase error is reduced. Since correction is necessary, there is a problem that the receiving apparatus is complicated and large.

  Therefore, an object of the present invention is to provide an OFDM reception method and OFDM reception capable of correctly estimating a transmission path and demodulating a carrier without error even when performing multiplex communication between transmission and reception via a plurality of paths by space division. Is to provide a device.

  The present invention provides an OFDM signal composed of a plurality of subcarriers orthogonal to each other including a data carrier to which transmission data is assigned and a pilot carrier to which a known phase and amplitude are assigned, transmitted from at least one transmission antenna. The present invention is directed to an OFDM receiving method and apparatus for receiving via a plurality of spatial paths using a receiving antenna. And in order to solve the said subject, the OFDM receiving method and apparatus of this invention are equipped with the following steps (structure).

  Each of OFDM signals received by a plurality of receiving antennas is subjected to OFDM demodulation and separated into a plurality of receiving subcarriers (a plurality of OFDM demodulating units), and propagation coefficients of a plurality of spatial paths are estimated from the plurality of receiving subcarriers, respectively. A step (transmission path estimation unit), a step of calculating an inverse matrix of a propagation matrix having an estimated propagation coefficient as an element (inverse matrix calculation unit), performing interference cancellation of a plurality of reception subcarriers using the inverse matrix, A step of estimating transmission subcarriers multiplexed in space (interference cancellation unit), a step of calculating reliability of an inverse matrix (reliability calculation unit), a pilot carrier is extracted from the estimated transmission subcarrier, and the extracted pilot Weighting carrier according to reliability (weighting calculation unit), and weighted pyro Based on the bets carrier, the step of correcting the error contained in the plurality of received subcarriers (fluctuation estimation section).

  Preferably, the step of correcting the error (variation estimation unit) uses a weighted pilot carrier to estimate the instantaneous variation of the demodulation error (instantaneous variation estimation unit) and the estimated instantaneous variation of the demodulation error. Accordingly, a step (carrier correction unit) for correcting the estimated transmission subcarrier is included. Alternatively, the step of correcting the error (variation estimation unit) is a step of estimating the time variation of the demodulation error related to the pilot carrier using the weighted pilot carrier (time variation estimation unit), and the estimated time variation is the frequency direction. And estimating the time variation of the demodulation error for each subcarrier (frequency direction interpolation unit), and correcting the estimated propagation coefficient based on the estimated time variation of the demodulation error for each subcarrier. A step (propagation coefficient correction unit) is included. Alternatively, all these configurations may be included. Note that, instead of the step of correcting the estimated transmission subcarrier (carrier correction unit), the step of correcting the estimated propagation coefficient (propagation coefficient correction unit) is estimated based on the instantaneous fluctuation of the estimated demodulation error. The propagation coefficient thus corrected may be corrected.

  Further, the present invention provides an OFDM signal composed of a plurality of subcarriers orthogonal to each other including a data carrier to which transmission data is assigned and a pilot carrier to which a known phase and amplitude are assigned, transmitted from a plurality of transmission antennas, respectively. The present invention is directed to an OFDM receiving method and apparatus for inputting via a plurality of spatial paths using a plurality of receiving antennas. And in order to solve the said subject, the OFDM receiving method and apparatus of this invention are equipped with the following steps (structure).

  Each of OFDM signals received by a plurality of receiving antennas is subjected to OFDM demodulation and separated into a plurality of receiving subcarriers (a plurality of OFDM demodulating units), and propagation coefficients of a plurality of spatial paths are estimated from the plurality of receiving subcarriers, respectively. A step (transmission path estimation unit), a step provided by the number of a plurality of reception subcarriers (a plurality of demodulation stages), and a step (reliability calculation unit) for calculating the reliability of the inverse matrix obtained in the plurality of demodulation stages ), Extracting a pilot carrier from transmission subcarriers estimated in a plurality of demodulation stages, weighting the extracted pilot carrier according to reliability (weighting calculation unit), and determining a demodulation error based on the weighted pilot carrier Estimate instantaneous fluctuations and temporal fluctuations of estimated demodulation errors for each subcarrier. Step of correcting the propagation coefficient estimated based on the instantaneous variation and time variation (fluctuation estimation section).

  The steps (plurality of demodulation stages) provided by the number of the plurality of reception subcarriers are respectively a step of calculating an inverse matrix of a propagation matrix having an estimated propagation coefficient as an element (inverse matrix calculation unit), an inverse matrix Is used to cancel interference of a plurality of reception subcarriers, to estimate a transmission subcarrier multiplexed in space (interference cancellation unit), to demodulate the estimated transmission subcarrier to obtain transmission data (data demodulation) Part), remodulating the demodulated transmission data to generate a plurality of subcarriers (data modulation part), multiplying the remodulated subcarriers by the corrected propagation coefficient to generate a replica signal Generating a subtraction signal obtained by subtracting the replica signal from a plurality of received subcarriers. As new plurality of received subcarriers, comprising the step (calculation unit) to be output to the interference canceller of the subsequent demodulation stage.

  According to the present invention, the pilot carrier is weighted with the reliability of the inverse matrix of the propagation coefficient. As a result, the estimation error due to noise enhancement can be suppressed, and the error between transmission and reception can be detected with high accuracy. Further, the separated data carrier is corrected based on the detected instantaneous fluctuation of the transmission / reception error. Thereby, demodulation errors can be reduced. Further, the inverse matrix is obtained after correcting the estimated propagation coefficient based on the detected temporal variation of the transmission / reception error. As a result, it is possible to follow the fluctuation of the propagation path and improve the signal separation accuracy by inverse matrix multiplication. Furthermore, by generating a replica signal using the corrected propagation coefficient, error correction for each stage necessary for iterative decoding can be performed collectively.

Hereinafter, an OFDM receiving apparatus of the present invention will be described with reference to the drawings. In each embodiment, an example will be described in which an OFDM receiving apparatus having two receiving antennas performs 2 × 2 MIMO-OFDM transmission with an OFDM transmitting apparatus having two transmitting antennas. The OFDM transmitter is the conventional OFDM transmitter 200 shown in FIG.
First, prior to description of the OFDM receiver according to each embodiment of the present invention, the OFDM transmitter 200 will be described.

(Transmission frame transmitted by OFDM transmitter)
The OFDM transmitter 200 transmits the transmission frame 1 and the transmission frame 2 as shown in FIG. From the transmission antennas 206 and 207, different data sequences are simultaneously transmitted to perform spatial multiplexing. A transmission frame 1 transmitted from the transmission antenna 206 includes a preamble 601, a training symbol 603, and a data symbol 605. A transmission frame 2 transmitted from the transmission antenna 207 includes a preamble 602, a training symbol 604, and a data symbol 606.

  The preambles 601 and 602 are known signals used for frame synchronization, frequency synchronization, clock synchronization, AGC, symbol synchronization, and the like, and the signal form is not particularly limited as long as they are suitable for these applications. The preambles 601 and 602 may be different signals for the transmission antennas 206 and 207, respectively.

Training symbols 603 and 604 are known signals for estimating the propagation coefficient h _{j, i} of each transmission path between the transmitting and receiving antennas in order to separate the spatially multiplexed transmission signals. In order to estimate the propagation coefficient h _{j, i} , symbols that are orthogonal to each other in terms of time, frequency, code, or a combination thereof may be used for the training symbols 603 and 604.
For example, the training symbol 603 is transmitted only from the transmission antenna 206 at time t1, and the training symbol 604 is transmitted from only the transmission antenna 207 at time t2. Thus, the OFDM receiver can estimate h _1,1 and h _2,1 from the training symbol 603 at time t1, and can estimate h _1,2 and h _2,2 from the training symbol 604 at time t2.

In this embodiment, since an OFDM signal is used as a training symbol, transmission is performed for each subcarrier. For example, the odd number subcarriers of the training symbol 603 are transmitted from the transmission antenna 206, and the even number subcarriers of the training symbol 604 are transmitted from the transmission antenna 207. Next, the order is changed and the even-numbered subcarriers of the training symbol 603 are transmitted from the transmitting antenna 206, and the odd-numbered subcarriers of the training symbol 604 are transmitted from the transmitting antenna 207. Thereby, in the OFDM receiver, h _{j, i} can be estimated independently for each subcarrier.

  Data symbols 605 and 606 are OFDM signals obtained by orthogonally multiplexing a plurality of subcarriers on the frequency axis. FIG. 23 shows an example of the OFDM signal. In FIG. 23, one data symbol is composed of a plurality (17) of subcarriers, and among these, a predetermined phase and amplitude are assigned to a predetermined subcarrier to be a pilot carrier 702. A subcarrier other than the pilot carrier 702 is assigned a phase and amplitude based on transmission data to be a data carrier 701.

FIG. 24 shows an example of subcarriers in a transmission frame. In FIG. 24, T _{i, x, y} represents a training carrier, D _{i, x, y} represents a data carrier, P _{i, x, y} represents a pilot carrier, and null represents a carrier with zero amplitude. i is a transmission antenna number, x is a symbol number, and y is a subcarrier number. T _{i, x, y} and P _{i, x, y} assign a known phase and amplitude, and D _{i, x, y} assign a phase and amplitude based on the transmitted data.

Training symbols 603 and 604 are composed of two OFDM symbols. In the first OFDM symbol, odd-numbered subcarriers (T _1,1,1 , T _1,1,3 ,..., T _1,1,17 ) are transmitted from the number 1 transmitting antenna. Even-numbered subcarriers (T _2,1,2 , T _2,1,4 ,..., T _2,1,16 ) are transmitted from the antenna. In the second OFDM symbol, even-numbered subcarriers (T _1,2,2 , T _1,2,4 ,..., T _1,2,16 ) are transmitted from number 1 to the number 2 transmission antenna. subcarriers odd numbers from the antenna _{(T 2,2,1, T 2,2,3, ...} , T 2,2,17) are transmitted respectively.

Data symbols 605 and 606 are composed of L OFDM symbols. Each OFDM symbol consists of 13 data carriers and 4 pilot carriers. In this example, the subcarrier number of the pilot carrier is always fixed (P _{i, x, 3} , P _{i, x, 7} , P _{i, x, 11} , P _{i, x, 15} ), but is shown in FIG. As such, it may be changed for each symbol.

  In FIG. 18, data modulation section 201 generates a time waveform of data symbols by orthogonally multiplexing a plurality of subcarriers by OFDM modulation sections 202 and 203 based on the data sequence transmitted from each transmission antenna 206 and 207. . For orthogonal multiplexing, inverse Fourier transform, inverse wavelet transform, inverse discrete cosine transform, or the like can be used. OFDM modulation sections 202 and 203 generate a transmission frame by adding a preamble and a training symbol to a data symbol sequence converted into a time waveform. The generated transmission frames are converted into radio frequencies by the frequency converters 204 and 205 and transmitted from the transmission antennas 206 and 207 at the same time.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of an OFDM receiving apparatus 100 according to the first embodiment of the present invention. In FIG. 1, an OFDM receiving apparatus 100 according to the first embodiment includes receiving antennas 101a and 101b, frequency conversion units 103a and 103b, OFDM demodulation units 105a and 105b, a transmission path estimation unit 107, and an inverse matrix calculation. Unit 109, interference cancellation unit 110, reliability calculation unit 111, weighting calculation unit 112, instantaneous fluctuation estimation unit 113, carrier correction unit 114, and data demodulation unit 115.
Hereinafter, the operation of each component of the OFDM receiver 100 and the OFDM reception method performed by the OFDM receiver 100 will be described in detail.

Signals (transmission frames 1 and 2) transmitted from the OFDM transmitter are received by the reception antennas 101a and 101b and input to the frequency conversion units 103a and 103b, respectively. The frequency converters 103a and 103b convert the signals (received frames 1 and 2) received by the receiving antennas 101a and 101b into time-axis signals in a frequency band suitable for subsequent signal processing, respectively. The arithmetic expression of the frequency-converted time axis signal can be expressed by the following expressions (4) and (5), and the image of the received frame can be expressed as shown in FIG. TS _{i, x} indicates a transmission OFDM symbol, and RS _{j, x} indicates a reception OFDM symbol. H _{j, i} represents the transmission path characteristic in time axis representation.
RS1 _{, x} = _H1,1 * TS1 _{, x} + _H1,2 * TS2 _{, x} (4)
RS _{2, x} = H _2,1 × TS _{1, x} + H _2,2 × TS _{2, x} (5)

The OFDM demodulation units 105a and 105b separate the frequency-converted time axis signal into subcarriers, respectively, and convert them into frequency axis signals. Specifically, the OFDM demodulating units 105a and 105b perform reception gain adjustment, frame synchronization, frequency synchronization, and symbol synchronization based on the preamble signal of the transmission frame, and subtract each OFDM symbol based on the detected symbol timing. Separate into carriers. For this processing, inverse transformation of transformation used for orthogonal multiplexing, for example, Fourier transformation, wavelet transformation, discrete cosine transformation, or the like can be used. This frequency axis signal arithmetic expression can be expressed by the following expressions (6) and (7). The image of the received frame can be expressed as shown in FIG. TSC _{i, x, y} indicates a transmission subcarrier, and SC _{j, x, y} indicates a reception subcarrier.
SC1 _{, x, y} = _{h1,1, y} × TSC1 _{, x, y} + _{h1,2, y} × TSC2 _{, x, y} (6)
SC2 _{, x, y} = _{h2,1, y} × TSC1 _{, x, y} + _{h2,2, y} × TSC2 _{, x, y} (7)

The transmission path estimation unit 107 estimates a propagation coefficient h _{j, i} for each transmission path between the transmission and reception antennas from the separated reception subcarriers SC _{j, x, y} . Here, it is assumed that the propagation coefficient for each subcarrier is h _{j, i, y} . Since the training carrier T _{i, x, y} is transmitted from only one antenna, the received training carrier, ie, the received subcarrier SC _{j, x, y ,} is divided by the known transmission training carrier T _{i, x, y.} Thus, the propagation coefficient h _{j, i, y} can be obtained. That is, the following equation (8) is obtained. An image of the received frame can be expressed as shown in FIG.
h _{j, i, y} = SC _{j, x, y} / T _{i, x, y} (8)

On the other hand, reception subcarriers (data carrier and pilot carrier) of data symbols are simultaneously transmitted from a plurality of antennas and multiplexed in space. As an algorithm for separating a spatially multiplexed signal, a ZF (zero forcing) algorithm or an MMSE (least square error) algorithm is used. In these algorithms, by obtaining the inverse matrix W of the propagation matrix H having the propagation coefficient h _{j, i} as an element and multiplying the received signal, the influence of the propagation coefficient on the signal to be separated and other interference components The influence of the signal can be canceled. There are various methods for obtaining the inverse matrix depending on the algorithm. For example, W = (H ^H H) ⁻¹ H ^H called a general inverse matrix is used as the inverse matrix in the ZF algorithm. Where ^H is the Hermitian transpose of the matrix. In the MMSE algorithm, W = (H ^* H ^T + zI) ⁻¹ H ^* is used. Here, ^* represents the complex conjugate, ^T represents the transpose, z represents the noise power, and I represents the unit matrix.

Inverse matrix calculation unit 109, the subcarrier y for propagation coefficients h _{j, i,} using the propagation matrix H _y to the _y components, inverse matrix W _y of propagation matrix H _y - Request (= H _y ¹⁾ . The propagation matrix H _y and the inverse matrix W _y are obtained by the following equations (9) and (10). The image of the received frame can be expressed as shown in FIG.

このように、伝搬行列Ｈが完全に推定できかつＨ―¹が存在すれば、Ｈ^-HＨ＝１の単位行列となって多重された信号を分離することができる。分離後の信号は、送信時の振幅が再生され伝送路特性の影響がなくなる。 The interference cancellation unit 110, received subcarrier SC j of data _{symbols, x,} perform interference cancellation using the inverse matrix W _y a _y, transmit spatially multiplexed subcarriers TSC _{'i, x,} and _y are separate. This separation is performed by multiplying a reception signal matrix having reception subcarriers SC _{j, x, y} as elements for each subcarrier by an inverse matrix W _y as shown in the following equation (11).

In this way, if the propagation matrix H can be completely estimated and H- ¹ exists, the multiplexed signal can be separated as a unit matrix of H ^-H H = 1. The separated signal reproduces the amplitude at the time of transmission and eliminates the influence of the transmission path characteristics.

  Hereinafter, a data symbol sequence including separated subcarriers is referred to as a stream. A data symbol sequence transmitted from the transmission antenna number i = 1 is referred to as stream 1, and a data symbol sequence transmitted from the transmission antenna number i = 2 is referred to as stream 2. An image of the data symbol can be expressed as shown in FIG.

However, the received signal includes a synchronization error between transmission and reception, a propagation coefficient estimation error, and noise. The synchronization error between transmission and reception includes, for example, carrier frequency error, clock frequency error, symbol timing error, phase noise, and the like, and includes errors that could not be detected or corrected in synchronization with the preamble. The estimation error of the propagation coefficient includes an estimation error in the training symbol, an error due to a change in the transmission path after estimation with the training symbol, and the like. When the propagation coefficient is h, the transmission / reception error is e, the transmission signal is s, and the noise is n, the reception signal R can be expressed by the following equation (12).
r = h × e × s + n (12)

Here, when the reception function R is multiplied by w as an inverse function that becomes w × h = 1 with respect to the propagation coefficient h, the following equation (13) is obtained.
w × R = w × h × e × s + w × n = e × s + w × n (13)
That is, the estimated transmission signal includes an error between transmission and reception and noise with an inverse function. Originally, the received signal has a constant noise level and the received signal level varies depending on the propagation environment. However, the transmission signal estimated to perform such an operation is the original transmission amplitude regardless of the level of the received signal. Normalized to the level of the signal, the noise will be amplified by an inverse function.

Here, the transmission / reception error e is divided into an instantaneous fluctuation component ea due to phase noise and the like, and a time fluctuation component eb that is slow with respect to a signal such as a clock frequency error and a transmission path fluctuation. (14).
e = eb × ea (14)
In the present invention, the above-described problem is solved by correcting one or both of the instantaneous fluctuation component ea and the time fluctuation component eb. In the first embodiment, an example of correcting only the instantaneous variation component ea is shown, and other examples are shown in the second and third embodiments.

For this reason, the amplitude of the subcarriers (data carrier and pilot carrier) output from the interference cancellation unit 110 is normalized, and noise is amplified based on the inverse matrix of the propagation matrix. When the difference between the estimated pilot carrier and the transmission pilot carrier is obtained in order to obtain the transmission / reception error e, a large error occurs due to the amplified noise, and there is a possibility that the transmission / reception error e cannot be estimated accurately.
Therefore, in the present invention, the transmission / reception error e estimated from the estimated pilot carrier is weighted according to the inverse function w, and the data carrier is corrected based on this weighting.

The reliability calculation unit 111 calculates the reliability q _{i, y} of the subcarrier TSC ′ _{i, x, y} separated based on the inverse matrix W for each subcarrier. As the reliability, SNR (signal-to-noise ratio), SINR (signal-to-interference noise ratio), or the like may be obtained. For example, when the inverse matrix in the ZF algorithm is used, if the elements of the inverse matrix W are w _1,1 , w _1,2 , w _2,1 and w _2,2 , the signal from the transmission antenna number i = 1. The reliability q _{1, y} for the signal and the reliability q _{2, y} for the signal from the transmission antenna number i = 2 may be expressed by the following equations (15) and (16). FIG. 7 is an image diagram of reliability output.
q _{1, y} = 1 / (| w _1,1 | ² + | w _1,2 | ² ) (15)
q _{2, y} = 1 / (| w _2,1 | ² + | w _2,2 | ² ) (16)

In addition, when the MMSE algorithm is used, the interference component is not completely canceled, and therefore SINR may be obtained. For example, using the inverse matrix W obtained from the propagation matrix H by the MMSE algorithm, the reliability q _{1, y} and q _{2, y} may be expressed by the following equations (17) and (18).
q _{1, y} = | (w _1,1 × h _1,1 + w _1,2 × h _2,1 ) /
(1-w _1,1 × h _1,1 + w _1,2 × h _2,1 ) | (17)
q _{2, y} = | (w _2,1 × h _1,2 + w _2,2 × h _2,2 ) /
(1-w _2,1 × h _1,2 + w _2,2 × h _2,2 ) | (18)

The reliability may be obtained based on the determinant or eigenvalue of the propagation matrix H. Depending on the condition of the propagation matrix H, the inverse matrix cannot be obtained, and the signal may not be separated. For example, if the determinant of the propagation matrix H is “0”, the inverse matrix of the propagation matrix H cannot be obtained. In such a case, the reliability after separation may be set to “0”. Alternatively, the reliability may be lowered as the determinant is smaller. For example, in the ZF algorithm, the reliability may be obtained based on the determinant of H ^H H. Alternatively, if the propagation matrix H can be converted into a diagonal matrix having eigenvalues as diagonal elements by eigenvalue decomposition, the transmission signal multiplied by the eigenvalue is received, and the eigenvalue is used as the reliability. It can also be used.

The weight calculation unit 112 extracts only specific subcarriers (typically pilot carriers) from the separated subcarriers TSC ′ _{i, x, y,} and the reliability q _{i, y} obtained by the reliability calculation unit 111. Based on the above, the extracted specific subcarriers are weighted. Then, the difference between the weighted specific subcarrier and the known specific subcarrier is obtained to estimate the transmission / reception error e _{i, x, y} . For example, subcarriers TSC ′ _{i, x, y} after separation of numbers y = 3, 7, 11, and 15 are pilot carriers P ′ _{i, x, y} , and reliability q _{i, y} is expressed in SNR. In this case, the weighted pilot carrier P ″ _{i, x, y} and the transmission / reception error e _{i, x, y} are expressed by the following equations (19) and (20). 8 and 9 are image diagrams of the received stream after the weighting process.
P ″ _{i, x, y} = P ′ _{i, x, y} × q _{i, y} (19)
e _{i, x, y} = P ″ _{i, x, y} / P _{i, x, y} (20)

The significance of this weighting process will be described with reference to FIGS. 10A to 10D.
Each pilot carrier P _{i, x, y} (FIG. 10A) transmitted from the OFDM transmitter side with a predetermined amplitude and phase changes in amplitude and phase depending on the transmission path characteristic h _{i, j} , and changes in reception level. Adds noise (FIG. 10B). For this reason, when the amplitude and phase are normalized by eliminating the components of the transmission line characteristics h _{i, j} using the inverse matrix calculation, the noise components are separated from each separated pilot carrier P by the reliability of the inverse matrix, the reception error, and the like. ' _{i, x, y} will be different (Fig. 10C). Therefore, the separated pilot carrier P ′ _{i, x, y} amplitude is weighted with the reliability of the inverse matrix W _y . Thereby, normalization in which the noise component included in each pilot carrier is constant can be realized (FIG. 10D).

The instantaneous fluctuation estimation unit 113 estimates the instantaneous fluctuation error ea of each data symbol from the weighted transmission / reception error ei _{, x, y of} each pilot carrier. Examples of the instantaneous fluctuation error ea include phase rotation due to phase noise that occurs in common to all subcarriers in a symbol, residual frequency error due to carrier frequency estimation error, and the like. For example, a weighted average of estimated transmission / reception errors for each symbol of each stream is set as an instantaneous variation error ea _{i, x, y} of all subcarriers in the symbol. The following equation (21) is an equation for obtaining the instantaneous variation error ea _{i, x, y} . FIG. 11 is an image diagram of instantaneous variation error.
ea _{i, x, y} = (e _{i, x, 3} + e _{i, x, 7} + e _{i, x, 11} + e _{i, x, 15} ) / 4 (21)

The carrier correction unit 114 corrects each separated subcarrier TSC ′ _{i, x, y} based on the instantaneous variation error ea _{i, x, y} for each symbol. For example, the data carrier D ″ _{i, x, y} is corrected as D ″ _{i, x, y} / ea _{i, x, y} . The corrected data carrier is demodulated into transmission data by the data demodulator 115.

  As described above, according to the OFDM receiving method and the OFDM receiving apparatus according to the first embodiment of the present invention, the pilot carrier is weighted with the reliability of the inverse matrix of the propagation coefficient. As a result, the estimation error due to noise enhancement can be suppressed, and the error between transmission and reception can be detected with high accuracy. Further, the separated data carrier is corrected based on the detected instantaneous fluctuation of the transmission / reception error. Thereby, demodulation errors can be reduced.

(Second Embodiment)
FIG. 12 is a block diagram showing a configuration of an OFDM receiving apparatus 120 according to the second embodiment of the present invention. In FIG. 12, the OFDM receiver 120 according to the second embodiment includes receiving antennas 101a and 101b, frequency converters 103a and 103b, OFDM demodulators 105a and 105b, a transmission path estimator 107, and propagation coefficient correction. Unit 128, inverse matrix calculation unit 109, interference cancellation unit 110, reliability calculation unit 111, weight calculation unit 112, time variation estimation unit 126, frequency direction interpolation unit 127, and data demodulation unit 115. Prepare.

As shown in FIG. 12, the OFDM receiving apparatus 120 according to the second embodiment replaces the instantaneous fluctuation estimating unit 113 and the carrier correcting unit 114 of the OFDM receiving apparatus 100 according to the first embodiment with time fluctuation estimation. The configuration includes a unit 126, a frequency direction interpolation unit 127, and a propagation coefficient correction unit 128. Other configurations of the OFDM receiving apparatus 120 according to the second embodiment are the same as those of the OFDM receiving apparatus 100 according to the first embodiment, and the same reference numerals are given and description thereof is omitted.
Hereinafter, the operation of each component of the OFDM receiver 120 and the OFDM reception method performed by the OFDM receiver 120 will be described in detail.

The time fluctuation estimation unit 126 estimates the time fluctuation error eb of each data symbol from the weighted transmission / reception error e _{i, x, y of} each pilot carrier. As the time fluctuation error eb, there are a phase error due to a clock frequency error and an error due to a transmission path fluctuation. Therefore, for example, the time variation error eb can be estimated by averaging each pilot carrier in the symbol direction (time direction). The time variation error eb _{i, x, y} of each pilot carrier (y = 3, 7, 11, 15) is obtained by the following equation (22). FIG. 13 is an image diagram of the time variation error of each pilot carrier. Note that the averaging period k may be set as appropriate in accordance with the situation of time fluctuation.
eb _{i, x, y} = (e _{i, x, y} + e _{i, x-1, y} + e _{i, x-2, y} + ... + e _{i, xk, y} ) / k (22)

The frequency direction interpolation unit 127 interpolates the time variation error eb _{i, x, y} (y = 3, 7, 11, 15) obtained for each pilot carrier in the frequency direction (linear interpolation, least square interpolation, spline interpolation). Etc.) or averaging to estimate the time variation error eb _{i, x, y} for each data carrier. For example, in the case of averaging, the time fluctuation error eb _{i, x, y} for each data carrier can be obtained by the following equation (23). FIG. 14 is an image diagram of the time variation error of each data carrier.
eb _{i, x, y} = (eb _{i, x, 3} + eb _{i, x, 7} + eb _{i, x, 11} + eb _{i, x, 15} ) / 4 (23)

The propagation coefficient correction unit 128 corrects the propagation coefficient h _{j, i} estimated by the transmission path estimation unit 107 with the time variation error eb _{i, x, y} to obtain a corrected propagation coefficient h ′ _{j, i} . In this example, as shown in the following equation (24), the propagation coefficient h _{j, i, y} for each subcarrier is multiplied by the time variation error eb _{i, x, y} to obtain a corrected propagation coefficient h ′ _{j, i. , y} is obtained.
h ′ _{j, i, y} = h _{j, i, y} × eb _{i, x, y} (24)

  As described above, according to the OFDM receiving method and OFDM receiving apparatus according to the second embodiment of the present invention, the pilot carrier is weighted with the reliability of the inverse matrix of the propagation coefficient. As a result, the estimation error due to noise enhancement can be suppressed, and the error between transmission and reception can be detected with high accuracy. Further, the inverse matrix is obtained after correcting the estimated propagation coefficient based on the detected temporal variation of the transmission / reception error. As a result, it is possible to follow the fluctuation of the propagation path and improve the signal separation accuracy by inverse matrix multiplication.

(Third embodiment)
FIG. 15 is a block diagram showing a configuration of an OFDM receiver 130 according to the third embodiment of the present invention. In FIG. 15, the OFDM receiver 130 according to the third embodiment includes receiving antennas 101a and 101b, frequency conversion units 103a and 103b, OFDM demodulation units 105a and 105b, a transmission path estimation unit 107, and propagation coefficient correction. Unit 128, inverse matrix calculation unit 109, interference cancellation unit 110, reliability calculation unit 111, weighting calculation unit 112, instantaneous variation estimation unit 113, carrier correction unit 114, time variation estimation unit 126, A frequency direction interpolation unit 127 and a data demodulation unit 115 are provided.

  The OFDM receiver 130 according to the third embodiment is configured to correct the instantaneous variation component ea described in the first embodiment and the configuration to correct the time variation component eb described in the second embodiment. It is the structure provided with both.

  Therefore, the OFDM receiving method and OFDM receiving apparatus according to the third embodiment of the present invention can reduce the demodulation error and correct the detected data carrier by correcting the separated data carrier based on the instantaneous fluctuation of the detected transmission / reception error. The inverse matrix is obtained after correcting the propagation coefficient estimated based on the time variation of the transmission / reception error, so that it is possible to follow the propagation path variation and improve the signal separation accuracy by inverse matrix multiplication.

(Fourth embodiment)
FIG. 16 is a block diagram showing a configuration of an OFDM receiver 140 according to the fourth embodiment of the present invention. In FIG. 16, the OFDM receiver 140 according to the fourth embodiment includes receiving antennas 101a and 101b, frequency converters 103a and 103b, OFDM demodulators 105a and 105b, a channel estimation unit 107, and an inverse matrix calculation. Unit 109, interference cancellation unit 110, reliability calculation unit 111, weight calculation unit 112, fluctuation estimation unit 131, and data demodulation unit 115. The fluctuation estimation unit 131 includes an instantaneous fluctuation estimation unit 113, a time fluctuation estimation unit 126, a frequency direction interpolation unit 127, a calculation unit 134, and a propagation coefficient correction unit 128.

In the OFDM receiving apparatus 130 according to the third embodiment, the instantaneous variation error ea is corrected for the data carrier D ″ _{i, x, y} for each reception stream, but the OFDM according to the fourth embodiment The receiving apparatus 140 corrects the instantaneous variation error ea by correcting the propagation coefficient h _{j, i, y} and then obtaining the inverse matrix W and separating it into the received stream.

In interference cancellation of data symbols, the interference cancellation unit 110 first separates received pilot carriers P ′ _{j, x, y} . Next, for each subcarrier of the received pilot carrier, an inverse matrix Wy is obtained from the propagation coefficient h _{j, i, y} , and interference of the received pilot carrier is canceled. Next, the instantaneous variation error ea _{i, x, y} and the time variation error eb _{i, x, y} are obtained using the separated pilot carrier P ″ _{i, x, y} as in the first embodiment. Based on this, the propagation coefficient h _{j, i, y} is corrected _, the inverse matrix of the data carrier is obtained from the corrected propagation coefficient h ′ _{j, i, y} , and the received data carrier D ′ _{j, x, y} Interference cancellation is performed to separate the data carrier D ″ _{i, x, y} for each stream.

  As described above, according to the OFDM receiving method and OFDM receiving apparatus according to the fourth embodiment of the present invention, the inverse matrix is obtained after updating the propagation coefficient for each symbol. For this reason, it is possible to reduce an inverse matrix calculation error and an interference cancellation error due to a propagation coefficient estimation error, and it is possible to perform stream separation more accurately.

(Fifth embodiment)
FIG. 17 is a block diagram showing a configuration of an OFDM receiver 150 according to the fifth embodiment of the present invention. In FIG. 17, an OFDM receiving apparatus 150 according to the fifth embodiment includes receiving antennas 101a to 101c, frequency conversion units 103a to 103c, OFDM demodulation units 105a to 105c, a transmission path estimation unit 107, and reliability calculation. Unit 111, weight calculation unit 112, fluctuation estimation unit 131, and demodulation stages 400a to 400c. Each demodulation stage 400a to 400c includes an inverse matrix calculation unit 109, an interference cancellation unit 110, a data demodulation unit 115, a data modulation unit 401, and a replica generation unit 402.

  As illustrated in FIG. 17, the OFDM receiver 150 according to the fifth embodiment includes each demodulation stage 400a to 400c that includes the inverse matrix calculation unit 109, the interference cancellation unit 110, the data demodulation unit 115, and new data. The modulation unit 401 and the replica generation unit 402 are different from the above embodiments. The configuration other than the data modulation unit 401 and the replica generation unit 402 of the OFDM receiver 150 according to the fifth embodiment is the same as the configuration described in each of the above embodiments, and the description is given with the same reference numerals. Omitted.

  In the fifth embodiment, a plurality of demodulation stages corresponding to the number of streams multiplexed on the transmission signal are prepared, and a reception performance improving function based on instantaneous fluctuation and time fluctuation of transmission / reception errors is provided with a plurality of demodulation stages. A configuration that is shared with each other is adopted. Hereinafter, assuming that a signal in which three streams are multiplexed is received by three antennas, processing performed by the OFDM receiver 150 will be described.

First, in the first first demodulation stage 400a, three streams multiplexed from each received signal are separated using an inverse matrix of a propagation matrix. One stream with the highest reliability is selected from the three separated streams, and a replica signal of the selected stream included in each received signal is generated using a propagation matrix. As the reliability, for example, SNR, SINR, whether there is a demodulation error, or the like can be used. Then, the component of the selected stream is removed by subtracting the generated replica signal from each received signal.
In the next second demodulation stage 400b, since one stream is removed in the first demodulation stage 400a, it is only necessary to separate two streams from each received signal. Then, a replica signal of the stream with the highest reliability of the two separated streams is generated and subtracted from each received signal to remove the stream component.
In the final third demodulation stage 400c, since two streams are removed in the first and second demodulation stages 400a and 400b, it is only necessary to separate one remaining stream from each received signal. With these processes, the reception SNR can be improved.

This will be specifically described. In the first demodulation stage 400a, three streams are separated from each received signal, as in the third embodiment. An error between transmission and reception is estimated using the pilot carrier P ″ _{i, x, y} of each stream, and h ′ _{j, i, y} is corrected so that the propagation coefficient h _{j, i, y} follows the error between transmission and reception. The data demodulating unit 115 demodulates the three separated streams to obtain transmission data, and the data demodulating unit 115 selects the stream i having the highest reliability among the three streams, and this stream. The transmission data of i is output as a demodulation result of the first demodulation stage 400a, and the data modulation unit 401 remodulates the transmission data output from the data demodulation unit 115 to _generate subcarriers D _{ri, x, y} and P _ri, Generate _{x, y} .

The replica generation unit 402 multiplies each subcarrier of the stream i re-modulated by the data modulation unit 401 by the corrected propagation coefficient h ′ _{j, i, y,} and then replicates the replica signal of the stream i for each reception antenna j. Is generated. For example, the replica data carrier RD _{j, x, y} and the replica pilot carrier RP _{j, x, y} can be expressed by the following equations (25) and (26).
RD _{j, x, y} = h ′ _{j, i, y} × D _{ri, x, y} (25)
RP _{j, x, y} = h ′ _{j, i, y} × P _{ri, x, y} (26)

The replica data carrier RD _{j, x, y} and replica pilot carrier RP _{j, x, y} generated in this way are subtracted from the received subcarrier subcarrier D ′ _{j, x, y} and P ′ _{j, x, y} , and the second demodulation is performed. The reception subcarrier of stage 400b is used (the following equations (27) and (28)).
D ′ _{j, x, y} = D ′ _{j, x, y} −RD _{j, x, y} (27)
P ′ _{j, x, y} = P ′ _{j, x, y} −RP _{j, x, y} (28)
In this way, the second and third demodulation stages 400b and 400c are sequentially processed to separate and demodulate all the streams i.

  As described above, according to the OFDM receiving method and the OFDM receiving apparatus according to the fifth embodiment of the present invention, since the propagation coefficient is updated so as to follow the error between transmission and reception and the fluctuation of the transmission path, the replica signal Can improve the accuracy. As a result, errors that occur when the replica signal is removed from the received signal can be reduced, and the streams can be separated more accurately. Furthermore, since it is not necessary to correct the error of the stream separated for each demodulation stage, the calculation amount and the circuit scale can be reduced.

The present invention is not limited to receiving and separating a MIMO signal, but also receives a signal in which a single OFDM signal and an interference signal overlap or a signal in which a MIMO signal and an interference signal overlap, It can also be used when suppressing.
In order to apply the present invention to the case where the interference signal is suppressed, for example, when the inverse matrix W is obtained from the propagation matrix H using the MMSE algorithm, the inter-antenna covariance matrix R _uu of the interference signal is used instead of zI. The inverse matrix W may be obtained by using W = (H ^* H ^T + R _uu ) ⁻¹ H ^* . For the process of correcting the error based on the pilot carrier, the method described in each embodiment can be used.

Note that all or some of the functional blocks constituting the OFDM receivers according to the first to fifth embodiments of the present invention are typically integrated circuits such as LSIs (ICs, system LSIs depending on the degree of integration). (Referred to as super LSI or ultra LSI). These may be individually made into one chip, or may be made into one chip so as to include a part or all of them.
Further, the method of circuit integration is not limited to LSI's, and implementation using dedicated circuitry or general purpose processors is also possible. Also, an FPGA (Field Programmable Gate Array) that can be programmed after manufacturing the LSI, or a reconfigurable processor that can reconfigure the connection and setting of circuit cells inside the LSI may be used. The calculation of these functional blocks can also be performed using, for example, a DSP or a CPU. Also, these processing steps can be processed by being recorded on a recording medium as a program and executed.
Furthermore, if integrated circuit technology comes out to replace LSI's as a result of the advancement of semiconductor technology or a derivative other technology, it is naturally also possible to carry out function block integration using this technology. There is a possibility of adaptation of biotechnology.

  INDUSTRIAL APPLICABILITY The present invention can be used for a wireless transmission / reception apparatus using the OFDM transmission method, and in particular, even when performing multiplex communication between transmission and reception via a plurality of paths by space division, the transmission path is correctly estimated and the carrier error This is useful when the user wants to demodulate without any problem.

The block diagram which shows the structure of the OFDM receiver 100 which concerns on the 1st Embodiment of this invention Image diagram of received frames output by frequency converters 103a and 103b Image diagram of received frames output from OFDM demodulation sections 105a and 105b The figure explaining the process performed in the transmission path estimation part 107 The figure explaining the process performed in the inverse matrix calculation part 109 The figure explaining the process performed in the interference cancellation part 110 The figure explaining the process performed in the reliability calculation part 111 The figure explaining the process performed in the weight calculation part 112 The figure explaining the process performed in the weight calculation part 112 Diagram explaining the significance of weighting processing Diagram explaining the significance of weighting processing Diagram explaining the significance of weighting processing Diagram explaining the significance of weighting processing The figure explaining the process performed in the instantaneous fluctuation estimation part 113 The block diagram which shows the structure of the OFDM receiver 120 which concerns on the 2nd Embodiment of this invention. The figure explaining the process performed in the time fluctuation estimation part 126 The figure explaining the process performed in the frequency direction interpolation part 127 The block diagram which shows the structure of the OFDM receiver 130 which concerns on the 3rd Embodiment of this invention. The block diagram which shows the structure of the OFDM receiver 140 which concerns on the 4th Embodiment of this invention. The block diagram which shows the structure of the OFDM receiver 150 which concerns on the 5th Embodiment of this invention The block diagram which shows the structure of the conventional OFDM transmitter 200 The block diagram which shows the structure of the conventional OFDM receiver 220 Diagram showing an example of a transmission frame The figure which shows an example of the subcarrier of a received frame The block diagram which shows the structure of the conventional OFDM receiver 230 Diagram showing an example of an OFDM symbol The figure which shows an example of the subcarrier of a transmission frame The figure which shows an example of arrangement | positioning of a pilot carrier

Explanation of symbols

100, 120, 130, 140, 220, 230 OFDM receivers 101a-101c, 206-209 Antennas 103a-103c, 204, 205, 210, 211 Frequency converters 105a-105c, 212, 213 OFDM demodulators 107, 214 Transmission Path estimation unit 109, 215 Inverse matrix calculation unit 110, 216 Interference cancellation unit 111 Reliability calculation unit 112 Weight calculation unit 113 Instantaneous fluctuation estimation unit 114 Carrier correction unit 115, 217 Data demodulation unit 601, 602 Preamble 603, 604 Training symbol 605 , 606 Data symbol 126 Time variation estimation unit 127 Frequency direction interpolation unit 128 Propagation coefficient correction unit 131 Variation estimation unit 134 Operation unit 200 OFDM transmitter 201, 401, 504 Data modulation unit 202, 203 O DM modulation unit 400a to 400c, 506-508 demodulation stage 402,505 replica generating unit 501 pilot extracting section 502 the phase error estimator 503 corrects unit 701 data carriers 702 pilot carrier

Claims (12)

An OFDM signal consisting of a plurality of subcarriers orthogonal to each other, including a data carrier to which transmission data is allocated and a pilot carrier to which a known phase and amplitude are allocated, is transmitted from at least one transmission antenna using a plurality of reception antennas OFDM receiving method for receiving via a plurality of spatial paths,
OFDM demodulating OFDM signals received by the plurality of receiving antennas, respectively, and separating the OFDM signals into a plurality of receiving subcarriers;
Estimating propagation coefficients of the plurality of spatial paths from the plurality of reception subcarriers, respectively.
Calculating an inverse matrix of a propagation matrix having the estimated propagation coefficient as an element;
Performing interference cancellation of the plurality of reception subcarriers using the inverse matrix and estimating transmission subcarriers multiplexed in space;
Calculating a reliability of the inverse matrix;
Extracting the pilot carrier from the estimated transmission subcarrier and weighting the extracted pilot carrier according to the reliability;
And correcting an error included in the plurality of reception subcarriers based on the weighted pilot carrier. The correcting step includes
Estimating instantaneous variation in demodulation error using the weighted pilot carrier;
2. The OFDM reception method according to claim 1, further comprising a step of correcting the estimated transmission subcarrier according to an instantaneous fluctuation of the estimated demodulation error. The correcting step includes
Using the weighted pilot carrier to estimate the time variation of demodulation error for the pilot carrier;
Interpolating the estimated time variation in the frequency direction to estimate the time variation of the demodulation error for each subcarrier; and
The OFDM reception method according to claim 1, further comprising: correcting the estimated propagation coefficient based on a time variation of the estimated demodulation error for each subcarrier. The correcting step includes
Estimating instantaneous variation in demodulation error using the weighted pilot carrier;
Correcting the estimated transmission subcarriers in response to instantaneous variations in the estimated demodulation error;
Using the weighted pilot carrier to estimate the time variation of demodulation error for the pilot carrier;
Interpolating the estimated time variation in the frequency direction to estimate the time variation of the demodulation error for each subcarrier; and
The OFDM reception method according to claim 1, further comprising: correcting the estimated propagation coefficient based on a time variation of the estimated demodulation error for each subcarrier. The correcting step includes
Estimating instantaneous variation in demodulation error using the weighted pilot carrier;
Using the weighted pilot carrier to estimate the time variation of demodulation error for the pilot carrier;
Interpolating the estimated time variation in the frequency direction to estimate the time variation of the demodulation error for each subcarrier; and
2. The OFDM according to claim 1, further comprising: correcting the estimated propagation coefficient based on the estimated instantaneous variation of the demodulation error and the estimated temporal variation of the demodulation error for each subcarrier. Reception method. An OFDM signal composed of a plurality of subcarriers orthogonal to each other including a data carrier to which transmission data is allocated and a pilot carrier to which a known phase and amplitude are allocated is transmitted from each of a plurality of transmission antennas. OFDM receiving method for inputting via a plurality of spatial paths,
A first step of OFDM-demodulating OFDM signals received by the plurality of reception antennas, respectively, and separating the OFDM signals into a plurality of reception subcarriers;
A second step of estimating propagation coefficients of the plurality of spatial paths from the plurality of reception subcarriers, respectively;
A third step of calculating an inverse matrix of a propagation matrix having the estimated propagation coefficient as an element;
A fourth step of performing interference cancellation of the plurality of reception subcarriers using the inverse matrix and estimating transmission subcarriers multiplexed in space;
A fifth step of calculating the reliability of the inverse matrix;
A sixth step of extracting the pilot carrier from the estimated transmission subcarrier and weighting the extracted pilot carrier according to the reliability;
Based on the weighted pilot carriers, instantaneous fluctuation of demodulation error and time fluctuation of the estimated demodulation error for each subcarrier are estimated, and the estimated propagation coefficient is corrected based on the instantaneous fluctuation and time fluctuation. A seventh step;
An eighth step of demodulating the estimated transmission subcarriers to obtain transmission data;
A ninth step of remodulating the demodulated transmission data to generate a plurality of subcarriers;
A tenth step of generating a replica signal by multiplying the re-modulated subcarriers by the corrected propagation coefficient;
An eleventh step of generating a subtraction signal obtained by subtracting the replica signal from the plurality of received subcarriers;
The OFDM reception characterized in that the sub-signal is used as a plurality of new reception subcarriers and the process from the third step to the eleventh step is repeated for the number of the reception subcarriers. Method. An OFDM signal consisting of a plurality of subcarriers orthogonal to each other, including a data carrier to which transmission data is allocated and a pilot carrier to which a known phase and amplitude are allocated, is transmitted from at least one transmission antenna using a plurality of reception antennas An OFDM receiver for receiving via a plurality of spatial paths,
A plurality of OFDM demodulation units for OFDM-demodulating OFDM signals received by the plurality of reception antennas, respectively, and separating the OFDM signals into a plurality of reception subcarriers;
A transmission path estimator for estimating propagation coefficients of the plurality of spatial paths from the plurality of reception subcarriers;
An inverse matrix calculation unit for calculating an inverse matrix of a propagation matrix having the estimated propagation coefficient as an element;
An interference cancellation unit that performs interference cancellation of the plurality of reception subcarriers using the inverse matrix and estimates transmission subcarriers multiplexed in space;
A reliability calculation unit for calculating the reliability of the inverse matrix;
A weight calculating unit that extracts the pilot carrier from the estimated transmission subcarrier and weights the extracted pilot carrier according to the reliability;
An OFDM receiver comprising: a fluctuation estimation unit that corrects an error included in the plurality of reception subcarriers based on the weighted pilot carrier. The fluctuation estimation unit
Using the weighted pilot carrier, an instantaneous fluctuation estimation unit for estimating an instantaneous fluctuation of a demodulation error;
The OFDM receiving apparatus according to claim 7, further comprising: a carrier correction unit that corrects the estimated transmission subcarrier according to an instantaneous fluctuation of the estimated demodulation error. The fluctuation estimation unit
Using the weighted pilot carrier, a time fluctuation estimation unit for estimating a time fluctuation of a demodulation error related to the pilot carrier;
A frequency direction interpolation unit that interpolates the estimated time variation in the frequency direction and estimates time variation of a demodulation error for each subcarrier;
The OFDM receiving apparatus according to claim 7, further comprising: a propagation coefficient correction unit that corrects the estimated propagation coefficient based on a time variation of the estimated demodulation error for each subcarrier. The fluctuation estimation unit
Using the weighted pilot carrier, an instantaneous fluctuation estimation unit for estimating an instantaneous fluctuation of a demodulation error;
A carrier correction unit that corrects the estimated transmission subcarrier according to the instantaneous fluctuation of the estimated demodulation error;
Using the weighted pilot carrier, a time fluctuation estimation unit for estimating a time fluctuation of a demodulation error related to the pilot carrier;
A frequency direction interpolation unit that interpolates the estimated time variation in the frequency direction and estimates time variation of a demodulation error for each subcarrier;
The OFDM receiving apparatus according to claim 7, further comprising: a propagation coefficient correction unit that corrects the estimated propagation coefficient based on a time variation of the estimated demodulation error for each subcarrier. The fluctuation estimation unit
Using the weighted pilot carrier, an instantaneous fluctuation estimation unit for estimating an instantaneous fluctuation of a demodulation error;
Using the weighted pilot carrier, a time fluctuation estimation unit for estimating a time fluctuation of a demodulation error related to the pilot carrier;
A frequency direction interpolation unit that interpolates the estimated time variation in the frequency direction and estimates time variation of a demodulation error for each subcarrier;
8. A propagation coefficient correction unit that corrects the estimated propagation coefficient based on the estimated instantaneous fluctuation of the demodulation error and the estimated temporal fluctuation of the demodulation error for each subcarrier. The OFDM receiver according to the description. An OFDM signal composed of a plurality of subcarriers orthogonal to each other including a data carrier to which transmission data is allocated and a pilot carrier to which a known phase and amplitude are allocated is transmitted from each of a plurality of transmission antennas. An OFDM receiver that inputs via a plurality of spatial paths,
A plurality of OFDM demodulation units for OFDM-demodulating OFDM signals received by the plurality of reception antennas, respectively, and separating the OFDM signals into a plurality of reception subcarriers;
A transmission path estimator for estimating propagation coefficients of the plurality of spatial paths from the plurality of reception subcarriers;
A plurality of demodulation stages provided by the number of the plurality of reception subcarriers;
A reliability calculation unit for calculating the reliability of the inverse matrix obtained in the plurality of demodulation stages;
A weight calculation unit that extracts the pilot carrier from transmission subcarriers estimated in the plurality of demodulation stages, and weights the extracted pilot carrier according to the reliability;
Based on the weighted pilot carriers, instantaneous fluctuation of demodulation error and time fluctuation of the estimated demodulation error for each subcarrier are estimated, and the estimated propagation coefficient is corrected based on the instantaneous fluctuation and time fluctuation. A fluctuation estimation unit,
Each of the plurality of demodulation stages is
An inverse matrix calculation unit for calculating an inverse matrix of a propagation matrix having the estimated propagation coefficient as an element;
An interference cancellation unit that performs interference cancellation of the plurality of reception subcarriers using the inverse matrix and estimates transmission subcarriers multiplexed in space;
A data demodulator that demodulates the estimated transmission subcarrier to obtain transmission data;
A data modulation unit that remodulates the demodulated transmission data to generate a plurality of subcarriers;
A replica generation unit that generates a replica signal by multiplying the re-modulated subcarriers by the corrected propagation coefficient;
A subtracting signal obtained by subtracting the replica signal from the plurality of reception subcarriers, and a calculation unit that outputs the generated subtraction signal as a plurality of new reception subcarriers to the interference cancellation unit of a subsequent demodulation stage; Including an OFDM receiver.

Images