JP4734080B2 - OFDM receiver for performing channel estimation correction - Google Patents

Description

  The present invention relates to digital signal transmission, and more particularly, to a channel equalization technique in an orthogonal frequency division multiplex transmission system (OFDM (Orthogonal Frequency Division Multiplex) transmission system).

  The orthogonal frequency division multiplexing (OFDM) scheme is widely used in the field of broadband mobile transmission such as terrestrial digital broadcasting and program material transmission because it is strong against multipath. In the OFDM method, data to be transmitted is allocated to a plurality of subcarriers arranged at equal intervals on the frequency axis, and individual subcarriers are phase-modulated (PSK) or quadrature amplitude modulation (QAM) according to the allocated data. The time axis information (valid symbol) is generated by performing inverse Fourier transform (IDFT, its high-speed version is IFFT).

  A major feature of the OFDM system is that a guard signal for multipath countermeasures is added before an effective symbol. The guard signal is a copy of the portion after the effective symbol. By adding such a guard signal, the connection between the guard signal and the effective symbol becomes continuous. When there are multipaths, inter-symbol interference occurs because the previous irrelevant effective symbol overlaps during the guard signal period. No interference occurs. For this reason, a transmission signal can be reproduced by extracting effective symbols and performing channel equalization.

  In general, a channel equalization method is a method in which a transmitter transmits a known signal, a receiver receives the signal, estimates a channel response from the received signal, and applies the inverse characteristic to the received signal for equalization. It is. As a method of inserting the known signal into the transmission signal, a specific subcarrier (pilot carrier) is modulated with information of the known signal. For example, in the transmission system of terrestrial digital broadcasting in Non-Patent Document 1, a pilot carrier called a scattered pilot (SP) embedded in 4 OFDM symbol periods is used for channel estimation. Further, in the method for program material transmission of Non-Patent Document 2, in order to support high-speed mobile transmission, a pilot carrier called a continuous pilot (CP) embedded at the same position for each OFDM symbol is used for channel estimation. Use.

  On the other hand, a MIMO (Multiple Input Multiple Output) transmission system that uses a plurality of transmission antennas and reception antennas to transmit different information at the same frequency has been actively studied in recent years. By using the OFDM scheme as the MIMO modulation scheme, a MIMO transmission scheme that is resistant to multipath can be realized. In this case, the channel estimation is performed by a transmitter transmitting a unique known signal that is different for each transmitter in a time division multiplexed manner on a pilot carrier (see Patent Document 1) or by code division multiplexing (see Patent Literature 1). (See Document 2), the receiver solves the multiplex and obtains the channel response of each unique known signal.

ARIB-STD B31, "Transmission method for digital terrestrial television broadcasting" ARIB-STD B33, “Portable OFDM digital wireless transmission system for transmitting television broadcast program material” JP 2005-57650 A JP 2005-124125 A

  In a SISO (Single Input Single Output) -OFDM system in which transmission and reception are one-sided, when channel estimation is performed using only a pilot carrier of one OFDM symbol, a channel response including a noise component added to the pilot carrier is included. Will be considered. For this reason, especially when the S / N ratio is small, data may be erroneously equalized by channel equalization, and the error rate may deteriorate. As a countermeasure, channel estimation is performed by averaging pilot carrier values for several OFDM symbols. Thereby, a noise component can be suppressed.

  On the other hand, in the MIMO-OFDM system, when the unique known signal of each transmitter is time-division multiplexed on the pilot carrier, channel estimation can be performed only when the known signal of the transmitter is the time-division multiplexed OFDM symbol. Absent. Therefore, during the OFDM symbol period until the next channel estimation can be performed, channel equalization must be performed using the same channel response matrix obtained by the channel estimation. In addition, when a unique known signal of each transmitter is code division multiplexed on a pilot carrier, it is usually code division multiplexed over a plurality of OFDM symbols. Requires a symbol.

  In this way, channel equalization of individual OFDM symbols is performed using channel responses obtained from a plurality of OFDM symbols, or channel equalization of a plurality of OFDM symbols is performed using a channel response obtained from one OFDM symbol. When performing, it is a premise that the channel state (propagation environment) can be regarded as the same among the plurality of OFDM symbols. Further, it is assumed that the frequency and phase of the transmission side are correctly tracked on the reception side by automatic frequency adjustment AFC (Automatic Frequency Control). When these assumptions are not satisfied, the constellation of the channel-equalized OFDM signal rotates and expands and contracts as shown in FIG.

  Accordingly, the present invention has been made to solve such a problem, and the object thereof is to correct the constellation when the constellation rotates or expands and suppresses the deterioration of the error rate characteristic. An object of the present invention is to provide an OFDM receiver. Another object of the present invention is to suppress deterioration of error rate characteristics by correcting constellation by signal processing when the number of multiplexing of time division multiplexing or code division multiplexing increases in a MIMO-OFDM system. As a result, an object of the present invention is to provide an OFDM receiver capable of increasing the number of multiplexing.

[Principle of the present invention]
The above problem is due to the fact that the channel response, which is the result of channel estimation, is not properly determined for each OFDM symbol that acts and equalizes it. The present invention solves the above problem by correcting the channel estimation result (channel response) as described below. As shown in FIG. 10, the present invention performs channel estimation once with a plurality of OFDM symbols (step S1001), and determines whether there is a pilot carrier modulated with known information for each OFDM symbol (step S1002). When there are pilot carriers that are roughly divided into two correction functions (first and second functions and third function) and are modulated with known information for each OFDM symbol, two more (first function) And second function) (steps S1003 to 1005). Then, the channel estimation result in step 1001 is corrected using the correction values obtained in steps 1003 to 1005 (step S1006).

  The first correction function (step S1003) performs channel estimation once again on the pilot carrier channel equalization result for each OFDM symbol to obtain a correction value. That is, the ratio (error) between the symbol points on the constellation of pilot carriers that have been channel equalized by the conventional method and the symbol points that should be originally detected is detected, and the errors for all subcarriers are interpolated from this. The correction value is obtained by processing. Then, the channel response is corrected by multiplying the channel response estimated by the conventional method by this correction value. The advantage is that a fine channel estimation deviation can be corrected at every pilot carrier interval on the frequency axis.

  The second correction function (step S1004) forms a group of adjacent subcarriers for each OFDM symbol, and averages the ratio (error) of the pilot carrier channel equalization result with the known symbol point for the correction value. Ask for. In the subcarrier grouping, as shown in FIG. 11, one group is formed for all subcarriers, or a plurality of groups are formed in units of a plurality of adjacent subcarriers. Since this correction function can correct a common channel estimation deviation within a group, a smaller channel estimation deviation on the frequency axis can be corrected by reducing the number of subcarriers within the group. However, as described later, since it is necessary to average the pilot carriers in the group in order to detect the channel estimation deviation, it is desirable that there are many pilot carriers in the group. In addition, since channel estimation deviation caused by frequency deviation and phase noise is a common value for all subcarriers, all subcarriers are excluded except when the propagation environment fluctuates and fine channel estimation deviation occurs on the frequency axis. It can be fully applied by creating one group. In this correction function, in addition to the processing being relatively simple, the equalization result is not subjected to noise enhancement.

  The third correction function (step S1005) is a function when there is no pilot carrier modulated with known information for each OFDM symbol, and the number of modulation multilevels such as TMCC (Transmission and Multiplexing Configuration Control) for each OFDM symbol. A hard decision is performed on the channel equalization result of a small subcarrier, and a correction value is obtained by averaging the ratio (error) between the equalization result and the hard decision result within the subcarrier group. Since the number of such subcarriers is generally small, the correction value is obtained by averaging substantially all such subcarriers.

但し、

[Explanation using mathematical formulas]
The principle of the present invention will be described using mathematical expressions by taking a MIMO-OFDM system with two transmission systems and four reception systems as an example. For the i-th subcarrier, if Y _i is a received signal vector, X _i is a transmitted signal vector, H _i is a channel response matrix, and N _i is a noise vector, this system can be expressed as equation (1). Can do.

However,

Here, each element of the vector and the matrix is as follows. As for Y _i , y _1i represents a received signal from the receiver 1, y _2i represents a received signal from the receiver 2, y _3i represents a received signal from the receiver 3, and y _4i represents a received signal from the receiver 4. For H _i , h _jki indicates the channel response between receiver j and transmitter k. Regarding X _i , X _1i indicates a transmission signal of the transmitter 1, and X _2i indicates a transmission signal of the transmitter 2. As for N _i , n _1i represents the noise component of the received signal 1, n _2i represents the noise component of the received signal 2, n _3i represents the noise component of the received signal 3, and n _4i represents the noise component of the received signal 4.

If the channel estimation is performed correctly (if H _i is estimated correctly) can be carried out in that manner, channel equalization multiplying H _i ^-1 from the left to the equation (1). The channel equalization result is as shown in equation (2). Here, the second term on the right side of the equation (2) is a noise component remaining after equalization.

に推定されたとすると、その逆行列を（１）式に左から掛けたチャンネル等化結果は、（４）式のようになり、コンスタレーション上のシンボル点がずれてしまう。

When channel estimation results (channel response matrix) obtained from a plurality of OFDM symbols are used for channel equalization of individual OFDM symbols, and conversely, channel estimation results obtained from one OFDM symbol are converted to a plurality of OFDM symbol channels. When there is an error in the channel estimation result, such as when used for equalization, channel equalization as in equation (2) cannot be performed. The channel response shift for the transmitter 1 is Δh _1i , the channel response shift for the transmitter 2 is Δh _2i , and the channel response matrix is

As a result, the channel equalization result obtained by multiplying the inverse matrix by (1) from the left is as shown in (4), and the symbol points on the constellation are shifted.

Here, when the elements of the inverse matrix of the channel response matrix H _i are expressed as in equation (5), each element in equation (4) becomes as in equations (6-1) and (6-2). .

In the first and second correction functions according to the present invention, Δh _1i and Δh _2i are detected using a pilot carrier. Since the values x _1i and x _2i are known for the pilot carrier, the equations (6-1) and (6-2) are respectively divided by x _1i and x _2i .

（第２の補正機能及び第３の補正機能について）

（第１の補正機能について）

（第２の補正機能及び第３の補正機能について）

By multiplying the estimated channel response matrix by the diagonal matrix (correction value matrix) of the equations (9-1) and (9-2) having the correction value as a diagonal element from the right, the equation (10-1) and As shown in equation (10-2), a substantially correct channel response matrix can be obtained. The channel response matrix corrected by Equations (10-1) and (10-2) is used for ZF (zero Forcing) and MLD (Maximum Likelihood Detection), which is close to the case where channel estimation is performed correctly. Demodulation results can be obtained.
(About the first correction function)

(About the 2nd correction function and the 3rd correction function)

(About the first correction function)

(About the 2nd correction function and the 3rd correction function)

（第１の補正機能について）

（第２の補正機能及び第３の補正機能について）

(About the first correction function)

(About the 2nd correction function and the 3rd correction function)

  Thus, it can be confirmed by mathematical formulas that the present invention can be applied to MIMO-OFDM.

Therefore , a MIMO-OFDM receiver according to the present invention realizes the correction function 2 and is known in a MIMO-OFDM receiver that equalizes an OFDM symbol using a channel response matrix estimated using the OFDM symbol. An equalization means for performing channel equalization using a received signal vector of a pilot carrier modulated by information and a channel response matrix estimated from the received signal vector, and outputting a channel equalization result, and known information of the pilot carrier And a plurality of adjacent pilot carriers are grouped, and the channel equalization result element output by the equalization means is divided by the known information element of the pilot carrier read from the storage means. Each division result is obtained, and the pilot carrier A correction value calculation means for averaging the division results obtained for each group to obtain a correction value, and generating a diagonal matrix having the correction values obtained by the correction value calculation means as diagonal elements; Correction means for correcting the channel response matrix by multiplying the channel response matrix estimated from the signal vector by the diagonal matrix.

Further, the MIMO-OFDM receiver according to the present invention realizes the correction function 3 and performs modulation in the MIMO-OFDM receiver that equalizes the OFDM symbol by the channel response matrix estimated using the OFDM symbol. An equalization means for performing channel equalization using a received signal vector of a subcarrier having a small multi-value number and a channel response matrix estimated from a received signal vector of a pilot carrier modulated by known information , and outputting a channel equalization result The channel equalization result output by the equalization means is hard-determined, the hard-decision means for outputting the hard-decision equalization result, and the adjacent subcarriers are grouped and output by the equalization means The element of the channel equalization result is the element of the hard decision equalization result output by the hard decision means. Each of the division results is calculated to obtain a division result, and the division result obtained for each subcarrier is averaged for each group to obtain a correction value, respectively, and a correction value obtained by the correction value calculation means. Correction means for generating a diagonal matrix as a corner element, multiplying the channel response matrix estimated from the received signal vector by the diagonal matrix, and correcting the channel response matrix is provided.

  According to the present invention, the correction value is obtained using the known information of the pilot carrier or the equalization result obtained by making a hard decision on the subcarrier having a small modulation multi-value number, and the channel response is corrected by the correction value. An appropriate channel response can be determined for the OFDM symbol. Therefore, when the constellation rotates or expands / contracts, it is possible to correct the constellation and suppress deterioration of the error rate characteristics. Further, in the MIMO-OFDM system, it is possible to correct the constellation and suppress the deterioration of the error rate characteristic. As a result, it is possible to increase the number of multiplexing of time division multiplexing and code division multiplexing.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Embodiments 1 to 3 of the present invention are examples of a receiver applied to a MIMO-OFDM system with 2 transmission systems and 4 reception systems, and Embodiments 4 to 6 are SISO-OFDM systems with one transmission and one reception. It is an example of the receiver applied to. Further, the first and fourth embodiments realize the first correction function, the second and fifth embodiments realize the second correction function, and the third and sixth embodiments realize the third correction function. In any of the embodiments, the channel estimation correction circuit provided in the receiver corrects the channel response which is the channel estimation result, thereby realizing the constellation correction and suppressing the deterioration of the error rate characteristic.

  The variables shown in FIGS. 1 to 6 for explaining the embodiments 1 to 6 are the same as those used in the means for solving the aforementioned problems. Further, the same number is assigned to the blocks of the common means.

  First, Example 1 will be described. The first embodiment is an example in which the first correction function is applied to a MIMO-OFDM system with two transmission systems and four reception systems. FIG. 1 is an overall block diagram of a channel estimation correction circuit provided in the MIMO-OFDM receiver according to the first embodiment of the present invention. The MIMO-OFDM receiver 100 includes channel estimation means 6 and a channel estimation correction circuit 1.

  This interpolation processing is processing for calculating a correction value of a subcarrier by linear approximation or polynomial approximation using a correction value of a pilot carrier nearby. Specifically, the correction value matrix calculation means 3 sets correction values for subcarriers other than the pilot carrier to zero in advance, and corrects the correction values for the low-pass filter (digital filter) in the order of the subcarriers arranged on the frequency axis. And a correction value for all subcarriers is obtained as its output. In this case, the correction values for all subcarriers, which are the output of the correction value matrix calculation means 3, are values that vary smoothly based on the correction values of the pilot carriers.

  As described above, according to the MIMO-OFDM receiver 100 of the first embodiment, since the channel estimation correction circuit 1 performs the correction process for all the subcarriers, the fine details for each pilot carrier interval on the frequency axis. The channel estimation shift can be corrected. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed. In addition, the number of time division multiplexing or code division multiplexing can be increased.

  Next, Example 2 will be described. The second embodiment is an example in which the second correction function is applied to a MIMO-OFDM system with two transmission systems and four reception systems. FIG. 2 is an overall block diagram of a channel estimation correction circuit provided in the MIMO-OFDM receiver according to the second embodiment of the present invention. The MIMO-OFDM receiver 200 includes a channel estimation circuit 6 and a channel estimation correction circuit 7.

  As described above, according to the MIMO-OFDM receiver 200 of the second embodiment, the channel estimation correction circuit 7 obtains an average correction value for each subcarrier group, and performs correction processing for each group. , It is possible to correct the deviation of the common channel estimation within the group. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed. In addition, the number of time division multiplexing or code division multiplexing can be increased.

  Next, Example 3 will be described. The third embodiment is an example in which the third correction function is applied to a MIMO-OFDM system with two transmission systems and four reception systems. FIG. 3 is an overall block diagram of a channel estimation correction circuit provided in the MIMO-OFDM receiver according to the third embodiment of the present invention. The MIMO-OFDM receiver 300 includes channel estimation means 6 and a channel estimation correction circuit 10.

  Here, the channel equalization means 11 recognizes in advance the number of a subcarrier transmitted by a modulation scheme with a small modulation multi-level number such as TMCC or BPSK, and performs channel equalization based on the number of the subcarrier. Identify subcarriers.

Here, the hard decision means 14, by hard decision equalization result, the modulation scheme to identify the closest symbol point X _i of the symbol points are known in advance to the received signal.

  As described above, according to the MIMO-OFDM receiver 300 of the third embodiment, the channel estimation correction circuit 10 performs a hard decision on the channel equalization result of a subcarrier with a small modulation multi-level number such as TMCC, and performs correction. Since the correction processing is performed by obtaining the value, it is possible to correct the channel estimation deviation even when there is no pilot carrier. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed. In addition, the number of time division multiplexing or code division multiplexing can be increased.

  As described above, the first to third embodiments are examples in which the first to third correction functions are applied to the MIMO-OFDM system. However, these correction functions are also applied to the SISO-OFDM system in which transmission and reception are one-facing. can do. Hereinafter, the case where the first to third correction functions are applied to the SISO-OFDM system will be described using mathematical expressions.

[Description of correction function in SISO-OFDM system]
With respect to the i-th subcarrier, if x _i is a transmission signal, y _i is a reception signal, h _i is a channel response, and n _i is a noise component, this system can be expressed by the following equation (12).

If the channel estimation is performed correctly (if H _i is estimated correctly) can be carried out in that manner, channel equalization dividing both sides of (12) with h _i. The channel equalization result is as shown in equation (13). Here, the second term on the right side of the equation (13) is a noise component remaining after equalization.

When channel estimation results (channel responses) obtained from multiple OFDM symbols are used for channel equalization of individual OFDM symbols, and conversely, channel estimation results obtained from one OFDM symbol are equalized to channels of multiple OFDM symbols. When there is an error in the channel estimation result, such as in the case of using the channel equalization, channel equalization as in equation (13) cannot be performed. The difference in channel response that should originally be h _i is Δh _i , and the channel response is

In the first and second correction function to detect this Delta] h _i using pilot carriers. Since the value x _i is known for the pilot carrier, equation (15) is divided by x _i .

ここで、Ｅ［ｃ_ｉ］はｃ_ｉの期待値を示す。（１７）式の右辺の第２項は、独立な確率変数の割り算の期待値なので、その期待値は各々の期待値の割り算になる。ここで、雑音成分ｎ_ｉの期待値はゼロであり、ｈ_ｉとｘ_ｉの期待値は非ゼロであるので、（１７）式の右辺の第２項は結局ゼロになる。したがって、グループ内で共通の補正値は、

になる。

Here, E [c _i ] indicates the expected value of c _i . Since the second term on the right side of the equation (17) is an expected value for division of independent random variables, the expected value is a division of each expected value. Here, since the expected value of the noise component n _i is zero and the expected values of h _i and x _i are non-zero, the second term on the right side of the equation (17) is eventually zero. Therefore, the common correction value within the group is

become.

In the third correction function, since there is no pilot carrier modulated with known information for each symbol, Equation (15), which is an equalization result for a specific subcarrier having a small modulation multilevel number such as TMCC, is hard-decision and, put the _{X i.} Also in this case, assuming that Δh _i is a common Δh in a group unit composed of a plurality of subcarriers, a common correction value within the group is obtained by the equations (16) to (18) similarly to the second correction function. Ask.

（第２、第３の補正機能について）

By multiplying the estimated channel response by a correction value, a substantially correct channel response can be obtained as shown in Equations (19-1) and (19-2). Therefore, Equations (19-1) and (19-2) By using the channel response corrected by the above equation for a demodulation method such as ZF or MLD, a demodulation result close to the case where channel estimation is correctly performed can be obtained.
(About the first correction function)

(Second and third correction functions)

（第１の補正機能に対して）

（第２、第３の補正機能に対して）

以下、第１〜３の補正機能を、送信及び受信が１対向のＳＩＳＯ−ＯＦＤＭシステムに適用した受信機の例について説明する。

(For the first correction function)

(For the second and third correction functions)

Hereinafter, an example of a receiver in which the first to third correction functions are applied to a SISO-OFDM system in which transmission and reception are one-sided will be described.

  First, Example 4 will be described. The fourth embodiment is an example in which the first correction function is applied to a SISO-OFDM system in which transmission and reception are one-facing. FIG. 4 is an overall block diagram of a channel estimation correction circuit provided in the SISO-OFDM receiver according to the fourth embodiment of the present invention. The SISO-OFDM receiver 400 includes channel estimation means 20 and a channel estimation correction circuit 15.

ここで、補間処理は、実施例１に示したものと同様である。

Here, the interpolation processing is the same as that shown in the first embodiment.

  As described above, according to the SISO-OFDM receiver 400 of the fourth embodiment, the channel estimation correction circuit 15 performs the correction process for all the subcarriers. The channel estimation shift can be corrected. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed.

  Next, Example 5 will be described. The fifth embodiment is an example in which the second correction function is applied to a SISO-OFDM system in which transmission and reception are opposed to each other. FIG. 5 is an overall block diagram of a channel estimation correction circuit provided in the SISO-OFDM receiver according to the fifth embodiment of the present invention. The SISO-OFDM receiver 500 includes channel estimation means 20 and a channel estimation correction circuit 21.

  As described above, according to the SISO-OFDM receiver 500 of the fifth embodiment, the channel estimation correction circuit 21 obtains an average correction value for each subcarrier group and performs correction processing for each group. , It is possible to correct the deviation of the common channel estimation within the group. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed.

  Next, Example 6 will be described. The sixth embodiment is an example in which the third function is applied to a SISO-OFDM system in which transmission and reception are one-facing. FIG. 6 is an overall block diagram of a channel estimation correction circuit provided in the SISO-OFDM receiver according to the sixth embodiment of the present invention. The SISO-OFDM receiver 600 includes channel estimation means 20 and a channel estimation correction circuit 24.

  Here, the channel equalization means 25 recognizes in advance the number of the subcarrier transmitted in the modulation scheme with a small modulation multi-level number, like the channel equalization means 11 shown in the third embodiment, and The subcarrier for channel equalization is specified by the carrier number.

Here, the hard decision unit 28, like the hard decision means 14 shown in Example 3, the modulation scheme to identify the closest symbol point x _i of the symbol points are known in advance to the received signal.

  As described above, according to the SISO-OFDM receiver 600 of the sixth embodiment, the channel estimation correction circuit 24 performs a hard decision on the channel equalization result of the subcarrier with a small modulation multi-level number such as TMCC, and the correction. Since the correction processing is performed by obtaining the value, it is possible to correct the channel estimation deviation even when there is no pilot carrier. As a result, when the constellation rotates or expands / contracts, the constellation can be corrected and deterioration of the error rate characteristic can be suppressed.

  As described above, according to the first to sixth embodiments, an appropriate channel response matrix or channel response can be obtained for each OFDM symbol in terms of mathematical expressions. FIGS. 7 and 8 show examples of constellations when applied to a MIMO-OFDM system with two transmission systems and four reception systems. FIG. 7 shows an example of constellation when channel equalization is performed by channel estimation using four OFDM symbols by the conventional method, and FIG. 8 applies the present invention grouped into one for all subcarriers. This is an example of constellation when channel equalization is performed by correcting channel estimation. 7 and 8, (a) shows a case where the frequency of the transmitter 1 is offset by +1 kHz with respect to the receiver, and (b) shows a case where the frequency of the transmitter 2 is offset by -1 kHz with respect to the receiver. This is an example of the constellation. MER (Modulation Error Ratio) in the figure is a parameter indicating how far the received symbol point is from the original symbol point due to the noise component on the constellation, and the larger this value, the better the error rate. 7 and 8 can be confirmed that, as shown in FIG. 8, the rotation of the constellation is corrected without deterioration of the SN ratio and the MER is improved by applying the present invention. .

  The present invention has been described with reference to the embodiments. However, the present invention is not limited to the first to sixth embodiments, and various modifications can be made without departing from the technical idea thereof. For example, although the first to third embodiments are examples applied to a MIMO-OFDM system with two transmission systems and four reception systems, the number of transmission systems and the number of reception systems are not limited thereto. There are an infinite number of combinations of the number of transmission systems and the number of reception systems in the MIMO-OFDM system, but elements such as a transmission signal vector, a reception signal vector, a channel response matrix, and a correction value matrix are applied by applying the first to third embodiments. Can be applied to a system having an arbitrary number of systems other than 2 transmission systems and 4 reception systems.

It is a whole block diagram of the channel estimation correction circuit with which the MIMO-OFDM receiver of Example 1 by this invention was equipped. It is a whole block diagram of the channel estimation correction circuit with which the MIMO-OFDM receiver of Example 2 by this invention was equipped. It is a whole block diagram of the channel estimation correction circuit with which the MIMO-OFDM receiver of Example 3 by this invention was equipped. It is a whole block diagram of the channel estimation correction circuit with which the SISO-OFDM receiver of Example 4 by this invention was equipped. It is a whole block diagram of the channel estimation correction circuit with which the SISO-OFDM receiver of Example 5 by this invention was equipped. It is a whole block diagram of the channel estimation correction circuit with which the SISO-OFDM receiver of Example 6 by this invention was equipped. It is a figure which shows the prior art example of the constellation inclined by the frequency shift of transmission and reception. It is a figure which shows the example of the constellation by which inclination was correct | amended. It is a figure which shows the example of rotation and expansion / contraction of a constellation. It is a figure which shows the correction function and flowchart of this invention. It is a figure explaining grouping of a subcarrier.

Explanation of symbols

1, 7, 10, 15, 21, 24 Channel estimation correction circuit 2, 16 Pilot carrier equalization means 3, 8, 12 Correction value matrix calculation means 4, 9, 13 Channel response matrix correction means 5, 19 Pilot information storage means 6, 20 Channel estimation means 11, 25 Channel equalization means 14, 28 Hard decision means 17, 22, 26 Correction value calculation means 18, 23, 27 Channel response correction means 100, 200, 300 MIMO-OFDM receivers 400, 500 , 600 SISO-OFDM receiver

Claims (2)

  In a MIMO-OFDM receiver for channel equalizing an OFDM symbol using a channel response matrix estimated using the OFDM symbol,
  Equalization means for performing channel equalization using a received signal vector of a pilot carrier modulated by known information and a channel response matrix estimated from the received signal vector, and outputting a channel equalization result;
  Storage means for storing known information of the pilot carrier;
  A plurality of adjacent pilot carriers are grouped, and the channel equalization result element output by the equalization means is divided by the known information element of the pilot carrier read from the storage means to obtain the division results, respectively, Correction value calculation means for averaging the division results obtained for each pilot carrier for each group to obtain correction values;
  Correction means for generating a diagonal matrix having the correction value obtained by the correction value calculation means as a diagonal element, multiplying the channel response matrix estimated from the received signal vector by the diagonal matrix, and correcting the channel response matrix And a MIMO-OFDM receiver.   In a MIMO-OFDM receiver for channel equalizing an OFDM symbol using a channel response matrix estimated using the OFDM symbol,
  Equalization that performs channel equalization using the received signal vector of the subcarrier with a small number of modulation multilevels and the channel response matrix estimated from the received signal vector of the pilot carrier modulated by known information, and outputs the channel equalization result Means,
  Hard decision means for making a hard decision on the channel equalization result output by the equalization means and outputting the hard decision equalization result;
  The adjacent subcarriers are grouped, and the channel equalization result elements output by the equalization means are divided by the hard decision equalization result elements output by the hard decision means to obtain division results, respectively. Correction value calculation means for obtaining a correction value by averaging the division results obtained for each subcarrier for each group;
  Correction means for generating a diagonal matrix having the correction value obtained by the correction value calculation means as a diagonal element, multiplying the channel response matrix estimated from the received signal vector by the diagonal matrix, and correcting the channel response matrix And a MIMO-OFDM receiver.

Images