JP2007028092A - Transmission apparatus, transmission method, receiving apparatus, and receiving method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transmission apparatus for enhancing arithmetic efficiency of peak reduction processing, without losing the orthogonality between subcarriers, in a mobile communications system adopting the MIMO-OFDM system. <P>SOLUTION: The transmission apparatus includes a plurality of transmission antennas and transmits a signal with a multicarrier system. The transmission apparatus includes a division inverse Fourier transform means that derives a plurality of signal sequence groups from transmitted information, applies Fourier transformation to each signal sequence group, and outputs a plurality of sets of sub sequence groups, including one or more sub sequences after the transform; a peak reduction control means that detects the peak power of the plurality of sub sequences after inverse Fourier transformation and outputs an instruction signal for designating an interleave pattern; and an interleaving means that replaces one or more sub sequences between the plurality of sub sequence groups; in accordance with the instruction signal. The transmission apparatus is provided with a plurality of means for converting a group of sub-sequences, including the sub sequences replaced by the interleave means into a signal transmitted wirelessly from one transmission antenna. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a multi-antenna system, a multi-carrier communication apparatus, and a communication method.

  In mobile communication, a signal to be propagated is left to a multipath propagation environment, so it is necessary to sufficiently suppress intersymbol interference. In the Orthogonal Frequency Division Multiplexing (OFDM) scheme, intersymbol interference can be effectively suppressed if propagation delays of various paths are within the guard interval. However, a signal modulated by the OFDM method (IFFT output) often has a very large amplitude value compared to the average amplitude. As shown in FIG. 1, when each of a large number of subcarrier signal components is synthesized in the same phase, the added output of the signal at a certain point becomes very large, and a large peak is generated with respect to the average output. is there. The ratio between the average power and the peak power is called a peak-to-average power ratio (PAPR). The maximum peak power that can be generated may reach the average power multiplied by the number of subcarriers (the maximum number of subcarriers used for signal transmission).

  FIG. 2 shows general input / output characteristics of the transmission amplifier. As shown in the figure, since the input / output characteristics include a linear region and a nonlinear region, a large signal exceeding the linear region is amplified nonlinearly to generate a distortion component. The distortion component of the signal causes problems such as deterioration of transmission quality and increase of radiated power outside the band. Although it is conceivable to greatly expand the linear region, this causes a problem that the amplification efficiency is lowered. Accordingly, it is desirable that the amplitude (power) distribution of the transmission signal does not generate a large amplitude (PAPR is small) compared to the average value.

  On the other hand, a multi-antenna system that performs signal transmission using a plurality of transmitting antennas and a plurality of receiving antennas is known from the viewpoint of increasing frequency utilization efficiency. The multi-antenna system is also called a MIMO (Multi Input Multi Output) system. A mobile communication system having both advantages by combining the OFDM scheme and the MIMO scheme has also been proposed.

  3 and 4 show a transmitter and a receiver used in such a mobile communication system. The transmitter has N transmit antennas. The pre-modulation signal generation unit 31 generates transmission symbols before modulation in the OFDM scheme by performing error correction coding, interleaving, symbol mapping and the like on the input information bit sequence. The transmission symbol is converted into a parallel signal by the serial-parallel converter 32, fast inverse Fourier transformed by the IFFT unit 33, converted to a serial signal by the parallel-serial converter 34, and a guard interval is added by the GI adder 35, An OFDM symbol is generated. Thereafter, symbols modulated by the OFDM scheme for each transmission antenna are transmitted from each transmission antenna. FIG. 5 shows input / output signals of the series-parallel converter and the parallel-serial converter. The receiver has M receiving antennas. The GI removal unit 41 removes the guard interval from the received signal and outputs it. The output signal is converted into a parallel signal by the serial-parallel converter 42 and fast Fourier transformed by the FFT unit 43. Thereafter, the signal separation unit 44 and the signal detection unit 45 perform signal separation and detection, and the signals transmitted from the transmission antennas are restored.

  There are several approaches to address the problem of excessive PAPR. For example, the frequency domain interleaving method and clipping filtering method described in Non-Patent Document 1 and the Partial Sequence Transmission Method (PTS) described in Non-Patent Document 2 and Non-Patent Document 3 The cyclic shift method as described is known. In the MIMO-OFDM scheme, as described in Non-Patent Document 4, a method of applying the PAPR reduction method in OFDM to each transmitter is known.

  FIGS. 6 and 7 show 2 × 2 MIMO-OFDM transmitters and receivers, which use frequency domain interleaving. As shown in FIG. 6, interleaving is performed on the output signal of the series-parallel converter. The interleaver prepares a plurality of interleave patterns and generates a plurality of interleave output signal sequences for the same information symbol sequence. IFFT conversion and peak power detection are performed for each of the information bit 1 and 2 sequences, and the one with the smallest peak power is determined and determined as the signal sequence to be actually transmitted. In order to appropriately perform deinterleaving at the receiver, the used interleaving pattern is notified to the receiver through a control channel or the like.

FIG. 8 and FIG. 9 show a 2 × 2 MIMO-OFDM transmitter and receiver, respectively, showing configurations when using the PTS method and the cyclic shift method when all subcarriers are divided into two. The division inverse Fourier transform unit does not generate a time signal including signals from subcarriers 0 to 3 and a time signal including signals from subcarriers 4 to 7 in the same manner. The divided inverse Fourier transform unit has a configuration as shown in FIG. In the PTS method, the time signal is phase-rotated and then added to a part of the time signal. In the cyclic shift method, cyclic shift is performed on a part of time signals, and then addition is performed. A plurality of types of phase rotation amounts and phase shift amounts are prepared, and a plurality of transmission signal sequences are generated for one transmission signal sequence. Among the plurality of transmission signal sequences, the one with the smallest peak power is determined as the signal sequence to be actually transmitted.
X. Li and L.L. J. et al. Cimini, “Effects of clipping and filtering on the performance of OFDM”, IEEE Commun. Lett. , Vol. 2, no. 5, pp. 131-133, May, 1998. L. J. et al. and N.N. R. Sollenberger, “Peak-to-Average power ratio of an OFDM signal using personal transmission sequence,” IEEE Commun. Lett. , Vol. 4, no. 3, pp. 86-88, March, 2000. G. Hill and M.M. Jaulkner, “Cyclic shifting and time inversion of partial transmit sequence to reduce the peak-to-average ratio in OFDM”, PIMRC2000, vol. 2, pp. 1256-1259, Sept. 2000. Seung Hee Han and Jae Hong Lee, “An overview of peak-to-average power ratio reduction technology for W e s ri e s e, E s”. 12, Issue 2, pp. 56-65, Apr. 2005.

  If the conventional peak reduction method is applied to a MIMO-OFDM mobile communication system, it is necessary to suppress peak power for each antenna at each transmitter. Although interleave processing itself in the frequency domain may be relatively easy, it is necessary to perform inverse Fourier transform for the number of prepared interleave patterns, and the amount of computation of IFFT processing becomes very large. There is. In this case, it may be considered to perform interleaving on the signal after the division IFFT, but this causes a problem that orthogonality between subcarriers is lost. Also, in the PTS method, the process of adding phase rotation is complex multiplication, and the phase rotation process itself is more complicated than interleaving, which is merely a rearrangement of signals. Furthermore, when peak reduction processing is performed for each transmission antenna, the peak reduction effect may be reduced due to the small number of transmit signal candidates that can be generated.

  The present invention has been made in order to address at least one of the above-described problems, and its problem is to perform peak reduction processing without destroying orthogonality between subcarriers in a MIMO-OFDM mobile communication system. It is to provide a communication device and a communication method that improve calculation efficiency.

  In the present invention, a transmitter having a plurality of transmission antennas and transmitting a multicarrier signal is used. The transmitter derives a plurality of signal sequence groups from the information to be transmitted, performs inverse Fourier transform for each signal sequence group, and outputs a plurality of sets of sub-sequence groups including one or more sub-sequences after the conversion. Means, peak reduction control means for detecting peak power for a plurality of sub-sequences after inverse Fourier transform and outputting an instruction signal designating an interleave pattern, and one or more between a plurality of sub-sequence groups according to the instruction signal Interleaving means for substituting the subsequences. The transmitter is provided with a plurality of means for converting a group of sub-sequences including the sub-sequence replaced by the interleaving means into a signal that is wirelessly transmitted from one transmission antenna.

  According to the present invention, in the MIMO-OFDM mobile communication system, it is possible to improve the calculation efficiency of peak reduction processing without breaking the orthogonality between subcarriers.

  According to one aspect of the present invention, a plurality of time signals after the division IFFT processing are interleaved between the same subcarrier components although the sequences are different. As a result, interleaving can be performed without breaking the orthogonality between the subcarriers. Since various combinations of sequences after interleaving can be prepared, the IFFT processing does not need to be performed for the number of interleave patterns. Also, it is not essential to perform complex operations. For this reason, peak reduction processing can be performed with a small amount of calculation.

For example, in the case of a 4-stream MIMO-OFDM communication apparatus, if the number of divisions is 4, the number of interleave patterns for a certain information signal is (4!) ^(4-1) = 1.4 × 10 ⁴ transmission signals Candidates can be generated, and the one having the smallest peak can be selected, and a very large peak reduction effect can be expected. Further, by combining with the PTS method and the cyclic shift method, it is possible to obtain a larger peak reduction effect than when performing peak reduction processing independently for each transmission antenna.

  Interleaving may be performed between a plurality of sub-sequences associated with the same time among outputs after the division inverse Fourier transform.

  In peak reduction control, a predetermined number of upper peak powers in one sub-sequence are detected, signal components of other sub-sequences contributing to peak power are specified, and peak power for the plurality of sub-sequences is determined. It may be performed by detecting. Focusing on the upper predetermined number of peak components, among them, the interleave pattern is determined so that the maximum value of the power after addition is minimized, and the same procedure is repeated for other sequences, so that an arbitrary number of An interleave pattern can be efficiently determined for a subsequence group.

  In response to a control signal related to peak reduction, weight adjustment means may be provided that weights one or more sub-sequences after inverse Fourier transform. The peak power of each sub-sequence is detected, and the position of the IFFT point that provides the peak power is aligned, and the interleave pattern can be determined so that it is transmitted from one transmission antenna.

  In response to the peak reduction control signal, cyclic shift means may be provided for cyclically changing the correspondence relationship between one or more non-interleaved subsequences and one or more interleaved subsequences. The interleave pattern can be determined such that a signal giving the peak power of the sub-sequence is detected and a signal sequence having an opposite phase to the signal is transmitted from one transmission antenna.

  Both the weight adjusting means and the cyclic shift means may be provided. The interleave pattern may be determined so that the difference between the peak powers of the plurality of sub-sequences after the inverse Fourier transform becomes small.

  One set of sub-sequence groups may be derived by interleaving and adding the sub-sequences included in the two sets of sub-sequence groups after the inverse Fourier transform. Thereby, peak reduction control (particularly, determination of an interleave pattern) for an arbitrary number of subsequence groups can be performed efficiently.

  Interleaves one set of subsequences derived by interleaving and adding the subsequences included in the two sets of subsequences after inverse Fourier transform and the subsequences included in the other two sets of subsequences Then, one set of sub-sequence groups may be derived by performing interleaving and addition with another set of sub-sequence groups derived by addition.

  The interleave pattern may be distinguished by a pilot symbol mapping pattern in a transmission symbol. Also, the first pilot symbol that is not interleaved and the second pilot symbol that is interleaved may be included in the transmission symbol.

FIG. 11 shows a transmitter according to an embodiment of the present invention. Two parallel streams (information bit sequences) are prepared in the transmitter, and MIMO-OFDM wireless transmission is performed. First, for each information bit sequence, a signal s _n (n = 1, 2) before being modulated by the OFDM method is generated by the pre-modulation signal generation unit, and is input to the division inverse Fourier transform unit. . Here, S _n ^(d) represents the d-th sub-sequence among the outputs from the division inverse Fourier transform unit for the n-th stream (n = 1, 2). S _n represents a transmission signal sequence modulated from the OFDM system and transmitted from the n-th antenna. For example, if the number of subcarriers is P _s , the signal s ₁ input to the division inverse Fourier transform unit of the first stream can be expressed as follows.

s ₁ = [s ₁ (0), s ₁ (1),. . . , S ₁ (P _s −1)].

Further, the two signal series groups S ₁ ⁽¹⁾ and S ₁ ⁽²⁾ output from the divided inverse Fourier transform unit can be expressed by the following equations.

S ₁ ⁽¹⁾ = [S ₁ ⁽¹⁾ (0), S ₁ ⁽¹⁾ (1),. . . , S ₁ ⁽¹⁾ (P _s -1)]
= F ^-1 (s ₁ (0), ..., s ₁ (P _{s /} 2-1), 0, ..., 0).

S ₁ ⁽²⁾ = [S ₁ ⁽²⁾ (0), S ₁ ⁽²⁾ (1),. . . , S ₁ ⁽²⁾ (P _s -1)]
= F ⁻¹ (0,..., 0, s ₁ (P _{s /} 2-1),..., _{S 1} (P _s −1)).
Here, F ⁻¹ represents the inverse Fourier transform.

Similarly, output signal sequences S ₂ ⁽¹⁾ and S ₂ ⁽²⁾ of the divided inverse Fourier transform unit of the second stream are also generated. Incidentally, in a conventional OFDM transmitter in which peak reduction processing is not performed as in this embodiment, a signal S ₁ = S ₁ ⁽¹⁾ + S ₁ ⁽²⁾ modulated by the OFDM method is transmitted from the first antenna. Then, the signal S ₂ = S ₂ ⁽¹⁾ + S ₂ ⁽²⁾ is transmitted from the second antenna.

In this embodiment, after the combination of sub-sequences output from the division inverse Fourier transform unit is appropriately changed, they are input to the addition unit, and a transmission signal is generated. That is, subsequences are interleaved across antennas. By providing the sub-sequence interleaver between the division inverse Fourier transform unit and the addition unit, it is possible to appropriately select the antenna that transmits the sub-sequence. For one information signal sequence, candidates for signal sequences S ₁ and S ₂ transmitted from antennas 1 and ₂ can be expressed as follows.

インターリーブのパターンは、加算後の信号のピーク電力が小さくなるように決定される。即ち、信号系列Ｓのピーク電力をｐ（Ｓ）とすると、
ｍａｘ［ｐ（Ｓ_１ ^（１）＋Ｓ_１ ^（２）），ｐ（Ｓ_２ ^（１）＋Ｓ_２ ^（２））］
ｍａｘ［ｐ（Ｓ_１ ^（１）＋Ｓ_２ ^（２）），ｐ（Ｓ_２ ^（１）＋Ｓ_１ ^（２））］
から求められる加算後の信号のピーク電力の小さい組み合わせが、インターリーブのパターンとして選択される。

The interleaving pattern is determined so that the peak power of the signal after addition becomes small. That is, if the peak power of the signal sequence S is p (S),
_{^{max [p (S 1 (1}} ) + S 1 (2)), p (S 2 (1) + S 2 (2))]
_{^{max [p (S 1 (1}} ) + S 2 (2)), p (S 2 (1) + S 1 (2))]
A combination having a small peak power of the added signal obtained from the above is selected as an interleaving pattern.

  The adding unit of the transmitter adds the subsequences input thereto and synthesizes the signal into one signal modulated by the OFDM method. A signal modulated by the OFDM method is parallel-to-serial converted, a guard interval (GI) is added thereto, and a generated transmission signal is output. Since the signal components related to the same subcarrier are exchanged between the transmission units related to each antenna, orthogonality between subcarriers of the OFDM signal generated by each transmission unit can be maintained. Note that the determined or selected interleave pattern is notified to the receiver through the control channel.

  FIG. 12 shows a receiver according to an embodiment of the present invention. The interleave (IL) information detection unit detects information (deinterleave information) related to the interleave pattern from the signal transmitted through the control channel. The deinterleave information is given to the subsequence deinterleaver. Processing such as guard interval removal, serial-parallel conversion, fast Fourier transform (FFT), signal separation, demapping, error correction, and the like is performed in the same manner as that performed by a normal MIMO-OFDM receiver.

13 and 14 show a transmitter and a receiver when the number of transmission antennas and the number of reception antennas are N and M, respectively. Assuming that the number of subcarriers (the number of FFT sampling points) is P _s and the number of divisions is D, the signal sequence before modulation of the OFDM system of the n-th stream is s _n (0),. . . s _n (P _s −1). The signal series Sn output from the adding unit can be expressed as follows.

ここで、ｄは分割逆フーリエ変換部から出力される個々のサブ系列を指定する番号（パラメータ）である。送信機のピーク低減制御部では信号系列Ｓ_ｎのピーク電力が最小になるように、サブ系列の可能な全ての組み合わせの中から１つの組み合わせパターンを選択する。ｎ番目のアンテナに関する送信部に関するサブ系列に対して、インターリーブパターンｘ_ｎ（ｄ）が適用される。

Here, d is a number (parameter) that designates each sub-sequence output from the division inverse Fourier transform unit. As the peak power of the signal sequence S _n is minimized by the peak reducing control part of the transmitter selects one combination pattern among all the possible combinations of sub-sequences. The interleave pattern x _n (d) is applied to the sub-sequence related to the transmission unit related to the n-th antenna.

In the first embodiment, peaks are estimated for a series of all transmission signal candidates (outputs of the addition unit), and these are compared. In the second embodiment of the present invention, a comparison is made for l of all input signals. For all the sub-sequences, the sequence having the largest peak power is represented as S _np ^(dp) for convenience. Among all the signals, the time points of the signals whose signal power is 1st to 1st largest are t _p1,. . . , T _pl . in this case,

で求められるＮ個のピーク電力の中で、ピーク電力が最も小さくなる組み合わせＳ_ｎ’ ^（ｄ’）とＳ_ｎｐ ^（ｄｐ）を同じ送信部で送信するように、インターリーブパターンが決定される。本実施例によれば、比較対象数が限定されるので演算量を少なくすることができる。

The interleaving pattern is determined so that the combination S _{n ′} ^{(d ′)} and S _np ^(dp) having the smallest peak power among the N peak powers obtained in ^{(1 )} are transmitted by the same transmitter. According to the present embodiment, since the number of comparison targets is limited, the amount of calculation can be reduced.

  The third embodiment of the present invention simplifies the peak reduction processing method. It is assumed that the number of streams is N (N> 2) and two sequences of transmission signals for each stream are generated. First, peak power is detected for all 2 × N OFDM modulated subsequences, and a transmission signal sequence having the maximum peak power is determined. Next, a sub-sequence transmitted by the same transmission antenna as this transmission signal sequence is determined. This sub-sequence may be determined based only on the maximum peak power before combining, or the highest l-th peak power may be considered. The interleave pattern may be determined by repeating the same processing for other signal sequences.

  FIG. 15 shows a method for determining a pattern in the stream direction in the case of four streams divided into two. In step 1, the sub-sequence with the highest peak power is determined among the sub-sequences of each transmitter (in the example shown, it is assumed that this is the sub-sequence of d = 1 in stream 1). Next, the sequence is added to the sub-sequence of d = 2 of the transmission unit, respectively, and signal sequence candidates are synthesized. The one with the lowest peak power is selected from all the combined sequences, and the subsequence for d = 2 is determined (in the example shown, stream 3 is selected as the subsequence with d = 2). In this way, the first pair (d = 1 stream 1, d = 2 stream 3) is determined. In step 2 and subsequent steps, the same procedure is performed for those other than the determined sub-sequence. In the illustrated example, in step 2, the second pair (d = 1 stream 4, d = 2 stream 1) is determined. In step 3, the third pair (d = 1 stream 2, d = 2 stream 4) is determined. In step 3, the remaining fourth pair (d = 1 stream 3, d = 2 stream 2) is also determined. In this way, a combination effective for reducing the peak power is efficiently derived.

In the fourth embodiment of the present invention, a pattern determination method when the number of divisions is 2 or more will be described. FIG. 16 is a diagram for explaining a method in the case of eight divisions. First, with respect to the first and second sub-sequences (d = 1, 2) output from the divided inverse Fourier transform units of all transmission units, the interleave pattern of sub-sequence 2 (d = 2) peaks by the already described method. It is determined by the reduction processing unit (the one on the left side in the figure). Next, the output of this peak reduction processing unit and the sub-sequence of d = 3 are input to the peak reduction processing unit in the next stage. In this peak reduction processing unit, an interleave pattern of a sub-sequence of d = 3 is determined, and a composite signal of two sequences is output. Thereafter, the same procedure is repeated, and the interleave patterns of all sequences are determined for the sequence S _n ^(d) (d> 3). In the illustrated example, only two sequences are input to the peak reduction processing unit, but more subsequences may be input, and input / output sequences of one peak reduction processing unit and another peak reduction processing unit. The number may be different.

  Also in the fifth embodiment of the present invention, a pattern determination method when the division number D is 2 or more will be described. FIG. 17 is a diagram for explaining a method in the case of eight divisions. The first and second sub-sequences (total of 4 sequences) of the two streams are input to the peak reduction processing unit at the top of the left column, and these interleave patterns are determined. Is output. Also, the (3,4) th, (5,6) th,. . . , (D-1, D) th sub-sequences are input. Each peak reduction processing unit determines an interleave pattern for each input signal sequence. As a result, D / 2 signal sequences are generated for each transmission antenna.

  The peak reduction processing unit in the upper row of the second column from the left (center column) receives the signal series output from the first and second peak reduction processing units in the left column, and has already been described. Is done. Similar processing is performed in the peak reduction processing unit in the lower row of the center row. The same procedure is performed for the peak reduction processing unit in the third column from the left. Finally, the number of signal sequences is one per transmission unit, and it is transmitted from the transmission antenna.

  In the illustrated example, only two sequences are input to the peak reduction processing unit, but three or more sub-sequences may be input. The number of input / output sub-sequences of the peak reduction processing unit may be the same or different in a plurality of peak reduction processing units.

  As described above, the interleave pattern applied on the transmission side may be notified to the reception side through the control channel. The notification of the interleave pattern may be notified by a method other than using the control channel. In the sixth embodiment of the present invention, when the same interleave pattern is applied to a plurality of OFDM symbols, the interleave pattern is transmitted using pilot symbols. FIG. 18 shows a frame configuration used in this embodiment. FIG. 19 shows a receiver used in this embodiment. As shown in FIG. 18, the first pilot symbol is inserted at the head of each frame for each antenna. In addition, a second pilot symbol having a symbol sequence (may be one symbol) unique to each stream is also prepared. Peak reduction processing is not performed for the first pilot symbol, but peak reduction processing is performed for the second pilot symbol in the same manner as other symbols belonging to the same block.

  In the receiver as shown in FIG. 19, channel estimation is performed using the first pilot symbol derived from the output signal from the FFT unit. Based on this channel estimation value, signal separation of the second pilot symbol is performed. Since the second pilot symbol is unique to each transmission stream, it can be determined based on the received symbol to which block the signal received by each antenna belongs. This determination is performed by an interleave (IL) information calculation unit. Deinterleave information regarding each subcarrier is input to a subsequence deinterleaver. The sub-sequence deinterleaver restores the order of the signals according to the deinterleave information and gives them to the signal detection unit.

  Note that it is not essential to perform peak reduction processing on all blocks. For example, peak reduction processing may be omitted in one block (for example, the first block) for each stream. It is not essential to insert a second pilot symbol in a block that is not subjected to peak reduction processing (rather, it is desirable to omit the second pilot symbol in that block and release resources for other data transmissions). .

  In the seventh embodiment of the present invention, a technique for reducing overhead due to pilot symbols is described. In the present embodiment, the second pilot symbol is not inserted. Each stream is distinguished by the mapping pattern of the first pilot symbol in the block. Further, it is desirable that the pilot symbols of each stream be mapped (arranged) so as to be orthogonal between the streams.

  FIG. 20 shows an example of such pilot symbol mapping. In the example shown in the figure, orthogonal codes with N code lengths of N are arranged every N symbols (N is the number of transmission units, and n is a number that designates transmission units). Further, the pilot of the transmission unit 1 (Tx1) and the pilot of the transmission unit 2 (Tx2) are transmitted at the same time, but are transmitted on different subcarriers. For example, when there are two transmitters, the pilot symbols of transmitter 1 are arranged every other subcarrier. The pilot symbols of the transmitter 2 are arranged every other subcarrier starting from the second subcarrier from the beginning in the frequency direction. When there are three transmitters, the pilot subcarriers of transmitter 1 are arranged every two subcarriers. Transmitter 2 has pilot symbols arranged in the frequency direction starting from the second subcarrier, and transmitter 3 has pilot symbols arranged every two subcarriers starting from the third subcarrier.

  FIG. 21 shows a receiver that can be used in this embodiment. Normal signal separation is performed after the FFT. Each transmitter inserts a unique pilot symbol into each block, performs interleaving, and transmits them. On the receiving side, interleaving between streams can be handled as channel fluctuation. Therefore, a signal sequence before interleaving can be obtained by performing normal channel estimation and signal separation. In addition, when noise reduction processing is performed when channel estimation value interpolation or channel estimation values of adjacent subcarriers are used, they need to be performed in the same block.

  In the eighth embodiment of the present invention, weight control is performed in addition to sub-sequence interleaving. FIG. 22 shows a transmitter that performs such processing. The transmitter performs interleaving between the streams and assigns a weight to the sub-sequence before addition. The weight may be expressed in amplitude, phase, or both. By combining interleaving and weight control, the peak reduction effect can be further promoted.

  For example, a plurality of transmission signals are generated for all interleave patterns and weight candidates related to a certain information bit sequence. Of these transmission signal candidates, the transmission signal having the smallest peak may be finally transmitted.

  Alternatively, a peak may be detected for a signal before synthesis, a transmission signal candidate may be generated only for a signal that contributes to the peak, and a candidate that minimizes the peak may be determined as the transmission signal. In this way, a transmission signal can be generated more easily.

Further, it is not essential that transmission signal candidates for all interleave pattern and weight candidates are generated. For example, when the number of divisions D is 2 and the number of transmission units is N, among the input 2 × N sub-sequences S _n ⁽¹⁾ and S _n ⁽²⁾ , the sequence with a large peak power S _np ^(dp) is determined. The signal component having the peak power is assumed to be S _np ^(dp) (t _p ). In this case, among the divided OFDM signal of another block d _{p ',} the largest series signal component of the time t _p is

が、Ｓ_ｎｐ ^（ｄｐ）と同一のアンテナから送信する分割ＯＦＤＭ信号に決定される。但し、Ｓ_ｎｐ ^（ｄｐ）に適用するウエイト（位相回転量θ）は、

Is determined to be a divided OFDM signal transmitted from the same antenna as S _np ^(dp) . However, the weight (phase rotation amount θ) applied to S _np ^(dp ) is

のように算出される。その結果、各系列は互いに弱めあうように合成され、合成後の系列のピークを低減することができる。

It is calculated as follows. As a result, the sequences are synthesized so as to weaken each other, and the peak of the sequence after the synthesis can be reduced.

  In the ninth embodiment of the present invention, sub-sequence interleaving and sub-sequence cyclic shift (cyclic shift) are combined. FIG. 23 shows a transmitter that performs such processing. After interleaving between the transmission units, the peak reduction effect can be further promoted by cyclically rearranging the sub-sequences in each transmission unit. Note that all interleave patterns and all cyclic shift candidates may be prepared, or the calculation burden may be reduced by paying attention to the signal corresponding to the peak signal before synthesis.

In the tenth embodiment of the present invention, sub-sequence interleaving, cyclic shift and weight control are combined. FIG. 24 shows a transmitter that performs such processing. Various methods already described can be used as the peak reduction processing method. As a simple method, a sequence S _np ^(dp) having the maximum peak power is detected among the sub-sequences input to the peak reduction control unit, and a signal having the peak power is represented by S _np ^(dp) (t _p ). Suppose that The interleave pattern is determined so that the sub-sequence S _{n ′} ^{(d ′)} (t ′) including the symbol of maximum power among the other signal sequences is transmitted from the same antenna as S _np ^(dp). . The subsequence S _{n ′} ^{(d ′)} (t ′) is expressed by the following equation.

このとき、サイクリックシフトのシフト量ｐ_{ｓｈｉｆｔ}とウエイト（位相回転量θ）はそれぞれ次式を満たすように選択される。

At this time, the shift amount p _shift and the weight (phase rotation amount θ) of the cyclic shift are selected so as to satisfy the following equations, respectively.

残りの系列についても同様な処理を行うことで適切なインターリーブパターンを決定することができる。また、上位ｌ番目までのピーク電力が使用される場合には第２実施例で説明済みの手法が使用されてもよい。

Appropriate interleaving patterns can be determined by performing similar processing for the remaining sequences. Further, when the peak powers up to the first l are used, the method described in the second embodiment may be used.

It is a figure which shows a mode that a peak-to-average power ratio becomes large excessively. It is a figure which shows the general input-output characteristic of a transmission amplifier. 1 is a diagram illustrating a transmitter used in a MIMO-OFDM mobile communication system. FIG. It is a figure which shows the receiver used with the mobile communication system of a MIMO-OFDM system. It is a figure which shows a serial-parallel conversion part and a parallel-serial conversion part. It is a figure which shows the transmitter of a 2 * 2 MIMO-OFDM system using the interleaving method of a frequency domain. It is a figure which shows the receiver of a 2 * 2 MIMO-OFDM system using the interleaving method of a frequency domain. It is a figure which shows the transmitter of a 2 * 2 MIMO-OFDM system using a PTS method and a cyclic shift method. It is a figure which shows the receiver of a 2 * 2 MIMO-OFDM system using a PTS method and a cyclic shift method. It is a figure which shows the structural example of the division | segmentation inverse Fourier-transform part in the case of 2 divisions. FIG. 3 shows a transmitter according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a receiver according to an embodiment of the present invention. FIG. 3 shows a transmitter according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a receiver according to an embodiment of the present invention. It is a figure which shows an example of the interleave pattern determination method by one Example of this invention. It is a figure which shows an example of the interleave pattern determination method by one Example of this invention. It is a figure which shows an example of the interleave pattern determination method by one Example of this invention. It is a figure which shows the example of mapping of a pilot symbol. FIG. 3 is a diagram illustrating a receiver according to an embodiment of the present invention. It is a figure which shows the example of mapping of a pilot symbol. FIG. 3 is a diagram illustrating a receiver according to an embodiment of the present invention. FIG. 3 shows a transmitter according to an embodiment of the present invention. FIG. 3 shows a transmitter according to an embodiment of the present invention. FIG. 3 shows a transmitter according to an embodiment of the present invention.

Explanation of symbols

S / P serial-parallel converter P / S parallel-serial converter FFT fast Fourier transform IFFT inverse fast Fourier transform GI guard interval

Claims (13)

A transmitter having a plurality of transmission antennas and transmitting a multicarrier signal,
A division inverse Fourier transform means for deriving a plurality of signal sequence groups from information to be transmitted, performing inverse Fourier transform for each signal sequence group, and outputting a plurality of sets of subsequence groups including one or more subsequences after the transformation;
Peak reduction control means for detecting peak power for a plurality of sub-sequences after inverse Fourier transform and outputting an instruction signal designating an interleave pattern;
Interleaving means for replacing one or more sub-sequences among a plurality of sub-sequence groups according to the instruction signal;
And a plurality of means for converting a group of sub-sequences including the sub-sequence replaced by the interleaving means into a signal wirelessly transmitted from one transmission antenna. 2. The transmitter according to claim 1, wherein the interleaving unit performs substitution among a plurality of sub-sequences associated with the same time among outputs of the divided inverse Fourier transform unit. The peak reduction control means detects the upper predetermined number of peak powers in one sub-sequence, specifies signal components of other sub-sequences that contribute to peak power, and determines peak power for the plurality of sub-sequences The transmitter according to claim 1, wherein the transmitter is detected. 2. The transmitter according to claim 1, further comprising weight adjusting means for weighting one or more sub-sequences after inverse Fourier transform in response to a control signal from the peak reduction control means. Cyclic shift means for cyclically changing the correspondence between one or more sub-sequences not input to the interleave means and one or more sub-sequences output from the interleave means in response to a control signal from the peak reduction control means The transmitter according to claim 1, further comprising: Weight adjusting means for weighting one or more sub-sequences after inverse Fourier transform in response to a control signal from the peak reduction control means;
Cyclic shift means for cyclically changing the correspondence between one or more sub-sequences not input to the interleave means and one or more sub-sequences output from the interleave means in response to a control signal from the peak reduction control means When,
The transmitter according to claim 1, further comprising: determining the interleave pattern such that a difference in peak power between a plurality of sub-sequences after inverse Fourier transform is small. The transmitter according to claim 1, wherein one set of sub-sequence groups is derived by interleaving and adding the sub-sequences included in the two sets of sub-sequence groups after inverse Fourier transform. Interleaves one set of subsequences derived by interleaving and adding the subsequences included in the two sets of subsequences after inverse Fourier transform and the subsequences included in the other two sets of subsequences The transmitter according to claim 1, wherein one set of sub-sequence groups is derived by performing interleaving and addition with another set of sub-sequence groups derived by adding together. . The transmitter according to claim 1, wherein the interleave pattern is distinguished by a mapping pattern of pilot symbols in a transmission symbol. The transmitter according to claim 1, wherein the first pilot symbol not subjected to interleaving and the second pilot symbol subjected to interleaving are included in a transmission symbol. A receiver having a plurality of Fourier transform means for receiving signals of a multi-antenna system and a multi-carrier system and performing Fourier transform on a signal received by one receiving antenna,
A signal separating means for individually separating the signals superimposed on the signal after Fourier transform;
Means for detecting interleave information identifying an interleave pattern performed on the transmission side;
Deinterleaving means for deinterleaving a plurality of subsequences after signal separation based on the interleave information;
And a plurality of means for detecting information bits to be transmitted from one or more sub-sequences after signal separation and sub-sequences after deinterleaving. A transmission method having a plurality of transmission antennas and transmitting a multicarrier signal,
Deriving a plurality of signal sequence groups from the information to be transmitted, performing an inverse Fourier transform for each signal sequence group, and outputting a plurality of sets of sub-sequence groups including one or more sub-sequences after conversion,
Detect peak power for multiple sub-sequences after inverse Fourier transform, and output an instruction signal specifying an interleave pattern,
Replacing one or more sub-sequences among a plurality of sub-sequence groups according to the instruction signal;
A transmission method comprising: converting a signal format of a group of sub-sequences including sub-sequences after interleaving and transmitting each of the plurality of transmission antennas by radio. A receiving method for receiving signals of a multi-antenna method and a multi-carrier method and performing Fourier transform on the received signals for each of a plurality of receiving antennas,
Separate the signals superimposed on the signal after Fourier transform,
Detect interleaving information that identifies the interleaving pattern performed on the sending side,
A plurality of subsequences after signal separation are deinterleaved based on the interleave information,
A receiving method comprising: individually detecting a plurality of information bits to be transmitted from one or more subsequences after signal separation and a subsequence after deinterleaving.

Images