JP2008228345A - Transmitter and receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To substantially expand cyclic prefix length, without having to change the data block size or the total size of a cyclic prefix and a data block. <P>SOLUTION: A transmitter has: a puncture section for thinning out a data bit from a plurality of data bits to obtain a data bit train; a block generation section for generating a plurality of data blocks by a data symbol train, where the data bit train is modulated; and an addition section for adding the copy of h data symbols, at the end of the data block to the tip of the data block as the cyclic prefix. K (k is an integer of not less than 1) data symbols prior to the h data symbols at the end of the first data block are the copies of k data symbols at the end of the second data block prior to the first one. At the puncture section, the larger the number of data symbols to be copied is, the larger the number of data bits to be thinned out becomes. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a transmitter and a receiver, for example, a transmitter and a receiver that perform single carrier communication using a cyclic prefix.

In the conventional method of extending the cyclic prefix (CP) length, the data block size is fixed and the number of symbols added as a cyclic prefix at the beginning of the data block is increased (first conventional extension). Method) and a method of reducing the data block size by fixing the total size of the cyclic prefix and the data block size and increasing the number of cyclic prefix symbols (second conventional extension method) There is.
Japanese Patent Laid-Open No. 2006-2222956

  In the first conventional extension method, the total size of the cyclic prefix and the data block size increases due to the extension of the cyclic prefix length (CP length), so that the timing of FFT processing at the receiver changes. There's a problem. The conventional second extension method has a problem that the FFT size at the receiver changes because the data block size is reduced.

  The present invention provides a transmitter capable of substantially extending a cyclic prefix length without changing any of the data block size and the total size of the cyclic prefix and the data block, and the transmitter. A receiver for communicating with a transmitter is provided.

A transmitter as a first aspect of the present invention includes a puncturing unit that thins out data bits from a plurality of data bits to obtain a data bit string,
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generator for generating a data block from a plurality of data symbols;
An adding unit that adds a copy of h data symbols (h is an integer of 1 or more) at the end of the data block as a cyclic prefix to the top of the data block to obtain a data block with a cyclic prefix;
A transmission unit for transmitting a signal of each data block with a cyclic prefix,
The k data symbols (k is an integer of 1 or more) preceding the h data symbols at the end of the first data block are the second data block preceding the first data block. A copy of the k data symbols at the end,
The puncturing unit increases the number of data bits to be thinned out from the plurality of data bits as the number of data symbols copied in the adding unit increases.

The transmitter as the second aspect of the present invention includes a puncturing unit that thins out data bits from a plurality of data bits to obtain a data bit string,
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generator for generating a data block from a plurality of data symbols;
An adding unit that adds a copy of h data symbols (h is an integer of 1 or more) at the end of the data block as a cyclic prefix to the top of the data block to obtain a data block with a cyclic prefix;
A transmission unit for transmitting a signal of the data block with the cyclic prefix,
K (k is an integer greater than or equal to 1) data symbols at the end of the first data block precede the h data symbols at the end of the third data block following the first data block. a copy of k data symbols,
The puncturing unit increases the number of data bits to be thinned out from the plurality of data bits as the number of data symbols copied in the adding unit increases.

A transmitter as a third aspect of the present invention includes a puncturing unit that thins out data bits from a plurality of data bits to obtain a data bit string,
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generation unit that generates a plurality of data blocks each including a plurality of data symbols using the data symbol sequence, and a part of the plurality of data blocks is a pilot block including a plurality of pilot symbols;
An adder that obtains a plurality of data blocks with a cyclic prefix by adding a copy of h data symbols (h is an integer of 1 or more) at the end of each data block as a cyclic prefix to the top of the data block. When,
A transmission unit for transmitting a signal of each data block with a cyclic prefix,
The k data symbols (k is an integer of 1 or more) preceding the h data symbols at the end of the first data block are the second data block preceding the first data block. A copy of the k data symbols at the end;
The k data symbols at the end of the first data block are copies of the k pilot symbols preceding the h pilot symbols at the end of the pilot block following the first data block. Yes,
The puncturing unit increases the number of data bits to be thinned out as the number of data symbols copied in the adding unit increases.

  According to the present invention, the cyclic prefix length can be substantially extended without changing any of the data block size and the total size of the cyclic prefix and the data block.

  FIG. 17 shows a configuration example of a conventional single carrier transmitter in the case where the length of a cyclic prefix (CP) is fixed.

  The data bit string to be transmitted data is converted into a data symbol string by performing data modulation such as QPSK modulation or QAM modulation in the modulation unit 101.

  The block generation unit 102 generates and outputs one data block from the plurality of data symbols output from the data modulation unit 101. An example in which one data block is generated from eight data symbols is shown in FIG. The data block A is composed of eight data symbols a1, a2, a3, a4, a5, a6, a7, a8, and the data block B is composed of eight data symbols b1, b2, b3, b4, b5, b6, b7. , B8. These two data blocks A and B are output from the block generator 102 in this order. Here, the number of data symbols included in the data block is defined as the data block size, and the time length of the data block is defined as the data block length. For example, the data block size of data block A is 8.

  The CP adding unit 103 adds one or more data symbols at the end of the data block to the data block input from the block generation unit 102 as a cyclic prefix at the head of the data block. FIG. 18B shows how a cyclic prefix is added to the data blocks A and B. For example, for data block A, the last two symbols (a7, a8) of data block A are copied and added to the beginning of data block A as a cyclic prefix. Here, the number of data symbols included in the cyclic prefix is defined as the CP size, and the time length of the cyclic prefix is defined as the CP length. For example, in the example of FIG. 18B, the CP size is 2.

  The data block to which the cyclic prefix is added is converted into an analog signal by the D / A converter 104 and transmitted from the antenna 106 via an IF (Intermediate Frequency) / RF (Radio Frequency) transmitter 105.

  FIG. 19 shows a configuration example of a conventional single carrier transmitter when the cyclic prefix length (CP length) is extended by the conventional first and second extension methods described in the background art section. The difference from the conventional transmitter when the CP length shown in FIG. 17 is fixed is that the control unit 107 outputs the data block size output from the block generation unit 102 and the CP size added by the CP addition unit 103. It is in the point that can be changed. In the figure, elements having the same names as those in FIG.

  20 (a) and 20 (b) show the output of the block generation unit 102 and the output of the CP addition unit 103 when the CP size is expanded from 2 to 3 using the conventional first extension method. It shows.

  The data block size remains 8, and the CP size is expanded from 2 to 3. For this reason, the total size combining the CP size and the data block size is larger than before the CP size is expanded. As a result, there arises a problem in the receiver that the timing of FFT for demodulating continuous data blocks changes compared to before the CP size is expanded.

  FIG. 21 shows the output of the block generation unit 102 and the output of the CP addition unit 103 when the CP size is expanded from 2 to 3 using the conventional second expansion method.

  The data block size is reduced to 7, and the CP size is expanded from 2 to 3. In this case, the total size of the CP size and the data block size remains the same, but since the data block size is reduced, the FFT size for demodulating the data block at the receiver is equal to the CP size. The problem that becomes smaller than before expansion occurs.

  Therefore, even when the cyclic prefix length is extended using any of the conventional first and second extension methods, a plurality of frequency-multiplexed single carrier signals or single carrier signals and OFDM (Orthogonal Frequency Division Multiplexing) are used. ) Signals are subjected to FFT (Fast Fourier Transform) processing at the same time by the receiver, causing interference between the signals.

  In the embodiment of the present invention, the cyclic prefix length can be substantially changed in units of data blocks while the data block size and the CP size are fixed. As a result, the CP size has been expanded while preventing the occurrence of interference between signals when FFT processing is performed on a plurality of frequency-multiplexed single carrier signals or single carrier signals and OFDM signals at the same time. The same effect can be obtained.

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

  FIG. 1 shows a first embodiment of a single carrier transmitter according to the present invention.

  A data bit string that becomes transmission data is converted into a data symbol string by being subjected to data modulation such as QPSK modulation or QAM modulation in the modulation unit 11 and input to the block generation unit 12. The modulation unit 11 has an input unit that inputs a data symbol sequence to the block generation unit 12.

  In the block generation unit 12, a data block is generated from the data symbols input from the modulation unit 11 and specific data symbols included in past data blocks stored in the storage unit 17. The generated data block is input to the storage unit 17 and the CP adding unit 13.

  The storage unit 17 stores the data block input from the data block generation unit 12.

  The CP adding unit 13 adds one or more data symbols at the end of the data block input from the block generating unit 12 to the head of the data block as a cyclic prefix.

  The data block to which the cyclic prefix is added is converted into an analog signal by the D / A converter 14 and transmitted from the antenna 16 via the IF / RF transmitter 15.

  Hereinafter, the operation of the block generation unit 12 will be described in detail.

When the CP size added by the CP adding unit 13 is Nc symbols, it is assumed that the CP size of a certain data block (for example, data block B) is substantially extended by Ne symbols. At this time, the block generation unit 12 uses the data block (for example, the data block) preceding the data block B stored in the storage unit 17 as Ne symbols from the Nc + 1th to the Nc + Neth counting from the end of the data block B. Use Ne symbols from the end of A). As other data symbols of the data block B, data symbols output from the modulation unit 11 are used as usual. The data block B generated in this way is added with a cyclic prefix as usual in the CP adding unit 13 subsequent to the block generating unit 12, so that the end of the previous data block (for example, data block A) is added. However, it functions as if it is a part of the cyclic prefix of the data block B, and the CP length for the data block B is substantially extended.
It is possible to substantially extend (change) the CP length by generating a data block using the method as described above, substantially extend (substantially change) the cyclic prefix, or CP size. (CP length) will be called substantially expanded (substantially changed).
Also, for clarity of discussion, the present invention is described with a constant symbol rate, so substantially extending the CP length and substantially expanding the CP size are used synonymously. .

  FIG. 2 is a diagram illustrating a specific example of the operation of the block generation unit 12. When the data block size is 8 (N = 8) and the CP size added by the CP adding unit 13 is 2 (Nc = 2), the CP size of the data block B is substantially extended by one data symbol (Ne). = 1) An example will be described.

  FIG. 2A shows data blocks A and B output from the block generator 12. As the third symbol from the end of the data block B, that is, the sixth symbol from the top, the data symbol a8 at the end of the data block A preceding the data block B is used. In other words, k data symbols preceding the data symbols b7 and b8 to be added as cyclic prefixes in the data block B (k is an integer of 1 or more) k symbols at the end of the data block A preceding the data block B A copy of the data symbol (one data symbol a8 in this example) is used. For the seven data symbols b1, b2, b3, b4, b5, b7, and b8 other than the sixth in the data block B, the output from the modulation unit 11 is used as it is. That is, in the data block B, the number of data symbols to be newly transmitted is reduced by the amount that the cyclic prefix is substantially expanded, and the transmission rate is reduced by that amount.

  If there is no data block preceding the data block whose cyclic prefix is to be expanded, the data symbol of the preceding data block is regarded as 0, and a part of the target data block is replaced with 0 by the above method. However, better characteristics may be obtained as a result when such treatment is not performed. This is because when there is no preceding data block, no inter-block interference occurs due to the preceding data block. The same applies to the case where the cyclic postfix described later is extended.

  FIG. 2B shows the output of the CP adding unit 13. Two data symbols a7 and a8 at the end of data block A are added as a cyclic prefix at the beginning of data block A, and two data symbols b7 and b8 at the end of data block B are cyclic at the beginning of data block B It is added as a prefix. Comparing the three data symbols a8, b7, b8 at the end of the data block B with the three data symbols preceding the data block B, it can be seen that they are the same. This means that the cyclic prefix length of data block B has been substantially extended to 3 data symbols. That is, for multipaths up to three data symbol lengths, the output of the FFT processing unit in the receiver prevents inter-data block interference or inter-subcarrier interference. Here, it is noted that the data block size and the CP size added by the CP adding unit 13 are not changed unlike the conventional first and second extension methods described above. This is because, among the three data symbols preceding the data block B, a8 actually belongs to the data block A, not the data block B. Data symbol a8 is transmitted in two consecutive data blocks (data block A and data block B) except for the cyclic prefix. More generally, when the CP size is substantially extended by Ne symbols, if the extension size Ne is equal to or smaller than the CP size Nc (Ne ≦ Nc), the Ne data at the end of each data block A symbol will be transmitted in two consecutive data blocks.

  FIG. 3 is a diagram for explaining an example of the operation of the block generation unit 12 when the extension size Ne> CP size Nc. An example in which the CP size added by the CP adding unit 13 is 2 (Nc = 2) and the CP size is substantially expanded by 3 symbols (Ne = 3) for the data block B and the data block C will be described. To do.

  FIG. 3A shows data blocks A, B, and C output from the block generator 12. Data symbols a6 and a7 in data block A are transmitted a total of two times in two data blocks A and B, and data symbol a8 is transmitted a total of three times in three data blocks A, B and C.

  FIG. 3B shows the output of the CP adding unit 13. It can be seen that the CP size is substantially expanded from 2 symbols to 5 symbols. More generally, when the data block size is N, the CP size is Nc, and the size (extended size) that substantially extends the CP is Ne, the nth (n is an integer between 1 and N) of a data block. Data symbols are transmitted in MAX (0, CEIL ((n−N + Ne) / Nc)) + 1 consecutive data blocks, excluding the cyclic prefix added by the CP adding unit 13. . However, MAX (x, y) means the maximum value of x and y, and CEIL (x) means the smallest integer greater than or equal to x.

  In the example of FIGS. 2 and 3, a cyclic prefix is added to the end of the data block, although the data block is periodically lengthened by adding a cyclic prefix to the top of the data block. It is clear that the idea of the present invention can be applied to the case where Cyclic Postfix (hereinafter abbreviated as CS) is added. Prior to describing this, an example of adding a cyclic postfix in a conventional transmitter will be described below.

  FIG. 4A shows an example of adding a cyclic postfix with two symbols (cyclic postfix with a CS size of 2) to the data block of FIG. 18A in a conventional transmitter (see FIG. 17). Show. The first two symbols a1 and a2 of the data block A are copied and added as a cyclic postfix at the end. Similarly, the top two symbols b1 and b2 of the data block B are copied and added as a cyclic postfix at the end. If an attempt is made to increase the cyclic post-fix length using the same idea as the conventional first and second extension methods described above, the same problem as the increase of the cyclic prefix length will occur. .

  In the following, with reference to FIGS. 4B and 4C, when adding a cyclic postfix instead of a cyclic prefix at the CP adding unit 13 in the transmitter of FIG. The operation of the expanding block generation unit 12 will be described.

When the CS size added by the CP adding unit 13 is Nc symbols, in order to substantially extend the CS size of a certain data block (for example, data block B) by Ne symbols, counting from the end of the data block B is performed. Nc-Ne + 1th from the Nc-th symbol of the data block (for example, data block A) preceding the data block B stored in the storage unit 17 from the Nc-th symbol to the top as Ne 1-th to Ne-th consecutive symbols Ne symbols that are consecutive up to this symbol are used. That is, as k data symbols at the end of the data block B, k from the end of the data symbol used as the cyclic postfix in the data block A preceding the data block B (data symbol a2 in this example) toward the start. A copy of data symbols (in this example, one data symbol a2) is used. An example in the case of Nc = 2 and Ne = 1 is shown in FIG. However, when Nc−Ne + 1 is 0 or less, it is handled that the immediately preceding symbol of the data block A is connected to the last symbol of the data block A. The other data symbols of the data block B are, as usual, the CP addition unit 13 for the data block generated in this way in the block generation unit 12 in which the data symbol output from the modulation unit 12 is used as it is. As shown in FIG. 4C, the cyclic postfix of the data block (eg, data block A) immediately before a certain data block (eg, data block A) is added. The last symbol a2 of (CS) can be regarded as a part of the data block B, and the CS length is substantially extended.

  As described above, the addition of the cyclic prefix and the addition of the cyclic postfix (CS) are both processes that make the data block longer in a periodic structure. Therefore, it is equivalent if the arrangement order of the data is changed. It is clear that there is. Therefore, in the following description, unless otherwise specified, a cyclic prefix is used as an example, but it is obvious that the present invention can be applied to the case of a cyclic postfix.

  FIG. 5 shows a second embodiment of a single carrier transmitter according to the present invention.

  In the first embodiment, the data symbol of the preceding data block is used as the data symbol at a specific position of the data block whose CP size is to be substantially expanded. In the second embodiment, the data symbol of the data block whose CP size is to be substantially expanded is used as the data symbol at the specific position of the data block preceding the data block whose CP size is to be substantially expanded. The second embodiment is suitable when the data block whose CP size is to be substantially expanded includes a fixed value data symbol such as a pilot symbol. The difference from the first embodiment shown in FIG. 1 is that the storage unit 17 is not provided, and a pilot data generation unit 19 that generates pilot data is provided on the input side of the block generation unit 18 instead. . Since the operation of each element after the CP adding unit 13 is the same as that of the first embodiment, description thereof will be omitted, and the operation of the block generation unit 18 will be described below.

  When the CP size added by the CP adding unit 13 is Nc symbols, it is assumed that the CP size of a certain data block (for example, data block B) composed of a plurality of pilot symbols is substantially extended by Ne symbols. At this time, Ne pilot symbols from Nc + 1 to Nc + Ne counted from the end of data block B are used as Ne symbols from the end of the data block preceding data block B (for example, data block A). In the first embodiment, the data block itself for which the CP size is to be expanded is operated. However, in the second embodiment, it is necessary to perform an operation on the data block preceding the data block whose CP size is to be expanded. To do.

  When the data block size is 8 (N = 8) and the CP size added by the CP adding unit is 2 (Nc = 2), the CP size of the data block B composed of pilot symbols is substantially equivalent to one symbol. An example of the operation of the block generation unit 18 in the case of extension to (Ne = 1) will be described with reference to FIG.

  FIG. 6A shows data blocks A and B output from the block generator 18. The sixth pilot symbol b6 of the data block B is used as the last symbol of the data block A preceding the data block B. In other words, the data symbols b7 and b8 added as cyclic prefixes in the data block B subsequent to the data block A as k data symbols (1 (= Ne) in this example) at the end of the data block A are preceded. K (in this example, 1 (= Ne)) data symbol copies are used.

  When such processing in the block generation unit 18 is performed, the output of the CP adding unit 13 outputs the three data symbols b6, b7, b8 at the end of the data block B and the data as shown in FIG. The three data symbols preceding block B are the same. That is, an effect equivalent to that in which the cyclic prefix length of the data block B is substantially expanded to 3 data symbols can be obtained.

  The operation of the block generation unit 18 when adding a cyclic prefix to the beginning of a data block has been described above. Hereinafter, the operation of the block generation unit 18 when adding a cyclic postfix to the end of a data block will be described. Will be described. When the data block size is 8 (N = 8) and the CS size added by the CP adding unit is 2 (Nc = 2), the CS size of the data block B composed of pilot symbols is substantially equal to one symbol. Shows an example of expansion (Ne = 1).

  FIG. 7A shows data blocks A and B output from the block generator 18. The eighth pilot symbol b8 of the data block B is used as the second symbol of the data block A preceding the data block B. That is, in the data block A as k data symbols (1 (= Ne) in this example) from the end (second symbol) to the top of the data symbol used as the cyclic postfix in the data block A A copy of the last k data symbols b8 (1 (= Ne) in this example) of the subsequent data block B is used. When Nc−Ne + 1 is 0 or less, it is assumed that the immediately preceding first symbol of the data block A is connected to the last symbol of the data block A.

  When such processing in the block generation unit 18 is performed, the output of the CP adding unit 13 outputs the last three data symbols b8 and b1 of the data block B with cyclic postfix as shown in FIG. 7B. , B2 and the three data symbols preceding data symbol b3 of data block B are the same. That is, assuming that the data block B is composed of 8 symbols from the last symbol b8 to b7 of the cyclic postfix (CS) of the data block A, the cyclic postfix length (CS length) of the data block B ) Is effectively expanded to 3 data symbols.

  FIG. 8 shows a third embodiment of a single carrier transmitter according to the present invention.

  The third embodiment is a combination of the first embodiment and the second embodiment. In the block generation unit 20 of the third embodiment, in addition to the data symbols output from the modulation unit 11, the data symbols in the data block generated in the past stored in the storage unit 17 and the data blocks A data block is generated using pilot symbols in the subsequent data block. Hereinafter, an operation example of the block generation unit 20 will be described.

  When there are three consecutive data blocks (data block A, data block B, and data block C), and data block C is composed of pilot symbols, the CP sizes of data block B and data block C are set as follows. Assume that it is desired to substantially extend (Ne = 1) by one symbol each. Each data block generated by the block generation unit 20 at this time is shown in FIG.

  Focusing on data block B, since subsequent data block C is composed of pilot symbols, first, in order to substantially expand the CP size of data block C, the sixth pilot symbol c6 of data block C is changed to data Used as the last symbol of block B. Next, in order to substantially extend the CP size of the data block B, the data symbol a8 at the end of the data block A is used as the sixth symbol of the data block B. As described above, when the subsequent data block is composed of pilot symbols, first, a process of substantially expanding the CP size of the subsequent data block is performed, and then the CP size of the target data block is set. It is preferable to perform processing that substantially expands. Here, in the single carrier transmitter according to the present invention, it is obvious that it is possible to substantially extend the cyclic prefix of different sizes for each data block, and therefore higher reception quality is required. For a data block composed of pilot symbols to be transmitted, CP extension can be performed larger than other data blocks.

  FIG. 10 shows a fourth embodiment of a single carrier transmitter according to the present invention. Since the elements after the modulator 11 are the same as those in FIG. 1, detailed description of these elements is omitted. It is also possible to combine each element up to the interleave unit 23 in FIG. 10 and each element after the modulation unit of the second embodiment or the third embodiment.

  In the fourth embodiment, the puncturing unit 22 punctures the encoded bits input to the modulation unit 11 in order to compensate for the data rate that is decreased by substantially expanding the CP size in the block generation unit 12 ( Characterized by points to be thinned out. In this example, the data rate is adjusted by the puncturing process, but other than that, a method of changing the coding rate itself of the coding unit 21, a method of changing the modulation multi-level number of the modulation unit 11, etc. It may be used.

  Hereinafter, operations of the encoding unit 21, the puncturing unit 22, and the interleaving unit 23 will be described.

  The data bit string that becomes the transmission data is subjected to error correction coding processing in the coding unit 21 to be a coded bit string to which redundancy is added. For example, when the coding rate is ½, two coded bits are output for one data bit input.

  The puncturing unit 22 performs a process of thinning out (removing) a certain number of coded bits from a plurality of consecutive coded bits. The number of encoded bits to be thinned out in the puncture unit 22 depends on the size of the CP generation unit 12 that substantially extends the CP size. Specifically, the number of transmission data symbols decreased due to transmission of the same data symbol in successive data blocks, or the number of transmission data symbols decreased due to transmission of a data block consisting of pilot symbols, The number of encoded bits corresponding to is punctured. For example, when the data block size is N and the CP size to be expanded is Ne, the puncture unit 22 averages and thins out the N coded bits. That is, as the number of data symbols copied from other data blocks increases, the number of encoded bits to be thinned out from the encoded data bit string corresponding to the data block is increased.

  The encoded bit string output from the puncturing unit 22 is input to the modulation unit 11 after the arrangement order of the encoded bits is changed by the interleaving unit 23 in the interleaving unit 23. As an interleaving method applied by the interleave unit 23, the CP size is substantially expanded by the block generation unit 12 and the position of the encoded bit punctured by the puncture unit 22 in addition to the block interleaving used conventionally. Therefore, rearrangement according to the position of the data symbol to be transmitted by a plurality of data blocks is also possible. That is, both the encoded bits thinned out by the puncture unit 22 and the data bits included in the data symbols to be transmitted in a plurality of data blocks due to the substantial expansion of the CP size are both included. It is also possible to perform rearrangement such that a plurality of encoded bits are output at the same state transition of convolutional encoding. By performing such rearrangement, among the plurality of bits output at the same state transition, the bits remaining after being thinned out by the puncture unit 22 are transmitted in the plurality of data blocks, and are combined at the receiver. Processing (described later) can be expected to minimize the degradation of the decoding characteristics of the error correction code due to puncturing.

  Hereinafter, a case where convolutional coding with a coding rate of ½ is used as error correction coding, and a modulation method (for example, QPSK modulation method) in which 2 bits are mapped to one symbol in the modulation unit 11 is used as an example. Operations up to the block generation unit 12 will be described with reference to FIG.

  When 8 bits (assuming to correspond to one data block) are input to the encoding unit 21, as shown in FIG. 11A, 16 encoded bits x1 to x16 are output. At this time, a pair of consecutive 2 bits (x1 and x2), (x3 and x4),..., (X15 and x16) are a plurality (in this case, 2) output by the same state transition in convolutional coding. Are the encoded bits. It is assumed that two bits x3 and x11 are excluded from these encoded bits in the puncturing unit 22, and the number of encoded bits is 14 bits (FIG. 11 (b)). This corresponds to a case where the data block size is 8 and the CP size is substantially expanded by one symbol. In this way, in the error correction encoding of the data bit string corresponding to the data block, some of the encoded bits obtained at the same state transition are thinned out in the puncture unit 22.

  If the CP size of the subsequent data block is substantially expanded by one symbol, the data symbol at the end of the data block is also transmitted in the subsequent data block. Therefore, the interleave unit 23 encodes the encoded data so that at least one of the encoded bits x4 and x12 output by the same state transition as the encoded bits excluded by the puncture unit 22 is included in the final data symbol of the data block. The rearranged bits are rearranged (FIGS. 11C and 11D). That is, the interleave unit 23 interleaves the encoded data bit string so that the remaining encoded bits of the plurality of encoded bits are located at the end. Here, the rearrangement is performed so that both the encoded bits x4 and x12 are included in the final data symbol. By performing such processing, the encoded bits x4 and x12 are transmitted twice together with the subsequent data block, and characteristic deterioration due to puncturing when demodulating the convolutional code is minimized. Can do.

However, when a cyclic postfix is added as shown in FIG. 4, the remaining encoded bits of the plurality of encoded bits are included at the end of the data symbol used as a cyclic postfix in the data block B. Next, the encoded data bit string is interleaved. In addition, when a cyclic prefix is added as shown in FIG. 6, the remaining encoded bits of the plurality of encoded bits precede the data symbol added as a cyclic prefix in the data block A. The encoded data bit string is interleaved so as to be included in the last data symbol among the data symbols (effective when the method of FIG. 6 is applied to the data block preceding data block A).
In addition, when adding a cyclic postfix as shown in FIG. 7, the encoded data bit string is such that the remaining encoded bits of the plurality of encoded bits are included in the last data symbol of the data block A. Are interleaved (effective when the method of FIG. 7 is applied to the data block preceding data block A). However, when the pilot symbols are copied and transmitted in a plurality of data blocks in the method of FIG. 6 and the method of FIG. 7, the pilot symbols are known data, and therefore, the encoded data bit string is specified by a specific method as described above. There is no need to interleave.

  Returning to FIG. 11, the block generation unit 12 generates data from the seven data symbols b1 to b5, b7, b8 generated from the 14 encoded bits and the data symbol a8 copied from the preceding data symbol. A block is generated (FIG. 11 (e)). The data symbol thus generated is added with a cyclic prefix in the CP adding unit 13 and then transmitted from the antenna 16 via the D / A converting unit 14 and the IF / RF transmitting unit 15.

  Next, a single carrier receiver for demodulating a signal from a single carrier transmitter according to an embodiment of the present invention will be described. In this single carrier receiver, the same data symbols that are transmitted in a plurality of consecutive data blocks due to the fact that the CP size is substantially expanded by the block generation unit of the single carrier transmitter are combined. It is characterized by performing processing.

  FIG. 12 shows a first embodiment of a single carrier receiver according to the present invention.

  The signal received by the antenna 31 is converted into a baseband signal by the IF / RF receiver 32 and then converted into a digital signal by the A / D converter 33.

  The CP removing unit 34 removes sample points (cyclic prefix or cyclic postfix) of the CP-length digital signal added by the CP adding unit of the transmitter.

  The FFT processing unit 35 performs a Fourier transform process with a time length corresponding to the transmitted data block, and decomposes the received signal into a plurality of orthogonal frequency components.

  The channel estimation unit 36 estimates the channel response of each frequency component based on a plurality of frequency component signals obtained from the data block composed of pilot symbols.

  The communication channel compensation unit 37 multiplies the complex number based on the estimated value of the transmission channel response of each frequency component estimated by the communication channel estimation unit 36 with respect to a plurality of frequency component signals output from the FFT processing unit 35. By doing so, distortion of each frequency component generated in the communication path is corrected (equalization processing).

  The equalized multiple frequency component signals output from the channel compensation unit 37 are transmitted by being subjected to inverse Fourier transform in an IFFT (Inverse Fast Fourier Transform) processing unit 38. It is converted into a time axis signal equal to the data block size. The output signal of the IFFT processing unit 38 corresponding to the transmitted data block will be referred to as a received data block. A sample value in the received data block corresponding to each data symbol in the transmitted data block is referred to as a received data symbol.

Next, for each received data symbol in the received data block, whether or not a past received data symbol corresponding to the same transmitted data symbol is stored in the recording unit 39 is checked by the combining unit 40 and stored. The combining unit 40 combines corresponding received data symbols. In other words, the synthesis unit 40 pre-designates the received data symbol at the first position designated in advance in the pre-designated received data block and the received data block following (or preceding) the pre-designated received data block. The received data symbol at the second position is combined to generate a combined received data symbol used in place of the received data symbol at the first position. The correspondence relationship between the reception data symbols between the reception data blocks is assumed to be set in advance by the receiver, for example, by acquiring information on the correspondence relationship from the transmitter as broadcast information.
As a combining method in the combining unit 40, equal gain combining or maximum ratio combining in which weights proportional to SINR (signal-to-noise interference power ratio) of each corresponding received data symbol are multiplied and combined with these received data. Such a method is conceivable.

  From each received data symbol subjected to the combining process as necessary, the demodulator 41 estimates the transmission data symbol and demodulates the data bit. That is, the demodulator 41 demodulates the received data symbol at each position different from the first position in the previously specified received data block and the combined received data symbol.

  Hereinafter, the operation after the IFFT processing unit 38 will be described in detail with reference to FIG.

  FIG. 13 shows the received data block A and the received data block B that are continuously output from the IFFT processing unit 38. These received data blocks A and B correspond to two consecutive data blocks shown in FIG. 2A in the transmitter. The eighth reception data symbol A8 of the reception data block A and the sixth reception data symbol A8 'of the reception data block B correspond to the same transmission data symbol (a8 in FIG. 2A). The storage unit 39 stores the received data symbol A8, and the received data symbol A8 is combined with the received data symbol A8 'by the combining unit 40. The demodulator 41 demodulates data from a value (combined reception data symbol) obtained by combining the reception data symbols A8 and A8 '. The other received data symbols A1 to A7 in the received data block A pass through the combining unit 40 without being performed, and the demodulation unit 41 performs data demodulation. Further, for the received data block B, the received data symbols B1 to B5, B7, and B8 pass through without being performed by the combining unit 40, and the demodulation unit 41 performs data demodulation.

  FIG. 14 shows a second embodiment of a single carrier receiver according to the present invention. The second embodiment is for demodulating a signal transmitted from the transmitter (fourth embodiment) in FIG. Reception processing corresponding to encoding, puncturing processing, and interleaving processing performed at the transmitter is added to the receiver shown in FIG. In FIG. 14, the operation of the antennas 31 to IFFT processing unit 38 is the same as that of the receiver of FIG.

  The likelihood of each bit corresponding to each received data symbol (soft decision bit data or bit likelihood data) in the bit metric calculation unit 42 from each received data symbol included in the received data block that is the output of the IFFT processing unit 38. Calculate

  For each bit likelihood data output from the bit metric calculation unit 42, the synthesis unit 44 checks whether past bit likelihood data corresponding to the same transmission data bit is stored in the recording unit 43, If stored, the corresponding bit likelihood data are combined by the combining unit 44. That is, the combining unit 44 follows the first bit likelihood data generated from the reception data symbol at the third position specified in advance in the reception data block specified in advance, and the reception data block specified in advance. The second bit likelihood data generated from the received data symbol at the fourth position designated in advance in the (or preceding) received data block is synthesized and used instead of the first bit likelihood data. Generated third bit likelihood data. The correspondence of bit likelihood data between received data blocks (or the correspondence of received data symbols) is set in advance in the synthesizing unit 44, for example, when the receiver acquires information on the correspondence from the transmitter as broadcast information. It shall be.

  The bit likelihood data obtained from the received data block and subjected to synthesis processing as necessary is arranged in the deinterleave unit 45 so that the arrangement order changed in the interleave unit of the transmitter is restored. The order is changed.

  In the depuncture unit 46, 0 (in the case of a log likelihood value) is inserted as the bit metric of the portion removed by the puncture unit of the transmitter.

  Finally, the error correction decoding process is performed in the decoding unit 47 from the bit string in which 0 is inserted in the depuncturing unit 46, and the decoding of the data bits is completed.

  Here, when demodulating a data block partially replaced with pilot symbols, such as a signal transmitted from the transmitter (second or fourth embodiment) shown in FIG. 5 or FIG. Data symbols that are data (for example, data symbol b6 of data block A in FIG. 6A) may be ignored, and other data symbols in the data block may be demodulated. A data block composed only of pilot symbols (for example, data block B in FIG. 6A) is used to estimate the frequency response of the communication channel by the communication channel estimation unit 36 of the receiver as described above. Used.

  Further, when the CP size of a data block composed of pilot symbols is substantially expanded, the transmitter (first embodiment) is not the method of the transmitter (second embodiment) of FIG. ), When a part of pilot symbols in a data block is replaced with data symbols of the preceding data block and transmitted, the channel is estimated from the data block as follows: do it. First, the data symbol of the preceding data block is demodulated, and the data symbol in the data block is treated as known data using the demodulation result. As a result, since all pilot symbols and data symbols in the data block can be regarded as known data, it is possible to perform channel estimation using the entire data block.

  FIG. 15 includes a terminal A having a function of a single carrier transmitter (single carrier transmitter 51) according to the present invention and a terminal B having a function of a single carrier receiver (single carrier receiver 53). 1 shows a wireless communication system.

  Depending on the state of the communication path from terminal A to terminal B, terminal B transmits a CP control signal for instructing to substantially expand (change) the CP size in single carrier transmitter 51 of terminal A. It transmits from the transmission part 54. Here, the meaning of the communication path includes the influence of both the multipath of the propagation path and the time response of the filter (not shown) applied in the transmission unit or the reception unit. The terminal A receives the CP control signal at the CP control signal receiving unit (symbol number information receiving unit) 52, substantially extends the CP size according to the instruction of the CP control signal, and transmits the CP from the single carrier transmitting unit 51 to the terminal B. Transmit a signal of substantially expanded size. The CP control signal includes a size (number of symbols to be copied) that substantially extends the CP size, and corresponds to symbol number information. The CP control signal receiving unit 52 corresponds to a symbol number information receiving unit.

  Instead of terminal A performing substantial expansion of the CP size in response to an instruction from terminal B, terminal A voluntarily expands the CP size according to the state of the communication path and transmits the information to terminal B Thus, the terminal B can also identify the substantial extension size of the data block transmitted from the terminal A. For example, terminal A and terminal B perform bidirectional communication at the same frequency using the TDD (time division duplex) method, and terminal A spontaneously substantially expands the CP size according to the condition of the communication path. It can be considered. In this case, the terminal A may include a changing unit (not shown) that performs a substantial change of the CP size (change of the number of symbols to be copied) according to the state of the communication path.

  Similarly, terminal A may substantially extend (change) the CP size according to the symbol rate to be transmitted. This is because when the shape of the filter (not shown) applied in the transmission unit or the reception unit (the tap coefficient of the FIR filter) is the same regardless of the transmission symbol rate, the lower the transmission symbol rate. This means that the time length of the time response of the filter becomes longer, so that the state of the communication path including the influence of the filter differs depending on the symbol rate to be transmitted. In such a case, in addition to the method using the CP control signal as described above and the method of transmitting the extension size, a method of defining a substantial CP size in advance according to the transmission symbol rate is used. Also good.

  Thus, in the embodiment of the present invention, the CP size can be substantially controlled for each terminal without changing the data block size and the CP size added by the CP adding unit. Therefore, it is obvious that the present invention can be applied not only to the one-to-one wireless communication system shown in FIG. 15 but also to a one-to-N wireless communication system such as a base station and a plurality of mobile terminals. . In this case, not only a terminal having a single carrier transmission unit but also an OFDM terminal can coexist as a mobile terminal. However, the relationship between the format of the signal from the OFDM terminal and the signal from the single carrier terminal is such that the CP lengths are equal as shown in FIG. Both need to be equal to the FFT size of the receiver. Also, in the embodiment of the present invention, the symbol rate is assumed to be constant for the single carrier transmitter. However, if the CP length and the data block length are equal, a pair of symbols having different symbol rates for each terminal. It is obvious that the present invention can be applied to N wireless communication systems.

  As described above, by using the single carrier transmitter according to the embodiment of the present invention, the cyclic prefix length can be changed without changing the length of the data block and the length of the cyclic prefix added by the CP adding unit. Can be substantially expanded. Further, by using the single carrier receiver according to the embodiment of the present invention, the cyclic prefix is synthesized by combining the same data symbols transmitted in the plurality of data blocks from the single carrier transmitter according to the embodiment of the present invention. Data demodulation with minimal loss due to the substantial extension of the length (CP length) becomes possible.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

The figure which shows 1st Embodiment of the single carrier transmitter of this invention. The figure which shows the operation example of the block production | generation part in the transmitter of FIG. The figure which shows the other operation example of the block production | generation part in the transmitter of FIG. The figure which shows the further another operation example of a block production | generation part in the transmitter of FIG. The figure which shows 2nd Embodiment of the single carrier transmitter of this invention. The figure which shows the operation example of the block production | generation part in the transmitter of FIG. The figure which shows the other operation example of the block production | generation part in the transmitter of FIG. The figure which shows 3rd Embodiment of the single carrier transmitter of this invention. The figure which shows the operation example of the block production | generation part of the transmitter of FIG. The figure which shows 4th Embodiment of the single carrier transmitter of this invention. The figure which shows the operation example of the transmitter of FIG. The figure which shows 1st Embodiment of the single carrier receiver of this invention. FIG. 13 is a diagram illustrating an operation example of the receiver of FIG. 12. The figure which shows 2nd Embodiment of the single carrier receiver of this invention. The figure which shows the radio | wireless communications system comprised from the single carrier transmitter and single carrier receiver by one Embodiment of this invention. The figure which shows the structural example of the format of the signal from the single carrier transmitter by one Embodiment of this invention, and the signal from an OFDM transmitter. The figure which shows the structural example of the conventional single carrier transmitter The figure explaining the operation | movement of the block production | generation part and CP addition part of the conventional single carrier transmitter The figure which shows the structural example of the conventional single carrier transmitter which can perform the conventional 1st and 2nd expansion method. The figure which shows the output of the block production | generation part in the case of performing the conventional 1st expansion method, and the output of CP addition part. The figure which shows the output of the block production | generation part in the case of performing the 2nd conventional expansion method, and the output of CP addition part.

Explanation of symbols

11: Modulation units 12, 18, 20: Block generation unit 13: CP addition unit 14: D / A conversion unit 15: IF / RF transmission unit 16: Antenna 17: Storage unit 19: Pilot data generation unit 21: Encoding unit 22: Puncture unit 23: Interleave unit 31: Antenna 32: IF / RF reception unit 33: A / D conversion unit 34: CP removal unit 35: FFT processing unit 36: Channel estimation unit 37: Channel compensation unit 38: IFFT Processing units 39, 43: Storage units 40, 44: Synthesis unit 41: Demodulation unit 42: Bit metric calculation unit 45: Deinterleave unit 46: Depuncture unit 47: Decoding unit 51: Single carrier transmission unit 52: CP control signal reception unit (Symbol number information receiver)
53: Single carrier receiver 54: CP control signal transmitter

Claims (15)

A puncture unit for thinning out data bits from a plurality of data bits to obtain a data bit string;
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generator for generating a data block from a plurality of data symbols;
An adding unit that adds a copy of h data symbols (h is an integer of 1 or more) at the end of the data block as a cyclic prefix to the top of the data block to obtain a data block with a cyclic prefix;
A transmission unit for transmitting a signal of the data block with the cyclic prefix,
The k data symbols (k is an integer of 1 or more) preceding the h data symbols at the end of the first data block are the second data block preceding the first data block. A copy of the k data symbols at the end;
The transmitter according to claim 1, wherein the puncturing section increases the number of data bits to be thinned out as the number of data symbols to be copied in the adding section increases. An encoding unit that obtains an encoded data bit string by performing error correction encoding on the plurality of data bits;
The transmitter according to claim 1, wherein the puncturing unit obtains the data bit sequence by thinning out encoded bits from the encoded data bit sequence. An interleaving unit for interleaving the data bit string;
The transmitter according to claim 1, wherein the modulation unit modulates the interleaved data bit sequence to obtain a data symbol sequence. The puncturing unit thins out some of the plurality of coded bits obtained at the same state transition in the error correction coding,
The interleaving unit performs interleaving so that the remaining encoded bits after thinning out of the plurality of encoded bits are included in the last data symbol of the second data block. The transmitter described. The transmitter according to any one of claims 1 to 4, wherein the number k of data symbols to be copied is changed according to a transmission symbol rate. A puncture unit for thinning out data bits from a plurality of data bits to obtain a data bit string;
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generator for generating a data block from a plurality of data symbols;
An adding unit that adds a copy of h data symbols (h is an integer of 1 or more) at the end of the data block as a cyclic prefix to the top of the data block to obtain a data block with a cyclic prefix;
A transmission unit for transmitting a signal of the data block with the cyclic prefix,
K (k is an integer greater than or equal to 1) data symbols at the end of the first data block precede the h data symbols at the end of the third data block following the first data block. A copy of the k data symbols;
The transmitter according to claim 1, wherein the puncturing section increases the number of data bits to be thinned out as the number of data symbols to be copied in the adding section increases. An encoding unit that obtains an encoded data bit string by performing error correction encoding on the plurality of data bits;
The transmitter according to claim 6, wherein the puncturing unit obtains the data bit sequence by thinning out encoded bits from the encoded data bit sequence. An interleaving unit for interleaving the data bit string;
The transmitter according to claim 6 or 7, wherein the modulation unit modulates the interleaved data bit sequence to obtain a data symbol sequence. The puncturing unit thins out some encoded bits among a plurality of encoded bits obtained at the same state transition in the error correction encoding of the data bit sequence,
In the interleave unit, the remaining encoded bits after thinning out of the plurality of encoded bits are converted into the last data symbol among the data symbols preceding the h data symbols at the end of the third data block. The transmitter according to claim 8, wherein interleaving is performed so as to be included. The transmitter according to any one of claims 6 to 9, wherein the number k of data symbols to be copied is changed according to a transmission symbol rate. A puncture unit for thinning out data bits from a plurality of data bits to obtain a data bit string;
A modulator for modulating the data bit sequence to obtain a data symbol sequence;
A block generation unit that generates a plurality of data blocks each including a plurality of data symbols using the data symbol sequence, and a part of the plurality of data blocks is a pilot block including a plurality of pilot symbols;
An adder that obtains a plurality of data blocks with a cyclic prefix by adding a copy of h data symbols (h is an integer of 1 or more) at the end of each data block as a cyclic prefix to the top of the data block. When,
A transmission unit for transmitting a signal of each data block with a cyclic prefix,
The k data symbols (k is an integer of 1 or more) preceding the h data symbols at the end of the first data block are the second data block preceding the first data block. A copy of the k data symbols at the end;
The k data symbols at the end of the first data block are copies of the k pilot symbols preceding the h pilot symbols at the end of the pilot block following the first data block. Yes,
The puncturing unit increases the number of data bits to be thinned out as the number of data symbols to be copied in the adding unit increases. An encoding unit that obtains an encoded data bit string by performing error correction encoding on the plurality of data bits;
The transmitter according to claim 11, wherein the puncturing unit obtains the data bit sequence by thinning out encoded bits from the encoded data bit sequence. An interleaving unit for interleaving the data bit string;
The transmitter according to claim 11 or 12, wherein the modulation unit modulates the interleaved data bit sequence to obtain a data symbol sequence. The puncturing unit thins out some encoded bits among a plurality of encoded bits obtained at the same state transition in the error correction encoding of the data bit sequence,
The interleaving unit is configured such that the remaining encoded bits after the thinning out of the plurality of encoded bits are the remaining encoded bits after the thinning out of the plurality of encoded bits are the h at the end of the pilot block. The interleaving is performed so as to be included in a last pilot symbol among pilot symbols preceding one pilot symbol or to be included in a last data symbol of the second data block. Transmitter as described in. The transmitter according to any one of claims 11 to 14, wherein the number k of data symbols to be copied is changed according to a transmission symbol rate.

Images