JP2006180386A - Receiving apparatus and equalizing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a receiving apparatus capable of increasing equalization accuracy, without increasing the number of pilot symbols. <P>SOLUTION: A pilot symbol PS is extracted by a pilot symbol extraction unit 14. At the same time, a data symbol which can approximate the pilot symbol, is determined by a determination means 13 from a sequence of data symbols DS, extracted by a data symbol extraction unit 12 to make the data symbol a pseudo pilot symbol PS'. Equalization of received signals are performed by using a transmission path response inferred by both the symbols (PS, PS'). <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a receiving apparatus, and more particularly to a receiving apparatus having an equalizer that equalizes a received signal (data symbol) using pilot symbols included in a transmission signal, and an equalization method.

  An equalization method is known in which pilot symbols are used to equalize a signal received by a receiving apparatus. As an example, for a terrestrial digital broadcast receiver, in the case of this terrestrial digital broadcast, an OFDM (Orthogonal Frequency Division Multiplexing) method is adopted as a modulation method of a transmission signal, and the data symbol group subjected to multicarrier phase modulation is used. Pilot symbols are inserted at predetermined intervals. Then, in the equalizer of the receiving device that receives the inserted pilot symbol, the channel characteristic, that is, the channel response (H) is estimated using these pilot symbols, and the data symbol is based on the estimated channel response. Equalize the group.

  In addition, as a well-known technique relevant to this invention, there exists the following [patent document 1], for example. However, this patent document 1 does not disclose a technical idea that “a pseudo pilot symbol is generated in an equalization unit and equalization is performed using the same”, which is a feature of the present invention. In this patent document 1, a pilot signal is demodulated together with a data signal from an OFDM signal, and the channel characteristic of the data signal demodulated from the received signal is estimated using the demodulated pilot signal.

JP 2001-237798 A

  As described above, the equalizer estimates the transmission line response (H) acting on the pilot symbols, and performs interpolation such as linear interpolation or filter interpolation on the response (H) for each of the symbol direction and the carrier direction. Then, transmission path responses (Hd) for all data symbols are obtained, and predetermined complex division is performed to perform equalization.

  Therefore, if more pilot symbols are included in the received signal, it is desirable that more accurate equalization can be performed. By doing so, even if the receiving apparatus is placed in a multipath fading environment, for example, it is possible to faithfully follow fluctuations in transmission path characteristics due to the fading and maintain high reception quality.

  On the other hand, inserting a lot of pilot symbols as described above conversely reduces the number of data symbols. For example, the original information (data) that the terrestrial digital broadcasting system should have The transmission capacity will be reduced. However, since the predetermined transmission capacity must be ensured, it is difficult to increase the number of pilot symbols described above. Eventually, in a high-speed multipath fading environment, it becomes impossible to follow rapid fluctuations in transmission path characteristics. As a result, there arises a problem that data errors in the received signal increase.

  Therefore, the present invention is capable of faithfully following the fluctuations in the transmission path characteristics even in the above-mentioned high-speed multipath fading environment, etc. with a limited number of pilot signals, that is, the occurrence of the above-described data error. It is an object of the present invention to provide a receiving apparatus and an equalization method that can be minimized.

FIG. 1 is a diagram showing a principle configuration of the present invention. In this figure, reference numeral 10 represents a receiving device. Of the receiving apparatus 10, the equalizer 11 is particularly relevant to the present invention.
In general, the equalizer 11 first estimates a transmission path response using a pilot symbol included in a received signal (Y) received from an antenna and detected. Further, equalization is performed by dividing the received signal by the estimated transmission line response.

  In such an equalizer 11, it is the indexing means 13 that particularly characterizes the present invention. The output is a pseudo pilot symbol PS ′, and the general equalizer 15 only receives a normal pilot symbol PS as an input. However, the present invention is characterized in that the number of normal pilot symbols PS is apparently increased. For this purpose, a pseudo pilot symbol PS ′ is obtained from the indexing means 13 and further added to the normal pilot symbol PS. To do.

  In order to obtain such a pseudo pilot symbol PS ′, the present invention focuses on a data symbol (DS) group in the vicinity of a normal pilot symbol PS. That is, among these data symbol groups, a data symbol having a channel response close to (or coincident with) the channel response estimated by the pilot symbol PS is determined, and the pseudo pilot symbol PS ′ described above is determined using the determined data symbol. It is. A method for determining the pseudo pilot symbol PS ′ will be described later.

  In summary, the receiving apparatus 10 according to the present invention receives a transmission signal including a data symbol DS and a pilot symbol PS that have been subjected to multicarrier phase modulation, and equalizes the received data symbol DS with the pilot symbol PS. The equalizer 11 includes an indexing means 13 for determining, as a pseudo pilot symbol, a data symbol that can be approximated as a pilot symbol from a series of data symbols. The data symbol adjacent to the data symbol is equalized using the transmission path response estimated by both the pilot symbol PS and the pseudo pilot symbol PS.

  The present invention also provides a novel equalization method in the receiving apparatus, as is apparent from the description of FIG.

FIG. 2 is a flowchart showing the basic steps of the equalization method according to the present invention. In this figure,
First step S11: These symbols are extracted from a received signal including a series of data symbols DS into which pilot symbols PS are inserted in a distributed manner.

  Second step S12: A data symbol that can approximate the pilot symbol PS is determined from the extracted series of data symbols DS, and is specified as a pseudo pilot symbol PS '.

  Third step S13: Equalization for a series of data symbols DS is performed using the transmission path response estimated by both the pilot symbol PS and the pseudo pilot symbol PS '.

  Of the first, second, and third steps (S11, S12, S13), the step that most characterizes the equalization method according to the present invention is the second step S12. The second step S12 will be described more specifically. The second step S12 includes steps S121, S122, and S123 described below, that is, a series of data symbols (PS1 and PS2) between a pair of adjacent pilot symbols (PS1, PS2). Step S121 for calculating the magnitude of the reception vector VC for each of DS1, DS2,..., The magnitude of the transmission path response H estimated between the pair of pilot symbols, and the magnitude of each of the calculated reception vectors. And a step S122 for obtaining a difference between the two and a step S123 for determining a data symbol having a small difference as a pseudo pilot symbol PS '.

  More preferably, the second step S12 includes the following steps S124 and S125, that is, a step S124 of equalizing and demodulating a data symbol DSp with a small difference regarded as a pseudo pilot symbol PS ′, and symbolizing it again. The data symbol DSp having a small difference is complex-divided by the re-symbolized data symbol, and the transmission path response for the data symbol is estimated.

  According to the present invention, one (or a plurality) of data symbols whose transmission line responses are approximated to such an extent that they can be substituted for the original pilot symbols in a series of data symbols are determined and used as pseudo pilot symbols. The equalization accuracy of each data symbol can be increased without substantially increasing the number of pilot symbols.

  In this case, the above alternative data symbols can be equalized and demodulated and re-symbolized to compensate for variations in phase and amplitude during transmission, and with these re-symbolized data symbols, If possible data symbols are complex-divided, a transmission path response with higher estimation accuracy, that is, high-precision equalization can be achieved.

  In order to improve the equalization accuracy in this way, it is usually necessary to increase the number of original pilot symbols. However, this reduces the transmission capacity in the terrestrial digital broadcasting system, for example, as described above. However, according to the present invention, since the original number of pilot symbols is kept as it is, the equalization accuracy can be further improved and the occurrence of data errors can be suppressed without reducing the transmission capacity.

  In order to clarify the effect brought about by the present invention, first, equalization in a conventional receiving apparatus will be described.

  FIG. 8 is a diagram showing an example of the format of the OFDM signal. The example of this figure is an example of the OFDM signal in the European DVB-T system or the like, and the vertical axis of this figure indicates the symbol direction, that is, the time direction, and the horizontal axis indicates the carrier direction, that is, the frequency direction.

  White circles in the figure represent data symbols DS, that is, data signals, and black circles represent pilot symbols PS, that is, pilot signals. Pilot symbols PS are transmitted by every 12 subcarriers, and are cyclically arranged so as to return to the same subcarrier after 4 symbols.

  The receiving apparatus 10 in the present invention pays particular attention to “equalization” performed on the frequency axis by correcting the distortion of the amplitude and phase that the series of data symbols arranged as described above receives due to multipath fading or the like. is there.

  FIG. 9 is a diagram showing a configuration example of a conventional equalizer 11. Throughout the drawings, similar components are denoted by the same reference numerals or symbols.

  The equalizer 11 of this figure is composed of the data symbol extraction unit 12, the pilot symbol extraction unit 13 and the equalization unit 15 already shown in FIG. 1, and the equalization unit 15 includes the constituent elements 21 to 25 shown in the figure. including.

  The pilot symbol generated by the pilot symbol generation unit 21 on the reception side and the pilot symbol from the pilot symbol extraction unit 13 received and extracted from the transmission side are input to the pilot symbol complex division unit 22, where By performing complex division, the variation of the received vector is calculated. By passing this variation through the symbol filter 23, pilot symbol interpolation (FIG. 10) is performed in the symbol direction, and when passing through the carrier filter 24, pilot symbol interpolation (FIG. 11) is performed in the carrier direction. Thus, this is the transmission path response H estimated by the pilot symbols.

  The data symbol complex division unit 25 performs complex division on the data symbol DS from the data symbol extraction unit 12 based on the estimated transmission line response H. Here, equalization is performed on the received data symbol DS. Further, the data symbol demodulator 26 obtains the required data symbol DS, that is, transmission data of “1, 0”.

FIG. 10 is a diagram showing a state of interpolation by the symbol filter 23.
FIG. 11 is a diagram showing a state of interpolation by the carrier filter 24.

  10 and 11, white circles (DS) and black circles (PS) are data symbols DS and pilot symbols PS as described above. In FIG. 10, an interpolated pilot symbol is generated by, for example, linear interpolation, and in FIG. 11, the interpolated pilot symbol is generated through, for example, an FIR filter. That is, the phase rotation and amplitude variation of the received vector in the time axis direction and the frequency axis direction are detected, and the transmission line response H is estimated.

  Referring to FIG. 8, FIG. 10 and FIG. 11, assuming that a transmission path response is estimated by a pilot symbol PS and equalization of the data symbol DS is required, it is required to increase the equalization accuracy. The pilot symbol PS was further increased. However, as described above, increasing the number of pilot symbols PS reduces the number of data symbols DS correspondingly, resulting in a disadvantage of reducing the transmission capacity.

  Therefore, the present invention introduces the above-described indexing means 13 and uses the specific data symbol DS as an apparent pilot symbol, that is, a pseudo pilot symbol, with the original number of pilot symbols PS being left as it is. This pseudo pilot symbol is shown as PS ′ in FIG. 10, for example. The indexing means 13 will be described more specifically.

  FIG. 3 is a diagram showing a schematic configuration of the indexing means 13. In this figure, the indexing means 13 includes a calculation function unit 31, a response estimation function unit 32, and a difference detection function unit 33.

  Here, the calculation function unit 31 calculates the magnitude of the received vector for each of a series of data symbols (DS1, DS2,...) Between a pair of adjacent pilot symbols (PS1, PS2). The unit 32 estimates the channel response between the pair of pilot symbols, and the difference detection function unit 33 compares the estimated channel response with the calculated magnitudes of the received vectors. Is obtained. Here, a data symbol having a small difference is determined as a pseudo pilot symbol PS ′.

  FIG. 4 is a diagram showing a more preferred schematic configuration of the indexing means 13 shown in FIG. That is, the indexing unit 13 further includes a re-symbolization function unit 34 that equalizes and demodulates the data symbol having a small difference, which is regarded as the pseudo pilot symbol PS ', and further symbolizes it.

  Further, the variation (phase and amplitude) of the reception vector of the data symbol is calculated by complex-dividing the data symbol having a small difference by the re-symbolized data symbol DS ′ from the re-symbolization function unit 34. A complex division unit 35 for generating a pseudo pilot signal PS ″ with higher accuracy.

  Here, the principle of channel response estimation according to the present invention, that is, the principle of equalization according to the present invention will be described with reference to the drawings.

  FIGS. 5A, 5B and 5C are diagrams for explaining the principle of operation of the indexing means 13 according to the present invention, that is, for explaining the principle of equalization of the present invention. . In this figure, (a) shows the estimation of the transmission path response that is normally performed by linear interpolation, for example, and the transmission path response H is first estimated by a pair of adjacent pilot symbols PS1 and PS2. Then, the transmission path response Hd estimated by the received vector relating to the series of data symbols DS1, DS2,... Between the pilot symbols PS1 and PS2 is obtained by linear interpolation of the illustrated transmission path responses, and these DS1, DS2,. Is divided by its Hd to equalize. In this case, since the estimated transmission line responses for the three data symbols DS1, DS2, and DS3 are obtained with the two pilot symbols PS1 and PS2, the accuracy is considerably deteriorated.

  Therefore, as shown in FIG. 5B, one data symbol having a transmission line response Hd closest to the transmission line response H is determined. That is, the magnitude of the received vector (the length of the arrow in the figure) is calculated for each series of data symbols, and the data symbol whose magnitude and transmission path response H approximate (or match) is specified. In FIG. 5B, the data symbol (DS2 in FIG. 5A) indicated by the alternate long and short dash line is specified. Therefore, this data symbol DS2 is used as a pseudo pilot symbol PS '.

  Thus, when the pseudo pilot symbol PS 'is determined, the original pilot symbol PS1 is used for each of the remaining data symbols (DS1 and DS3 in FIG. 5C) using the symbol PS'. Along with PS2 and PS2, the transmission line response is estimated by linear interpolation. That is, for data symbol DS1, transmission path response Hd1 is estimated based on linear interpolation between original pilot symbol PS1 and pseudo pilot symbol PS ', and for data symbol DS3, original pilot symbol PS2 and pseudo pilot symbol PS are estimated. The transmission line response Hd2 is estimated on the basis of the linear interpolation with '.

  Here, an example of the overall configuration of the receiving apparatus 10 shown in FIG. 1 is shown. In this case, an example in which the indexing means 13 of FIG. 1 has the configuration of FIGS. 3 and 4 described above is shown.

  FIG. 6 is a diagram showing a specific configuration example of the receiving apparatus according to the present invention. In the receiving apparatus 10 of this figure, the upper components 12, 25 and 26 are the same as the conventional components 12, 25 and 26 shown in FIG. 9, and the lower components 13, 22 and 21 are also the same. These are the same as the components shown in FIG. Furthermore, the components 24, 23B and 23A shown in these intermediate stages substantially correspond to the components 24 and 23 shown in FIG.

The calculation function unit 31 in FIG. 3 is represented as a “data symbol size” detection unit 31 at the left end of FIG.
The response estimation function unit 32 in FIG. 3 is shown as the components 13, 22, 21, 23A and 41 in FIG.
The difference detection function unit 33 in FIG. 3 is represented as an “error comparison” unit 33 in the center portion of FIG.
The symbol filter 23 of FIG. 3 is represented in FIG. 6 as a first “symbol interpolation (linear approximation)” portion 23A and a second “symbol interpolation (linear approximation)” portion 23B.

Further, the re-symbolization function unit 34 in FIG. 4 includes a “data symbol complex division” unit 51 (corresponding to 25), a “data symbol demodulation” unit 52 (corresponding to 26), and a “re-symbolization ( Mapping) "part 53,
The complex division unit 35 in FIG. 4 is represented as a “complex division” unit 35 at the right end of FIG.

  In FIG. 6, the symbol filter 23 </ b> A generates a transmission path response (a variation ΔPS of the reception vector PS) estimated for the original pilot symbol PS, and the output thereof is further “a magnitude of the transmission path response”. Input to the detection unit 41. This produces H in FIG.

  Here, the “error comparison” unit 33 calculates the “data symbol size” calculated by the “data symbol size” detection unit 31 (the length of the dotted arrow (DS) in FIG. 5; That is, the phase rotation amount) and the magnitude of the transmission line response by the pilot symbol PS from the “transmission line response magnitude” detecting unit 41 are input, and the magnitude comparison is made for each data symbol (DS1, DS2,...). The data symbol having the smallest difference is identified as the pseudo pilot symbol PS ′.

  Since this pseudo pilot symbol PS ′ contains some phase fluctuations (rotations) and amplitude fluctuations, these fluctuations are preferably removed, and normal pseudo pilot symbols having no fluctuations are re-symbolized. "Mapping" section 53, and performs complex division (35) with PS 'from "error comparison" section 33 to calculate the variation ΔPS' of the pseudo pilot symbol. By adding the variation ΔPS ′ to the variation ΔPS and inputting it to the “symbol interpolation (linear approximation)” section 23B, linear interpolation shown in FIG. 5C is executed. After that, equalization of data symbols is performed using the output through the carrier filter 24 as in the prior art.

The above equalization operation will be described in more detail using mathematical expressions.
The pilot signal arrangement of European and Japanese OFDM systems is as shown in FIG. 8, where K is a carrier number and 1 is a symbol number. The pilot symbol is a symbol known on the receiving side, and the magnitude Y of the reception vector of the pilot symbol is
Y = H ・ C + N
It becomes. Here, H is a transmission path response, C is a pilot symbol value, and N is noise.

Dividing Y by the pilot symbol value C known at the receiving side, H ′ = Y / C = H + N / C
When the noise is small (N≈0), the transmission path characteristics (transmission path response) can be estimated with H ′ = H. By estimating the channel characteristics for all pilot symbols and interpolating in the symbol direction and the carrier direction, respectively (see FIGS. 10 and 11), the channel characteristics for all data symbols become clear. The demodulator can demodulate the correct data by dividing the value of the received data symbol by this transmission path characteristic value.

  Note that the above-mentioned channel response interpolation estimated for the pilot symbols is performed by filtering interpolation using an ideal low-pass filter that passes a delayed wave within the guard interval period in the carrier direction and the maximum Doppler frequency in the symbol direction. It can be carried out. In general, an FIR filter is used as a low-pass filter in the carrier direction, but step interpolation or linear interpolation is performed in the symbol direction. However, if the transmission path characteristics fluctuate in the time direction, such as reception by a mobile unit, if the pilot symbol interval in the symbol direction is large, the fluctuation of the transmission path characteristics cannot be tracked in a high-speed multipath fading environment. Data errors increase.

  On the other hand, if a large number of pilot symbols are injected, fluctuations in transmission path characteristics can be followed even in a high-speed multipath fading environment. However, as described above, the transmission capacity of the system will be reduced this time.

  Therefore, as described above, the present invention obtains a transmission path response of a series of data symbols between pilot symbols in the symbol direction, and uses data symbols that can be used as pilot symbols as further pilot symbols. Increase the apparent number (see Figure 5).

  When the subcarrier is phase-modulated, the amplitude is proportional to the magnitude of the transmission line response. Therefore, the transmission line response Hd for the data symbol is obtained by linear approximation between two pilot symbols (PS1-PS2). Data symbol Cd is then received as follows.

Yd = H · Cd + N
This Yd is equalized by the estimated transmission path response Hd as shown in the following equation. In other words, if the noise can be ignored,
Cd ′ = Yd / Hd = H · Cd / Hd
Thus, the value of the data symbol is extracted. If Hd = H,
Cd ′ = Cd
It becomes.

However, when Hd≈H, Cd ′ undergoes phase rotation and amplitude fluctuation with respect to Cd due to H / Hd. Therefore, Cd ′ is once decoded and symbolized again (see FIGS. 4 and 6). FIG. 7 is a diagram showing how re-symbolization is performed. Since the phase rotation is added to the above-mentioned Cd ′ as shown in the left coordinate of the figure, it is restored to the normal phase by re-symbolization as shown in the right coordinate of the figure. Thus, even when Hd≈H,
Cd ′ = Cd
It can be. When the received signal Yd is divided by this Cd ′,
Hd ′ = Yd / Cd ′ = H · Cd / Cd ′ = H
Thus, the transmission line response H for the data symbol Cd is estimated.

  Therefore, the difference detection function unit 33 compares the magnitudes of a series of all data symbols Cd between pilot symbols and the magnitudes of the corresponding transmission path responses H, and detects data symbols having a small difference between them. Further, using the transmission path response for the detected data symbol and the transmission path response for the original pilot symbol, the “symbol interpolation (linear approximation)” unit 23B (FIG. 6) again performs linear approximation to Estimate the channel response of data symbols. Thereby, estimation accuracy can be improved.

It is a figure which shows the principle structure of this invention. 4 is a flowchart illustrating a method according to the present invention. It is a figure which shows schematic structure of an indexing means. It is a figure which shows the more preferable schematic structure of an indexing means. It is a figure for demonstrating the principle of operation of an indexing means. It is a figure which shows the specific structural example of the receiver which concerns on this invention. It is a figure which shows the mode of re-symbolization. It is a figure which shows an example of the format of an OFDM signal. It is a figure which shows an example of the conventional equalizer. It is a figure which shows the mode of the interpolation by a symbol filter. It is a figure which shows the mode of the interpolation by a carrier filter.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Receiving device 11 Equalizer 12 Data symbol extraction part 13 Indexing means 14 Pilot symbol extraction part 15 Equalization part 21 Pilot symbol production | generation part 22 Pilot symbol complex division part 23 Symbol filter 24 Carrier filter 25 Data symbol complex division part 26 Data Symbol demodulation unit 31 Calculation function unit 32 Response estimation function unit 33 Difference detection function unit 34 Re-symbolization function unit 35 Complex division unit

Claims (8)

In a receiving apparatus having an equalizer that receives a transmission signal including multi-carrier phase-modulated data symbols and pilot symbols, and equalizes the received data symbols with the pilot symbols,
The equalizer includes indexing means for determining, as a pseudo pilot symbol, a data symbol that can be approximated as the pilot symbol from a series of the data symbols, and a data symbol adjacent to the data symbol is determined as the pilot symbol. A receiving apparatus that performs equalization using a channel response estimated by both a symbol and the pseudo pilot symbol.   The indexing means estimates a channel response between the pair of pilot symbols and a calculation function unit that calculates the magnitude of the received vector for each of the series of data symbols between a pair of adjacent pilot symbols. A response estimation function unit, and a difference detection function unit that compares the estimated magnitude of the transmission path response with the calculated magnitude of each received vector to obtain a difference between the two, and the difference is small. The receiving apparatus according to claim 1, wherein a data symbol is determined as the pseudo pilot symbol.   3. The indexing unit further comprises a resymbolization function unit that equalizes and demodulates the data symbol with the small difference regarded as the pseudo pilot symbol and further resymbolizes the data symbol. Receiver.   A complex division unit that calculates a variation of a received vector of the data symbol by complex dividing the data symbol having a small difference with the re-symbolized data symbol from the re-symbolization function unit; The receiving apparatus according to claim 3. A first step of extracting these symbols from a received signal including a series of data symbols into which pilot symbols are inserted dispersedly;
A second step of determining a data symbol capable of approximating the pilot symbol from the extracted series of data symbols and specifying the symbol as a pseudo pilot symbol;
A third step of equalizing the series of data symbols using a channel response estimated by both the pilot symbols and the pseudo pilot symbols;
An equalization method comprising: The second step includes
Calculating the magnitude of the received vector for each of the series of data symbols between a pair of adjacent pilot symbols;
Comparing the magnitude of the channel response estimated between the pair of pilot symbols and the magnitude of each of the calculated received vectors to obtain a difference between the two,
6. The equalization method according to claim 5, further comprising the step of determining the data symbol having a small difference as the pseudo pilot symbol.   7. The method according to claim 6, wherein the second step further includes a step of equalizing and demodulating the data symbol having the small difference regarded as the pseudo pilot symbol, and further re-symbolizing it. Method.   8. The equalization method according to claim 7, further comprising a step of complex-dividing the data symbol having a small difference by the re-symbolized data symbol and estimating a transmission line response for the data symbol.

Images