DESCRIPTION
Title of Invention: CARRIER WAVE REPRODUCTION DEVICE AND
CARRIER WAVE REPRODUCTION METHOD
Technical Field
[OOOl]
[Description about related applications]
The application claims priority based 011 Japanese Patent Application
No. 2012-099167 (filed on April 24, 2012), tlie disclosure of which is
hereby incorporated by reference in its entirety.
The present invention relates to a carrier wave reproduction device
and a carrier wave reproduction method and in particular, relates to a
carrier wave reproduction device which can be suitably applied to a
demodulation device for multilevel trans~nissioni ncluding error correction
and a carrier wave reproduction method.
Backgroulid Art
[0002]
In digital communication, as a modulation and de~nodulation
niethod for efficiently transmitting and receiving data, the quadrature
amplitude modulation (QAM) method in which both phase infornlation and
amplitude information is used for data discrimination is known. At
present, with tlie increase of the demand of a large capacity wireless
conlliiunication system, it is required to increase the modulation
multi-level number.
[0003]
However, when the lllodulation multi-level number is increased, a
problem in which a transmission error rate due to noise increases and noise
imlnunity decreases occurs. In particular, a phase noise generated by a
reference oscillator (Local Oscillator; LO) provided in a transmissio~l
device or a reception device increases indeterminacy of phase information
and deteriorates a bit error rate (BER) characteristic. Accordingly, for
example, in order to perform highly reliable data comtnunication by using
a multi-level QAM method in which 256 or more signal points are used, a
phase error produced by the phase noise needs to be compensated for at
high accuracy. Further, at the same time, im~nunityto error due to other
factors such as thermal noise and the like needs to be improved.
[0004]
Fig. 14 is a block diagram showing a configuration of a
denlodulation device according to a related techtlology in a reception
device of a digital wireless device syste~n. Referring to Fig. 14, the
demodulation device includes a reference oscillator 121, a detector 122, an
analogldigital (AID) converter 123, a carrier wave reproductio~lp hase lock
loop (Phase Lock Loop; PLL) in which a phase rotator 124, a phase error
detector 125, a loop filter 126, and a n~t~llericcaol ntrol oscillator 127 are
connected in a loop, a QAM symbol deu~appi~lugn it 128 which converts a
received symbol into a bit string, and an error co~rectiond ecoder 129.
[0005]
The reference oscillator 121 outputs a reference signal having a
predetermined fixed frequency. The detector 122 performs quadrature
detection of an input signal by using the reference signal and generates an
Ich (In-phase channel) baseband signal and a Qch (Quadrate-phase
Channel) baseband signal. The generated baseband signal is converted
into a digital signal through the AID converter 123.
[0006]
The phase rotator 124 corrects the phase error by rotating the phase
of the received symbol that corresponds to the Ich digitalized baseband
signal and the Qch digitalized baseband signal according to output
information of the numerical control oscillator 127. The output signal of
the phase rotator 124 is inputted to the phase error detector 125. The
phase error detector 125 detects the phase error which remains in the
received symbol and outputs it to the loop filter 126. The loop filter 126
removes an unnecessary high frequency component included in the phase
error and outputs it to the nu~l~ericaclo ntrol oscillator 127. The
nulnerical control oscillator 127 generates phase error infor~nationw hich
specifies a phase rotation anlount in the phase rotator 124 from the output
of the loop filter 126 and outputs it.
[0007]
As described above, by the operation of the carrier wave
reproduction PLL in which the phase rotator 124, the phase error detector
125, the loop filter 126, and the numerical control oscillators 127 are
connected in a loop, a stable phase locked state can be realized. Whereby,
the phase error can be compensated for.
[OOOS]
The received sytnbol to which the phase noise correction is
performed by the phase rotator 124 is inputted to the phase error detector
125 and also, inputted to the QAM symbol demapping unit 128. The
QAM symbol delnapping unit 128 calculates a received bit string
corresponding to the received sy~nbol from the received symbol and the
error correction decoder 129 perforrns an error correction process and
outputs the received bit string. When a soft decision decoder which
receives likelihood infor~natiorli ndicating a certainty of each received bit
and performs the c o r r e c t i o ~p~ro cess is used for the error correction decoder
129, the QAM symbol de~napping unit 128 outputs the bit string in which
the likelihood infornlation is reflected. For example, the QAM s y ~ l ~ b o l
de~napping unit 128 which outputs the bit string in which the likelihood
information is reflected is described in patent literature 1.
[0009]
As described above, the demodulation device according to the
related technology compensates for the phase error by the carrier wave
reproduction PLL and realizes improvement in error imnlunity by the error
correction process performed in a later stage. However, there is a case in
which when the accuracy of the phase error detector 125 decreases by the
phase noise included in the baseband signal outputted by the detector 122,
a ther~naln oise, and the like, the satisfactory BER performance cannot be
obtained. In this case, a technology in which the phase error
compensation accuracy is improved by adaptively adjusting a bandwidth of
the loop filter 126 in the carrier wave reproduction PLL is disclosed in
patent literature 2, patent literature 3, and patent literature 4. However,
there is a case in which even when these technologies are used, a sufficient
effect cannot be obtained.
[OO lo]
As a method other than the method using the carrier wave
reproduction PLL, thereis a method in which a known signal (a pilot signal)
is inserted in the transmission signal and the phase error is compensated
for by using this known signal. Fig. 15 is a block diagram showing a
configuration of the demodulation device using this method. Referring to
Fig. 15, the denlodulatioti device includes a reference osciilator 131, a
detector 132, an AID converter 133, a QAM sy~nbol demapping unit 137,
and an error correction decoder 138 like the demodulation device shown in
Fig. 14. Further, the de111odulation device includes an interpolation filter
135 which estimates the phase error between the received pilot sy~nbols
from the received pilot symbol corresponding to the pilot signal, a delay
circuit 134 which delays a signal by the number of synlbols corresponding
to a delay incurred by the interpolation process, and a phase rotator 136
which corrects the estimated phase error.
[OOll]
The demodulation device shown in Fig. 15 performs quadrature
detection of tlie input signal in the detector 132 by using tlie reference
signal with the fixed frequency outputted by the reference oscillator 13 1
and generates the Icli (In-phase channel) baseband signal and the Qch
(Quadrate-phase Channel) baseband signal like the demodulation device
sliown in Fig. 14. Tlie generated baseband signal is converted into the
digital signal through the AID converter 133.
[0012]
The received symbol that corresponds to the Ich digitalized
baseband signal and the Qch digitalized baseband signal is inputted to the
delay circuit 134. Here, only when tlie received symbol is the pilot
symbol corresponding to the known pilot signal, the received sy~iibol is
also inputted to the interpolation filter 135. Tlie interpolation filter 135
estimates the phase error in the received symbol between the pilot symbols
fro111 a plurality of the pilot symbols by the interpolation process. The
phase rotator 136 rotates the pliase of tlie received symbol based on the
pliase error infor~ilation outputted by the interpolation filter and corrects
the pliase error in the received symbol.
[00 131
The received synlbol to which tlie phase error is compensated for by
the interpolation filter 135 is inputted to tlie QAM symbol de~nappingu nit
137 like the demodulation method perfor~iied by the de~iiodulation device
shown in Fig. 14 and converted into the (soft decision) received bit string.
After this process, the error correction process is performed in the error
correction decoder 138 and it is outputted as output data.
[00 141
For example, the denlodulation method using the interpolation filter
wliicli uses the pilot symbol is described in non-patent literature 1 and
non-patent literature 2.
Citation List
Patent Literature
[00 151
Patent literature 1 : International Publication 201 11068 119
Patent literature 2: Japanese Patent Application Laid-Open No.
2000-1 01666
Patent literature 3: Japanese Unexainined Patent Application
Laid-Open No. 2003-53 1523
Patent literature 4: Japanese Patent Application Laid-Open No.
2011-101177
Non Patent literature
[00 161
Non Patent literature 1: Arnaldo Spalvieri, Luca Barletta,
"Pilot-Aided Carrier Recovery in the Presence of Phase Noise", IEEE
Transactions on Comtnunications, July 1, 201 1, pp.1966 to 1974
Non Patent literature 2: Volker Simon, Andreas Senst, Michael
Speth, Heinrich Meyr, "Phase Noise Estimation via Adapted Interpolation",
IEEE Global Telecon~munications Conference, November 25, 2001, pp.
3297 to 3301
Summary of Invention
Technical Proble~n
[0017]
The following analysis is performed by the inventor of the present
invention.
[OO 181
The demand of a large capacity wireless coniiilunication system is
greatly increasing and it is required to increase the modulation multi-level
number. However, when the phase noise of a LO signal which has a large
influence on a transmission characteristic is reduced, the cost greatly
increase. Further, in order to inlprove the perforlnance of tlie error
correction function, the denlodulator is required to stably operate even
when a carrier to noise ratio (CIN) is low. The effect of the phase noise
conipensation by the carrier wave reproduction PLL according to the
related technology is limited with respect to this problem. For example,
when the level of the phase noise included in the baseband signal outputted
by the detector is great compared with the signal multi-level nu~nbero f the
QAM method, the satisfactory BER perfor~i~anccea nnot be obtained and it
is difficult to perfor111 a large capacity and high-quality data
communication.
[0019]
On the other hand, a de~nodulation niethod in which the inserted
pilot signal is used as a clue and the phase error is conipensated for by
passing through the interpolation filter is basically used for a niodulatio~l
method such as QPSK or the like in which the number of signal points is
small. Therefore, it is difficult to obtain tlie characteristic required for a
multi-level modulation method based on only this demodulation method.
Further, a problenl ill which when an insertion ratio of the pilot signal
increases, the transnlission capacity decreases and the effect of a
multi-level modulation decreases occurs.
[OOZO]
Accordingly, it is desired to improve a bit error rate characteristic
witliout decreasing tlie trans~nissionc apacity.
An object of the present invention is to provide a carrier wave reproduction
device and a carrier wave reproduction method which contribute to the
demand.
Solution to Problenl
[0021]
A carrier wave reproduction device according to a first
aspect of the present invention includes
an interpolation filter which estimates a phase error of a
received symbol based on a pilot symbol included in the received
symbol,
a first phase rotator which rotates the phase of the received
symbol according to the phase error estimated by the interpolation
filter and outputs the sy~nboal s a first output symbol,
a phase error compensation unit which compensates for the
phase error which remains in the first output synlbol and outputs the
symbol as a second output symbol,
a QAM symbol de~napping unit which calculates a first bit
string corresponding to the first output symbol and calculates a
second bit string corresponding to the second output symbol, and
an error correction decoder ~vhichp erforms error correction
of a bit error in the first bit string and outputs the first bit string
after error correction. Wl~erein the phase error co~npensation unit
compensates for the phase error which remains in the first output
synlbol by referring to the first bit string after error correction.
[0022]
A carrier wave reproduction method according to a second aspect of
the present invention includes
a step of estimating a phase error of a received symbol based on a
pilot symbol included in the received symbol,
a step of rotating the phase of the received symbol according to the
estimated phase error and outputting the symbol as a first output symbol,
a step of calculating a first bit string corresponding to the first
output synlbol,
a step of performing error correction of a bit error in the first bit
string,
a step of compensating for the phase error which remains in the first
output symbol by referring to the first bit string after error correction and
outputting the symbol as a second output symbol,
a step of calculating a second bit string corresponding to the second
output sytnbol, and
a step of performing error correction of a bit error in the second bit
string.
Advantageous Effects of Invention
[0023]
By using the carrier wave reproduction device and the carrier wave
reproduction method according to the present invention, a bit error rate
characteristic can be improved without decreasing a transmission capacity.
Brief Description of Drawings
[0024]
[Fig. 11 Fig. 1 is a block diagram showing a configuration of a
carrier wave reproduction device according to a first exemplary
embodiment as an example.
[Fig. 21 Fig. 2 is a block diagram showing a configuration of an
interpolation filter in a carrier wave reproduction device according to a
first exemplary embodiment as an example.
[Fig. 31 Fig. 3 is a block diagram showing a configuration of a tap
coefficient generation unit in a carrier wave reproduction device according
to a first exemplary embodiment as an example.
[Fig. 41 Fig. 4 is a figure showing a structure of a pilot symbol and
a payload symbol in a first exemplary embodiment as an example.
[Fig. 51 Fig. 5 is a flowchart showing operation of a carrier wave
reproduction device according to a first exemplary embodime~it as an
example.
[Fig. 61 Fig. 6 is a flowchart showing operation of an
interpolation filter in a carrier wave reproduction device according to a
first exemplary embodiment as an example.
[Fig. 71 Fig. 7 is a block diagram showing a configuration of a
carrier wave reproduction device according to a second exemplary
embodi~llenta s an example.
[Fig. 81 Fig. 8 is a figure showing a configuration of a loop filter
in a carrier wave reproduction device according to a second exe~nplary
embodiment as an example.
[Fig. 91 Fig. 9 is a block diagram showing a configuration of a
~iumerical control oscillator in a carrier wave reproduction device
according to a second exemplary embodiment as an example.
[Fig. 101 Fig. 10 is a flowchart showing operation of a carrier
wave reproductio~id evice according to a second exemplary enlbodime~~ast
an example.
[Fig. 111 Fig. 11 is a block diagram showing a configuration of a
carrier wave reproduction device according to a third exemplary
embodiment as an example.
[Fig. 121 Fig. 12 is a block diagram showing a configuration of a
low path filter in a carrier wave reproduction device according to a third
exemplary en~bodime~aist an example.
[Fig. 131 Fig. 13 is a flowchart showing operation of a carrier
wave reproduction device according to a second exe~llplarye ~nbodi~neanst
an example.
[Fig. 141 Fig. 14 is a block diagram showing a configuration of a
carrier wave reproduction device according to a related technology.
[Fig. 151 Fig. 15 is a block diagram showing a configuration of a
carrier wave reproduction device according to a related technology.
Description of E~nboditnents
[0025]
First, an outline of an exemplary embodiment will be described.
Further, a drawing reference code described in figures is shown as an
exa~ilplei n order to help the understanding and it is not intend to limit the
present invention to the embodiment shown in the drawing.
[0026]
Fig. 1 is a block diagram showing a configuration of a
deniodulation device equipped with a carrier wave reproductio~i device
according to one exemplary embodiment as an example.
Referring to Fig. 1, a carrier wave reproduction device (10) includes an
interpolation filter (15) which estimates a phase error of a received 'symbol
based on a pilot syrnbol included in the received symbol, a first phase
rotator (17) which rotates tlie phase of the received sy~nbol according to
the phase error estimated by tlie ir~terpolation filter (15) and outputs the
symbol as a first output symbol, a phase error compensation unit (19)
which compensates for the phase error which remains in the first output
symbol and outputs the sy~nbola s a second output symbol, a QAM symbol
demapping unit (3 1) which calculates a first bit string corresponding to the
first output sylllbol and calculates a second bit string corresponding to the
second output symbol, and an error correction decoder (32) which performs
error correction of a bit error in the first bit string and outputs the bit
string. The phase error co~npensationu nit (19) colnpensates for the phase
error which remains in the first output sy~nbolb y referring to the first bit
string after error correction.
[0027]
Fig. 7 and Fig. 11 are block diagralns sl~owing a detailed
configuration of the phase error con~pensation unit (19) shown in Fig. 1 as
an example. Referring to Fig. 7 and Fig. 11, a carrier wave reproduction
device (a carrier wave reproduction device 10a shown in Fig. 7, a carrier
wave reproduction device lob sl~own in Fig. 11) includes a buffer (18)
which temporarily holds the first output symbol outputted from the first
phase rotator (17) and a phase error con~pensation unit (a phase error
colnpensation unit 19a shown in Fig. 7, a phase error compensation unit
19b shown in Fig. I l ) rnay include a second phase rotator (a second phase
rotator 55 shown in Fig. 7, a second phase rotator 61 shown in Fig. 11)
which rotates the phase of the first output symbol outputted from the buffer
(18) and generates the second output symbol.
[0028]
Referring to Fig. 7 and Fig. 11, the phase error colnpensation unit
(the phase error conipensation unit 19a shown in Fig. 7, the phase error
colnpensation unit 19b shown in Fig. 11) may include a QAM symbol
mapping lullit (a QAM symbol mapping unit 51 shown in Fig. 7, a QAM
symbol mapping unit 62 sllown in Fig. 11) which generates the sy~nbol
according to the bit string after error correction as a third output symbol.
[a0291
Referring to Fig. 7 and Fig. 11, the phase error con~pensation unit
(the phase error colnpensation unit 19a shown in Fig. 7, the phase error
colnpensation unit 19b shown in Fig. 11) lnay compensate for the phase
error which remains in the first output sylnbol by detecting a phase
difference between at least either the first output symbol or the second
output syn~boal nd the third output symbol.
[0030]
Referring to Fig. 7, the phase error co~npensation unit (19a)
includes a phase error detector (52) which detects the phase difference
between the second output symbol and the third output sytnbol and a
numerical control oscillator (54) which calculates a phase rotation amount
based on the phase difference from which a high frequency component
included in the phase difference is removed and the second phase rotator
(55) may geilerate the second output symbol by rotating the phase of the
first output symbol outputted from the buffer (18) according to the phase
rotation amount.
[003 11
Referring to Fig. 7, the phase error compensation unit (19a) may
include a loop filter (53) whicl~ removes the high frequency coiiiponent
included in the phase difference and output the phase difference after
removal of the high frequency component to the nunlerical control
oscillator (54).
[0032]
Referring to Fig. 11, the phase error compensation unit (19b) may
include a phase error detector (63) which detects the phase difference
between the first output syrnbol and the third output synlbol and a low pass
filter (64) which perforins the averaging of the phase difference and
calculates the phase rotation amount. The second phase rotator (61) rnay
rotate the phase of the first output symbol outputted from the buffer (18)
according to the phase rotation alnount and generate the second output
symbol.
[0033]
As described above, in the carrier wave reproduction device (10)
according to one exe~nplary embodiment, the interpolation filter (15)
corrects the phase error in the received symbol from the received pilot
synlbol by the interpolation process and the phase error conlpensation unit
(19) further corrects the phase error which remains in the received symbol
after correction. Here, the phase error conlpensation unit (19) performs
error correction of a result of a first stage correction performed by the
interpolation filter (15) through the denlapping process and estimates the
phase error wit11 high accuracy by using the result as the reference data.
By repeatedly performing a series of the processes: a demapping process,
an error correction process, and a phase error compensation process, the
carrier wave reproduction device which estimates the phase error with high
accuracy without increasing the pilot symbol and has the excellent bit error
rate characteristic can be realized.
LO0341
The carrier wave reproduction device according to one exemplary
embodiment has an excellent bit error rate characteristic even under severe
noise environment which is created by a phase noise, a tllermal noise, and
the like and can provide large capacity and high-quality data
communication service. By using such carrier wave reproduction device,
even when the phase noise level is great conlpared with the modulation
multi-level number, the bit error rate characteristic of the reception device
can be intproved witl~outr emarkably reducing the transnlission capacity.
[0035]
Further, in the present invention, the following embodiment can be
realized.
[Embodiment I]
A carrier wave reproduction device is the same as the carrier wave
reproduction device according to the first aspect mentioned above.
[Embodiment 21
A carrier wave reproduction device may include a buffer which
temporarily holds the first output symbol outputted from the first phase
rotator and the phase error compensation unit may include a second phase
rotator which rotates the phase of the first output symbol outputted from
the buffer and generates the second output symbol.
[Embodiment 31
The phase error compensation unit may include a QAM symbol
mapping unit which generates the symbol corresponding to the bit string
after error correction as a third output symbol.
[Embodiment 41
The phase error compensation unit may compensate for the phase
error which remains in the first output symbol by detecting a phase
difference between at least either the first output sy~ilbol or the second
output symbol and the third output symbol.
[E~nbodiment 51
The phase error colnpensation unit may include a phase error
detector which detects the phase difference between the second output
symbol and the third output symbol and a numerical control oscillator
which calculates a phase rotation amount based on the phase difference in
wl~icll the hig11 frequency component included in the phase difference is
renioved and the second phase rotator may generate the second output
syn~bolb y rotating the phase of the first output sylnbol outputted from the
buffer according to the phase rotation amount.
[Embodiment 61
The phase error co~npensatiou unit may include a loop filter which
removes the high frequency component included in the phase difference
and outputs the phase difference after removal of the high frequency
coniponent to the numerical control oscillator.
[Embodiment 71
The phase error compensatio~l unit may include a phase error
detector which detects the phase difference between the first output symbol
and the third output symbol and a low pass filter which performs the
averaging of the phase differe~lcea nd calculates the phase rotation amount.
The second phase rotator may rotate the phase of the first output sy~nbol
outputted fron~ the buffer according to the phase rotation amount and
generate the second output symbol.
[Embodiment 81
The interpolation filter may be a finite impulse response type filter
including a plurality of registers which hold the pilot symbol, a plurality of
~nultipliers whose number is equal to the number of registers, and a
plurality of adders whose number is equal to the number of registers.
[Embodiment 91
The interpolatio~f~il ter may determine an initial value of a tap
coefficient and a step width used when the tap coefficient is updated
according to a phase noise level and a carrier to noise ratio and may update
the tap coefficient the number of times that is equal to the number of
intervals between the pilot symbols included in the received syn~bol.
[Embodi~nent 101
A demodulation device based 011 the quasi-synchronous detection
method may include the above-mentioned carrier wave reproduction
device.
[Embodiment 1 I]
A carrier wave reproduction method is the same as the carrier wave
reproduction method according to the second aspect mentioned above.
[Embodiment 121
A carrier wave reproduction method may repeat
a step for compensating for a phase error which remains in a 11-th
output symbol by referring to a n-th bit string after error correction (n is a
natural number) and outputting the sy~nbola s a (n+l)th output synlbol,
a step for calculating a (n+l)th bit string correspo~lding to the
(n+l)th output symbol, and
a step for performing error correction of a bit error in the (n+l)th
bit string until n is equal to a predetermined number while increlnenting n
or all the errors are corrected in the error correction.
[Embodilnent 131
A carrier wave reproduction device based on a quasi-synchronous
detection deniodulation method which de~nodulatesth e data signal including
a known pilot signal may include
an interpolation filter which estimates a phase error to all the
received symbols including the received pilot sp~nbol from the received
pilot synlbol,
a first phase rotator which corrects the phase error by rotating the
phase of the received sylnbol by the phase error amount estimated by the
interpolation filter,
a phase error co~npensation unit which compensates for a residual
phase error which renlains in the output sylnbol of the first phase rotator,
an error correction decoder which generates the data inputted to the
phase error compensation unit as a reference, and
a QAM sy~nbol deinapping unit which converts the output of the
phase error co~npensation unit and the output symbol of the first phase
rotator into the input bit string inputted to the error correctio~l decoder.
The phase error co~npensationm ay be performed two times: the phase error
compensation perfornled by the interpolation filter and the phase error
conlpensation perfornled by the phase error compensation unit.
[Embodiment 141
A carrier wave reproduction device may include a buffer which
te~nporarilyh olds the output of the first phase rotator.
[Enlbodilnent 151
The phase error conlpensation unit may include
a second phase rotator which rotates the phase of the output sylnbol
of the buffer,
a QAM symbol mapping unit which generates a replica of a
transn~ission symbol sequence corresponding to the output data from the
output data of the error correction decoder,
a phase error detector which detects a phase difference between the
output sylnbol of the second phase rotator and the replica,
a loop filter which removes the high frequency component included
in the output of the phase error detector, and
a numerical control oscillator wl~ich calculates a phase rotation
alnount from the output of the loop filter and may perform a second-stage
phase error con~pensation by the second phase rotator that is connected
with the phase error detector, the loop filter, and the numerical control
oscillator so as to form a loop.
[Embodiment 161
The phase error compensation unit may include
a second phase rotator which rotates the phase of the output sy~nbol
of the buffer,
a QAM synibol mapping unit which generates a replica of the
trans~nission sy~nbol sequence corresponding to the output data from the
output data of the error correction decoder,
a phase error detector which detects a phase difference between the
output sy~nbolo f the buffer and the replica, and
a low pass filter which performs the averaging of the output value
of the phase error detector and may perform the second-stage phase error
compensation by rotating the phase of the output sy~nbolo f the buffer by
the second phase rotator by using the output of the low pass filter.
[Embodiment 171
The interpolation filter may be a finite ilnpulse response (FIR) type
filter including registers which receive only the pilot sylnbol and hold a
fixed number of the received pilot synlbols, tllultipliers whose number is
equal to the number of the registers, and cunlulative addition devices
whose number is equal to the number of the registers and tlie tap
coefficient of the interpolation filter may be updated the number of times
that is equal to the number of intervals between the received pilot synlbols.
[Embodiment 181
A carrier wave reproduction method for a quasi-synchronous
detection demodulation system which denlodulates the data signal
including a known pilot signal may include
an interpolation process step in which an interpolation process
which estimates a phase error to all the received sy~nbols including a
received pilot symbol by interpolation from the received pilot symbol is
performed,
a first-time phase error compensation step in which the estimated
phase error is corrected,
a de~napping process step in which demapping of the received
sylllbol to which the phase error is cotnperlsated for is performed to obtain
a bit string,
an error correctidn process step in'which the error correction is
performed to the bit string obtained by demapping, and
a step in which when the number of phase error conlpensation times
is smaller than the number of times set in advance, a second-time phase
error compensation and the successive-time phase error cotnpensation are
performed.
In a second-time and successive-time phase error compensation
processes, the data to which the error correction process is performed may
be used as a reference and
the second-time and successive-time phase error compensation
process step, the denlapping process step, and tlie error correction process
step may be repeatedly perfor~ned until the number of phase error
conlpensatio~i times is equal to the number o f times set in advance or all
the errors are corrected by the error correction process.
[Embodiment 191
The second-time and successive-time phase error compensation
process step may include
a step in which a replica o f the transmission symbol sequence is
generated by mapping the data to which the error correction process is
performed to a QAM symbol,
a step in which the phase error is detected by using the replica as a
reference,
a step in which the high frequency component included in the phase
error is removed,
a step in which phase rotation information is generated by
converting the phase error inforlnation from which the high frequency
component is removed, and
a step in which the phase error compensation is performed by the
phase rotation information and
the phase error may be detected by feeding the received sy~nbol to
which the phase error compensation is performed back to the step which
detects the phase error.
[Embodiment 201
The second-time and successive-time phase error compensation
process step may include
a step in which a transmission symbol replica is generated by
mapping the data to which the error correction process is performed to the
QAM symbol,
a step in which the phase error is detected by using the transmission
symbol replica as a reference, and
a step in which the averaging of the detected phase error is
performed by passing through a low pass filter and a phase rotation signal
is generated and
the phase error compensation may be performed based on the phase
rotation signal.
[Embodiment 211
The interpolation process step may include
a step in which an initial value of a tap coefficient is set,
a step in which a product-sum operation process is performed,
and
two kinds of tap coefficient update steps in which the tap
coefficient is updated by the cutnulative addition and
the product-sum operation process step and the tap coefficient
update step may be repeated the nu~nber of times that is equal to the
nu~nbero f intervals between the received pilot symbols.
[0036]
Next, the exetnplary embodiment will be described in detail with
reference to the drawing. In the following exenlplary embodiment, a
~l~odulationm ethod in which phase information is used for data
identification is used. In the followi~lge xemplary embodiment, the QAM
lnodulation method is taken as an example. However, when another
~llodulation method which uses the phase inforniation for data
identification is used, this exelnplary enlbodiment can be applied to the
another nlodulation method. 111 this exemplary embodiment, as a
detection method of a demodulation method, the quasi-synchronous
detection method in which the detection is perfornled by using a periodic
signal with a fixed frequency generated by a demodulation device itself
and the phase error is removed after the detection is used. When this
method is used, it is not necessary to generate a signal which completely
synchronizes with the carrier wave.
[0037]
(Exemplary embodiment 1)
A carrier wave reproduction device according to a first exemplary
embodiment will be described with reference to the drawing. Fig. 1 is a
block diagram showing a configuration of a delnodulation device 20
equipped with the carrier wave reproduction device 10 according to the
exemplary enlbodilnent as an example.
[0038]
The demodulation device 20 is a device which delliodulates a data
signal including a known pilot signal. Referring to Fig. 1, the
delnodulation device 20 includes a reference oscillator 11, a detector 12, an
AID converter 13, and the carrier wave reproduction device 10. The
carrier wave reproduction device 10 includes the delay circuit 14, the
interpolation filter 15, a tap coefficient generation unit 16 that relates to
the interpolation filter 15, the phase rotator 17, the phase error
compensation unit 19, the QAM symbol detnapping unit 3 1, and the error
correction decoder 32.
[0039]
The reference oscillator 11 outputs a reference signal with a fixed
frequency to the detector 12. The detector 12 performs quadrature
detection of an IF (Intermediate Frequency) input signal and generates an
Ich baseband signal and a Qch baseband signal. The generated baseband
signal is converted into the digital signal by passing through the AID
converter 13.
[0040]
When the number of signal points of the QAM method that is a
lnodulation and demodulation means is 21n (m is a positive integer), the
transtnitted bit string is divided into sections, each of which is composed
of nl bits at a transmission side and each of m bits is mapped to one signal
point anlong 2111 signal points. The mapped signal point call be expressed
by a complex value and this is called a trans~nission symbol. Tlie Ich
digitalized baseband signal and the Qch digitalized baseband signal that
are obtained by the AID converter 13 are the received sy~nbols
corresponding to the trans~nissions ymbols. The received symbol can be
expressed by a co~nplex value whose real part corresponds to the Ich
baseband signal and whose imaginary part corresponds to the Qch baseband
signal. Hereinafter, this coniplex value is called a "received symbol" and
this is inputted to tlie carrier wave reproduction device 10. On the other
hand, the output of the carrier wave reproduction device 10 is conlposed of
a bit string correspondi~lg to the transmission symbol estimated from the
received symbol.
[0041]
The delay circuit 14 receives the received symbol, delays it by a
fixed ti~ile interval according to a processing time required by the
interpolation filter 15 described later, and outputs it to tlie phase rotator
17.
[0042]
The interpolation filter 15 receives only the known pilot symbol
inserted at the transmission side in the received synlbol composed of the
Ich digitalized baseband signal and the Qch digitalized baseband signal,
calculates an estimation value of the phase error included in the received
sy~nbol from a plurality of pilot synlbols with respect to the received
sy~nbolb etween the respective pilot symbols, and outputs it to the phase
rotator 17.
[0043]
The tap coefficient generation unit 16 generates the data used for
the process for estinlating the phase error in the interpolation filter 15 from
statistical infornlation about a phase noise and a thermal noise, and
supplies it to the interpolation filter 15.
[0044]
The phase rotator 17 receives the output of the delay circuit 14 and
the output of the interpolation filter 15, rotates the phase of the received
symbol that is the output of the delay circuit 14 based on the output data of
the interpolation filter 15, and outputs the received synlbol to the buffer
18.
[0045]
The phase error conlpensation unit 19 receives the output synlbol of
the phase rotator 17 via the buffer 18, estimates the phase error which
renlains in the output sy~nbol of the phase rotator 17, and performs a
correction. In order to estinlate the phase error which remains in the
output symbol, the phase error cornpensation unit 19 uses output data of
the error correction decoder 32.
[0046]
The QAM synlbol detnapping unit 31 receives the output symbol of
the buffer 18 and the output synlbol of the phase error cornpensation unit
19 via the selector 30, calculates the bit string corresponding to the output
sytnbol from the output symbol of the selector 30, and outputs the bit string
to the error correction decoder 32 described later.
[0047]
Further, in this exemplary embodiment, as an example, a case in
which a soft decision decoder which receives likelihood information to
each transmission bit and performs the correction process is used for the
error correction decoder 32 will be explained. In this case, the QAM
symbol denlapping unit 31 outputs the bit string in which the likelihood
information indicating a certainty of the received bit is reflected. As an
example, a unit described in patent literature 1 can be used for the QAM
symbol delnapping unit 3 1 .
[0048]
The error correction decoder 32 receives the output data of the QAM
syrnbol delnapping unit 31, estimates the bit error included in the output
sylnbol of the buffer 18 or the phase error con~pensation unit 19, and
performs correction.
[0049]
Fig. 2 is a block diagram showing a configuration of the
interpolation filter 15 in the carrier wave reproduction device 10 according
to the exemplary embodiment as an example. Referring to Fig. 2, the
interpolation filter 15 is a finite impulse response (FIR) type filter and
includes registers 41 and 44, selectors 42 and 45, a nlultiplier 43, and
adders 46 and 47.
[OOSO]
Fig. 3 is a block diagranl showing a configuration of the tap
coefficient generation unit 16 as an example. Referring to F i g . 3, the tap
coefficient generation unit 16 includes a ROM table 38 which generates an
initial value of a tap coefficient h, and a ROM table 39 which generates a
step width A, used when the tap coefficient hJ is updated. The ROM table
38 receives a phase noise level (PNL) and a carrier to noise ratio (CNR)
and outputs the initial value of the tap coefficient 11, according to these
values. On the other hand, the ROM table 39 receives the PNL and the
CNR and outputs the step width A, used when the tap coefficient 11, is
updated according to these values.
[005 11
Next, the operation of the carrier wave reproduction device 10 will
be described. The carrier wave reproduction device 10 receives the
received symbol that corresponds to the Ich digitalized baseband signal
and the Qch digitalized baseband signal obtained by passing through the
AID converter 13, compensates for the phase error included in the received
symbol, and corrects the bit error doe to disturbance such as residual phase
noise, thermal noise, or the like through a process for decoding an error
correction code. The carrier wave reprodnctio~l device 10 estimates the
transmission bit string by performing the above-mentioned process and
output it.
[0052]
Fig. 4 is a figure showing a structure of the input symbol string of
the carrier wave reproduction device 10 as an example.
Referring to Fig. 4, the input symbol string is corilposed of a transmission
data maill body (payload) and a known pilot symbol determined in advance.
In the following explanation, for ease of explanation, it is assumed that one
pilot symbol is inserted for every M input symbols at a fixed interval.
Here, M is a positive integer and indicates a pilot symbol interval.
[0053]
The input symbol string to the carrier wave reproduction device 10
is inputted to the delay circuit 14. At the same time, only the received
pilot symbol corresponding to the pilot symbol that is iuserted for every M
syinbols is inputted to the interpolation filter 15. Namely, an input event
to the interpolation filter 15 occurs once for every M times. Because the
pilot spnlbol is the k11ow11 symbol deter~nined in advance, the phase error
included in the reception pilot can be extracted by comparing the received
pilot symbol with the symbol determined in advance. On the other hand,
with respect to the payload symbol, because the transmission symbol is not
known, an error occurs when the phase error is extracted.
[0054]
The i~lterpolatio~fill ter 15 is an FIR filter with (2t + 1) taps (t is a
positive integer) and estimates the phase error included in (M - 1) payload
symbols and one pilot symbol from (2t + 1) input pilot symbols. A
relation between the input pilot synlbol and the payload symbol to estimate
the phase error will be described in detail below.
[OOSS]
It is assumed that the nunlber of symbols inputted to the carrier
wave reproduction device 10 is one synlbol per one unit time, (2tM + 1)
input sy~nbols are represented by So, SI, S2, ... ,and S2tM and in particular,
the pilot sy~nbolsi nserted for every M symbols are represented by So, SM,
S ~ MS,3 M,. .., and S2tM (2t + 1) pilot sy~nbolsa re successively inputted to
the interpolation filter 15 and the interpolation filter 15 successively
estin~ates the phase error included in the (M - 1) payload sy~nbols
expressed by the following equation 1 and the pilot sy~nbol StM in the M
unit tinle until a next pilot sy~nbol S(2t + I)M is inputted after the last pilot
sy~nbolS ztM is inputted.
[0056]
[Equation 11
, , , , : J . .... S,.,, _,. S,, ,,.,. S,,,, ,, S, ,, 2. .". S ,,,,., [.,,'?J
[0057]
Further, in Equatioii 1, [MI21 indicates a maxi~num integer among
integers equal to or smaller than M/2. Accordingly, when M is even,
[MI21 is equal to MI2 and when M is odd, [MI21 is equal to (M - 1)/2,
This is similarly applied to [(M - 1)/2].
[OOSS]
The interpolation filter 15 estimates the phase error included in (M
- 1) payload sy~nbols by updating the tap coefficient of the filter
sequentially. The information required to update the tap coefficient is
supplied from the tap coefficient generation unit 16. Accordingly, the
delay a~noonto f the delay circuit 14 is equal to or Inore than (tM + [(M -
1)/2]) unit time.
[0059]
The phase rotator 17 rotates the phase of the received symbol that is
the output of the delay circuit 14 based on the output data of the
interpolation filter 15 and outputs the received symbol. The received
synlbol is expressed by the complex value whose real part corresponds to
the Ich baseband signal and whose imaginary part corresponds to the Qch
baseband signal and similarly, the output data of the interpolation filter is
also expressed by the complex value. Accordingly, as an exalnple, the
phase rotator 17 can be realized by a complex value multiplier. However,
the phase rotator 17 is not limited to this.
[0060]
The output symbol of the phase rotator 17 is the received symbol
string in which the phase noise is corrected by the interpolation filter 15.
However, when the phase noise level is high or when a pilot symbol
interval M is large, a problem of the residual phase error occurs.
Accordingly, the output sy~nbolo f the phase rotator 17 is co~npensatedf or
by the phase error colnpensation unit 19 and the error correction decoder
32.
[0061]
The phase error colnpensation unit 19 receives the output symbol of
the phase rotator 17 via the buffer 18, estimates the phase error which
re~nainsin the output symbol, and corrects it.
[0062]
The symbol used as the reference when the phase error
colnpensation unit 19 esti~natesth e phase error which re~nainsi n the output
sy~nbol is generated by using the error correction decoder 32. First, a
process for correcting a bit error included in the output sylnbol of the
phase rotator 17 is perfor~ned by the QAM sylnbol denlapping unit 31 and
the error correction decoder 32.
[0063]
The QAM symbol demapping unit 31 receives the output sy~nbolo f
the phase rotator 17 via the buffer 18 and the selector 30, generates the bit
string i11,which the likelihood information is reflected to each bit of which
the output sy~llbol is composed, and outputs it to the error correction
decoder 32.
[0064]
The error correction decoder 32 receives the bit string, performs a
first-time erior correction decoding process, estimates the trans~nissionb it
string, and outputs the estimated transmission bit string to the phase error
compensation unit 19.
[0065]
The phase error compensation unit 19 uses the received estimated
transmission bit string to estimate the residual phase error included in the
output symbol of the phase rotator 17 and performs a second-time phase
error co~npensation.
[0066]
The received symbol to which the second-time phase error
compensation is applied in the phase error compensation unit 19 is inputted
to the error correctio~di ecoder 32 via the selector 30 and the QAM synlbol
demapping unit 31. The error correction decoder 32 perfor~ns a
second-time error correction decoding process and estinlates the
transmission bit string.
[0067]
Subsequently, a third-time phase error compelisation process and a
third-time error correction process are perfornled in a similar manner.
Further, a fourth-time phase error con~pensation process, a fourth-time
error correction process, a fifth-time phase error compensation process,
and a fifth-time error correction process can be repeated. After the phase
error compensation process and the error correction process are repeated a
predetermined number of times, the carrier wave reproduction device 10
outputs the output data of the error correction decoder 32 as the output data
of the carrier wave reproduction device 10.
[0068]
It is expected that a certainty of the transmission bit estimated by
the error correction decoder 32 and an accuracy of the phase error
estimated by the phase error conlpensation unit 19 by using the estimated
transmission bit are gradually improved when the phase error
compensation process and the error correctioll process are repeated many
times.
[0069]
Fig. 5 is a flowchart showing the operation of the carrier wave
reproduction device 10 as an example. Namely, it shows the operations
performed until the output data is generated after the received symbol is
inputted.
[0070]
Referring to Fig. 5, tlle carrier wave reproduction device 10
receives the received symbol outputted fro111 the AID converter 13 (Step
Al).
[0071]
The interpolation filter 15 uses only the pilot symbol in the
received symbol string and estimates the phase error included in the
payload sylllbol between the pilot symbols by the interpolation process
(Step A2).
100721
Next, the phase rotator 17 performs the first-time phase error
compensation (Step A3).
100731
The QAM symbol demapping unit 31 performs demapping of the
received sy~nbolto which the phase error compensation is performed to tlie
bit string (Step A4) and the error correctioll decoder 32 performs an error
correction using the error correctiotl code (Step A5).
[0074]
When the number of phase error conlpensation times k is equal to or
greater than the number of times T set in advance (k and T are positive
integers) or when all the errors are successfully corrected by the error
correction (Step A5), the data to which the error correction is performed
(Step A5) is used as the output of the carrier wave reproduction device 10
(Step A6).
[0075]
On the other hand, when the ~ l u n ~ b eorf phase error compensation
times k is smaller than tlie nunlber of times T set in advance, a (k +
1)th-time phase error compensation is performed (Step A7). Further, in
the phase error compensation (Step A7), the output obtained through the
error correction decoding (Step A5) is used as the reference data.
Therefore, the phase error compensation (Step A7) is different from tlie
phase error compensation (Step A3) performed by the phase rotator 17
based 011 the interpolation between pilots (Step A2).
[0076]
The operation of the interpolation filter 15 shown in Fig. 2 will be
described by using the flowchart shown in Fig. 6.
As shown in Fig. 4, it is assunled that the input symbol string is composed
of the payload syn~bol and the pilot symbol and one pilot symbol is
inserted for every M input syl~lbolsa t a fixed interval. At this time, the
input event to the interpolation filter 15 occurs once for every M times.
[0077]
Referring to Fig. 6, the received pilot symbol is inputted to the
interpolation filter 15 (Step B1) and set to the leftmost register among
registers 41 shown at an upper side of Fig. 2.
Further, the initial value is set to each register 44 which holds the tap
coefficient (Step B2).
[0078]
When the inputted received pilot symbol is represented by S n(n~ is
an integer), the received pilot sylnbol string held in tlie registers 41 is
represented by S n ~S,( ,,.I)M.,. ., and S(,,.Z~)Min order from the leftrnost
register to the rightmost register. The received pilot sy~ilbols are
multiplied by the tap coefficients ho, hl ..., and hzt by the multipliers 43,
respectively and the sum is calculated by the adder 47 (Step B3).
[0079]
The interpolation filter 15 outputs the calculated sum as the
interpolation data (Step B4). The interpolation data is expressed by the
following equation 2.
[OOSO]
[Equation 21
The interpolation filter 15 repeats the above-li~entioned
product-sum operation shown by equation 2 M times while updating the tap
coefficients ho, 111, ..., and hzt. The update of the tap coefficient is
perfornied by a method to update the tap coefficient (step B5) and a method
to update the tap coefficient (Step B6) shown in Fig. 6. When the tap
coefficient is updated in the steps B5 and B6, the step widths (Aj , Aj+,)
that are values set in advance are added to the latest tap coefficient.
However, the step width used in the step B5 is different from the step width
used in tlie step B6.
[0082]
After a (k)th-time product-sum operation is performed, if k <= [(M
- 1)/2], the interpolation filter 15 performs the tap coefficielit update
process of step B5 and if k > [(M - 1)/2], the illterpolation filter 15
performs the tap coefficient update process of Step B6. The interpolation
filter 15 updates the tap coefficient by using the registers 44, the adders 46,
and the selectors 45 connected to the adders 46 shown in Fig. 2. The
selector 45 selects one of the tap coefficielit update process of step B5 and
the tap coefficient update process of Step B6.
[0083]
Further, the initial value of the tap coefficient h, set in Step B2 and
the step width A, used when the tap coefficient is updated in Steps B5 and
B6 are determilied based on the phase noise level (PNL) and the carrier to
noise ratio (CNR) by the thermal noise and supplied from the tap
coefficie~itg elieratioll unit 16.
[0084]
(Exemplary embodiment 2)
A carrier wave reproduction device according to a second
exemplary embodiment will be described with reference to the drawing.
Fig. 7 is a block diagram showing a collfiguratioli of a demodulatioli device
20a equipped with the carrier wave reproduction device 10a according to
the exemplary embodiment as an example.
[0085]
The reference oscillator 11, the detector 12, and the A/D converter
13 that are used for the demodulation device 20 (Fig. 1) accordiug to the
first exemplary embodiment are also used for the demodulation device 2023.
Further, the delay circuit 14, the interpolation filter 15, the tap coefficient
generation device 16, the phase rotator 17, the buffer 18, the selector 30,
the QAM symbol demapping unit 31, and the error correction decoder 32
that are used for the carrier wave reproduction device 10 (Fig. 1) accordilig
to the first exemplary embodinlent are also used for the carrier wave
reproduction device 10a.
[0086]
Referring to Fig. 7, the phase error conlpensation unit 19a includes
the QAM symbol nlapping unit 51, the phase error detector 52, the loop
filter 53, the numerical control oscillator 54, and the phase rotator 55.
[0087]
The QAM synlbol niapping unit 51 generates a replica of the
transtnissioa symbol sequence corresponding to the output data from the
output data of the error correction decoder 32 and outputs it to the phase
error detector 52.
[OOSS]
The phase error detector 52 detects the phase difference between
the output sytnbol of the phase rotator 55 and the output sytnbol of the
QAM symbol mapping unit 51 and outputs it to the loop filter 53.
[0089]
The loop filter 53 removes the high frequency noise included in the
output of the phase error detector 52 and outputs it to the numerical control
oscillator 54.
[0090]
The ntinierical control oscillator 54 calculates the phase rotation
a~nount from the output of the loop filter 53 and outputs it to the phase
rotator 55.
[0091]
~ k r et,h e phase error detector 52, the loop filter 53, the nulnerical
control oscillator 54, and the phase rotator 55 are connected so as to form a
loop.
100921
Next, the operation of the carrier wave reproduction device 10a
sliowtl in Fig. 7 will be described. Here, the delay circuit 14, the
interpolation filter 15, the tap coefficient generation unit 16, the phase
rotator 17, the buffer 18, the selector 30, the QAM symbol demapping unit
31, and the error correction dccodcr 32 that are used for the carrier wave
reproduction device 10 (Fig. 1) according to the first exemplary
embodiment are also used for the carrier wave reproduction device 10a.
Accordingly, the operation of inputting the received symbol to the carrier
wave reproduction device 10a, the operations of the delay circuit 14, the
interpolation filter 15, the tap coefficient generation unit 16, and the phase
rotator 17, and the operations of the QAM symbol demapping unit 31 and
the error correctio~d~e coder 32 are the same as those of the carrier wave
reproductio~l device 10 (Fig. 1) according to the first exemplary
embodiment. Therefore, an explanation of these operations is omitted.
LO0931
The QAM symbol mapping unit 51 generates the replica of the
transmission symbol by mapping the output data of the error correction
decoder 32 to the signal point in the QAM method.
When the 2"'-QAM neth hod is used as the modulation method, each replica
of the tra~ls~iiissiosny ~nbolo utputted by the QAM symbol mapping unit 51
corresponds to one of the co~nplexv alues indicating 2"' signal points of the
QAM method.
[0094]
Further, when the error correction decoder 32 is a soft output
decoder which outputs likelihood information indicating a certainty of the
data after correction together with the data after correction, the symbol
outputted by the QAM symbol mapping unit 51 is a value obtained by
su~nming the transmission symbols that are weighted by using the
likelihood information. In this case, the output of the QAM sy~nbol
mapping unit 51 does not necessarily correspond to one of the co~nplex
values representing 2"' signal points o f the QAM method.
[0095]
The phase error detector 52 receives the replica of the transmission
sylnbol generated by the QAM symbol inapping unit 5 1 . Moreover, the
phase error detector 52 receives the received symbol whose phase error is
compensated for by the interpolation filter process via the phase rotator 55.
The received sy~nbol is represented by a complex number a and the replica
of the transmission symbol generated by the QAM symbol mapping unit 51
is represented by a complex number P . When a real part of a colnplex
number x is represented by Re(x) and an imaginary part of a complex
number x is represented by Im(x), the output value of the phase error
detector 52 is expressed by equation 3.
[0096]
[Equation 31
[0097]
When an argument of the complex number u is represented by 0,
and an argu~itent o f the con~plexn umber P is represented by OD, Equation 3
is rewritten to Equation 4.
[0098]
[Equation 41
[0099]
Accordingly, the output value of the phase error detector 52 that is
calculated by Equation 3 is a value corresponding to the phase difference
between two input symbols. The phase error detector 52 outputs the
output value to the loop filter 53.
[O 1 001
Fig. 8 is a figure showing a configuration of the loop filter 53 as an
example. Referring to Fig. 8, the loop filter 53 includes a multiplier 91.
The loop filter 53 multiplies the input value expressed by Equation 3 by a
constant number y and outputs a calculation result. The constant nulnber
y can be adaptively changed according to discriniination between the pilot
sylnbol and the payload symbol, the number of phase error colnpensation
times, or the like. The loop filter 53 outputs the output value to the
numerical control oscillator 54.
[OlOl]
Fig. 9 is a block diagram showing a configuration of the nunierical
control oscillator 54 as an example. Referring to Fig. 9, the nulnerical
control oscillator 54 includes a cun~ulativea dder composed of an adder 34
and a register 35 and a ROM table 36 which converts a cun~ulativea ddition
result 0 into a value of cos(6) and a value of sin(0).
[O 1021
The numerical control oscillator 54 adds the data held in the
register 35 to the input data and updates the data held in the register 35
with the addition result. The addition result is converted into a cosine
function value and a sine function value and these values are outputted to
the phase rotator 55 by the ROM table 36.
[0 1031
The phase rotator 55 outputs the sylnbol a obtained by rotating the
phase of a sy~nboal' by an angle of -0 according to Equation 5 based on the
output symbol a' of the buffer 18 and the output values (cos(0) and sin(0))
of the numerical control oscillator 54.
[0104]
[Equation 51
Rc(n) .: cos(0). Rc(z1') +sin(O). Im(nl)
IIII(U) - cos(0). I111(u') - sin(0). Kc(crl)
[0105]
The phase rotator 55, the phase error detector 52, the loop filter 53,
and the nulnerical control oscillator 54 are connected so as to for111 a loop
as shown in Fig. 7.
I0 1061
The received sylnbol to which the second-time phase error
compensation is applied by the phase rotator 55 is inputted to the error
correction decoder 32 via the selector 30 and the QAM syn~bold enlapping
unit 3 1.
[0107]
The error correction decoder 32 performs the second-time error
correction decoding process and estimates the transmission bit string.
The transn~ission bit string is supplied to the phase error detector 52
through the QAM symbol napping unit 51 ill a similar way to that
previously used. Tlle output symbol of tlie buffer 18 is supplied to tlie
phase error detector 52 via the phase rotator 55 as the other input. Not
only the way of supplying the output sy~nbol but also the symbol value
supplied from the buffer 18 is the same as that of the previous time.
[0108]
It is expected that the number of errors included in the replica of
the transmission symbol supplied through the QAM sylnbol mapping unit
51 is smaller than the nu~nber of errors included in the replica of the
transmission synlbol at the last time. By this effect, the phase error
compensation can be performed with high accuracy compared to the
second-time phase error compensation. Subsequently, the same process
is performed. Therefore, the phase noise can be compensated for with
more high accuracy by repeatedly updating the replica of the transmission
symbol.
[O 1091
Fig. 10 is a flowchart showing the operation of the carrier wave
reproduction device 10a according to the exemplary embodiment equipped
with the phase error compensation unit 19a as an example. Namely, it
shows the operations performed until the output data is obtained after the
received symbol is inputted.
[OllO]
Referring to Fig. 10, the carrier wave reproduction device 10a
receives the received synlbol outputted from the AID converter 13 (Step
Cl).
[Olll]
The interpolation filter 15 uses only the pilot symbol in the
received symbol .string and estimates the phase error included in the
payload synlbol between the pilot symbols by the interpolation process
(Step C2).
[0112]
Next, the phase rotator 17 performs the first-time phase error
compensation (Step C3).
[0113]
The QAM symbol demapping unit 31 performs demapping of the
received symbol to which the phase error conlpensation is performed to the
bit string (Step C4) and the error correction decoder 32 performs the error
correction using the error correction code (Step C5).
[0114]
When the number of phase error compensation times k is equal to or
greater than the nunlber of times T set in advance (k and T are positive
integers) or when all the errors are successfully corrected by the error
correction (Step C5), the data to which the error correction is performed
(Step C5) is used as the output of the carrier wave reproduction device 10a
(Step C6).
[0115]
On the other hand, when the number of phase error compensation
times k is smaller than the number of titnes T set in advance, the QAM
symbol mapping unit 51 creates the replica of the transmission symbol by
mapping the data to which the error correction is performed (Step C5) to
the QAM symbol (Step C7).
[0116]
Next, the phase error detector 52 uses the replica of the
transmission symbol as the reference symbol and detects the phase error
(Step C8).
[0117]
Next, the loop filter 53 removes the high frequency component of
the detected phase error (Step C9).
[0118]
Next, the numerical control oscillator 54 converts the phase error
information from which the high frequency component is removed and
generates phase rotation information (Step ClO).
[0119]
Next, the phase rotator 55 performs the (k + 1)th-time phase error
compensation according to the phase rotation information (Step C11).
[0120]
Further, as shown in Fig. 7, in the (k + 1)th-time phase error
compensation, when the operation is viewed in synlbol unit, the feedback is
perfornled for the phase error detection performed by the phase error
detector 52.
[0121]
(Exemplary embodiment 3)
A carrier wave reproduction device according to a third exemplary
embodiment will be described with reference to the drawing. Fig. 11 is a
block diagram sl~owing a configuration of a de~nodulation device 20b
equipped with the carrier wave reproduction device lob according to the
exemplary embodiment as an example.
[O 1221
Referring to Fig. 11, the reference oscillator 11, the detector 12,
and the AID converter 13 that are used for the de~nodulationd evice 20 (Fig.
1) according to the first exeinplary embodiment are also used for the
demodulation device 20b.
The delay circuit 14, the interpolation filter 15, the tap coefficient
generatio~ld evice 16, the phase rotator 17, the buffer 18, the selector 30,
the QAM symbol demapping unit 31, and the error correction decoder 32
that are used for the carrier wave reproduction device 10 (Fig. 1) according
to the first exemplary embodinlent are also used for the carrier wave
reproduction device lob.
[0123]
Referring to Fig. 11, the phase error coinpensation unit 19b
includes the phase rotator 61, the QAM symbol mapping unit 62, the phase
error detector 63, and the low pass filter 64.
[0 1241
The phase rotator 61 receives the output of the phase rotator 17 via
the buffer 18 and rotates the phase thereof by using the output of the low
pass filter 64 described later.
The output of the phase rotator 61 is inputted to the QAM sy~llbol
demapping unit 31 via the selector 30.
[0125]
The QAM symbol mapping unit 62 generates the replica of the
transmission sy~nbol sequence corresponding to the output data from the
output data of the error correction decoder 32 and outputs it to the phase
error detector 63.
[0 1261
The phase error detector 63 detects the phase difference between
the output symbol of the buffer 18 and the output symbol of the QAM
synlbol mapping unit 62 and outputs it to the low pass filter 64.
[O 1271
The low pass filter 64 perfor~ns the averaging of the output of the
phase error detector 63 and outputs it to the phase rotator 61.
[0128]
Next, the operation of the carrier wave reproductioli device lob will
be described. The delay circuit 14, the interpolation filter 15, the tap
coefficient generation device 16, the phase rotator 17, the buffer 18, the
selector 30, the QAM synibol demapping unit 31, and the error correction
decoder 32 that are used for the carrier wave reproduction device 10 (Fig.
1) according to the first exelnplary embodiment are also used for the
carrier wave reproduction device lob. Accordingly, the operation of
inputting the received synlbol to the carrier wave reproduction device lob,
the operations of the delay circuit 14, the interpolation filter 15, the tap
coefficient generation unit 16, and the phase rotator 17, and the operations
of the QAM synlbol mapping unit 62 and the error correction decoder 32
are the sanle as those of the carrier wave reproduction device 10 (Fig. 1)
according to the first exelnplary emboditnent. Therefore, the explanation
of these operations is omitted.
[0129]
The QAM symbol mapping unit 62 generates the replica of the
transmission sylnbol by mapping the output data of the error correctioll
decoder 32 to the signal point in the QAM method.
The QAM sylnbol lnapping unit 62 is the same as the QAM syn~bol
mapping unit 51 provided in the phase error compensation unit 19a of the
carrier wave reproduction device 10a (Fig. 7) according to the second
exemplary embodiment.
[0130]
The phase error detector 63 receives the replica of the transmission
sylilbol generated by the QAM symbol mapping unit 62 as one of two
inputs like the phase error detector 52 provided in the carrier wave
reproduction device 10a (Fig. 7) according to the second exemplary
embodiment. In the second exemplary embodiment, the phase error
detector 52 receives the output symbol of the phase rotator 55 as the other
input. In this exelilplary embodiment, the phase error detector 63
receives the output syliibol of the buffer 18 which holds the received
sytiibol that is corrected by the interpolation filter 15 as the other input.
[0131]
When the output symbol of the buffer 18 is represented by the
coliiplex number u and the replica of the tralisliiission symbol outputted by
the QAM symbol tilapping unit 62 is represented by the complex number P,
the output value of the phase error detector 63 is expressed by Equation 3
like the second exeniplary embodiment.
[0132]
Fig. 12 is a block diagram showing a configuration of the low pass
filter 64 as an example. Referring to Fig. 12, the low pass filter 64
includes an averaging filter 65 which performs the averaging of the output
of the phase error detector 63 and a ROM table 66 which converts the
output value 0 of the averaging filter 65 into a set of a cosine function
value and a sine function value (cos(8) and sin(6)) and outputs it.
[0133]
The output data of the ROM table 66 is used for the phase rotation
of the output symbol of the buffer 18 in the phase rotator 61. When the
output syliibol of the buffer 18 is represented by a' arid the output of the
averaging filter 65 is represented by 8, the phase rotator 61 rotates the
phase according to equation 4 and outputs the synlbol u.
[0134]
The received symbol to which the second-time phase error
compensation is applied by the phase rotator 61 is inputted to the error
correctio~ld ecoder 32 via the selector 30 and the QAM symbol de~napping
unit 31 like the carrier wave reproductio~l device 10a shown in Fig. 7.
[0135]
The error correction decoder 32 performs the second-time error
correction decodiug and estimates the tra~ismission bit string. The
transmission bit string is supplied to the phase error detector 63 through
the QAM symbol niapping unit 62 in a similar way to that previously used.
The output symbol of the buffer 18 is supplied to the phase error detector
63 as the other input. Not only the way of supplying the output symbol
but also the sylilbol value supplied from the buffer 18 is the same as that of
the previous time.
[0136]
It is expected that the number of the errors included it1 the replica
of the transmission symbol supplied through the QAM symbol mapping
unit 62 is smaller than the number of the errors included in t11k replica of
the transmission sy~nbol at the last time. By this effect, the phase error
compensation can be performed with high accuracy compared to the
second-time phase error compensation. Subsequently, the same process
is performed. Therefore, the phase noise can be compensated for with
more high accuracy by repeatedly updating the replica of the transmission
symbol.
[0137]
Fig. 13 is a flowchart showing the operation of the carrier wave
reproduction device lob according to the exemplary embodiment equipped
with the phase error compensation unit 19b as an example. Namely, it
shows the operations performed until the output data is obtained after the
received symbol is inputted.
[0138]
Referring to Fig. 13, the carrier wave reproduction device lob
receives the received symbol outputted from the AID converter 13 (Step
Dl).
[0139]
The interpolation filter 15 uses only the pilot symbol in the
received symbol string and estimates the phase error included in the
payload symbol between the pilot symbols by the interpolation process
(Step D2).
[0140]
Next, the phase rotator 17 performs the first-time phase error
compelisation (Step D3).
[0141]
The QAM symbol delnappilig unit 31 perforlns denlapping of the
received symbol to which the phase error colnpensation is performed to the
bit string (Step D4) and the error correction decoder 32 performs an error
correctioli using the error correction code (Step D5).
[0142]
When the number of phase error compensation times k is equal to or
greater than the number of times T set in advance (k and T are positive
integers) or when all the errors are successfully corrected by the error
correctioli (Step DS), the data to which the error correction is perfornied
(Step D5) is used as the output of the carrier wave reproduction device lob
(Step D6).
[0 1431
On the other hand, when the number of phase error compensation
times k is smaller than the number of times T set in advance, the QAM
synlbol mapping unit 62 creates the replica of the transmission symbol by
mapping the data to which the error correction is performed (Step D5) to
the QAM syrnbol (Step D7).
[0 1441
Next, the phase error detector 63 uses the replica of the
tra~lsmission symbol as the reference symbol and detects the phase error
(Step D8).
[0145]
Next, the low pass filter 64 performs the averaging of the detected
phase error and generates a phase rotation signal (Step D9).
[0 1461
Next, the phase rotator 61 performs the (k + 1)th-time phase error
compensation by using the phase rotation signal (Step D10).
[0147]
The exemplary embodiment of the present invention has been
explained above with reference to the drawing. The configuration
described in the drawing is shown as an example. Therefore, the present
invention is not limited to the configuration described in the drawing.
[0148]
The carrier wave reproduction device according to the present
invention can be suitably applied to a digital wireless communication
device including a portable ter~ninal device and a backbone wireless
device.
[0 1 491
Further, the disclosure of the prior art document such as the
above-mentioned patent literature or the like is incorporated by reference.
Modification and adjustment of the exemplary embodiment can be made
within the scope of the overall disclosure (including claims) of the present
invention, and based on the basic technical concept of the invention.
Moreover, various combinations or selections of the various disclosed
elements (including each element of each claim, each element of each
exelnplary embodiment, and each element or the like of each drawing) are
possible within the scopc of the claims of thc present invention. Namely,
modification and adjustment that can be conceived by those skilled in the
art can be made within the scope of the overall disclosure including claims
of the present invention, and based on the basic technical concept of the
invention. With respect to a range of the value described in this
docun~ent, it is interpreted that even when an arbitrary value or a small
range included in the range is not specifically described, it is specifically
described.
Reference signs List
[0150]
10, 10a, and lob carrier wave reproduction device
1 1 2 a d 3 reference oscillator
12, 122, and 132 detector
13, 123, and 133 AID converter
14 and 134 delay circuit
15 and 135 interpolation filter
16 tap coefficient generation unit
17, 55, 61, 124, and 136 phase rotator
18 buffer
19, 19a, and 19b phase error colnpensation unit
20, 20a, and 20b demodulation device
30, 42, and 45 selector
31, 128, and 137 QAM sylnbol delnapping unit
32, 129, and 138 error correction decoder
34, 46, and 47 adder
35, 41, and 44 register
36, 38, 39, and 66 ROM (read only memory) table
43 and 91 multiplier
51 and 62 QAM symbol mapping unit
52 and 63 phase error detector (with reference signal input) 53 and 126
loop filter
54 and 127 nu~nericalc ontrol oscillator
64 low pass filter
65 averaging filter
125 phase error detector
CLAIMS
[Claim 11
A carrier wave reproduction device including
an interpolation filter which estimates a phase error of a received
symbol based on a pilot symbol included in the received symbol,
a first phase rotator which rotates the phase of the received symbol
according to the phase error estimated by the interpolation filter and
outputs the received symbol as a first output symbol,
a phase error compensation unit which co~upensates for the phase
error which remains in the first output syn~bola nd outputs the first output
symbol as a second output symbol,
a QAM sy~nbol demapping unit which calculates a first bit string
corresponding to the first output symbol and calculates a second bit string
corresponding to the second output symbol, and
an error correction decoder which perfornis error correction of a bit
error in the first bit string and wherein the phase error compensation unit
compensates for the phase error which remains in the first output synlbol
by referring to the first bit string after error correction.
[Claim 21
The carrier wave reproduction device described in claim 1 wherein
the carrier wave reproduction device includes a buffer which
temporarily holds the first output symbol outputted froni the first phase
rotator and
the phase error compensation unit includes a second phase rotator
which rotates the phase of the first output symbol outputted from the buffer
and generates the second output symbol.
[Claim 31
The carrier wave reproduction device described in claim 2 wherein
the phase error compensation unit includes a QAM symbol mapping unit
which generates a symbol correspondi~lg to the bit string after error
correction as a third output symbol.
[Claim 41
The carrier wave reproduction device described in claim 3 wherein
the phase error compensation unit detects a phase difference between at
least either the first output sy~nbol or the second output symbol and the
third output sy~nbol and compensates for the phase error which re~nainsi n
the first output symbol.
[Claim 51
Tlie carrier wave reproduction device described in claim 4 wherein
the phase error compensation unit includes
a phase error detector which detects the phase difference between
the second output symbol and the third output symbol and
a nu~nerical control oscillator which calculates a phase rotation
amount based on the phase difference from which the high frequency
component included in the phase difference is removed and the second
phase rotator rotates the phase of the first output symbol outputted from
the buffer according to the phase rotation anlount and generates the second
output symbol.
[Claim 61
The carrier wave reproduction device according to claim 5 wherein
the phase error compensation unit includes a loop filter which removes the
high frequency co~ilponent included in the phase difference and outputs it
to the gunierical control oscillator.
[Claim i [ ]
The carrier wave reproduction device described in claim 4 wherein
the phase error compensation unit includes
a phase error detector which detects the phase difference between
the first output symbol and the third output symbol and
a low pass filter which performs the averaging of the phase
difference and calculates the phase rotation amount and
the second phase rotator rotates the phase of the first output symbol
outputted from the buffer according to the phase rotation amount and
generates the second output symbol.
[Claim 81
The carrier wave reproduction device described in any one of claim
1 to claim 7 wherein the interpolation filter is a finite impulse response
type filter including
a plurality of registers which hold the pilot symbol,
a plurality of multipliers whose number is equal to the number of
plural registers, and
a plurality of adders whose number is equal to the number of plural
registers.
[Claim 91
The carrier wave reproduction device described in claim 8 wherein
the interpolation filter determines an initial value of a tap coefficient and a
step width used when the tap coefficient is updated according to a phase
noise level and a carrier to noise ratio and updates the tap coefficient the
number of times that is equal to the number of intervals between the pilot
symbols included in the received syn~bol.
[Claim 101
A demodulation device based on a quasi-synchronous detection
method which includes the carrier wave reproduction device described in
any one of claim 1 to claim 9.
[Claim 111
A carrier wave reproduction method including
a step of estimating a phase error of a received synlbol based 011 a
pilot symbol included in the received symbol,
a step of rotating the phase of the received symbol according to the
estimated phase error and outputting the received symbol as a first output
symbol,
a step of calculating a first bit string corresponding to the first
output symbol,
a step of performing error correction of a bit error in the first bit
string,
a step of compensating for the phase error which remains in the first
output symbol by referring to the first bit string after error correction and
outputting the first output symbol after phase error conlpensation as a
second output symbol,
a step of calculating a second bit string corresponding to the second
output symbol, and
a step of performing error correction of a bit error in the second bit
1
slring.
[Claim 121
The carrier wave -reproduction method described in claim 11 in
which
a step for conlpensating for a phase error which remains in a n-th
output symbol by referring to a 11-th bit string after error correction (n is a
natural number) and outputting the symbol as a (n+l)th output symbol,
a step for calculating a (n+l)th bit string corresponding to the
(n+l)th output symbol, and
a step for perfornii~lg error correction of a bit error in the (n+l)th
bit string are repeated until n is equal to a predetermined number ~vhile
incre~nentingn or all the errors are corrected in the error correction.