Title of Invention

INTERLEAVE ADDRESS GENERATION APPARATUS, TURBO CODING APPARATUS, TURBO DECODING APPARATUS, COMMUNICATION TERMINAL APPARATUS, BASE STATION APPARATUS AND INTERLEAVE ADDRESS GENERATION METHOD

Abstract The column numbers and row numbers of a two-dimensional array are incremented by a counter control unit (101) a block interleave represented by the two-dimensional array and outputted as read address values. The read address values are inputted to a bit inversion unit (102) by which bit inversion is conducted. A row conversion unit (103) outputs address values corresponding to the bit inversion output values and the row numbers from the counter control unit (101) as row conversion values. A shift register (104) bit-shifts the output values of the bit inversion unit (102) and outputs the bit-shifted values as address offset values. An adder (105) adds the address offset values to the row conversion values. A value comparator (106) compares the sum values with the interleave size and outputs data within the interleave size as address values.
Full Text DESCRIPTION
INTERLEAVE ADDRESS GENERATION APPARATUS
Technical Field
The present invention relates to an interleave
address generation apparatus and interleave address
generation method, and more particularly, to an
interleave address generation apparatus and interleave
address generation method ideally applicable to a
communication terminal apparatus or base station
apparatus.
Background Art
One of conventional interleave address generation
apparatuses and interleave address generation methods
is described in the Unexamined Japanese Patent
Publication No.HEI 7-212250.
In accordance with the trend of the present times
toward the worldwide standardization of a third-
generation communication system, standardization plans
on interleave are being proposed and GF interleave is
one of the interleave methods currently being studied.
This GF interleave is one of block interleaves that
perform processing on a two-dimensional array with the
number of rows N=2r and the number of columns M=2C. The
GF interleave sequentially rearranges bits (hereinafter

referred to as "column exchange") of a bit string of M
bits in length delimited row by row from the first row
to Nth row in different sequences, and generates an
interleave address pattern by reading the matrix
subjected to the row exchange carried out according to
the sequence based on a bit inversion method from the
first row of the first column from top to bottom to the
Nth row of the Mth column.
In the block-size matrix two-dimensional array
above, an example of generating an interleave address
pattern by calculating a row conversion pattern  i ( j )
will be explained.
FIG.l is a block diagram showing a configuration
of a column exchange apparatus used for a conventional
GF interleave.
In FIG.l, memory 11 outputs vector α 10
corresponding to row number i (0 logical sum calculator 13. Memory 12 outputs vector
corresponding to column number j (0 exclusive logical sum calculator 13. Exclusive logical
sum calculator 13 calculates an exclusive logical sum
of α 10 and αj and outputs calculation result β to memory
14 .
Memory 14 outputs column conversion pat tern i(j)
for the ith row based on calculation result β . i (j)
is calculated from following expression (1).


Furthermore, conversions in memory 11, memory 12
and memory 14 are carried out using the conversion table
shown in FIG.2.
FIG.2 shows a conversion table used for the GF
interleave.
The conversion table in FIG. 2 is a table that shows
correspondences between power in a Galois field power
expression and a vector expression with the Galois field
polynomial basis.
The vector expression consists of vectors output
from memory 11 and memory 12 and vectors input to memory
14. Power loga /3 in a power expression is a value input
to memory 11 or memory 12 and a value output from memory
14.
Here, the column conversion pattern on the ith row
is obtained through the operation shown below. By
calculating parameter iO corresponding to row number i
in memory 11, vector Oi 10 corresponding to parameter iO
is output. Exclusive logical sum calculator 13
calculates an exclusive logical sum of a10 and a3 output
from memory 11 and memory 12 and memory 14 outputs log
corresponding to calculation result ]3 .
By fixing address value i of memory 11 and
incrementing address value j of memory 12 from 0, column
conversion pattern Ki(j ) corresponding to the ith row

is generated.
Then, an example of interleave address generation
is shown below.
FIG.3A, FIG.3B, FIG.3C and FIG.3D illustrate the
process of generation of interleave addresses.
An example of creating an interleave address
pattern of size 30 on an 8X4 two-dimensional array will
be explained below.
First, the interleave address generation apparatus
stores an interleave address pattern with addresses 0
to 7 rearranged in the column direction in memory (i=0,
j = 0 to 7) .
Likewise, the interleave address generation
apparatus stores an interleave address pattern with
addresses 0 to 7 rearranged in different sequences from
the next rows onward in memory (i = l to 3, j = 0 to 7 ) . The
storage result is shown in FIG.3A.
Then, the interleave address generation apparatus
performs rearrangement processing in row units. More
specifically, row i=l and row i=2 are switched round.
The switching result is shown in FIG.3B.
Then, the interleave address generation apparatus
adds an offset address to the stored value in row units.
More specifically, a value obtained by multiplying the
value of i by the number of column components is added
to the value of i. For example, value 16 obtained by
multiplying 2, the value of i, by 8, the number of
components, is added to values on the 2nd column

respectively. Value 8 obtained by multiplying 1, the
value of i, by 8, the number of components, is added to
values on the 3rd column respectively. Value 24 obtained
by multiplying 3, the value of i, by 8, the number of
components, is added to values on the 4th column
respectively. The addition result is shown in FIG.3C.
Then, the interleave address generation apparatus
extracts addresses in the column direction from memory
and outputs only addresses smaller than the size of the
interleave address pattern to be created. More
specifically, in FIG.3C, value 7 stored in i=0, j=0 is
output and then value 20 stored in i=2, j=0, value 14
stored in 1=1, j=0 and value 29 stored in i=3, j=0 are
output. Then, value 3 stored in 1 = 0, j = 1 is output and
then value 22 stored in i = 2, j = l, value 12 stored in i = l,
j = l and value 26 stored in i = 3, j = l are output. Likewise,
the interleave address generation apparatus extracts
values stored in memory in the column direction and
outputs the extracted values as an interleave address
pattern. FIG.3D shows the interleave address pattern
output.
However, the conventional interleave address
generation method develops an interleave address pattern
created in predetermined units in memory and then
performs row rearrangement and an addition of offset
addresses, and thus the conventional interleave address
generation method has a problem that a considerable/
memory space and considerable processing time are

required to generate an interleave address pattern.
Disclosure of Invention
It is an object of the present invention to provide
an interleave address generation apparatus and
interleave address generation method capable of
generating an interleave address pattern with a small
memory space and short processing time.
This object is attained by simultaneously carrying
out row rearrangement processing and column
rearrangement processing during generation of
interleave addresses and continuously carrying out this
row rearrangement processing and column rearrangement
processing and offset address addition processing to
create interleave addresses.
Brief Description of Drawings
FIG.l is a block, diagram showing a configuration
of a column exchange apparatus used in a conventional
GF interleave;
FIG. 2 illustrates a conversion table used in the
GF interleave;
FIG.3A illustrates a process of generation of
interleave addresses;
FIG.3B illustrates a process of generation of
interleave addresses;
FIG.3C illustrates a process of generation of
interleave addresses;

FIG.3D illustrates a process of generation of
interleave addresses;
FIG.4 is a block diagram showing a configuration
of an interleave address generation apparatus according
to Embodiment 1 of the present invention;
FIG. 5 illustrates a data configuration example in
an exclusive logical sum calculation in the embodiment
above;
FIG.6 illustrates an example of generation of
interleave addresses;
FIG. 7 is a block diagram showing a configuration
example of an interleave address generation apparatus
according to Embodiment 2 of the present invention;
FIG.8 is a block diagram showing a configuration
example of an interleave address generation apparatus
according to Embodiment 3 of the present invention;
FIG . 9 illustrates a table example stored in memory;
FIG.10 illustrates a table example stored in
memory;
FIG.11A illustrates an example of row number i and
column number j output from counter control section 201;
FIG. 11B illustrates an example of the converted row
number and column number;
FIG.lie illustrates an example of an addition
result output from an adder;
FIG.11D illustrates an example of interleave
address output from interleave address apparatus 200;
FIG.12 is a block diagram showing a configuration

of an interleave apparatus according to Embodiment 4 of
the present invention;
FIG.13 is a block diagram showing a configuration
of a turbo coding apparatus according to Embodiment 5
of the present invention;
FIG.14 is a block diagram showing a configuration
of a turbo decoding apparatus according to Embodiment
6 of the present invention;
FIG.15 is a block diagram showing a configuration
of a communication terminal apparatus according to
Embodiment 7 of the present invention; and
FIG.16 is a block diagram showing a configuration
of a base station apparatus according to Embodiment 8
of the present invention.
Best Mode for Carrying out the Invention
With reference now to the attached drawings,
embodiments of the present invention will be explained
below.
(Embodiment 1)
The interleave address generation apparatus
according to Embodiment 1 carries out row rearrangement
processing and column rearrangement processing in
parallel.
FIG. 4 is a block diagram showing a configuration
of the interleave address generation apparatus according

to Embodiment 1 of the present invention.
Interleave address generation apparatus 100 shown
in FIG. 4 is mainly constructed of counter control section
101, bit inversion apparatus 102, column conversion
apparatus 103, shift register 104, adder 105 and size
comparison section 106.
Furthermore, column conversion apparatus 103 is
mainly constructed of memory 110, memory 111, memory 113
and exclusive logical sum calculator 112.
In FIG.4, counter control section 101 outputs row
number i (0^i inversion apparatus 102 and column number j (0S*j a two-dimensional array to memory 111.
For example, when addresses on a 22X23 two-
dimensional array are output, counter control section
101 outputs row number i=0, column number j=0, then
outputs row number i = l, column number j = 0. Then, counter
control section 101 outputs row number i = 2, column number
j=0, then outputs row number i=3, column number 3=0.
Then, counter control section 101 outputs row
number i=0, column number j=l. In this way, counter
control section 101 increases the value of column number
j every time the value of row number i exceeds a maximum
value of 3, sets and outputs i = 0 and continues to output
up to a combination of row number i = 3 and column number
j = 7.
Bit inversion apparatus 102 carries out bit
inversion of row number i output from counter control

section 101 in a binary state and outputs bit - inverted
row number i' to memory 110 and shift register 104. More
specifically, bit inversion apparatus 102 switches
between the higher digit and the lower digit of the binary
coded row number. That is, bit inversion apparatus 102
switches the highest digit and lowest digit, then
switches between the 2nd highest digit and the 2nd lowest
digit . And this same switching between a higher digit
and lower digit is continued.
Memory 110 stores values resulting from assigning
different i0's to α10 on respective rows, calculates 10
corresponding to i' input and outputs α10 corresponding
to 10 to exclusive logical sum calculator 112.
Memory 111 stores values resulting from assigning
different j's to aJ (0^j outputs ai corresponding to j input to exclusive logical
sum calculator 112.
Exclusive logical sum calculator 112 calculates an
exclusive logical sum of a i0 output from memory 110 and
aj output from memory 111 and outputs the calculation
result to memory 113.
Memory 113 stores the column conversion pattern
corresponding to the calculation result of exclusive
logical sum calculator 112 and outputs the column
conversion pattern corresponding to the calculation
result input to adder 105.
Shift register 104 bit-shifts the value output from
bit inversion apparatus 102 and outputs this to adder

105 as an address offset value.
Adder 105 adds up the output from shift register
104 and the output from column conversion apparatus 103
and outputs the addition result to size comparison
section 106.
Size comparison section 106 compares the addition
result output from adder 105 with the interleave size
and outputs the addition result which is not greater than
the interleave size as an address value.
Then, the data processing in the interleave address
generation apparatus of this embodiment will be
explained.
The following explanations will describe a case
assuming that interleave sizeL = 30, N (=2r)XM (=2C) block
size. r = 2, C=3, primitive polynomial of degree 3
expressed in Galois field GF (23) is x3 + x+l and the root
of the primitive polynomial is a . All the originals in
Galois field GF (23) can be expressed in power of (X .
FIG.5 illustrates a data configuration example in
an exclusive logical sum calculation of this embodiment.
Memory 110 stores parameter 10 corresponding to row
number i in correspondence with 3-bit vector a x0 and
outputs 3-bit vector ai0 corresponding to 10 input.
Memory 111 stores parameter j corresponding to row
number i in correspondence with aj and outputs 3-bit
vector a3 corresponding to j input.
Exclusive logical sum calculator 112 carries out
an exclusive logical sum of vector a ±0 and vector a J and

outputs calculation result β to memory 113.
Memory 113 outputs the column-rearranged data
corresponding to calculation result β .
Then, the data processing of the interleave address
generation apparatus according to this embodiment will
be explained.
FIG.6 shows an example of creation of interleave
addresses.
In FIG.6, i and j denote a row number and column
number output from counter control section 101 and i'
denotes a row number output from bit inversion apparatus
102. Furthermore, αi0 and αj denote vector data output
from memory 110 and memory 111 and αi0+αj denotes the
calculation result in exclusive logical sum calculator
112 . Log,, αi0+αj denotes data output from memory 113 and
the offset addition result denotes the result of the
addition of an offset address output from shift register
104 .
Data processing is carried out on one row at a time
starting from the top row.
First, counter control section 101 outputs row
number i=0, column number j=0. Bit inversion apparatus
102 outputs row number i after switching between the
higher digit and lower digit of the row number bits in
a binary state. Row number i=0 is expressed as "00" in
a two-bit binary number and when the higher bit and lower
bit are switched round, the row number is "00" and row
number i'=0 is output.



and then, processing with i = 0 , j=l is carried out.
As shown above, the interleave address generation
apparatus controls so that i is reset every time i exceeds
a maximum value and in this way sequentially outputs
addresses for a 22X23 two-dimensional array in the column
direction. Furthermore, the interleave address
generation apparatus uses output i' from bit inversion
apparatus 102 as the read address value for memory 110,
and can thereby carry out row exchange on the
aforementioned two-dimensional array simultaneously.
Thus, the interleave address generation apparatus
according to Embodiment 1 outputs row numbers and column
numbers individually and converts numbers individually,
and can thereby carry out row rearrangement processing
and column rearrangement processing in parallel,
allowing an interleave address pattern to be generated
in a small space and short processing time.
The explanation above describes the case of
interleave address generation when the size of an
interleave address pattern is 30 and the block size is
22X23, but it is possible to carry out block interleave
of N(2r)XM(2c) for an arbitrary number of data L by
changing the memory data to be stored and setting the
number of shifts of shift register 104 to c bits.
(Embodiment 2)
FIG. 7 is a block diagram showing a configuration
example of an interleave address generation apparatus

according to Embodiment 2. However, the configuration
common to that in FIG.4 is assigned the same reference
numerals as those in FIG.4 and detailed explanations
thereof are omitted.
Interleave address generation apparatus 150 in
FIG.7 is different from FIG.4 in that Interleave address
generation apparatus 150 is equipped with storage cell
array 151 and adds up offset addresses according to the
output timing from memory 113.
In FIG. 7, storage cell array 151 temporarily stores
row number i' output from bit inversion apparatus 102
and then outputs row number i' to shift register 104.
For example, storage cell array 151 is constructed
of a two-stage storage cell array to adjust the timings
of the output from column conversion apparatus 103 and
the output from shift register 104 to the timing of the
output value i' from bit inversion apparatus 102.
Then, output value i' from bit inversion apparatus
102 is temporarily stored in storage cell array 151,
sequentially output in synchronization with the addition
timing of adder 105, input to shift register 104 and a
value resulting from shifting 3 bits is output as an
address offset value for the i'th row.
Thus , the interleave address generation apparatus
according to Embodiment 2 outputs row numbers and column
numbers individually and converts numbers individually,
and can thereby carry out row rearrangement processing
and column rearrangement processing in parallel.

allowing an interleave address pattern to be generated
in a small space and short processing time.
Furthermore, the interleave address generation
apparatus according to Embodiment 2 can adjust the output
timing of the adder by delaying the output timing of an
offset address value using a temporary storage circuit,
making it possible to generate an interleave address
pattern even when the speed of generation of an offset
address value is different from that of a column exchange
pattern.
(Embodiment 3)
FIG. 8 is a block diagram showing a configuration
example of an interleave address generation apparatus
according to Embodiment 3.
In FIG.8, counter control section 201 outputs row
number i of a two-dimensional array to memory 202 and
outputs column number j of a two-dimensional array to
memory 203.
Memory 202 stores N(i) corresponding to input i and
outputs N(i) corresponding to i output from counter
control section 201 to multiplier 204.
Memory 203 stores M( j) corresponding to input j and
outputs M(j) corresponding to j output from counter
control section 201 to adder 205.
Multiplier 204 multiplies N(i) output from memory
202 by the number of columns M and outputs the
multiplication result to adder 205.

Adder 205 adds the multiplication result output
from multiplier 204 to M(j) output from memory 203 and
outputs the addition result to size comparison section
206 .
When the addition result output from adder 205 is
smaller than the size of the required interleave address,
size comparison section 206 outputs the addition result
as the interleave address and when the addition result
is equal to or greater than the size of the required
interleave address, size comparison section 206 does not
output the addition result.
Then, the conversion operation of memory 202 will
be explained.
FIG. 9 shows a table example stored in memory 202.
In FIG.9, N(i) is an output corresponding to input
i and i has a one-to-one correspondence with N(i) and
N(i)s corresponding to different i's take mutually
different values.
When i=0 is input, memory 202 outputs N(i)=2. When
i=l is input, memory 202 outputs N(i)=3. When i=2 is
input, memory 202 outputs N(i)=0. When i=3 is input,
memory 202 outputs N(i)=l.
Then, the conversion operation of memory 203 will
be explained.
FIG.10 illustrates a table example stored in memory
203 .
In FIG.10, M(j) is an output corresponding to input
j and j has a one-to-one correspondence with M(j) and

M(j)s corresponding to different j's take mutually
different values.
When j = 0 is input, memory 203 outputs M( j ) = 3 . When
j=l is input, memory 202 outputs M(j)=6. When j=2 is
input, memory 202 outputs M(j)=4. When j=3 is input,
memory 202 outputs M( j ) =2 . Likewise, when i = 4 to 7, M( j )
corresponding to j is output according to the table in
FIG.10.
Then, an example of generation of interleave
addresses will be shown.
FIG.11A illustrates an example of row number i and
column number j output from counter control section 201.
When addresses of a 22X23 two-dimensional array is
output, counter control section 201 outputs row number
i=0, column number j=0 and then row number i=l, column
number j=0. Then, counter control section 201 outputs
row number i=2, column number j=0, and row number i=3,
column number j=0.
Then, counter control section 201 outputs row
number i=0, column number j=l. Thus, counter control
section 201 increases the value of column number every
time the value of row number i exceeds a maximum value
of 3, sets and outputs i=0, and continues to output up
to a combination of row number i = 3 and column number j = 7.
FIG.11B illustrates an example of converted row
numbers and column numbers.
Row number i output from counter control section
201 is converted to N(i) according to the table in FIG.9

and column number J is converted to M(j) according to
the table in FIG.10.
For example, when counter control section 201
outputs row number i=0, column number j = 0, memory 202
outputs N(i)=2 and memory 203 outputs M(j)=3.
FIG.11C illustrates an example showing the
addition result output from adder 205.
Adder 205 outputs a value obtained by adding M(j)
to a multiplication result of multiplying N(i) by the
number of columns.
For example, when N(i)=2, M(j)=3, adder 20 5 outputs
19, a value obtained by adding M( j) to a multiplication
result of multiplying the number of columns 8 by N(i).
Size comparison section 206 outputs addition
results among other addition results in FIG.11C which
are smaller than the requested interleave address size
as the interleave addresses.
FIG.11D illustrates an example of interleave
addresses output from interleave address apparatus 200.
For example, when the size of the requested
interleave address is 30, addition results whose value
is 29 or smaller are output as interleave addresses and
addition results equal to or greater than 30 are not
output .
Thus, the interleave address generation apparatus
according to Embodiment 3 carries out row rearrangement
processing and column rearrangement processing in
parallel, and carries out this row rearrangement

processing and column rearrangement processing and
offset address addition processing continuously,
allowing an interleave address pattern to be generated
in a small space and short processing time.
Furthermore, the interleave address generation
apparatus according to Embodiment 3 can generate
interleave addresses in a simpler configuration than
Embodiment 1 or Embodiment 2 by using a same column
exchange pattern on respective rows.
(Embodiment 4)
FIG.12 is a block diagram showing a configuration
of an interleave apparatus according to Embodiment 4 of
the present invention.
In FIG.12, interleave address generation apparatus
301 outputs an interleave address pattern to memory 303
according to an input instruction by which data is input
to memory.
Address counter 302 outputs addresses one by one
starting from the start address in memory to memory 303
according to a data output instruction by which data is
output.
Memory 303 stores data sequentially at addresses
output from interleave address generation apparatus 301,
and after storing a predetermined amount of data, outputs
address data output from address counter 302
sequentially.
Thus, the interleave apparatus of this embodiment

can perform high-speed interleave processing with small
memory by rearranging information series using an
interleave address pattern generated by the interleave
address generation apparatus of Embodiment 1 or
Embodiment 2.
The interleave apparatus of Embodiment 4 stores
data in memory of addresses output from interleave
address generation apparatus 301 and reads data from
memory of addresses output from address counter 302, but
the present invention is not limited to this and it is
also possible to store data in memory of addresses output
from address counter 302, read data from memory of
addresses output from interleave address generation
apparatus 301 and rearrange data.
(Embodiment 5)
FIG.13 is a block diagram showing a configuration
of a turbo coding apparatus according to Embodiment 5
of the present invention.
In FIG.13, turbo coding apparatus 400 is mainly
constructed of recursive convolutional coder 401,
interleaver 402 and recursive convolutional coder 403.
Recursive convolutional coder 401 performs
convolutional code coding on an information series input
and outputs the coded information series to the outside.
Interleaver 402 is constructed of the interleave
apparatus of Embodiment 4, performs interleave
processing on the information series input and outputs

the interleaved information series to recursive
convolutional coder 403.
Recursive convolutional coder 403 carries out
convolutional code coding on the information series
output from interleaver 402 and outputs the coded
information series to the outside.
Then, the operation of turbo coding apparatus 400
will be explained.
The input information series is subjected to
convolutional coding by recursive convolutional coder
401 and the coded information series is output.
Furthermore, the input information series is
subjected to data rearrangement by interleaver 402 and
the rearranged information series is subjected to
convolutional coding by recursive convolutional coder
403 and the coded information series is output.
That is, regarding the information series to be
coded, 3 bits are output as a code series corresponding
to 1 information series bit: the output of the
information series itself; the output from recursive
convolutional coder 401 that performs coding of
convolutional codes using the information series as an
input; and the output of recursive convolutional coder
403 that performs coding of convolutional codes using
data rearranged by interleaver 402 as an input which is
the data written in memory before being input to
recursive convolutional coder 403 using the information
series as an input.

Through the operation above, in response to the
input of the information series, turbo coding apparatus
400 outputs the information series input, the
information series subjected to convolutional coding and
the information series subjected to data rearrangement
and convolutional coding.
Thus, the turbo coding apparatus of this embodiment
can speed up processing by rearranging information
series using the interleave apparatus of Embodiment 4,
and can thereby improve the error correcting
performance.
For example, in turbo coding apparatus 400 of
Embodiment 5, it is possible to implement turbo coding
apparatus 400 with improved error correcting performance
for decoding of a code series on the receiving side by
applying the interleave apparatus of Embodiment 4
according to the GF interleave system to interleaver 402.
Furthermore, the turbo coding apparatus of this
embodiment can speedily generate interleave addresses
with small memory and perform interleave by rearranging
the information series using the interleave apparatus
of Embodiment 4, and can thereby perform turbo coding
with small memory.
[Embodiment 6]
FIG.14 is a block diagram showing a configuration
of a turbo decoding apparatus according to Embodiment
6 of the present invention.

In FIG.14, turbo decoding apparatus 500 is mainly
constructed of soft decision output decoder 501,
interleaver 502, soft decision output decoder 503 and
deinterleaver 504.
Soft decision output decoder 501 decodes the code
series input and outputs the input decoded code series
to interleaver 502.
Interleaver 502 rearranges the code series output
from soft decision output decoder 501 and outputs to soft
decision output decoder 503.
Soft decision output decoder 503 decodes the code
series output from interleaver 502 and outputs to
deinterleaver 504.
Deinterleaver 504 rearranges the code series
output from soft decision output decoder 503 and outputs
the obtained code series to soft decision output decoder
501 and the outside.
Then, the operation of turbo decoding apparatus 500
will be explained.
In the initial operation, the code series subjected
to convolutional coding by the turbo coding apparatus,
etc. of Embodiment 5 is decoded by soft decision output
decoder 501 and the obtained soft decision output is
output to interleaver 502.
The soft decision output from soft decision output
decoder 501 is subjected to data series rearrangement
by interleaver 502 and output to soft decision output
decoder 503.

The data series output from interleaver 502 is
decoded together with the information series received
by soft decision output decoder 503 and the soft decision
output obtained is output to deinterleaver 504.
The soft decision output from soft decision output
decoder 503 is subjected to data series rearrangement
by deinterleaver 504 and the rearranged data series is
output to soft decision output decoder 501 and the
outside•
The data series output from deinterleaver 504 is
output to soft decision output decoder 501 and used for
the second and subsequent turbo decoding processing as
reliability information.
In the second and subsequent turbo decoding
processing, the code series subjected to convolutional
coding is decoded by soft decision output decoder 501
using the data series output from deinterleaver 504 as
reliability information and the soft decision output
obtained is output to interleaver 502.
As shown above, the turbo decoding apparatus of this
embodiment can speed up processing by rearranging the
information series using the interleave apparatus of
Embodiment 4, and can thereby improve the error
correcting performance.
For example, in turbo coding apparatus 500 of
Embodiment 6, it is possible to implement turbo decoder
500 with improved error correcting performance by
applying the interleave apparatus of Embodiment 4 to

interleaver 502 and interleaver 504.
Furthermore, the turbo decoding apparatus of this
embodiment can reduce memory required for processing by
rearranging information series using the interleave
apparatus of Embodiment 4, and can thereby perform turbo
coding with small memory.
(Embodiment 7)
FIG.15 is a block diagram showing a configuration
of a communication terminal apparatus according to
Embodiment 7 of the present invention.
In FIG.15, communication terminal apparatus 600 is
mainly constructed of antenna 601, reception section 602,
transmission section 603, demodulation section 604,
modulation section 605, decoding processing section 606,
coding processing section 607, speech CODEC section 608,
data input/output section 609, speaker 610 and
microphone 611.
Decoding processing section 606 is constructed of
deinterleave section 614, rate matching section 615 and
error correcting/decoding section 616.
Coding processing section 607 is constructed of
error correcting/coding section 617, rate matching
section 618 and interleave section 619.
Here, error correcting/coding section 617 is
constructed using the interleave apparatus of Embodiment
4 or turbo coding apparatus 400 of Embodiment 5.
On the other hand, error correcting/decoding

section 616 is constructed using the interleave
apparatus of Embodiment 4 or turbo decoding apparatus
500 of Embodiment 6 for non-speech data.
On the other hand, deinterleave section 614 and
interleave section 619 are constructed using the
interleave apparatus of Embodiment 4.
Antenna 601 transmits and receives signals.
Reception section 602 carries out radio processing
on the reception signal from antenna 601 and outputs the
reception signal subjected to radio processing to
demodulation section 604.
Transmission section 603 carries out radio
processing on a transmission signal output from
modulation section 605 and transmits to antenna 601.
Demodulation section 604 demodulates the reception
signal output from reception section 602 using
despreading apparatus 612 and outputs the demodulated
signal to deinterleave section 614.
Modulation section 605 modulates the transmission
signal output from interleave section 619 using
spreading apparatus 613 and outputs the modulated
transmission signal to transmission section 603.
Deinterleave section 614 carries out data
rearrangement processing on the demodulated signal
output from demodulation section 604 and outputs the
rearranged data to rate matching section 615.
Rate matching section 615 adjusts the length of the
data output from deinterleave section 614 to a length

that allows error correcting processing and outputs the
data of the adjusted length to error correcting/decoding
section 616.
Error correcting/decoding section 616 corrects
errors of the data output from rate matching section 615
and outputs the error-corrected data to speech CODEC
section 608.
Error correcting/coding section 617 carries out
error correcting/coding on the transmission data output
from speech CODEC section 608 and outputs to rate
matching section 618.
Rate matching section 618 adjusts the transmission
data output from error correcting/coding section 617 to
a length required for interleave processing and outputs
to interleave section 619.
Interleave section 619 applies rearrangement
processing to the transmission data output from rate
matching section 618 and outputs to modulation section
605.
Speech CODEC section 608 carries out coding on a
speech signal output from microphone 611 and outputs to
error correcting/coding section 617 as transmission
data.
Furthermore, speech CODEC section 608 decodes the
reception data output from error correcting/decoding
section 616 and outputs the decoded speech data to
speaker 610.
Microphone 611 outputs an input speech to speech

CODEC section 608 as speech data.
Speaker 610 outputs the speech data output from
speech CODEC section 608 as speech.
Then, the operation of communication terminal
apparatus 600 during transmission will be explained.
When speech is transmitted, the speech is analog-to
digital-converted (hereinafter referred to as "AD
conversion" ) to a speech signal by microphone 611, output
to speech CODEC section 608, coded by speech CODEC
section 608, subjected to convolutional coding by error
correcting/coding section 617 and output to rate
matching section 618 as transmission data.
Furthermore, when non-speech data is transmitted,
the non-speech data is sent through data input/output
section 609 to error correcting/coding section 617 where
the non-speech data is subjected to turbo coding and
convolutional coding according to the data transfer rate
and output to rate matching section 618 as transmission
data.
The transmission data is adjusted to a length
required for interleave processing by rate matching
section 618, rearranged by interleave section 619,
digital-modulated and digital-to-analog-converted
(hereinafter referred to as "DA conversion") by
modulation section 605, subjected to radio processing
by transmission section 603 and transmitted via antenna
601.
Then, the operation of communication terminal

apparatus 600 during reception will be explained.
A reception signal is received via antenna 601,
subjected to radio processing and AD conversion by
reception section 602, digital-demodulated by
demodulation section 604 and output to deinterleave
section 614 as reception data.
The reception data is subjected to rearrangement
processing by deinterleave section 614, adjusted to a
data length appropriate for error correction by rate
matching section 615 and output to error
correcting/decoding section 616.
In the case where the reception data is a speech
signal, the reception data is subjected to Viterbi
decoding by error correcting/decoding section 616,
subjected to speech decoding and DA conversion by speech
CODEC section 608 and output from speaker 610 as speech.
In the case where the reception data is a non-speech
signal, the reception data is subjected to turbo decoding
by error correcting/decoding section 616 according to
the data transfer rate and output to the outside via data
input/output section 609.
Thus, the communication terminal apparatus of this
embodiment can transmit/receive non-speech
communications with a communication characteristic of
high transmission quality at a lower bit-to-error rate
by applying the interleave apparatus of Embodiment 4 to
the error correcting/coding apparatus and error
correcting/decoding apparatus for non-speech data.

Furthermore, the interleaver included in turbo
coding/decoding is constructed of an interleave
apparatus that allows high-speed processing and with a
reduced amount of memory, and therefore can implement
communication terminal apparatus 600 allowing high-
speed interleave processing with a reduced amount of
memory.
This embodiment describes the case where the
communication terminal apparatus is applied to a CDMA
communication, but the communication system is not
limited to this, and the present invention can also be
applied to other communication systems by replacing
spreading apparatus 613 in modulation section 605 and
despreading apparatus 612 in demodulation section 604
with a modulation apparatus and demodulation apparatus
according to the relevant communication system.
(Embodiment 8)
FIG.16 is a block diagram showing a configuration
of a base station apparatus according to Embodiment 8
of the present invention.
Base station apparatus 700 shown in FIG. 16 is mainly
constructed of antenna 701, reception section 702,
transmission section 703, demodulation section 704,
modulation section 705, decoding processing section 706,
coding processing section 707 and data input/output
section 708.
Decoding processing section 706 is constructed of

deinterleave section 709, rate matching section 710 and
error correcting/decoding section 711.
Coding processing section 707 is constructed of
error correcting/coding section 712, rate matching
section 713 and interleave section 714.
Error correcting/coding section 712 is constructed
using interleave address generation apparatus 100 or 200
of Embodiment 1 or turbo coding apparatus 400 of
Embodiment 4.
Error correcting/decoding section 711 is
constructed using the interleave apparatus of Embodiment
4 or turbo decoding apparatus 500 of Embodiment 5 for
non-speech data.
Deinterleave section 709 and interleave section
714 are constructed using the interleave apparatus of
Embodiment 4.
Antenna 701 transmits/receives signals.
Reception section 702 carries out radio processing
on the reception signal from antenna 701 and outputs the
reception signal to demodulation section 704.
Transmission section 703 carries out radio
processing on a transmission signal output from
modulation section 705 and transmits to antenna 701.
Demodulation section 704 demodulates the reception
signal output from reception section 702 using
despreading apparatus 715 and outputs the demodulated
signal to deinterleave section 709.
Modulation section 705 modulates the transmission

signal output from interleave section 714 using
spreading apparatus 716 and outputs the modulated
transmission signal to transmission section 703.
Deinterleave section 709 carries out data
rearrangement processing on the demodulated signal
output from demodulation section 704 and outputs the
rearranged data to rate matching section 710.
Rate matching section 710 adjusts the length of the
data output from deinterleave section 709 to a length
that allows error correcting processing and outputs the
data of the adjusted length to error correcting/decoding
section 711.
Error correcting/decoding section 711 carries out
decoding and error correction on the data output from
rate matching section 710 and outputs the error-
corrected data to data input/output section 708.
Error correcting/coding section 712 carries out
error correcting/coding on the transmission data output
from data input/output section 708 and outputs to rate
matching section 713.
Rate matching section 713 adjusts the transmission
data output from error correcting/coding section 712 to
a length required for interleave processing and outputs
to interleave section 714.
Interleave section 714 applies rearrangement
processing to the transmission data output from rate
matching section 713 and outputs to modulation section
705 .

Data input/output section 708 outputs the data to
be transmitted to error correcting/coding section 712
and outputs the reception data output from error
correcting/decoding section 711 to the outside.
Then, the operation of base station apparatus 700
during transmission will be explained.
The transmission data is sent through data
input/output section 708 to error correcting/coding
section 712 where the transmission data is subjected to
turbo coding and convolutional coding according to the
transfer rate or type of data and output to rate matching
section 713 as transmission data.
The transmission data is adjusted to a length
required for interleave processing by rate matching
section 713, rearranged by interleave section 714,
digital-modulated and D/A-converted by modulation
section 705, subjected to radio processing by
transmission section 703 and transmitted via antenna
701.
Then, the operation of base station apparatus 700
during reception will be explained.
A reception signal is received via antenna 701,
subjected to radio processing and AD conversion by
reception section 702, digital-demodulated by
demodulation section 704 and output to deinterleave
section 709 as reception data.
The reception data is subjected to rearrangement
processing by deinterleave section 709, adjusted to a

data length appropriate for error correction by rate
matching section 710 and output to error
correcting/decoding section 711.
The reception data is subjected to turbo decoding
by error correcting/decoding section 711 according to
the data transfer rate and output to the outside via data
input/output section 708.
Thus, the base station apparatus of this embodiment
can transmit/receive with a communication
characteristic of high transmission quality at a lower
bit-to-error rate by applying the interleave address
generation apparatus of Embodiment 1 to the error
correcting/coding apparatus and error
correcting/decoding apparatus.
For example, according to base station apparatus
700 of Embodiment 8, it is possible to obtain base station
apparatus 700 with a communication characteristic with
lower BER and of high transmission quality for non-speech
communications by applying turbo coding apparatus 400
of Embodiment 5 to error correcting/coding section 712
and turbo decoding apparatus 500 of Embodiment 6 to error
correcting/decoding section 711.
Furthermore, the configuration of the interleaver
included in turbo coding and decoding allows high-speed
processing and by configuring the apparatus using an
interleave apparatus with a reduced amount of memory,
it is possible to obtain base station apparatus 700
allowing high-speed interleave with a reduced amount of

memory.
By the way. this embodiment describes the case where
the base station apparatus is applied to a CDMA
communication, but the communication system is not
limited to this, and the present invention can also be
applied to other communication systems by replacing
spreading apparatus 716 in modulation section 705 and
despreading apparatus 715 in demodulation section 704
with a modulation apparatus and demodulation apparatus
according to the respective communication systems.
As is apparent from the explanations above, the
present invention can generate interleave address
patterns with a smaller memory space and shorter
processing time.
This application is based on the Japanese Patent
Application No.HEI 11-286981 filed on October 7, 1999,
entire content of which is expressly incorporated by
reference herein.

We Claim
1. An interleave address generation apparatus of a block interleave system,
generating a read address or a write address of data arranged in the form
of a matrix two dimensional array, said interleave address generation
apparatus comprising:
a counter that outputs a row number and a column number on the
two dimensional array in the block interleave system;
a bit inversion section that inverts a bit of the row number;
a column exchange section that outputs an address value
corresponding to the bit inverted row number and the column number as
a column conversion value;
a shift register that shifts the bit of the bit inverted row number
and outputs the shifted bit as an address offset value;
an adder that adds the address offset value and the column
conversion value; and
a size comparison section that compares the addition value to an
interleave size and outputs the addition value within the interleave side as
an address value.
2. The interleave address generation apparatus as claimed in claim 1,
wherein said bit inversion section switches a higher bit and a lower bit
based on the row number outputted from said counter.
3. The interleave address generation apparatus as claimed in claim 1,
comprising a storage cell array that stores the bit inverted row number on
a temporary basis and thereafter outputs said row number to said shift
register.
4. The interleave address generation apparatus as claimed in claim 1,
wherein said interleave address generation apparatus is used in a Galois
field interleave system.

5. The Interleave address generation apparatus as claimed In claim 1,
wherein said column exchange section comprises:
first storing means that stores a unique constant value on a per
row basis based on the bit inverted row number;
second storing means that stores a unique constant value on a per
column basis;
an exclusive logical sum calculator that performs an exclusive
logical sum calculation of the stored row number and column number; and
third storing means that stores the exclusive logical sum calculation
value as the address value.
6. A turbo coding apparatus comprising a recursive convolutional coder that
performs a convokitional coding of an information series, and an
interleaver that comprises the interleave address generation apparatus as
claimed in claim 1.
7. A turbo decoding apparatus comprising:
a soft decision output decoder that decodes a code series;
an interleaver that comprises the interleave address generation
apparatus as claimed in claim 1 that performs interleave processing of
output from said soft decision output decoder;
a soft decision output decoder that decodes the code series in
which a sequence of input data has been changed by said interleaver; and
a deinterleaver that comprises the interleave address generation
apparatus as claimed in claim 1 that performs deinterleave processing of
the output from said soft decision output decoder,
8. A communication terminal apparatus comprising:
a decoding processing apparatus having the turbo decoding
apparatus as claimed in claim 7 that decodes a demodulated received
signal; and

a coding processing apparatus having the turbo coding apparatus
as claimed in claim 6 that codas a transmission signal.
9. A base station apparatus comprising:
a decoding processing apparatus having the turbo decoding
apparatus as claimed in claim 7 that decodes a demodulated received
signal; and
a coding processing apparatus having the turbo coding apparatus
as claimed in claim 6 that codes a transmission signal.
10.An interleave address generation method of a block interleave system,
generating a read address or a write address of data arranged in the form
of a two dimensional matrix, said interleave address generation method
comprising:
counting up and outputting a row number and a column number on
a two dimensional array in the block interleave system;
inverting the row number;
determining an address value corresponding to the bit inverted row
number and the column number as a column conversion value;
determining an address offset value by shifting bit of the inverted
row number;
adding the address offset value and the column conversion value;
comparing the addition value to an interleave size; and,
outputting the addition value within the interleave size as an
address value.

The column numbers and row numbers of a two-dimensional array are incremented by a counter control unit (101) a block interleave represented by the two-dimensional array and outputted as read address values. The read
address values are inputted to a bit inversion unit (102) by which bit inversion is
conducted. A row conversion unit (103) outputs address values corresponding to the bit inversion output values and the row numbers from the counter control unit (101) as row conversion values. A shift register (104) bit-shifts the output values of the bit inversion unit (102) and outputs the bit-shifted values as address offset values. An adder (105) adds the address offset values to the row conversion values. A value comparator (106) compares the sum values with the interleave size and outputs data within the interleave size as address values.

Documents:

in-pct-2001-577-kol-abstract.pdf

in-pct-2001-577-kol-claims.pdf

in-pct-2001-577-kol-correspondence.pdf

in-pct-2001-577-kol-description (complete).pdf

in-pct-2001-577-kol-drawings.pdf

in-pct-2001-577-kol-examination report.pdf

in-pct-2001-577-kol-form 1.pdf

in-pct-2001-577-kol-form 18.pdf

in-pct-2001-577-kol-form 2.pdf

in-pct-2001-577-kol-form 3.pdf

in-pct-2001-577-kol-form 5.pdf

in-pct-2001-577-kol-gpa.pdf

in-pct-2001-577-kol-granted-abstract.pdf

in-pct-2001-577-kol-granted-claims.pdf

in-pct-2001-577-kol-granted-correspondence.pdf

in-pct-2001-577-kol-granted-description (complete).pdf

in-pct-2001-577-kol-granted-drawings.pdf

in-pct-2001-577-kol-granted-examination report.pdf

in-pct-2001-577-kol-granted-form 1.pdf

in-pct-2001-577-kol-granted-form 18.pdf

in-pct-2001-577-kol-granted-form 2.pdf

in-pct-2001-577-kol-granted-form 3.pdf

in-pct-2001-577-kol-granted-form 5.pdf

in-pct-2001-577-kol-granted-gpa.pdf

in-pct-2001-577-kol-granted-priority document.pdf

in-pct-2001-577-kol-granted-reply to examination report.pdf

in-pct-2001-577-kol-granted-specification.pdf

in-pct-2001-577-kol-granted-translated copy of priority document.pdf

in-pct-2001-577-kol-priority document.pdf

in-pct-2001-577-kol-reply to examination report.pdf

in-pct-2001-577-kol-specification.pdf

in-pct-2001-577-kol-translated copy of priority document.pdf


Patent Number 235411
Indian Patent Application Number IN/PCT/2001/577/KOL
PG Journal Number 27/2009
Publication Date 03-Jul-2009
Grant Date 01-Jul-2009
Date of Filing 30-May-2001
Name of Patentee MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD.
Applicant Address 1006, OAZA KADOMA, KADOMA-SHI, OSAKA 571-8501
Inventors:
# Inventor's Name Inventor's Address
1 IKEDA TETSUYA 31-26, FUJIMIGAOKA, TSUZUKI-KU, YOKOHAMA-SHI, KANAGAWA 224-0051
2 YAMANAKA RYUTARO 1-33-15-201, KAMIOOKA - HIGASHI, KONAN-KU, YOKOHAMA-SHI, KANAGAWA 233-0001
PCT International Classification Number H03M 13/27
PCT International Application Number PCT/JP2000/06974
PCT International Filing date 2000-10-06
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 11-286981 1999-10-07 Japan