Method and apparatus for termination of iterative turbo decoding
View Patent ↗In the method for termination of turbo decoding, a plurality of first LLR values (L a i (k)) of a-priori information and a plurality of second LLR values (L e i (k)) of extrinsic information are called up. A value is determined for a decision variable which is characteristic of the number of mathematical sign discrepancies between the first L a i (k)) and the second L e i (k)) values. The turbo decoding is terminated if the number of mathematical sign discrepancies is less than or at least equal to a first number or if the number is greater than or at least equal to a second number.
1. A method for termination of turbo decoding of received data, wherein said turbo decoding is carried out iteratively comprising two decoding elements per iteration, the method comprising:
reading first values from a memory of the corresponding decoding element, said first values being either a plurality of first log-likelihood ratio values of a-priori information or a variable which is associated with the a-priori information;
calling up second values as a function of output values from the same decoding element, said second values being either a plurality of second log-likelihood ratio values of extrinsic information or a variable which is associated with the extrinsic information for one of the two decoding elements, said first and second values being associated with the same systematic bits of the received data;
determining a value of a decision variable which is characteristic of the number of mathematical sign discrepancies between the first and the second values, wherein a mathematical sign discrepancy is present when the mathematical sign of a first value and the mathematical sign of a second value are different for the same systematic bit of the received data;
terminating the turbo decoding when the value of the decision variable indicates that the number of mathematical sign discrepancies is:
(a) less than or alternatively at least equal to a first number, or
(b) greater than or alternatively at least equal to a second number,
wherein the second number is greater than the first number.
2. The method according to claim 1 ,
wherein each of the first values is read from the memory before it is overwritten in the memory by the respective second value for the same systematic bit.
3. The method according to claim 1 , wherein the first and the second values of a binary decision are used in the determining step.
4. The method according to claim 1 , wherein the determining step comprises calculating the value of the decision variable by:
E
=
c
·
∑
k
=
1
K
f
(
k
)
where
f
(
k
)
=
{
0
if
sgn
(
L
1
(
k
)
)
=
sgn
(
L
2
(
k
)
)
1
if
sgn
(
L
1
(
k
)
)
≠
sgn
(
2
(
k
)
)
wherein E represents the decision variable, K represents the number of systematic bits of the received data, k represents the index of the systematic bits, c is a constant factor, L 1 (k) describes the first values, and L 2 (k) describes the second values.
5. The method according to claim 1 , wherein the second values are obtained either from the sum of the extrinsic information and systematic information or from the sum of the extrinsic information weighted by a factor and from the systematic information.
6. The method according to claim 1 , wherein the method is executed for both the first and for the second of the two decoding elements.
7. The method according to claim 1 , further comprising:
performing a cyclic redundancy check based on values of a-posteriori information of one decoding element, wherein the received data comprises redundancy information for the cyclic redundancy check; and
terminating the turbo decoding if the cyclic redundancy check is correct.
8. The method according to claim 7 , wherein the step of performing a cyclic redundancy check is based solely on the values of the a-posteriori information of the first decoding element.
9. The method according to claim 1 , wherein the method is suitable for use in a UMTS mobile radio receiver.
10. A turbo decoding method which is terminated by means of the method according to claim 1 .
11. The turbo decoding method according to claim 10 , wherein the turbo decoding of the first and the second decoding element is performed by a single decoder element using time-division multiplexing, and
wherein the extrinsic information of the first decoding element or a variable which is associated with the extrinsic information, and the extrinsic information of the second decoding element or a variable which is associated with the extrinsic information being stored, are both stored in a single memory.
12. An apparatus for terminating turbo decoding of received data, wherein said turbo decoding is carried out iteratively comprising two decoding elements per iteration, said apparatus comprising:
means for reading first values from a memory of the corresponding decoding element, said first values being either a plurality of first log-likelihood ratio values of a-priori information or a variable which is associated with the a-priori information;
means for calling up second values as a function of the result of the same decoding element, said second values being either a plurality of second log-likelihood ratio values of extrinsic information or a variable which is associated with the extrinsic information for one of the two decoding elements, said first and second values being associated with the same systematic bits of the received data;
means for determining a value of a decision variable, which is characteristic of the number of mathematical sign discrepancies between the first and the second values, wherein a mathematical sign discrepancy is present when the mathematical sign of a first value and the mathematical sign of a second value are different for the same systematic bit of the received data; and
means for terminating the turbo decoding when the value of the decision variable indicates that the number of mathematical sign discrepancies is:
(a) less than or alternatively at least equal to a first number, or
(b) greater than or alternatively at least equal to a second number, wherein the second number is greater than the first number.
13. The apparatus according to claim 12 ,
wherein the means for determining the value of the decision variable comprises a means for binary decision for the first and the second values.
14. The apparatus according to claim 12 , wherein
the second values are obtained either from the sum of the extrinsic information and systematic information or from the sum of the extrinsic information weighted by a factor and from the systematic information.
15. The apparatus according to claim 12 ,
wherein the apparatus receives the first and the second values for both decoding elements.
16. The apparatus according to claim 12 , wherein the apparatus comprises:
means for performing a cyclic redundancy check based on values of a-posteriori information of one of the two decoding elements, wherein the received data comprises redundancy information for the cyclic redundancy check; and
means for terminating the turbo decoding if the cyclic redundancy check is correct.
17. A UMTS mobile radio receiver comprising the apparatus according to claim 12 .
18. A turbo decoder comprising the apparatus according to claim 12 .
19. The Turbo decoder according to claim 18 , wherein
a single decoder element comprises both decoding elements and is configured to use time-division multiplexing, and
a single memory configured to store the extrinsic information of the first decoding element or a variable which is associated with the extrinsic information, and configured to store the extrinsic information of the second decoding element or a variable which is associated with the extrinsic information.