Turbo decoder with reduced-size branch metric cache
View Patent ↗A method and an apparatus is provided for producing branch metrics in a LogMAP turbo decoding operation. During a forward recursion of a trellis, a set of primary branch metrics is generated. The primary branch metrics are stored in receiver form in a relatively small memory cache module and corresponding secondary branch metrics are produced by negating the primary branch metrics. The primary branch metrics and the secondary branch metrics constitute all possible branch metrics for a given state in the trellis. During a backwards recursion of the trellis, the stored primary branch metrics are retrieved from the memory cache module and the secondary branch metrics are regenerated by negating the retrieved primary branch metrics.
1. A method for decoding using branch metrics generated using a decoding trellis, the method comprising:
generating, during a forward recursion of the decoding trellis, a first subset of a plurality of forward branch metrics of the decoding trellis from a primary combination of extrinsic, parity and information data associated with at least one symbol, the first subset having a property that a second subset of the plurality of forward branch metrics can be determined based on a symmetry of the plurality of forward branch metrics;
storing the first subset of the plurality of forward branch metrics and discarding the second subset of the plurality of forward branch metrics;
upon commencing a reverse recursion of the decoding trellis, retrieving the stored first subset and determining the second subset using the symmetry of the plurality of forward branch metrics, the first and second subsets being applied in the further decoding operation;
generating, during the reverse recursion of the decoding trellis, a plurality of reverse branch metrics using the plurality of forward branch metrics; and
decoding said at least one symbol using the forward and reverse branch metrics.
2. The method of claim 1 , wherein at least one of the branch metrics in the second subset is the negation of a corresponding one of the stored branch metrics.
3. The method of claim 2 , wherein the step of generating the second subset comprises:
negating the retrieved first subset of branch metrics.
4. The method of claim 3 , wherein the forward recursion and the backwards recursion are part of a trellis algorithm.
5. The method of claim 1 , wherein generating the first subset of the plurality of branch metrics of the decoding trellis comprises retrieving the extrinsic data from a first memory having a first size and retrieving the parity and information data from at least one second memory having a second size.
6. The method of claim 5 , wherein the first size and the second size are sufficiently large to store data and parity information for each block that is being decoded.
7. The method of claim 6 , wherein storing the first subset of the plurality of primary branch metrics comprises storing the first subset of the plurality of primary branch metrics in a cache memory having a third size that is smaller than either the first size or the second size.
8. The method of claim 7 , wherein the third size is approximately equal to a window size associated with the decoding trellis.