Viterbi Signal Decoding Using CRC-Based Initial State Selection

The patent segments the set of all possible initial states into multiple candidate groups based on CRC bit patterns. Instead of decoding all initial states, the system divides them into manageable groups (e.g., first candidate group, second candidate group) and performs selective decoding on relevant groups only, thereby reducing computation complexity while maintaining decoding reliability through systematic coverage of likely candidates.

The patent applies partial action by performing Viterbi decoding only on selected candidate groups rather than all possible initial states. The system determines which candidate groups are most likely to contain the correct codeword based on CRC checks and performs decoding exclusively on those groups, achieving sufficient reliability without the excessive computation required for exhaustive decoding of all initial states.

The patent segments the initial states into candidate groups and performs decoding only on selected groups, directly reducing the number of computational operations required. This segmentation approach maintains decoding accuracy by ensuring that the correct initial state is included in the decoded candidate groups while significantly reducing power consumption through avoided computations on irrelevant initial states.

The system performs partial decoding action by limiting Viterbi decoding to only those candidate groups that pass preliminary CRC checks or are most likely to contain valid codewords. This partial action approach preserves decoding accuracy for the relevant subsets while reducing overall power consumption by eliminating unnecessary decoding operations on unlikely candidates.

The patent performs preliminary CRC checks on candidate groups before performing full Viterbi decoding. By using CRC bits to pre-screen and identify promising candidate groups, the system prepares in advance which groups warrant full decoding attention. This preliminary action reduces computation complexity by filtering out unlikely candidates while maintaining reliability through systematic verification of selected groups.

The system uses CRC check results as feedback to guide the selection of candidate groups for Viterbi decoding. The CRC outcomes provide information about which initial state groups are more likely to produce valid codewords, and this feedback mechanism allows the system to adaptively focus computational resources on the most promising candidates, thereby maintaining decoding reliability while reducing overall computation complexity.