We use viterbi decoders in multiple applications, so were looking forward to support for it in the IPP. As I understand, viterbi decoders have a 'state' parameter and a traceback 'depth' parameter. But during integration, it appears that the example is always setting the 'depth' equal to the 'state' (16, 32 or 64), thereby making the depth essentially non-configurable.
I'm trying to re-work the example to make depth configurable, but am starting to wonder if this (depth = state) is a requirement of the associated IPP viterbi functions (GetVarPointDV, CalcStatesDV, BuildSymblTableDV4D, UpdatePathMetricsDV).
So my questions are:
Is there a way to make 'depth' configurable in the example / using the existing IPP functions?
If not, is this a 'bug' in these functions? (Will it be fixed? Should I submit this as a support issue?)
-Thanks in advance for any help
if you have some specific functionality requests it would be nice to submit them to Intel Premier Support, so they will be used as a input at the next IPP version planning stage.