An implementation for maximally-flat FIR filters is proposed that requires a much smaller number of multiplications than a direct form structure. The values of the multiplier coefficients in the implementation are conveniently small, and do not span a huge dynamic range, unlike in a direct form implementation.
No takes yet. Share an insight, caveat, or question.
P. P. Vaidyanathan (1984) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: