Architecture for power of two coefficient FIR filter
AssigneeNorth American Philips Corporation
InventorsArup K. Bhattacharya, Michael G. Cristofalo, David Koo, Amihai Miron, Imran A. Shah
Life of this patent
Estimating the term…
Loading patent…
AssigneeNorth American Philips Corporation
InventorsArup K. Bhattacharya, Michael G. Cristofalo, David Koo, Amihai Miron, Imran A. Shah
Estimating the term…
An architecture for a very large scale integrated (VLSI) implementation of a finite imprise response (FIR) digital filter having no multipliers and a coefficient space limited to powers of two. The filter structure includes a data bus, a coefficient bus and a sum-in bus to each coefficient tap. Each tap has a coefficient and control word register which is loaded during an initialization phase of the filter. Multiplication is provided by a shifter which provides the correct power of two weighting of an input data sample. The weighted data sample at each tap is added to the output of the previous tap. This architecture results in a regular, modular structure which can be cascaded and which is programmable for various data word lengths and coefficient spaces.
A non-recursive finite impulse response (FIR) digital filter of n taps, where 1.ltoreq.n, comprising:
Dependencies are inferred from the claim text.