FIR Filter Library

mikroC PRO for dsPIC30/33 and PIC24 includes a library for finite impulse response (FIR) filter. All routines work with fractional Q15 format.

A finite impulse response (FIR) filter is a type of a digital filter, whose impulse response (the filter's response to a delta function) is finite because it settles to zero in a finite number of sample intervals.

Library Routines

FIR_Radix

FIR_Radix

Prototype	unsigned FIR_Radix(unsigned FilterOrder, const unsigned ptrCoeffs, unsigned* BuffLength, unsigned ptrInput, unsigned* Index); // For dsPIC33EP family of MCUS : unsigned FIR_Radix(unsigned FilterOrder, far const unsigned ptrCoeffs, unsigned* BuffLength, far unsigned ptrInput, unsigned* Index);
Description	This function applies FIR filter to `ptrInput`.
Parameters	`FilterOrder:` order of the filter + 1 `ptrCoeffs:` pointer to filter coefficients in program memory `BuffLength` number of input samples `ptrInput:` pointer to input samples `Index:` index of current sample
Returns	with : `N` - buffer length `k` - current index
Requires	Nothing.
Example	const unsigned BUFFFER_SIZE = 32; const unsigned FILTER_ORDER = 20; const COEFF_B[FILTER_ORDER+1] = { 0x0000, 0x0048, 0x0133, 0x02D3, 0x052B, 0x0826, 0x0BA0, 0x0F62, 0x1329, 0x16AA, 0x199A, 0x16AA, 0x1329, 0x0F62, 0x0BA0, 0x0826, 0x052B, 0x02D3, 0x0133, 0x0048, 0x0000 }; ydata unsigned input[BUFFFER_SIZE]; // Input buffer unsigned inext; // Input buffer index ... unsigned CurrentValue; CurrentValue = FIR_Radix(FILTER_ORDER+1, // Filter order COEFF_B, // b coefficients of the filter BUFFFER_SIZE, // Input buffer length input, // Input buffer inext); // Current sample
Notes	Input samples must be in Y data space.

Want more examples and libraries?
Find them on