interpolation in FPGA

Hello!

I like to know if anybody have done any interpolation (e.g quadrature) in FPGA or where I can get application note or any literature about it.

I know delta (slope) of signal (pulse) and duration which FPGA must generate. Intermediate points must be calculate with quadrature or polynomial functions. I must send data from FPGA on 16 bit parallel bus trough an DAC. It is not possible to store this signals inside FPGA and then regenerate because shapes of these signals is changed dynamically (slope, duration, shape..)

thank you and regards,

Amir

Reply to
amko
Loading thread data ...

Are you sure you can't use FIR?

regards

Jerzy Gbur

Reply to
jerzy.gbur

Take a look at the paper writte by Julius Smith from Stanford centre for computer/music research (google it). It is THE definitive paper on arbitrary sample rate conversion of band-limited signals.

I'd be intererested to hear if anyone has attempted an FPGA implementation of sample rate conversion based on his algorithm...

Good luck, PeterC.

Reply to
peter

This sounds more like curve fitting than filtering (rate conversion).

If the sample rate is high compared to your FPGA speed grade, you need to build the polynomial multiplier and scaling blocks directly and "wire" them up with adders. If your FPGA can run significantly faster than the sample rate, you can multiplex the same logic over multiple computations to save area.

Alternately, you can take the AWG approach and compute the waveform into a large bank of memory, then play it back repeatedly. In this case, you can get away with using a processor IP to do the calculations, an external (or internal, for a large enough device) RAM to store the waveform, and logic to read out this waveform to the DAC.

Reply to
jimgeorge at gmail dot com

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.