Hi Jason: right, 32bit floats I & Q each; in fact, GNU Radio doesn't really care, it just uses the native float type, but as far as machines that you'd use GNU Radio on are concerned, that should be IEEE754, indeed. Scaling is up to the application, so you can't presume [-1;1]. However, for the USRPs, +-1 is full scale, indeed. Typical case where this matters is OFDM, where you might be putting a symbol vector with each symbol being within these boundaries, but the (I)FFT might have significantly higher peaks.
Best regards, Marcus On 02.10.2015 18:55, Jason Matusiak wrote: > I just wanted to make sure I have this right (because I always seem to > confuse myself); the complex data type is 64bits, 4 bytes of I and 4 > bytes of Q, right? > > Secondly, those 4B are of type signed floats from [-1,1], right? I > assume that these are of the IEEE 754 type? > > TIA! > > _______________________________________________ > Discuss-gnuradio mailing list > [email protected] > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio _______________________________________________ Discuss-gnuradio mailing list [email protected] https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
