BPSK vs. GMSK

Ludanov, Sergei FISLUDAN@dof.ca.gov
Mon, 2 Feb 1998 08:21:05 -0800


Hello everyone,


>
>BPSK is robust but not very spectrum efficient. When you start with only a
>couple  kHz, that's an important consideration. GMSK and some variants
>don't have this problem. 
Theoretically speaking MSK (or GMSK) should give 6dB performance
increase
over BPSK but only using coherent demodulation, which is much more
difficult
to do than in case with BPSK, that is why most if not all existing MSK 
receivers use non-coherent decoding which introduces 10 dB loss in SNR.
I am currently working on algorithm for real time coherent BPSK decoding
using DSP board from Analog Devices and it almost (not quite really)
push
33 MOPS processor to its limits. MSK will require to increase software
complexity about 4 times, but I will be willing to try it.

73 de Sergei KD6CJI
>
>