Hello all. There is a question about sample rate in “56Gbps Coherent 16QAM_B2B_detail with DSP”.
In the Universal DSP, there are 262144 samples before [Resampling]. After [Resampling] sps = 2. Because there are 16384 symbols, the samples should be 32768 after [Resampling]. Why it is 32756? We can see this phenomenon in “Report->DSP for 16QAM->Constellation After Resampling”
In the help document of Universal DSP we can see”Note: Between algorithm stages, the signals are up-sampled to their original rate (for the case of 16-QAM, 8 x Samples per bit). For 2 samples/symbol, the first half of the sampled signals are set to a and the second half to b. For 1 sample/symbol all sampled signals are set to c.” And there are 32768 points on the constellation of [QI Compensation][Timing Recovery][Adaptive Equalizer]. Is every algorithm upsample the data first and downsample the data before output to the next algorithm, and the next do the same thing?
I find a phenomenon that the Universal DSP will make the signal power rise. Every algorithm except [Filter] will make the signal power rise. We can see the phenomenon by comparing the [RF Spectrum Analyzer] of the input and output of the DSP signal. I don’t know why. I use optisystem 15.
Thanks for your help.
.
