You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the code that generates S0 symbols for ofdm frames, line64: liquid-dsp/src/multichannel/src/ofdmframe.common.c
I see, that the M-sequence is generated not only for occupied carriers, but also for unoccupied ones. Is this really required? I mean could the sequence been drawn only for used carriers?
Background:
To me this matters, because as I am trying to use ofdm with devices having different sampling rates, i.e. 2.4MHz and 4MHz. The planned bandwith is 2MHz which is covered well by both rates. I tried to rectify the different rates by employing two sets of subcarriers, 1000 and 600 respectively and different number of guards amounting to the same number of usable carriers within the bandwidth. I was hoping that either way the signals are generated the analog signals would be be the same. However the current way of generating the M sequence enforces a dependency on the total number of subcarriers, while I would have expected that the used number of carriers is used, so that the analog signal would not depend on how it was generated.
The text was updated successfully, but these errors were encountered:
In the code that generates S0 symbols for ofdm frames, line64:
liquid-dsp/src/multichannel/src/ofdmframe.common.c
I see, that the M-sequence is generated not only for occupied carriers, but also for unoccupied ones. Is this really required? I mean could the sequence been drawn only for used carriers?
Background:
To me this matters, because as I am trying to use ofdm with devices having different sampling rates, i.e. 2.4MHz and 4MHz. The planned bandwith is 2MHz which is covered well by both rates. I tried to rectify the different rates by employing two sets of subcarriers, 1000 and 600 respectively and different number of guards amounting to the same number of usable carriers within the bandwidth. I was hoping that either way the signals are generated the analog signals would be be the same. However the current way of generating the M sequence enforces a dependency on the total number of subcarriers, while I would have expected that the used number of carriers is used, so that the analog signal would not depend on how it was generated.
The text was updated successfully, but these errors were encountered: