You are not logged in.
Hi,
I'm trying to have OFDM PHY implementation in WARPLAB. From the postings, it looks like currently there is no such implementation.
1. Are there any particular difficulties in implementing the OFDM PHY on WARPLAB.
2. I want to inter-operate between the current OFDM design and the one to be implemented on WARPLAB (like, one node with WARPLAB OFDM transmitting and other node with current OFDM receiving) . Do you see any issue with such kind of a setup.
3. Can you suggest an efficient way to implement. Like, can I take the simulink model in the present OFDM design and convert into a matlab code (I'm new to simulink, excuse my ignorance). I would like the WARPLAB implementation as close to the present OFDM as possible.
On a different question, I was trying to experiment with BER performance by changing Tx power. In two posts I found two different calibrations for the power. One post says
All four values are linear-in-dB; the increment of each value is different:
TxGainBB: 1.5dB/step
TxGainRF: ~1dB/step
RxGainBB: ~2dB/step
RxGainRF: 15dB/step
and in other
The Tx RF gain covers about 30dB, with digital 0 mapping to the lowest gain. The full span of [0,63] is mapped linearly-in-dB over the 30dB range. See the middle plot on pg. 17 of the datasheet for details. The Tx BB gain covers just 5dB range, with 0 mapping to the lowest gain. See the two LSB of register 0x9 (Table 17 on pg. 36 of the datasheet) for details.
If Tx RF gain covers about 30dB in a range of [0,63], shouldn't it be ~0.5dB/step. Can you please confirm other gains also.
Thanks
Offline
There is no automatic way to generate m-code for the real-time OFDM PHY. A few suggestions:
-Implement the transmitter first. The real-time PHY's preamble structure and training symbol values are stored in its m-code init scripts. The actual modulation process is straightforward in m-code (qammod for bits->symbols and ifft for symbols/subcarriers->time domain waveform). You'll need to implement 4x interpolation in your m-code (WARPLab processes waveforms at 40MHz, while the PHY's over-the-air bandwidth is just 10MHz).
This sounds like an interesting project- please keep us posted!
vijay wrote:
If Tx RF gain covers about 30dB in a range of [0,63], shouldn't it be ~0.5dB/step. Can you please confirm other gains also.
You're right- it's ~0.5dB per step (the curve saturates a little towards the min/max). The other gains are correct. All of these are specified in the MAX2829 datasheet.
Offline