waosai
Member level 2
ook ber
Hello everyone, recently I am building a OOK transceiver system in simulink to
simulate the BER performance under different AWGN channel SNR. The
following is a brief of my system:
1) Digital baseband signal is a 10-bit 2Gb/s binary pseudo-random
sequence; Carrier (center) frequency is 40GHz with -10dBm signal
power.
2) The above two signals are mixed (multiplied, modeling Up Mixer),
amplified (modeling PA) and band-pass filtered (modeling antenna).
3) Output Signals are fed into an AWGN channel (SNR is to be sweeped
in simulation) with free-space path loss (18dB).
4)At receiver part, input signal is amplified (modeling LNA), mixed
(modeling Down Mixer), low-pass filtered and buffered.
5) Output signals in (4) are sampled and quantized into digital form
and then sent to "Error Rate Calculation" module to get BER results.
Since the center frequency is 40GHz, the sampling frequency of the
system is set to 102.4GHz to meet Nyquist theory. I run the simulation
under different AWGN channel SNR (0~20dB) and do get the BER curves.
The problem, however, is the mininum BER is still larger than 1e-4 at
about 10dB. The other values are all 0. I want to see even smaller BER
value (not zero) at larger SNR. How should I do? Is there any mistake
in my system setup?
Hello everyone, recently I am building a OOK transceiver system in simulink to
simulate the BER performance under different AWGN channel SNR. The
following is a brief of my system:
1) Digital baseband signal is a 10-bit 2Gb/s binary pseudo-random
sequence; Carrier (center) frequency is 40GHz with -10dBm signal
power.
2) The above two signals are mixed (multiplied, modeling Up Mixer),
amplified (modeling PA) and band-pass filtered (modeling antenna).
3) Output Signals are fed into an AWGN channel (SNR is to be sweeped
in simulation) with free-space path loss (18dB).
4)At receiver part, input signal is amplified (modeling LNA), mixed
(modeling Down Mixer), low-pass filtered and buffered.
5) Output signals in (4) are sampled and quantized into digital form
and then sent to "Error Rate Calculation" module to get BER results.
Since the center frequency is 40GHz, the sampling frequency of the
system is set to 102.4GHz to meet Nyquist theory. I run the simulation
under different AWGN channel SNR (0~20dB) and do get the BER curves.
The problem, however, is the mininum BER is still larger than 1e-4 at
about 10dB. The other values are all 0. I want to see even smaller BER
value (not zero) at larger SNR. How should I do? Is there any mistake
in my system setup?
