Software defined radio, ADC, AGC, NF
时间:04-08
整理:3721RD
点击:
Hi,
I'm in position to choose ADC for SDR operating in HF frequency range.
RX part of SDR should use harmonic sampling of bandpass signal.
From simple calculations and fact that i want to use max
ratio (RF/sample_rate) = 4, implies min_sample_rate = 7.5Mhz.
Bandwidth of signal will be narrow compared to
carrier frequency.
The issue is if we want high dynamic range about 100dB and want to use
12bit adc which has dynamic range about 72dB or even 10bit adc with dynamic
range 60dB the analog part should provide 30dB to 40dB of dynamic range.
Gain of analog part will be controlled by AGC. Friis equation implies
(Ftotal = Fanalog + Fadc/Ganalog) that overall noise figure is mainly dependent
on first stage. But in literature i found that noise figure of ADC is in range
20 to 30 dB. So to summarize it if gain of analog part is low (in presence of
strong signal) then overall noise figure will be worse than in presence of weak signal.
Is my reasoning correct? And if it is, can worsened noise figure impact radically reception of strong signals? What other parameters should I consider? (interception
points, desensitization ...)
I'm in position to choose ADC for SDR operating in HF frequency range.
RX part of SDR should use harmonic sampling of bandpass signal.
From simple calculations and fact that i want to use max
ratio (RF/sample_rate) = 4, implies min_sample_rate = 7.5Mhz.
Bandwidth of signal will be narrow compared to
carrier frequency.
The issue is if we want high dynamic range about 100dB and want to use
12bit adc which has dynamic range about 72dB or even 10bit adc with dynamic
range 60dB the analog part should provide 30dB to 40dB of dynamic range.
Gain of analog part will be controlled by AGC. Friis equation implies
(Ftotal = Fanalog + Fadc/Ganalog) that overall noise figure is mainly dependent
on first stage. But in literature i found that noise figure of ADC is in range
20 to 30 dB. So to summarize it if gain of analog part is low (in presence of
strong signal) then overall noise figure will be worse than in presence of weak signal.
Is my reasoning correct? And if it is, can worsened noise figure impact radically reception of strong signals? What other parameters should I consider? (interception
points, desensitization ...)
- How is the Passband defined for SAW filters?
- problem in importing a defined signal from matlab to ADS
- User-defined model generation for Cadence ADE
- Looking for a radio module for software defined radio.
- "The port has no terminals defined on it"
- S4100 Multi-Function Reader Module [Evaluation Software]
