|
Pilot1
- response to NF degradation due to alias noise
|
(no rating)
2008-05-08 |
I believe you're asking whether the NF degradation due to noise in 15 alias bands should be calculated as 10*log(16) instead of 10*log(15). The value in the parenthesis is a factor, or in this case, 15 times more noise power than when alias noise is totally filtered. For this case, I believe the 10*log(15)is correct.
For your second question, the minimum attenuation target of 16dB for noise power in 'each' of the adjacent alias bands was selected to limit the total NF degradation due to alias noise = 0.2dB. This is calculated by adding noise powers in dB. That is, if I assume a reference power (X dBm) and add power that is 13dB lower than the reference, the total power is (XdBm 0.2dBm). Therefore, if noise power from the two adjacent alias bands is attenuated to a level of -16dB each relative to the ADC noise power in the desired bandwidth, the total added noise due to aliasing is 13dB lower than the desired, which will cause 0.2dB degradation.
I hope this answers your questions.
|
|
|
