d****6
发帖数: 971 | | x***y
发帖数: 830 | 2
less flicker noise
【在 d****6 的大作中提到】
: 谢谢
: 搜了下没搜到啥很可靠的回答
| b*********y
发帖数: 830 | 3 my 2 cents. don't know if it is rite.
1, the magnitude of the fundamental freq component of a squre wave has a pi/
2 relationship with the amplitude of the square wave, which means higher
conversion gain for square wave LO.
2, in current switching mixer like gilbert cell, sine LO will result in more
noise since the both of RF differential pair maybe on at some time so they
are not degenerated by the tail current source transistor controlled by the
LO. for sqaure wave LO, all the time, only one transistor of the RF
differential pair is on, which is always degerated by the tail transistor
controlled by the LO, so the diff pair will contribute no noise up to medium
frequency since it acts as a cascode device rather than amplifier device. | d*****o
发帖数: 2868 | 4 actually, i was reading about this couple days ago. Basically, when you
using a square wave to its LO input, it alternatively multiplies the linear
input of the RF-port signal by +/-1, this will prevent low intermodulation
and harmonic distortion; all it has to do is preserve the fast rise and fall
times othe switching signal(mixer) |
|
