请教一下大家为什么CT quadrature BP SD ADC 在2002年才有发表的论文呢？

vdslafe

你是说第一个发表CT quadrature, 那有可能

hezudao

CT quadrature bandpass modulator

CT is more difficult than DT(historically), need clean sampling clock, and multibit-DAC, so research on on-chip LO and DAC mismatch-shaping are prerequisites.

Quadrature signal processing normally needs good IMR, the higher the f0, the more stringent of requirement.

bandpass SDM shift the problem to aaf (less sensitive to low-frequency noise but need high Q aaf)

so CT quadrature bandpass modulator has no obvious advantage over other structures.

kuxuanxinzai

回复 12# hezudao


   

敢问仁兄这段话是哪里的？

hezudao

回复 13# kuxuanxinzai


    it's my understanding.....

kuxuanxinzai

回复 14# hezudao


    但是并不是所有的都要多比特的DAC，现在已有的大都是单比特的。
    quadrature 可以有效的提升OSR,而且在镜像通带引入适当的噪声传输函数零点，也可以将mismacth 的影响降低。这个如果应用在接收机中，可以避免直流失调和闪烁噪声。而且，在一般的时数调制器结构中，负频率的噪声整形零点在一定程度上实际上是浪费的。

   所以说quadrature 完全没有优势还是不很准确的，要看应用场合了。

hezudao

回复 15# kuxuanxinzai


    但是并不是所有的都要多比特的DAC，现在已有的大都是单比特的。
    quadrature 可以有效的提升OSR,而且在镜像通带引入适当的噪声传输函数零点，也可以将mismacth 的影响降低。这个如果应用在接收机中，可以避免直流失调和闪烁噪声。而且，在一般的时数调制器结构中，负频率的噪声整形零点在一定程度上实际上是浪费的。

   所以说quadrature 完全没有优势还是不很准确的，要看应用场合了。
   

I just want to explain why bandpass quadrature takes off relatively late. And I never say bandpass quadrature has no advantage at all. Read my post carefully, what I said is that 'at the very beginning', bandpass quadrature has no 'obvious' advantage.

但是并不是所有的都要多比特的DAC，现在已有的大都是单比特的。
if you work in industry, you will find the majority is multi-bit, rather than one-bit right now.
when you look at those papers, you need pay attention to the maximum input swing, clock used(some assume perfect off-chip clock).

quadrature 可以有效的提升OSR,
compared with what? if bandpass, yes, if low pass no

而且在镜像通带引入适当的噪声传输函数零点，也可以将mismacth 的影响降低。
This would help. But it's not only about quantization noise. When you have strong interference as in GSM, still quite a lot of image leakage not matter how good NTF you have.

这个如果应用在接收机中，可以避免直流失调和闪烁噪声。
for sure, this is the benefit of low IF architecture, and not special to quadrature ADC

而且，在一般的时数调制器结构中，
I don't understand what do you mean by 时数调制器
负频率的噪声整形零点在一定程度上实际上是浪费的。
you are right. That's the beauty of quadrature. But it does not mean it's easy to make use of it.

Again, I would never claim bandpass quadrature is inferior to other structures, and I would not be surprised that bandpass quadrature SDM has better performance than its lowpass counterpart in certain applications.

kuxuanxinzai

回复 16# hezudao


    You are very professional，Thank you very much for your advice.
    之前的回复中“时数调制器结构”是笔误，我的意思是实数结构的调制器，相对于quadrature 结构的调制器而言。
    另外你的观点“quadrature 可以有效的提升OSR, compared with what? if bandpass, yes, if low pass no”我觉得有误。我觉得提升OSR是相对于所有实数结构的的调制器吧，不管是带通还是低通。原因如下“For a rea) (non-quadrature) system, signals beyond fs/2 suffer from aliasing,whereas for a quadrature system the corresponding limits are ±fs/2. The total alias-free bandwidth is thus fs. In order for OSR = 1 to correspond to no oversampIing, the OSR of a quadrature system is defined as OSR = fs/fb. In other words, for a given signal bandwidth and sampling rate, a quadrature modulator has an OSR that is twice that of a real modulator.”引自"Understanding delta sigma data converters" P144, L3-L8.

vdslafe

NTF high-frequency corner 需要远小于fs, 要不modulator 不稳定。
但是NTF hf corner 还需要大大于IF+fb/2，要不就需要高阶系统
所以比较适合用于IF 频率比较低的应用，要不时钟就要跑得飞快

ralphtw

quad sd adc !

fireworker

learn quad bp adc