Figure 12.39 Analog-to-digital conversion. - PowerPoint PPT Presentation

PPT – Figure 12.39 Analog-to-digital conversion. PowerPoint presentation | free to download - id: 56e339-ZDk3Y The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
Title:

Figure 12.39 Analog-to-digital conversion.

Description:

Figure 12.39 Analog-to-digital conversion. Figure 12.40 The DAC output is a staircase approximation to the original signal. Filtering removes the sharp corners. – PowerPoint PPT presentation

Number of Views:232
Avg rating:3.0/5.0
Slides: 13
Provided by: MWAH3
Category:
Tags:
Transcript and Presenter's Notes

Title: Figure 12.39 Analog-to-digital conversion.

1
Figure 12.39 Analog-to-digital conversion.
2
Figure 12.40 The DAC output is a staircase
approximation to the original signal. Filtering
removes the sharp corners. (Note In addition to
smoothing, the filter delays the signal. The
delay is not shown.)
3
7?
6?
5?
4?
3?
2?
1?
0?
Figure 12.49 Output versus input for a 3-bit
flash A/D converter
4
DNL (differential nonlinearity) and INL (integral
nonlinearity)
5
Figure 12.41 Circuit symbol for a
digital-to-analog converter.
6
Figure 12.42 DACs can be implemented using a
weighted-resistance network. (Note If di 1,
the corresponding switch is to the right-hand
side. For di 0, the i th switch is to the
left-hand side.)
7
Figure 12.43 An R -- 2R ladder network. The
resistance seen looking into each section is
2R. Thus, the reference current splits in half at
each node.
8
Figure 12.44 An n-bit DAC based on the R2R
9
Figure 12.45 Parallel, simultaneous, or flash
A/D conversion.
10
11
Initially, all bits are set to 0 In step 1, the
control logic sets MSB to 1 and if the comparator
output is high, MSB is set back to 0, otherwise
MSB remains 1 The process is repeated for the
next bit. After n steps, the process is complete,
and the input to the DAC is the digital code for
12
Oversampling A/D converters
E(n)Y(n)-X(n) is defined as quantization noise,
Y(n) is the quantized output and X(n) is the
input. E(n) is between (-?/2, ?/2) Where ? is the
quantization level. E(n) is typically
approximated as an independent uniformly
distributed white noise and its power spectral
density is k , fs is the sampling
frequency. Therefore, increase the fs relative to
the signal bandwidth will give higher resolution
than Nyquist sampling converters. Even further,
if oversampling is combined with noise shaping,
such as in a Sigma-Delta A/D converter, then the
resolution could be even higher.