oversampling

W H Powell (bill@mrv5.demon.co.uk)
Sun, 4 Jan 1998 15:51:21 +0000


On Sat, 3 Jan 1998, KC5TJA <kc5tja@topaz.axisinternet.com> wrote
>
>I'm sorry, but you cannot use a 1-bit A/D converter for this application.
>The cost of the circuitry and software development time would make it
>completely unworthy of design for high throughput and error-free
>operation.
>
>Besides, you have different requirements for signal input than signal
>output. A 1-bit D/A converter is NEVER used in commercial-grade
>telecommunications equipment, and for good reason.
ITT, in its UK STL Labs., pioneered the use of 1 bit codecs in the mid
to late 1970s for use in 64 kbit/s telephone codec. The A/D ran at
approx 4 MHz sampling rate. The speech signal is only to 3.3 KHz or so.

I worked on a follow-on telecom project with sampling running at 100
MHz. It was the only way we could build an A/D that was sufficiently
linear to avoid generation of intermodulation products where the wanted
signal was much smaller than interfering signals.

One bit A/D is very simple in principle, but needs some good _analogue_
design. 
Also the resulting 1 bit code has to be digitally 'decimated' to make
any use in micro or DSP device. Every bit is of equal weight. Only
sampling time distinguishes one bit from another. It does not turn into
'big or little Endion' byte data until you've done this as well.

The higher the oversampling is the better a 1 bit A/D can be. It works
very like FM deviation, the more the deviation or sampling rate the
better. Indeed what you see on a spectrum analyser if you look at the 1
bit data stream has great similarity to Frequency Modulation.
>
>> As to the A/D, remember that the spectrum of the phone signal is band
>> limited. So oversampling is again the key word.
Very true.

-- 
Bill Powell.  ( MIME, UU )

  Atherstone, Warks., CV9 3AR.  | Tel: +44 1827-718 945
  <whpowell@iee.org>   | Fax:  -714 884

.