🚀 go-pugleaf

In article <4528@manta5.UUCP>, wang@motcid.UUCP (Jerry Wang) writes:
>    length of the sequence.  This is why it's called 'spread' spectrum.  The
>    key advantage of spread spectrum is 'processing gain' over white noise, as
>    evidenced by Shannon's channel capacity theory (Shannon said,
>    you can get any BER performance you want given a S/N, as long
>    as you have the 'bandwidth' to do it.  So the bandwidth increase
>    contributes to 'processing gain directly).  Another advantage
>    is 'anti-jamming' capability due to the spectrum spreading effect.

I don't think this is quite true. By itself, spread spectrum does not
give you any power advantage over non-spread signals in the presence
of gaussian ("white" or thermal) noise. That is, if it takes 10 watts
to send 10 kb/s over a given path without spread spectrum, spreading
will not change this figure. What *will* change this figure is forward
error correction (FEC) coding, which spends bandwidth to save power.
But, unlike spread spectrum, this excess bandwidth cannot be shared
with other stations.

The so-called "processing gain" in spread spectrum is simply an
artifact of the "effective" (information) bandwidth of the signal
being much smaller than its actual (occupied) bandwidth. If you have a
signal that ordinarily fits in 10 KHz and you spread it to 1 MHz (a
ratio of 100:1 or 20 dB) then your receiver will still perform as
though its noise bandwidth were 10 KHz even though its front end
bandwidth is 1 MHz.  So the "processing gain" in this case is a wash.

Where processing gain *does* buy you something is in the presence of
narrowband interference in your signal passband ("narrowband" == much
narrower than the spreading bandwidth). An interfering narrowband
signal will be *spread* by the despreading process at the receiver to
the full spreading bandwith of the system, so for our example its
effect on the wanted signal will be reduced by the 20-dB processing
gain of the system.

So "processing gain" would be better described as spread spectrum's
"unwanted signal rejection ratio", analogous to the "adjacent channel
rejection" figures given for conventional narrowband receivers. Since
the latter figures are often 60 dB or more, you can see that spread
systems are actually at a disadvantage when the signals are not all
closely matched in level. This is the well-known "near-far" problem,
and solving it requires careful transmitter power control.

So why use spread spectrum at all? There are several key advantages:

1. Jam resistance. This is primarily of military interest, but it can
also help in civilian environments where the "jamming" may be more
accidental than intentional. It may also eliminate the need for the
bureaucratic overhead of frequency coordination. (I see the Part 15 SS
rules at 902-928 MHz as a "regulatory experiment" to see if spread
spectrum can indeed work this way).

2. Multipath resistance. Spread spectrum receivers can track just one
of several reflected signals, rejecting the others as interference.
This is very attractive in a mobile environment, especially in cities.

3. Suppression of narrowband interference. This is especially useful
in making use of "garbage" spectrum that would not otherwise be useful
for communications, e.g., the ISM bands.

Phil