+- 90 phase shifter revisited

Jun 22, 2013 65 Replies

The dielectric constant of air changes when you compress it, and acoustic n oise can change it tolerably fast, but I somehow doubt that thunderclaps ar e going to be a troublesome noise source in this context. Multiple path eff ects could be more of a problem, but they can be really evil, and are bette r tackled by highly directional receivers or multiple receivers.

I've got a good memory, but not that good.

But it does double as an advertising arena, as John Larkin demonstrates fro m time to time. That particular example struck me as more impressive than i nteresting - but I didn't do the work, and I have learned that anything I'v e worked on always looks a lot more interesting to me than it does to other people.

Bill Sloman, Sydney

Sure did. It is used for a whole lot of things, a lot of it is induced by ATM adaption layers. At your best did you really understand DQLB? I couldn't quite follow the statistical math at the time.

Yes, that is called Asyncronous Transfer Mode (ATM).

You are probably already familiar with what is in this:

formatting link

How familiar are you with these:

formatting link

formatting link

How well do you understand the ability to dynamically assign void pointers in the headers? How about virtual tributaries? Understand that every header pointer refers to one or more octets (bytes) never a single bit; thus no bit stuffing.

CHANGES, how fast changes you can detect is set by the loop filter:

output

------->------ VCO (= voltage controlled oscillator) ----

|
|
|

-----

incoming signal:

controlled oscillator) ----->-- phase shifter carrier relative to input

|
|
|

carrier relative to input

formatting link

It's called a pointer movement. ATM is just one of many payloads that can be carried by SONET, although these days the few new SONET installations typically do not carry ATM. We're still selling new SONET gear (that I designed almost a decade ago), but only into old installations, and not very often. Our ATM sales stopped years ago.

In any case, none of this refutes Miso's statement that justification is used to account for frequency differences.

Pretty well.

Byte stuffing (as in SONET) or bit stuffing (as in many PDH systems), in either case justification is being used to account for frequency differences.

I must thank you for the deep belly laugh I got from "I happen to have read various [...] standards."

Allan

That

to

for

T-1,

pointer

induced

I

Really? The whole system is plesiocronous.

AMI has problems with long sequences of zeroes, so for DS-1 they use B8ZS which "stuffs bits" to improve data integrity, not to correct for clock differences. They also use B3ZS for DS3 for the same reason. Clock synchronization for such systems uses other methods than simple bit stuffing.

Glad to have made you happy.

?-)

Plesiocronous means that there can be small, constrained frequency differences.

What a coincidence! Here is a B3ZS encoder and decoder that I designed on the back of an envelope sometime last century, then finally converted to VHDL and made open source in about 2003:

formatting link

That particular design never made it into a product, but related ones I did around the same time ended up in various bits of Fujitsu telecoms gear.

Regards, Allan

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required