What is serial 'NRZ' digital data?
king-chung, Ng
Does anyone know what does serial 'NRZ' digital data look like?
What is different between normal digital data and serial 'NRZ' digital
data . What is advantage of producing it? Does it same as normal digital
data between 0V to +5V?
If anyone have any information, please tell me.
Thank You.
Dave Bushong
>Hi!
> Does anyone know what does serial 'NRZ' digital data look like?
>What is different between normal digital data and serial 'NRZ' digital
>data . What is advantage of producing it? Does it same as normal digital
>data between 0V to +5V?
NRZ stands for "Non-Return to Zero" and it is a way of encoding data
so that you continue to have zero-crossings even if the data stays
constant at 0 or 1. The reason for this is so that the clock can be
recovered from the data stream, and you don't have to encode data and
clock as two separate data streams.
It is independent of what a logic '0' and logic '1' are, as far as
what voltage it happens to be, but is a data format instead.
It is commonly used digital tape drives (streaming backups).
Dave
--
Dave Bushong
OPEN/image Recognition Products
Sam Goldwasser
>Hi!
> Does anyone know what does serial 'NRZ' digital data look like?
>What is different between normal digital data and serial 'NRZ' digital
>data . What is advantage of producing it? Does it same as normal digital
>data between 0V to +5V?
> NRZ stands for "Non-Return to Zero" and it is a way of encoding data
Yes.
> so that you continue to have zero-crossings even if the data stays
> constant at 0 or 1.
I do not believe so.
NRZ simply codes 0s as lows and 1s as highs without regard to run length.
Most encoding techniques have strived to be self clocking.
> The reason for this is so that the clock can be
> recovered from the data stream, and you don't have to encode data and
> clock as two separate data streams.
For example, the codes used on hard disks, tapes, CDs, communications networks.
Simple NRZ is NOT self clocking since there is no specified limit
on the length of a run of 0s or 1s. A simple encoding technique that is
self clocking is called 'phase modulation' or PM. With PM, each '0'
bit is encoded with a 0-1 transition in the middle; each '1' bit with a
1-0 transition in the middle. Adjacent 0s or 1s will have an additional
transition at their common boundary.
An Example:
Data 0 1 0 0 1 1 1 1 0 1
_____ _______________________ _____
NRZ _____| |___________| |_____| |
_____ __ _____ __ __ __ _____
PM __| |_____| |__| |__| |__| |__| |_____| |__
A PLL or even simpler circuit can easily recover the clock from the PM data
stream. Very simple logic can then reconstruct the data and generate what
amounts to a NRZ bitstream.
A disadvantage of PM is the higher channel bandwidth required due to the
additional transitions.
I don't recall the exact details, but MFM and RLL of PC hard disk fame,
and the EFM (eight to forteen modulation) code used on audio CDs and CDROMs
are self clocking. They are examples of run length limited codes
designed to facilitate clock recovery and control data stream bandwidth.
--- sam
Mike Nauss
Well to stay out of any arguments I will just quote something
out of Sams Data Communications book.
Unipolar Code:
The waveform of binary signals normally used in computers and
terminals is called unipolar;that is, the voltage representing the bits
varies between 0V and +5V.(This code is called Non-Return to Zero, because
the voltage does not return to zero between adjacent 1 bits.)
What I beleive the first person that replied was refering to was
NRZI. Now I am looking at my ZILOG Comm data book. NRZI is a one is represented
by a transition at the beginning of a bit cell. A zero is represented by the
absence of a transition. By insuring that there are enough transitions the
receiving device can lock to the incomming data. To further enhance the code
a minimum ones density must be used. If to many zeros are sent then the
receiver will lose lock and a frame slip will occur. If a string of zeros
are required to be sent then things like Bit seven stuffing are used.
Brian Mork
> Hi!
> Does anyone know what does serial 'NRZ' digital data look like?
> What is different between normal digital data and serial 'NRZ' digital
> data . What is advantage of producing it? Does it same as normal digital
> data between 0V to +5V?
I want to know the same thing. I know what the letters stand for:
"Non-Return to Zero" and it uses a transition to represent a 1 and a
non-transition to represent a 0. But what confuses me is that people
claim the time duration of a bit is 1/2 as much. Doesn't seem to
make sense!?
Please forward any responses you get!
--
Brian J. Mork. Voice (509) 244-3764 Data (509) 244-9260
UUCP bm...@opus-ovh.spk.wa.us / ARO ka9snf@ka7fvv.#ewa.wa.usa
USMail 6006-B Eaker, Fairchild, WA 99011
Mike Nauss
Whats with the BZZZZT!!!! WRONG!!!!! If you insist on being a
know-it-all, you should at least be right. The following is taken from
page 10 of the 1991 ZILOG Datacom IC databook.
NRZ: In NRZ a 1 is represented by a high level for the duration
of the bit cell and a 0 is represented by a low level for the duration
of the bit cell.
NRZI: In NRZI a 1 is represented by a transition at the beginning
of the bit cell. That is,the level present in the preceding bit cell is
reversed. A 0 is represented by the absense of a transition at the
beginning of the bit cell.
Well that seems fairly clear. Now go argue with ZILOG!!!
>In article <SAM.94Se...@colossus.stdavids.picker.com> s...@colossus.stdavids.picker.com (Sam Goldwasser) writes:
>>NRZ simply codes 0s as lows and 1s as highs without regard to run length.
>^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>BZZZZT!!!! WRONG!!!!!
>NRZ comes in two types: NRZ1 and NRZ0
>The state of the logic gate driving the head toggles for either a "1" or
>"0" and remains the same for a "0" or a "1" depending on the type.
>A J-K flipflop was the usual method for incoding data.
>The usual check bit is a odd parity. This forces one transition at least
>so there is always something to lock your circuits to.
>Ken
Kenneth Roy Smith
Andrew Ingraham
(king-chung, Ng) writes:
> Does anyone know what does serial 'NRZ' digital data look like?
>What is different between normal digital data and serial 'NRZ' digital
>data .
None. NRZ is just ordinary, unencoded binary data that you're already
familiar with.
Clearly, there are differences in opinion; one person's "NRZ" might not be the
same as another's. Especially when it comes to data encoding or combining
with clock. But this is how I learned it.
Andy Ingraham
Andrew Ingraham
s...@colossus.stdavids.picker.com (Sam Goldwasser) writes:
>... A simple encoding technique that is
>self clocking is called 'phase modulation' or PM. With PM, each '0'
>bit is encoded with a 0-1 transition in the middle; each '1' bit with a
>1-0 transition in the middle. Adjacent 0s or 1s will have an additional
>transition at their common boundary.
Sounds to me like what's called "Manchester encoding."
Andy Ingraham
Vincent C. Gaw
coding. NRZ is the used to conceptualize digital representation. But when we
go into actual implementation, we *may* decide to use other methods to name
some of a few: Unipolar RZ, Bipolar, Return to Zero (RZ), Bipolar Phase Encoded
(Manchester), Delay Modulation, Dicode Binary (multi-level), etc.... for
purposes such as synchronization and/or noise immunity.
Vincent