Re: Pseudorandom Hashing
From: Jim Thompson (thegreatone_at_example.com)
Date: 09/24/04
- Next message: Zorpetus: "Re: Tap V.90 modem line - possible?"
- Previous message: xray: "Re: [OT]: Ping Kevin Aylward - re your "scientific paper""
- In reply to: John Fields: "Re: Pseudorandom Hashing"
- Next in thread: Rich Grise: "Re: Pseudorandom Hashing"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 23 Sep 2004 17:02:01 -0700
On Thu, 23 Sep 2004 18:39:10 -0500, John Fields
<jfields@austininstruments.com> wrote:
>On Thu, 23 Sep 2004 14:40:28 -0700, Tim Wescott
><tim@wescottnospamdesign.com> wrote:
>
>>I am having trouble coming up with the right keywords to do a web
>>search, so help me out here:
>>
>>There is a technique where, to significantly reduce the probability of
>>getting a long string of zeros, a message is run through a CRC
>>generator, and the output bits are taken off. The transmitted message
>>is thoroughly hashed, yet it is a simple matter of a shift register and
>>some XOR gates to decode the message on the other end.
>>
>>I thought I knew how to do this, yet in trying to actually make it work
>>I find that over half of my brain cells appear to be attending a
>>management seminar.
>>
>>So, know where I can find out how to do this right? "pseudorandom" and
>>"hash" get me tons of cryptography, but not what I'm looking for.
>
>
>---
>Is this what you had in mind?
>
> A-----------+
> +------------------------Y EXOR |
> | B--+ |
> +--A | |
> EXOR Y--[D Q]---[D Q]---[D Q]-+-[D Q]--+--->DATA OUT
>DATA IN>-- B
>
>Each [D Q] is a stage of a shift register, and as each new data bit is
>presented to the first EXOR, it's EXORed with the EXOR of older bits
>which have propagated down the chain and been EXORed with each other.
>
>The result is that the DATA OUT stream of bits is scrambled and bears
>no resemblance, on the surface, to DATA IN.
>
>In order to recover DATA IN, the DATA OUT stream is presented as DATA
>IN to an identical shift register / EXOR circuit and, if they're both
>synced up (the circuits) what comes out of the receiver will be
>exactly what went into of the transmitter.
>
>Depending on what goes into the transmitter input, there may be long
>strings of ones or zeroes which come out of the output. This can be a
>bad thing for synchronous data modems which use data transitions to
>keep their their 'dot clocks' in sync with incoming data, so what's
>done to circumvent that is that a counter is placed in the transmitter
>circuitry which counts ones and/or zeroes, and if a certain number of
>them occur, in a row, then a one or a zero is stuffed into the data
>stream (or a bit is flipped) to break up the sequence.
I posted...
Newsgroups: alt.binaries.schematics.electronic
Subject: Pseudorandom Hashing (S.E.D) - NRZ-AddTransitions.pdf
Message-ID: <9mo6l0tpq9jn3q07fsrfpkd1lbot33330i@4ax.com>
which is INCORRECT. John has it right... that's what I was trying to
remember from eons ago.
...Jim Thompson
-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | | http://www.analog-innovations.com | 1962 | I love to cook with wine. Sometimes I even put it in the food.
- Next message: Zorpetus: "Re: Tap V.90 modem line - possible?"
- Previous message: xray: "Re: [OT]: Ping Kevin Aylward - re your "scientific paper""
- In reply to: John Fields: "Re: Pseudorandom Hashing"
- Next in thread: Rich Grise: "Re: Pseudorandom Hashing"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|