Browse Prior Art Database

Sequence Decoder

IP.com Disclosure Number: IPCOM000056788D
Original Publication Date: 1988-Feb-01
Included in the Prior Art Database: 2005-Feb-14

Publishing Venue

IBM

Related People

Authors:
Martin, GN Schneider, RC [+details]

Abstract

This article describes a novel method of decoding (1,k) code data by looking at sequences of bits rather than one bit at a time. This results in greater resistance to noise and other signal improvements. Fig. 1 shows a typical (1,k) code data pattern equalized to class IV partial response. This results in a three-code bit wide pulse for each ONE as shown. Data is sampled at the edge of each code bit cell by an analog-to-digital (A/D) converter. Data is detected by an all-digital decoder that implements the equations shown as Table 1. Note, Vn is the value of the voltage sampled at time n, Vn+j is the value of the voltage sampled at time N+j, where j can be 1, 2 or 3. Examination of Fig.