Browse Prior Art Database

JITTER-BASED ADAPTIVE EQUALIZATION

IP.com Disclosure Number: IPCOM000007937D
Original Publication Date: 1996-Nov-01
Included in the Prior Art Database: 2002-May-07

Publishing Venue

Motorola

Related People

Authors:
David Feldbaumer Rimon Shookhtim Mark Weaver

Abstract

In jitter-based adaptive equalization, the distri- the prior art, adaptive equalizers rely on signal bution of edge transitions of the received data is amplitude to direct the adaptation. Our technique used to select the correct equalizer settings. The uses time-based digital information to determine technique is used in conjuction with a digital phase- proper adaptation for optimum signal to noise ratio locked loop (DPLL) which samples the incoming (Fig. 1). waveform multiple times during each symbol. In Signal Adaptive Analog D Equalizer Noise Edge Jitter Analyzer Relative Edge Error Fig. 1 Block Diagram of Jitter-Based Adaptive Equalizing In the preferred embodiment, an edge jitter the lock position. This provides a quantified meas- analyzer (Fig. 3) operates on information provided ure of relative edge error in time to a known refer- by a DPLL. The DPLL performs clock recovery on ence point. The distribution of this relative error is the incoming digital data which contains jitter. Once used to quantify the equalization and direct the the DPLL is in lock, each subsequent data edge tran- adaptation. sition occurs at a known position in time relative to Lock Position DPLL lime Samples 2 3 4 ' (Time) -1 Edge Tran.&ion Fll. 2 Relative Error Example showing a jitter edge crossing with a relative error of-l ' ' 0 Mncdn. Inc. 19% 171 November1996 MOFOROLA Technical Developments 8 The edge jitter analyzer consists ofan edge coun- the distribution of the edge crossings. The accumu- ter and an error accumulator. The relative error is lated value is the "Root Sum of Squares" of the jit- accumulated over a known number of edge cross- ter distribution. Its similarity to the standard devia- ings, producing a value which is very sensitive to tion is shown below.