Browse Prior Art Database

Recursive Tracking Threshold

IP.com Disclosure Number: IPCOM000052077D
Original Publication Date: 1981-Apr-01
Included in the Prior Art Database: 2005-Feb-11
Document File: 2 page(s) / 37K

Publishing Venue

IBM

Related People

Hartley, JD: AUTHOR [+2]

Abstract

In detecting NRZI data where a pulse represents a one and no pulse represents a zero, a threshold is used in the detection circuitry. Signals above the threshold are detected as ones and signals below the threshold are detected as zeros, as shown in Fig. 1. The signal may be full wave rectified, or two thresholds of opposite polarity may be used.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 74% of the total text.

Page 1 of 2

Recursive Tracking Threshold

In detecting NRZI data where a pulse represents a one and no pulse represents a zero, a threshold is used in the detection circuitry. Signals above the threshold are detected as ones and signals below the threshold are detected as zeros, as shown in Fig. 1. The signal may be full wave rectified, or two thresholds of opposite polarity may be used.

Certain high density flexible media systems (1600 bpi, 6250 bpi, etc.) use a form of detection where signal peaks are converted to zero crossings. These methods often require low frequency boost. This is undesirable at high densities where a separate erase head is not used, since low frequency boost tends to amplify the unwanted residual signal from the previous write passes.

An improved equalization method is described in connection with Fig. 1 where the readback pulses are narrowed and a different detection method is used. The technique involves the use of a variable threshold and is implemented by a simple circuit.

The variable threshold voltage is stored on a capacitor, and the voltage value is updated with each read pulse (detected one). The new voltage at time (N) is the old voltage at time (N-1) plus a voltage change that is proportional to the difference between the signal peak voltage and the capacitor voltage. Hence, when there is zero difference between the signal and capacitor voltage, the capacitor voltage is correct and no change is needed.

The circuit of Fig. 2 is an analog cir...