Browse Prior Art Database

Multiple Sampling Magnetic Film Memory

IP.com Disclosure Number: IPCOM000075751D
Original Publication Date: 1971-Nov-01
Included in the Prior Art Database: 2005-Feb-24
Document File: 2 page(s) / 43K

Publishing Venue

IBM

Related People

Sowter, BR: AUTHOR

Abstract

Random or thermal noise is overcome by detecting several signals from a storage location or element of a nondestructive memory having a magnetic film 10, as shown in Fig. 1.

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 54% of the total text.

Page 1 of 2

Multiple Sampling Magnetic Film Memory

Random or thermal noise is overcome by detecting several signals from a storage location or element of a nondestructive memory having a magnetic film 10, as shown in Fig. 1.

A bit stored in an element of film 10 is selected in a conventional manner by applying from word driver 12 word drive pulses 13, indicated in Fig. 2, to a desired word line 14 in conjunction with the operation of low-level switch 16. The sense signal produced in a corresponding sense line 18, after passing through switch 16, is amplified in amplifier 20 and shaped in shaper 22 before being applied to AND gate 24. A reference signal of predetermined polarity produced in reference line 26 is amplified in amplifier 28 and shaped in shaper 30 before being applied to AND gate 24. As indicated in Fig. 2, the output of the AND gate 24 is a train of pulses representing a 1 bit of information, and the absence of a pulse representing a 0 bit of information. The presence of thermal noise causes some of the samples to be incorrect but by utilizing, e.g., a scale of 16 or divided- by 16 binary counter 32, it is only necessary to have a minimum one more correct sample than incorrect samples before a correct pulse is applied to data latch 34. Assuming a signal-to-noise ratio of 2:1, total samples per read operation n is 31 and the maximum permissible incorrect samples k is 15, the error rate is 0.112x10/-17/ per read operation, as given by

(Image Omitted)

where, p = 1-q, which is equal to the probability of an incorrect sample, a function of si...