Browse Prior Art Database

METHOD and APPARATUS for EFFICIENT ENCODING of START/STOP or START/DURATION NUMERICAL CONTROL DATA

IP.com Disclosure Number: IPCOM000034432D
Original Publication Date: 1989-Feb-01
Included in the Prior Art Database: 2005-Jan-27
Document File: 5 page(s) / 119K

Publishing Venue

IBM

Related People

Ryan, PM: AUTHOR

Abstract

This article suggests a means to reduce the number of bits required to transmit X start/stop or X start/displacement data for numerically controlled (N/C) tools. It is proposed to code X start and W X values needed to specify a rectangle in such a way that one less bit is needed per (X,W X) pair. The extra bit is helpful in keeping N/C data on fixed data boundaries. In N/C tools used in semiconductor operations, such as an E-beam lithography tool, it may be necessary to specify in some coordinate system a start and stop point, or a start point and displacement in one or more dimensions. For instance, a rectangle to be exposed (Fig. 1) will be expressed in terms of X,W X and Y,W Y. Each coordinate axis will generally be addressed by an n-bit code, providing 2n addressable points along the axis (0, 1, 2, ....2n -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 5

METHOD and APPARATUS for EFFICIENT ENCODING of START/STOP or START/DURATION NUMERICAL CONTROL DATA

This article suggests a means to reduce the number of bits required to transmit X start/stop or X start/displacement data for numerically controlled (N/C) tools. It is proposed to code X start and W X values needed to specify a rectangle in such a way that one less bit is needed per (X,W X) pair. The extra bit is helpful in keeping N/C data on fixed data boundaries. In N/C tools used in semiconductor operations, such as an E-beam lithography tool, it may be necessary to specify in some coordinate system a start and stop point, or a start point and displacement in one or more dimensions. For instance, a rectangle to be exposed (Fig. 1) will be expressed in terms of X,W X and Y,W Y. Each coordinate axis will generally be addressed by an n-bit code, providing 2n addressable points along the axis (0, 1, 2, ....2n -1). Typically, certain constraints are observed: X + delta X & 2n (start/displacement case), or (start/stop case)

X start & X stop Normally, X and W X or start and X stop are expressed in the n-bit code noted above. This requires 2n bits to describe a line along either axis. However, the total number of distinct lines which may be drawn along an axis is: 1/2(2n) (2n-1) = (2n-1) (2n-1) This is illustrated in Fig. 2 for the case n=3. The (2n-1) (2n-1) combinations may be represented in a code of (n-1) + n = (2n-1) bits, rather than the 2n bits required in the natural form of specification.

(Image Omitted)

The effect of the constraints for the case n=3 is illustrated in Fig. 3. The table indicates the value of X stop for only the legal combinations of X start and W X. Code compression will be accomplished by "folding over" the upper right triangular portion of the table to coincide with the shaded triangle at the bottom of the table, and appropriately modifying row and column labels. The results of the foregoing modifications are illustrated in Fig. 4, also for the case n = 3. The W X columns have been relabeled to run from 0 to 2 n-1-1 so it may be specified with n-1 bits. The coded Xstart rows run from 0 to 2 n-1, requiring n bit specification. Where METHOD AND APPARATUS FOR EFFICIENT ENCODING OF START/STOP OR START/DURATION
NUMERICAL CONTROL DATA - Continued the sum (Xstart +W X) produces a carry into bit (n+1) of the sum, the bits of both the sum and row label are reinterpreted for output as shown. This choice of codes permits a very simple hardware implementation of decoding logic. This is illustrated for the general (n, n-1) case in Fig. 5.

Generation of the code is straightforward as indicated in Fig. 6, whether the original specification is in (Xstart W X) format or (Xstart'W X stop). In Fig. 6, the overbar indicates the ones complement inversion of all bits. Two examples illustrate the

1

Page 2 of 5

application of this technique:

(Image Omitted)

EXAMPLE 1:...