Patent US4833635A successfully refreshed from Google Patents.
US4833635A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Bit-slice digital processor for correlation and convolution
Assignee: UK Secretary of State for Defence
•
Filing: 1987-03-05
•
Granted: 1989-05-23
•
Expires: 2007-03-05
Abstract
Abstract A bit-slice digital processor for performing an N-point correlation or convolution of N single-bit coefficients with a bit parallel, word serial, bit-staggered stream of M bit data words. The processor includes an N row, M column array of one-bit gated full adders with rows extended by half adders to accommodate word growth. Intercell connections incorporating clock activated latches provide for data and result flow unidirectionally down columns, one at twice the rate of the other. Carries and coefficients move unidirectionally along array rows at the faster (data or result) rate. Complex computations are executed by arrays of processors each with output delaying means and an adder to sum separate processor contributions. The processor may include data and result bypass connections subdivided by clocked latches for bypassing without operating speed penalties, as required for fault-tolerant processor array construction.
Inventors
- 👤 John V. McCanny
- 👤 Richard A. Evans
- 👤 John G. McWhirter
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-28 23:35
CPC Classifications
G; G06; G06F; G06F17/00; G06F17/10; G06F17/15; G06F15/00; G06F15/76; G06F15/78; G06F15/7896; G06F15/80; G06F15/8007; G06F15/8023
Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
219 words
A bit-slice digital processor for executing correlation and convolution operations of N single bit coefficients with a bit parallel, word serial, bit staggered data stream of M bit words, where N and M are positive integers greater than unity, and wherein: (1) the processor includes an array of N rows and M columns of logic cells; (2) each logic cell is arranged to: (a) input a data bit, a carry bit and a cumulative sum bit, (b) output the data bit, and (c) produce an output cumulative sum bit and an output carry bit corresponding to addition of the input cumulative sum bit and input carry bit to a product arising from multiplication of the input data bit by a coefficient bit corresponding to a respective row of cells; (3) cell interconnection lines are arranged to provide for carry bit propagation along array rows and for data flow and cumulative sum generation to proceed unidirectionally down array columns; and (4) the cell interconnection lines include clock activated delaying means arranged to provide for data bits and cumulative sum bits to move down array columns at respective rates one of which is twice the other, and for carry bits to move along array rows at the rate of the faster of these in the direction of increasing data bit significance.
Claim 2
Dependent Claim
219 words
A bit-slice digital processor for executing correlation and convolution operations of N single bit coefficients with a bit parallel, word serial, bit staggered data stream of M bit words, where N and M are positive integers greater than unity, and wherein: (1) the processor includes an array of N rows and M columns of logic cells; (2) each logic cell is arranged to: (a) input a data bit, a carry bit and a cumulative sum bit, (b) output the data bit, and (c) produce an output cumulative sum bit and an output carry bit corresponding to addition of the input cumulative sum bit and input carry bit to a product arising from multiplication of the input data bit by a coefficient bit corresponding to a respective row of cells; (3) cell interconnection lines are arranged to provide for carry bit propagation along array rows and for data flow and cumulative sum generation to proceed unidirectionally down array columns; and (4) the cell interconnection lines include clock activated delaying means arranged to provide for data bits and cumulative sum bits to move down array columns at respective rates one of which is twice the other, and for carry bits to move along array rows at the rate of the faster of these in the direction of increasing data bit significance.
Claim 3
Dependent Claim
39 words
A processor according to claim 1 including additional cell interconnection lines and clock activated delaying means arranged to provide for coefficient bits to move along respective array rows at the rate and in the direction of carry bit movement.
Claim 4
Dependent Claim
56 words
A processor according to claim 1, wherein array columns have respective final cells, said final cells having cumulative sum outputs connected via respective programmable clock activated delaying means to respective first inputs of a programmable clock activated parallel adder arranged to add signals at said first inputs to signals at respective second inputs of said adder.
Claim 5
Dependent Claim
26 words
A processor according to claim 3 including bypass connections for input data and for said second adder inputs, the bypass connections being subdivided by clock-activated latches.
Claim 6
Dependent Claim
34 words
A processor according to claim 3, wherein the width of the parallel adder is sufficient to accommodate a relative difference in bit significance between signals applied to said first and second adder inputs respectively.
Claim 7
Dependent Claim
26 words
A processor according to claim 5, wherein the width of the parallel adder is sufficient to accommodate a relative difference in bit significance of M bits.
Claim 8
Dependent Claim
40 words
A processor according to claim 1 including log 2 n half adders extending the nth row of logic cells, where n is equal to 1 to N and log 2 n is rounded up to a whole number if necesssary.
Claim 9
Dependent Claim
79 words
A processor according to claim 1 including log 2 (n-1) half adder cells extending the nth row of logic cells, and wherein each row has a respective final cell to which carry bits flow and the (n-1)th row has a final cell carry output connected via delaying means to a sum input of an appropriate nth row half adder, where n is equal to 2 to N and log 2 (n-1) is rounded up to an integer if necessary.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1992-05-22 | 1992-11-21 | 1993-05-23 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1996-05-22 | 1996-11-21 | 1997-05-23 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2000-05-22 | 2000-11-21 | 2001-05-23 | $7700.00 | Window Closed / Lapsed |