Patent US5956342A successfully refreshed from Google Patents.
Patent US5355345A successfully refreshed from Google Patents.
← Back to Patent Explorer
US5355345A Expired / Lapsed US ⏱️ 0.0 years remaining

Fully scalable memory apparatus

Assignee: Nokia Bell Labs
•
Filing: 1993-10-04
•
Granted: 1994-10-11
•
Expires: 2013-10-04
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A memory is partitioned into rows and columns of memory blocks comprised of latches, sense amplifiers, and logic circuitry that form independent pipelines through which flow a) input addresses for memory access requests and b) data to be written into a specific memory cell within a memory block. The memory allows multiple data access requests in consecutive clock cycles to be pipelined in the rows and columns of memory blocks such that the memory clock speed is equal to the clock speed of a single memory block, independently of the memory size.

Inventors

  • 👤 Alexander G. Dickinson
  • 👤 Christopher J. Nicol

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-08 21:16
CPC Classifications
G; G11; G11C; G11C8/00; G11C8/12

Claims Analysis

Total Claims: 11
Claim 1
Independent Claim 119 words
A memory, comprising: an array partitioned into rows and columns of memory blocks, each one of the memory blocks being partitioned into a plurality of rows and columns of memory cells; means for decoding addresses of the memory blocks and the memory cells into separate block column, block row, cell column, and cell row address signals; means for addressing memory cells within the array using the block and cell address signals; and means for pipelining into the blocks of the array a plurality of memory access requests in consecutive clock cycles using the decoding and addressing means to provide memory cell access such that the memory operates at a clock speed equal to a single memory block clock speed.
Claim 2
Dependent Claim 119 words
A memory, comprising: an array partitioned into rows and columns of memory blocks, each one of the memory blocks being partitioned into a plurality of rows and columns of memory cells; means for decoding addresses of the memory blocks and the memory cells into separate block column, block row, cell column, and cell row address signals; means for addressing memory cells within the array using the block and cell address signals; and means for pipelining into the blocks of the array a plurality of memory access requests in consecutive clock cycles using the decoding and addressing means to provide memory cell access such that the memory operates at a clock speed equal to a single memory block clock speed.
Claim 3
Dependent Claim 83 words
The invention of claim 1 wherein the pipelining means include means for propagating a) the block column, and cell column address signals through the columns of the array and b) the block row and cell row address signals through the rows of the array, said block and cell address signals being propagated one block per clock cycle to permit simultaneous arrival of the row and column address signals at a desired memory block identified by the block row and block column address signals.
Claim 4
Dependent Claim 27 words
The invention of claim 1 wherein address signals and data are propagated through data paths formed by latches, sense amplifiers and logic circuitry associated with the blocks.
Claim 5
Dependent Claim 16 words
The invention of claim 1 further comprising means for retrieving data from a selected memory cell.
Claim 6
Dependent Claim 31 words
The invention of claim 4 wherein said retrieving means includes means for recognizing a small voltage swing read by a sense amplifier associated with the block of the selected memory cell.
Claim 7
Dependent Claim 52 words
The invention of claim 4 further comprising means for sending data retrieved from at least one memory cell through the blocks of the array, one block per cycle, until the retrieved data reaches an output path of the memory such that pipelined memory access requests are processed in a first-in first-out order.
Claim 8
Dependent Claim 37 words
The invention of claim 1 further comprising a content addressable memory which stores addresses of defective subsections in the memory and maps the addresses of the defective subsections to different addresses of operative subsections in the memory.
Claim 9
Dependent Claim 84 words
The invention of claim 1 wherein each block in the array comprises: latches which store in addition to data, valid row and column addresses of memory cells within the block; control logic circuitry which receives input from the latches and generate control signals to a) propagate received input data and b) write received input data to a particular memory cell; and sense amplifiers which receive a column address as input from one of the latches and transfer the address to a vertically adjacent block.
Claim 10
Dependent Claim 30 words
The invention of claim 1 further comprising means for writing input data to a selected memory cell when input data are propagated with address signals of the selected memory cell.
Claim 11
Dependent Claim 27 words
The invention of claim 1 wherein said writing means further includes means for applying full voltage swings on a selected bit line of the selected memory cell.

References Cited (Backward Citations)

US5199033A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 1997-10-10 1998-04-11 1998-10-11 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2001-10-10 2002-04-11 2002-10-11 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2005-10-10 2006-04-11 2006-10-11 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag