Binary data memory design with data stored in low-power sense

• US 6,507,887 B1
• Filed: 04/18/2000
• Issued: 01/14/2003
• Est. Priority Date: 01/13/1998
• Status: Expired due to Term

• Current Assignee: Callahan Cellular LLC (Intellectual Ventures LLC)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Ehmann, Gregory E., Pontius, Timothy A.
• Primary Examiner(s): Hudspeth, David
• Assistant Examiner(s): TZENG, FRED
• Application Number:
  US09/551,087
• Time in Patent Office: 1,001 Days
• Field of Search: 711/103, 711/102, 711/104, 711/170, 711/105, 326/40, 365/105, 365/49, 365/185.21, 365/203, 365/230.06, 365/233.5
• US Class Current: 711/102
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: G06F 1/3275 Power saving in memory, e.g. RAM, cache
• CPC Class Codes: G11C 17/12 using field-effect devices
• CPC Class Codes: G11C 2207/2227 Standby or low power modes
• CPC Class Codes: G11C 7/06 Sense amplifiers; Associated circuits , e.g. timing or triggering circuits
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method of designing a mask-programmable random-access read-only memory device begins with a step of assigning weightings to addresses according to their expected frequency of access. These weighting are used in a second step of determining for each sense amplifier, what is the low-power sense (inverted or uninverted) of the stored bits using that sense amplifier as an output. The third step involves storing the data in the low-power sense. The fourth step involves inverting the outputs for the data that is stored inverted. This can involve using sense inverting sense amplifiers for inverted data and sense preserving amplifiers for uninverted data. The method can result in memories in which some outputs are sense inverting while others are sense preserving. The result is a memory device with reduced power consumption relative to a comparable design not taking advantage of the relationship between data values and power consumption.

Claims

1. A read-only random-access memory design method comprising:

determining a low-power sense for binary data having bit values;

storing said data in said low-power sense in a row-addressable memory array having rows and columns, each of said bit values being stored at a respective row and a respective column, said data being stored so that all bit values stored in a first column are inverted and all bit values stored in a second column are not inverted; and

inverting data output from one of said first and second columns and not inverting data output from the other of said columns.

2. A method as recited in claim 1 wherein said determining and storing steps are performed for each column in dependently of the other columns.

3. A method as recited in claim 1 further comprising a preliminary step of weighting rows of said memory, said determining step applying the weightings assigned in said preliminary step to determine which of said columns is to have inverted bit values.

4. A method as recited in claim 3 wherein said preliminary step involves weighting addresses as a function of expected frequency of access.

5. A random access memory resulting from the method of claim 1.

6. A read-only random access memory comprising:

plural bit-value storage locations arranged in an array with rows and columns, each of said rows having an address;

an address input for receiving said addresses; and

a data output for outputting bit values stored in a row when its address is received by said address input, said data output including means for inverting some but not all of the bit values of said row.

7. A random access memory as recited in claim 6 further comprising bias means for pre-biasing said data output to the sense corresponding to a majority of the bit values stored in said array.

8. A random access memory as recited in claim 6 wherein said rows have associated weightings.

9. A random access memory as recited in claim 8 further comprising pre-charge means for pre-biasing said data output to the sense corresponding to a weighted majority of the bit values stored in said array, said weighted majority being a function in part of said weightings.

10. A random access memory as recited in claim 6 wherein said data output includes programmable sense amplifiers.