Structures and methods for data reading apparatus and reading out non-volatile memory using referencing cells

Abstract

The structures and methods of a data reading apparatus and reading out semiconductor Non-Volatile Memory (NVM) using referencing cells are disclosed. The data reading apparatus comprises: -a sensing amplifier, which is used for sensing the sensing amplifier. -a first carry-in terminal and -a second carry-in terminal. The voltage difference between the first and second carry-in terminal. -a first capacitor and a second capacitor, connected with the first and second carry-in terminal respectively. The first capacitor and the second capacitor make an electrical discharge by reading a NVM unit and referencing a NVM unit when a grid bias is inflicted across control grids of a reading NVM unit and a referencing NVM unit simultaneously. And capacitance value of the first capacitor and the second capacitor is same materially. The new invented scheme can reduce large current consumption from the direct current biasing in the conventional scheme.

Description

technical field [0001] The present invention relates to an integrated circuit (integrated circuit) for sensing information stored in a semiconductor non-volatile memory (non-volatile memory, NVM), in particular to an integrated circuit for sensing non-volatile memory using a reference memory cell (referencing memory cell). A circuit for storing information of a volatile memory and an operation method thereof, in particular to a data reading device, a nonvolatile memory device and a reading method thereof. Background technique [0002] Semiconductor non-volatile memory (NVM) and especially electrically erasable programmable read-only memory (Electrically Erasable, Programmable Read-Only Memories, EEPROM) are widely used in the field of electronic equipment (equipment), from computers, electronic communications Hardware to consumer appliances. The way EEPROM cells store data is by injecting charge carriers into the charge storage layer (charge storage) above the channel regio...

AI Technical Summary

Problems solved by technology

[0004] The traditional current sensing architecture has a disadvantage: two currents must pass through two memory cells M rf and M C And the two currents are amplified by two current mirror amplifiers 120 to maintain a stable voltage state at the two input terminals of the differential voltage sense amplifier 110

The traditional current sensing architecture has the disadvantage of high power consumption due to the DC current path generated by the pull-up element 130 to the NVM cell and is mostly a DC current path generated by an amplified mirrored current.

In practical applications, the disadvantage of high power consumption of conventional current sensing circuits imposes a critical limitation on semiconductor NVM circuit design when a large number of NVM cells are read in parallel.

Images