An analog-to-digital conversion circuit for complementary code quantization of rram storage and calculation integrated chip

Abstract

The present invention relates to an analog-to-digital conversion circuit, in particular to an analog-to-digital conversion circuit for complementary code quantization of an RRAM storage and calculation integrated chip, including a current and voltage conversion module, eight sampling switches, two asymmetric capacitor arrays, a comparator device and logic control module, in which the current-voltage conversion module converts the output current of the RRAM array into a voltage, and connects to an asymmetric capacitor array through a sampling switch, and the remaining seven sampling switches are connected to another asymmetric capacitor array, two asymmetric The capacitor array is connected to the two input terminals of the comparator, and the output terminal of the comparator is connected to the logic control module. The logic control module outputs the comparator control clock and the capacitor array control signal, and outputs the quantization result. Through this new complement quantization analog-to-digital converter, the problem of complement quantization of multi-bit weights in the RRAM storage and calculation integrated chip used in array multiplication and addition operations can be solved, its operation speed and parallelism can be improved, and the chip area can be saved.

Description

technical field [0001] The invention relates to an analog-to-digital conversion circuit, in particular to an analog-to-digital conversion circuit used for complementary code quantization of an RRAM memory-computing integrated chip. Background technique [0002] The current mainstream computer architecture is facing the von Neumann bottleneck due to the separation of storage and computing modules. The in-memory computing technology based on RRAM can realize the integration of storage unit and logic unit, break through the von Neumann bottleneck, and promote the development of artificial intelligence and integrated circuit technology. When the RRAM-based memory-computing integrated chip is used in artificial intelligence applications, the most critical link is the multiplication and addition operation of the array, that is, the multiplication and addition operation of the input signal and the storage weight, and the readout of the result is quantization, so the design is reaso...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides an analog-to-digital conversion circuit for complementary code quantization of an RRAM integrated chip to solve the problem of complementary code quantization of RRAM arrays in multiplication and addition operations. A PN-DAC asymmetrical SAR ADC structure is adopted, wherein P-DAC receives the sign bit output in the RRAM array, N-DAC receives the digital bit output in the RRAM array, and adjusts the weight of each bit through charge redistribution, and finally realizes the complement quantization of the output result of the RRAM array

Images