4-2 approximate compressor and approximate multiplier

Abstract

An approximate 4-2 compressor and an approximate multiplier belong to the field of integrated circuit technology. The approximate 4-2 compressor includes a first OR gate, a second OR gate, a third ORgate, a first AND gate, a second AND gate, and a first AND gate and a first input of the first AND gate The end is a first input of the approximate 4-2 compressor, and their second input is connectedas a second input of the approximate 4-2 compressor, and their outputs are connected to the two inputs of the third AND gate; The first input of the second or second AND gate is used as the third input of the approximately 4-2 compressor, and their second input is the fourth input of the approximately 4-2 compressor; the third or the gate The input ends are respectively connected to the output ends of the first AND gate, the second AND gate and the third AND gate, and the output end thereof is an output terminal of the approximate 4-2 compressor. The approximate compression module of the approximate multiplier adopts the approximate 4-2 compressor proposed by the present invention, and is supplemented by an error correction unit, which has the characteristics of high precision, low power consumption and low delay.

Description

technical field [0001] The invention belongs to the technical field of integrated circuits, and relates to an approximate 4-2 compressor and an approximate multiplier designed based on the approximate 4-2 compressor. Background technique [0002] The vast majority of computing units in modern computer systems are composed of high-precision and high-reliability digital integrated circuits, but in many current applications, such as multimedia processing, image processing, neural networks, Internet of Things, etc., there are certain In the case of errors, the system can still obtain meaningful calculation results. Therefore, the precise calculation unit is not so suitable and effective for such an arithmetic circuit. Thus, approximate computing came into being. The approximate circuit sacrifices a certain degree of accuracy to reduce circuit complexity, circuit delay and power consumption. Since the development of integrated circuits, in nanometer-scale digital signal proces...

AI Technical Summary

Problems solved by technology

The output result of the traditional precision multiplier is completely correct, but its circuit consumes a lot of hardware resources, and has high delay and power consumption

The perfection of precision consumes a lot of resources

And this is not worth it in some application scenarios where a certain error can be tolerated

Images