Author:
Kattekola Naresh,Majumdar Shubhankar
Abstract
Purpose
This paper aims to implement a novel design of approximate comparator which can be suitable for image processing applications.
Design/methodology/approach
Here, the N-bit approximate comparator is implemented by taking reference of N as 2-, 4- and 8-bit. The design analyses the fractional change in error to bit in several bit formats. The final implementation of approximate comparator design application compares the structural similarity index, colour test and extraction of an image to the results.
Findings
The novel approximate comparator was designed using 2-, 4- and 8-bit to explore N-bit comparator expressions. The implementation, computations, evaluation of errors, applications and the design constraints were executed using Python and Synopsys, respectively. The computations, evaluation of errors, applications and the design constraints were executed using Python and Synopsys, respectively.
Originality/value
This paper presents the N-bit accurate and approximate comparator which is novel over the existing design of comparators.
Subject
Electrical and Electronic Engineering,Industrial and Manufacturing Engineering
Reference23 articles.
1. Low-power approximate multipliers using encoded partial products and approximate compressors;IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2018
2. Development of approximate compressor based hybrid dadda multiplier for image de-noising applications,2019
3. Approximate adder with reduced error,2019
4. Systematic design of an approximate adder: the optimizer lower part constant-OR adder;IEEE Transactions on Very Large Scale Integration (VLSI) Systems,2018
5. Approximate multipliers based on new approximate compressors;IEEE Transactions on Circuits and Systems I: Regular Papers,2018