scispace - formally typeset
Proceedings ArticleDOI

Low power divider using vedic mathematics

Dalal Rutwik Kishor, +1 more
- pp 575-580
TLDR
The validated designs were implemented using industry standard Cadence® software using 45nm technology library and show appreciable reduction in circuit latency and in Look-Up-Table (LUT) utilization using proposed Vedic divider than the conventional divider.
Abstract
Divider is an inevitable and basic hardware module employed in advanced and high speed digital signal processing (DSP) units of high precision. It is widely used in radar technology, communication, industrial control systems and linear predictive coding (LPC) algorithms in speech processing. This paper proposes a fast, low power and cost effective architecture of a divider using the ancient Indian Vedic division algorithm. The merits of the proposed architecture are proved by comparing the gate count, power consumption and delay against the conventional divider architectures. The validation of the proposed architecture has resulted in 52.93 percent reduction in power dissipation against comparison with conventional divider using repeated subtraction. The designs were implemented using industry standard Cadence® software using 45nm technology library. The design has been validated on FPGA Spartan-3E kit. The validation results show appreciable reduction in circuit latency and in Look-Up-Table (LUT) utilization using proposed Vedic divider than the conventional divider.

read more

Citations
More filters
Proceedings ArticleDOI

Design of Approximate Dividers for Error Tolerant Applications

TL;DR: This paper explores architectural designs of approximate unsigned integer dividers based on restoring and non-restoring algorithms as the subtractor cell is the basic building of a divider and the approximate restoring dividers are performing superior to non-Restoring dividers both in-terms of design parameters and error metrics.
Proceedings ArticleDOI

Design Of High Performance Digital Divider

TL;DR: Computation using Vedic mathematics decreases considerable extent of iterations resulted in minimized delay and power compared to the mostly used non – Vedic architectures.
Proceedings ArticleDOI

Division operation based on Vedic mathematics

TL;DR: The proposed work focuses on division operation which is an important operation in areas such as image processing, networking, signal processing, computer graphics, numerical application, scientific applications and in processor implementation.
Proceedings ArticleDOI

Low latency divider using ensemble of moving average curves

TL;DR: This paper proposes a fixed-point divider based on an ensemble of moving average (EnMA) curves that guaranteed 13-bit precision in division and can be effectively used for image processing applications.
Journal ArticleDOI

Performance analysis of Vedic mathematics algorithms on re-configurable hardware platform

TL;DR: It has been observed that the Vedic algorithms operate faster, consume less power, and occupy less area on a targeted hardware platform.
References
More filters
Journal ArticleDOI

Fast division using accurate quotient approximations to reduce the number of iterations

TL;DR: A class of iterative integer division algorithms is presented based on look-up table and Taylor-series approximations to the reciprocal, which naturally produce an exact remainder, which is very useful for implementing precise rounding specifications.
Proceedings ArticleDOI

Novel binary divider architecture for high speed VLSI applications

TL;DR: Novel Binary divider architecture for high speed VLSI application using such ancient methodology using Boolean logic with ancient Vedic mathematics resulted in substantial amount of iteration being eliminated that resulted in ~46% reduction in delay and ~27% reduced in power compared with the mostly used (Repetitive subtraction method).
Journal ArticleDOI

Design of a fast radix-4 SRT divider and its VLSI implementation

TL;DR: This paper presents a fast radix-4 SRT division architecture that takes 247 ns for a double precision division (56 bits for fraction part), where the 2 /spl mu/m CMOS technology in MAGIC is employed and simulated.
Proceedings ArticleDOI

Revisiting SRT quotient digit selection

TL;DR: This work presents expressions for the number of bits needed for the truncated remainder and divisor, thus eliminating the need for a search through the truncation parameter space for validation, and presents simple algorithms to properly map truncated negative divisors and remainders into nonnegative values.
Related Papers (5)