Paper Details: Downloads: 566
Serial Number: P1180625004
Title: A Novel Reconfigurable Processor for Multi way Circuit partitioning Using Genetic Algorithm
Authors: Rajaram Sivasubramanian and Dhinesh Joseph and Abhaikumar Varadhan
Abstract: Genetic Algorithms (GAs) are robust techniques based on natural selection that can be used to solve a wide range of problems including VLSI layout optimization, Circuit partitioning and Boolean satisfiability. This paper proposes ‘CIRPART’ – a novel architecture for implementing Genetic Algorithm (GA) used for circuit Multiway Partitioning in VLSI physical design automation. CIRPART is built upon ‘user programmable’, Field Programmable Gate Array (FPGA) device and employs a combination of Pipelining and Parallelization. CIRPART provides flexibility and also achieves speedups over software based GA. The design uses four modules and three external memories. The proposed design was coded in VHDL and was functionally verified by writing a test bench and simulating it using ModelSim. The design was synthesized on Xilinx Virtex V100CS144 FPGA chip using VHDL. CIRPART achieves more than 100% improvement in processing speed as compared to the software implementation while the quality of results remains unaltered.
Keywords: Genetic algorithm, Field Programmable Gate array, Multi-way circuit partitioning, VHDL.
Journal/Conference: International Journal of Digital Signal Processing
Volume: 6
Issue: 1
Submission Date: 7/1/2006 12:00:00 AM
Review Date: 8/1/2006 12:00:00 AM
Publishing Date: 9/1/2006 12:00:00 AM
Article Downloads: 566
Download:

Facebook