Abstract
This paper presents a novel distributed genetic algorithm (GA) architecture for the design of vector quantizers. The design is based on a multi-core architecture, where each island of the GA is associated with a hardware accelerator and a softcore processor for independent genetic evolutions. An on-chip RAM with a mutex circuit is adopted for the migration of genetic strings among different islands. This allows a simple and flexible migration for the implementation of hardware distributed GA. Experimental results shows that the proposed architecture has significantly lower computational time as compared with its software counterparts running on multicore processors with multithreading for GA-based optimization.
Highlights
Genetic algorithms (GAs) [1] are a class of general-purpose search algorithms for solving optimization problems by simulating natural evolution over populations of candidate solutions
The objective of this paper is to present a VLSI architecture for distributed GA
As compared with its software counterparts running on multicore processors with multithreading, numerical results reveal that the proposed field programmable gate array (FPGA)-based GA architecture attains higher performance with significantly lower training time for vector quantization (VQ) design
Summary
Genetic algorithms (GAs) [1] are a class of general-purpose search algorithms for solving optimization problems by simulating natural evolution over populations of candidate solutions. The algorithms have been found to be effective for solving problems in engineering, science and business. When they are applied to complex problems, the computational complexity may become very high. The distributed GA may be able to converge at faster rate while finding good solutions. To find near optimal solutions, large population size may still be desired. This will still result in long computation time
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have