Konferensartikel

Verifying an Implementation of Genetic Algorithm on FPGA-SoC using SystemVerilog

Hayder Al-Hakeem
Department of Electrical and Energy Engineering, University of Vaasa, Finland

Suvi Karhu
Department of Electrical and Energy Engineering, University of Vaasa, Finland

Jarmo T. Alander
Department of Electrical and Energy Engineering, University of Vaasa, Finland

Ladda ner artikelhttp://dx.doi.org/10.3384/ecp171421095

Ingår i: Proceedings of The 9th EUROSIM Congress on Modelling and Simulation, EUROSIM 2016, The 57th SIMS Conference on Simulation and Modelling SIMS 2016

Linköping Electronic Conference Proceedings 142:161, s. 1095-1101

Visa mer +

Publicerad: 2018-12-19

ISBN: 978-91-7685-399-3

ISSN: 1650-3686 (tryckt), 1650-3740 (online)

Abstract

In this paper we show how an ef?cient implementation of genetic algorithms can be done on Field Programmable Gate Array i.e. on programmable hardware using the latest hardware design language aiding veri?cation. A four-way number partitioning problem of 128 unsigned 16-bit integers is used as a test case of the implementation. However, other similar problems could be solved using the proposed approach. The design was implemented using a combination of reusable veri?ed intellectual property cores for arithmetic operations and VHDL to describe the genetic algorithm operators in register transfer level. The register transfer level components were veri?ed in ModelSim using SystemVerilog assertions and cover groups. Test results show signi?cant improvements in performance compared to C language implementation running on a core i-7 desktop computer.

Nyckelord

genetic algorithms, veri?cation, FPGA, System on Chip (SoC)

Referenser

Inga referenser tillgängliga

Citeringar i Crossref