计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (33): 54-57.DOI: 10.3778/j.issn.1002-8331.2010.33.015

• 研发、设计、测试 • 上一篇    下一篇

群智能算法可并行性分析及其FPGA实现

许 林,须文波,柴志雷   

  1. 江南大学 信息工程学院,江苏 无锡 214122
  • 收稿日期:2010-05-19 修回日期:2010-07-27 出版日期:2010-11-21 发布日期:2010-11-21
  • 通讯作者: 许 林

Parallel analysis of swarm intelligence algorithm and FPGA implementation

XU Lin,XU Wen-bo,CHAI Zhi-lei   

  1. School of Information Engineering,Jiangnan University,Wuxi,Jiangsu 214122,China
  • Received:2010-05-19 Revised:2010-07-27 Online:2010-11-21 Published:2010-11-21
  • Contact: XU Lin

摘要: 分析了量子行为的粒子群QPSO算法和粒子间相互协作的CQPSO算法结构的可并行性,并结合FPGA技术可并行处理信息的特点,说明了在并行运算模式下粒子的收敛性能。实验验证了QPSO和CQPSO算法的可并行性,并得到粒子收敛的相关数据,数据表明CQPSO算法粒子的收敛精度要远优于QPSO算法,但是粒子的收敛速度上面要远低于QPSO算法。

关键词: 现场可编程门阵列(FPGA), 可并行性分析, 收敛速度, 收敛精度

Abstract: The structure about the Quantum-behaved PSO algorithm QPSO and an improved hybrid QPSO algorithm with cooperative method between particles CQPSO,whose structure can be parallel is analysed.Then combing with FPGA technology characters which can be parallel processing of information,indicates the convergence of particles performance in parallel operation mode.Experiments verify the QPSO and CQPSO algorithm which can be parallel and get the related data about the convergence performance of the particles.The data show that CQPSO algorithm is far superior to QPSO algorithm in the convergence accuracy of particles,but CQPSO algorithm is far lower than the QPSO algorithm in the convergence speed of the particles.

Key words: Field-Programmable Gate Array(FPGA), parallel analysis, convergence speed, convergence accuracy

中图分类号: