Multi-objective q-bit coding genetic algorithm for hardware-software co-synthesis of embedded systems

  • Authors:

  • Wei Wen-long;Li Bin;Zou Yi;Zhuang Zhen-quan

  • Affiliations:

  • Nature Inspired Computation and Applications Laboratory, University of Science and Technology of China, Hefei, China;Nature Inspired Computation and Applications Laboratory, University of Science and Technology of China, Hefei, China;Nature Inspired Computation and Applications Laboratory, University of Science and Technology of China, Hefei, China;Nature Inspired Computation and Applications Laboratory, University of Science and Technology of China, Hefei, China

  • Venue:
  • SEAL'06 Proceedings of the 6th international conference on Simulated Evolution And Learning
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

One of the key tasks in Hardware-Software Co-design is to optimally allocate, assign, and schedule resources to achieve a good balance among performance, cost, power consumption, etc. So it's a typical multi-objective optimization problem. In this paper, a Multi-objective Q-bit coding genetic algorithm (MoQGA) is proposed to solve HW-SW co-synthesis problem in HW-SW co-design of embedded systems. The algorithm utilizes the Q-bit probability representation to model the promising area of solution space, uses multiple Q-bit models to perform search in a parallel manner, uses modified Q-bit updating strategy and quantum crossover operator to implement the efficient global search, uses an archive to preserve and select pareto optima, uses Timed Task Graph to describe the system functions, introduces multi-PRI scheduling strategy and PE slot-filling strategy to improve the time performance of system. Experimental results show that the proposed algorithm can solve the multi-objective co-synthesis problem effectively and efficiently.