Fast Implementation of Genetic Algorithm Based on Software/Hardware Co-design Method

Genetic algorithm (GA) has been widely used since it was proposed. However, due to the complexity of itself, fact that actual problems have huge data and scale, needs meet certain real-time requirements, acceleration genetic become a hot spot in both academia industry field. Previous work done on GA, which primarily focuses software aspect, while existing hardware aspect lacks adaptability for different industrial scenarios. Therefore, this paper puts forward novel architecture based software/hardware co-design method, produces random numbers, hardware, taking full account characteristics sequential circuit, is implemented hierarchical pipeline architecture. This accelerates providing flexibility. completes simulation MATLAB Vivado, total running time 987.4.u. Compared with pure implementation, speed increased by 73.9 times.