摘要
提出了利用基于BP(Back Propagation)神经网络的遗传算法来设计FIR数字滤波器的方法。针对遗传算法很难实现全局最优和搜索速度比较慢的缺陷,提出了改进算法,该算法充分利用了遗传算法的全局搜索功能强和BP神经网络的搜索效率高,优化了搜索时间,提高了算法性能,对于解决大规模多极值优化问题特别有效。最后,以设计低通滤波器的实例验证算法的可行性。
FIR filter is designed by the genetic algorithm based on BP neural network.Aimed at the difficulty of global optimization and the slow computational rate,an improved method is given.The algorithm takes advantage of the global search capabilities of genetic algorithm and the strong search efficiency of BP neural network.It optimizes the search time and improves the performance of the algorithm.It is particularly effective for solving large-scale optimization problems.The example of low-pass filter designed by the improved algorithm shows its feasibility.
出处
《价值工程》
2011年第17期37-38,共2页
Value Engineering
