提出了一种用于认知无线电线性加权协作频谱感知的改进混合蛙跳算法(shuffled frog leaping algorithm,SFLA)的群体初始化技术,提出在SFLA初始群体中包含基于修正偏差因子所得的解,从而改进算法初期性能.仿真结果表明相比于传统群体初始化技术,本文所提出的群体初始化技术能够以更快的速率得到期望解,从而节约计算时间,更有利于实时应用 A group initialization technique for improved shuffled frog leaping algorithm (SFLA) for cognitive radio-linear weighted cooperative spectrum sensing is proposed, and a solution based on a modified bias factor is included in the initial population of SFLA, so as to improve The initial performance of the algorithm.The simulation results show that compared with the traditional group initialization technology, the group initialization technology proposed in this paper can get the expected solution at a faster rate, which can save computation time and is more beneficial to real-time applications