针对多径衰落信道中幅相调制信号的类型识别问题,提出了一种基于自适应马尔可夫链蒙特卡洛(MCMC)——自适应Metropolis的改进算法.该算法利用Markov链的历史信息自动调整建议分布函数的协方差矩阵,使其不断地逼近目标分布,从而避免了传统Metropolis算法中建议分布函数选取这一难题,并能有效地提高采样效率.仿真实验表明,算法具有很好的识别精度,证明了算法的有效性. Aiming at the problem of type identification of amplitude and phase modulated signals in multi-path fading channels, an improved MCMC-adaptive Metropolis algorithm based on adaptive Markov chain is proposed. The algorithm uses the history information of Markov chain automatically The covariance matrix of the proposed distribution function is adjusted to keep it close to the target distribution so as to avoid the problem of selecting the recommended distribution function in the traditional Metropolis algorithm and to effectively improve the sampling efficiency.The simulation results show that the algorithm has good recognition Accuracy, proved the validity of the algorithm.