自适应扰动暂态混沌神经网络MPSK盲检测

Blind Detection for MPSK with adaptive perturbation Transiently Chaotic Neural Network

摘要:

为提高MPSK信号盲检测算法的性能,针对CHNN_APHM算法易陷入局部最优的缺点,本文引入暂态混沌神经网络模型,使用新的模拟退火策略,加入扰动因子和混沌,提出带扰动的幅值相位型离散幅值多电平暂态混沌神经网络算法。算法使用暂态混沌神经网络提高抗噪性能,并在起始时刻使用混沌初始化获得原始信号,选取与发送信号相关性高的微小扰动因子使算法跳出局部最优解。实验仿真结果证明,带扰动的幅值相位型离散幅值多电平暂态混沌神经网络MPSK信号盲检测算法需要较少的起点个数,能在更小的信噪比和更短的数据长度下收敛,有效提高了抗干扰性能。

Abstract:

In order to improve the performance of blind detection algorithm for MPSK “multiple phase shift keying(MPSK)”, a chaotic neural network model is introduced with a new annealing model, and the perturbation factor and chaos are added. Disturbed Complex Transiently Chaotic Neural Network with Amplitude-Phase-type Hard-Multistate-activation-function is proposed. The algorithm uses transient chaotic neural network to improve the anti-noise performance, and the chaotic initialization to obtain the original signal at the initial moment, and selects the small perturbation factor with high correlation with the transmitted signal to make the algorithm jump out of the local optimal solution. Experimental simulation results show that the MPSK blind detection algorithm with perturbation amplitude phase-phase discrete amplitude multi-level transient chaotic neural network requires fewer starting points, can converge under smaller signal-to-noise ratio and shorter data length.