基于混沌神经网络的盲检测改进新算法

Novel improved blind detection algorithms based on chaotic neural networks

针对Hopfield神经网络的多起点问题,提出了一种新的基于混沌神经网络的盲信号检测算法,实现了二进制移相键控信号盲检测.据此进一步提出双sigmoid混沌神经网络模型,构造了新的能量函数,且证明了该模型的稳定性,并对网络参数进行配置.仿真实验表明:混沌神经网络能够避免局部极小点且具备较强的抗噪性能,双sigmoid混沌神经网络则继承了其所有的优点,且其收敛速度更快,仅需更短的接收数据即可到达全局真实平衡点,从而降低了算法的计算复杂度,减少了运行时间.

In this paper we apply the transiently chaotic Hopfield neural networks （TCHNN） to the blind signal detection algorithm with BPSK signals and solve multi-start problem of Hopfield neural networks （HNN）. And in this paper we propose an improved algorithm of double sigmoid transiently chaotic Hopfield neural networks （DS-TCHNN） on the basis of TCHNN, construct a new energy function of DS-TCHNN, and prove the stability of DS-TCHNN in asynchronous update mode and synchronous update mode. Simulation results show that TCHNN can skip local minima and has better anti-noise performance than HNN. While, DS-TCHNN inherits all the advantages of TCHNN and its speed of convergence is fast. Besides, DS-TCHNN needs shorter data to reach a global true equilibrium point so that the computational complexity is reduced and the running time is shortened.