论文部分内容阅读
水声通信信道带宽非常有限。与自适应均衡算法相比,盲均衡算法不需要训练序列而节省了带宽,特别适合于高速水声通信和多点通信。Bussgang 类盲均衡算法计算简单,但收敛速度慢,且对权向量的初始化很敏感。不恰当的初始化会造成算法收敛到局域极小值。多谱类盲均衡算法收敛速度快,且是全局最优的,但计算量大,不利于硬件实现。本文利用倒三谱,根据倒三谱累积量方程,由一般数据计算得到倒谱参数,从而直接估计出均衡器参数,对Bussgang类算法进行初始化,大大加快了算法的收敛速度,保证了算法收敛到最优。通过对浅海信道均衡的计算机仿真,证明了该方法的有效性。
Acoustic communication channel bandwidth is very limited. Compared with the adaptive equalization algorithm, the blind equalization algorithm saves the bandwidth without the training sequence and is especially suitable for high speed underwater acoustic communication and multi-point communication. The Bussgang-type blind equalization algorithm is simple to calculate, but its convergence speed is slow, and it is sensitive to the initialization of weight vector. Improper initialization can cause the algorithm to converge to a local minima. The multispectral blind equalization algorithm has the advantages of fast convergence and global optimization, but its computational complexity is not good for hardware implementation. In this paper, the inverted spectrum is used to calculate the cepstrum parameters from the general data according to the inverse cumulative spectrum equation, which directly estimates the equalizer parameters and initializes the Bussgang class algorithm, which greatly accelerates the convergence rate of the algorithm and ensures convergence of the algorithm To the best. Through the computer simulation of shallow channel channel equalization, the effectiveness of the method is proved.