一种基于高阶累积量的回波对消方法

Higher Order Cumulants in Echo Cancellation

提出一种基于高阶累积量的回波对消新方法，以期在信噪比较低的情况下，优于基于自适应滤波技术的回波对消方案．该方法的实质是高阶累积量能够自动抑制高斯有色或白色噪声，先用回波数据来计算其４阶累积量，再利用部分累积量估计回波通路的滤波器系数，最后利用可能得到的全部累积量来对滤波器系数进行细估计．在信噪比分别为２０、１５、１０、５ｄＢ的条件下，各进行１０次随机试验，仿真结果显示：在不高于１５ｄＢ的低信噪比情况下。

Echo cancellation (EC) is fundamental to full duplex data transmission. It has been used as adaptive filtering techniques. The method is not effective when the signal to noise ratio is low. An improved method developed in this paper considers higher order cumulants (HOC) that can suppress additive Gaussion noise (AGN) automatically. The two step procedure of Mo and Shafai is used to calculate the cumulants of echo and reconstruct finite impulse response (FIR) of the echo path. Comupter simulation shows that the present scheme is superior to the least mean square (LMS) algorithm for the FIR structure when the signal to noise ratio is no more than 15 dB.