自适应频域块LMS算法在抵消干扰方面的研究

Adaptive Frequency Block LMS Algorithm Research in Aspect of Interference Cancellation

分析了无线通信系统中自适应滤波算法中的LMS算法,并提出了利用自适应抵消输入信号中的干扰而获取有用信号的在频域实现的干扰抵消方法,该方法基于时域中的数据块最小均方误差LMS算法和该块LMS算法中存在线性相关和线性卷积的过程,通过1/2重叠保留法的快速傅立叶变换FFT,在频域以直接相乘的计算方式实现快速相关和快速卷积,利用自适应滤波器在频域实现LMS算法。通过实验仿真比较可以看出,该频域块LMS算法不仅保证了与时域自适应滤波算法有相同的收敛性,而且由于利用了快速FFT技术,使运算量大大减少。

After analyzing LMS algorithms of the adaptive filtering algorithms in wireless telecommunication,the interference cancellation method was proposed which can gain the useful signals in frequency by using adaptive to counteract the interference from the input signals. This algorithm is based on least mean square(LMS) adaptive filtering algorithm and the processes of linear correlation and linear convolution of the block LMS.By using 1/2 overlap-save method fast fourier transform(FFT),realized the fast correlation and fast convolution by the calculation method that multiply directly in frequency domain.It can be concluded through comparing simulation results that the frequency block LMS algorithm not only has the same astringency with time - domain adaptive filtering algorithms,but also can reduce computation greatly by the use of fast FFT technology.