采用余弦调制滤波器组的多径衰落信号子带合成

Sub-band Combining for Multipath Fading Signals Using Cosine-modulated Filter Banks

本文关注的是多路多径衰落信号的最佳合成问题。分集技术是对抗多径衰落的有效方法,其中采用最大比合并的方式是最佳的分集合成方案。但是传统分集技术中多径信号合成采用的是实数合成权值,不能使合成信号所有频率范围内的信号分量均取得最大信噪比。本文根据多径衰落信号的最佳合成结构,提出了一种子带合成算法。该算法采用余弦调制滤波器组将宽带衰落信号分为若干带宽较小的子带信号。通过对应子带的最佳合成与合成信号的完美重构,使得合成输出信号具有最佳的信噪比。仿真结果表明通过该子带合成算法可以有效提高合成信号的解调质量,降低解调的误符号率。

This paper focuses on the optimal combining of multiple multi-path fading signals. Diversity is a classical technology combating the multi-path fading. The maximum ratio combining is the optimal combining schemes. Real weights are used for multi-path signals combining in the traditional diversity system. This may not make all frequencies in the spec- trum of combined signal to achieve maximum signal-to-noise ratio. Based on the best combining structure of multi-path fa- ding signals, a sub-band combining algorithm is proposed in this paper. The cosine modulated filter banks are used to di- vide the width band fading signals into several narrowband sub-signals. Optimal combining of corresponding sub-band sig- nals and perfect reconstruction techniques make the optimal combined output signal. Simulation results show that the sub- band combining algorithm can effectively improve the quality of the demodulation of the combined signal. And it evidently reduces the symbol error rate of combined signal.