中文摘要：

随机共振（SR）作为一种能够利用噪声来提高系统输出信噪比的工具，近来得到越来越多的关注和研究．一个很大的问题是如何设计一个简单高效的SR系统．针对常见的微弱信号检测问题，我们分析了二值量化器和二值量化器阵列，推导出最优二值量化检测器和渐近最优阵列SR检测器，给出渐近最优阵列SR检测器的设计准则．根据理论分析结果，给出了鲁棒阵列SR检测器及基于噪声样本集合的参数选择算法，能够在未知背景噪声情形中获得较好的检测性能．

英文摘要：

Stochastic Resonance（SR） is getting more and more attention in recent few years, as it can be used as a tool of using noise to enhance output signal-to-noise-ratio. One of the key problems is how to design a simple and efficient SR system.As for the weak signal detection, we analyzed two-state quantizers and two-state quantizer arrays,deduced the optimal two-state quantization detector and asymptotic optimal array SR detector, and gave out the designing criterion of asymptotic optimal array SR detector. According to the theoretical analysis, we designed a robust array SR detector and parameter selection algorithm based on noise sam- ple collection, which can achieve good detection performance even the ambient noise is unknown.