Analog-Domain Mitigation of Outlier Noise in the Process of Analog-to-Digital Conversion

摘要

In the process of analog-to-digital (A/D) conversion the signal bandwidth is commonly reduced by linear analog filtering, which affects the temporal and amplitude structure of the incoming noise and interference. For example, reduction of the bandwidth of broadband impulsive noise makes it appear less impulsive. This diminishes the effectiveness of the subsequent digital noise removal techniques. In this paper, we focus on outlier (impulsive) noise mitigation, and perform impulsive noise filtering of the analog input signal by analog nonlinear filters ahead of the A/D conversion. This enables more effective mitigation of various types of in-band non- Gaussian noise and interference, especially that of technogenic origin, including broadband impulsive interference. In addition to improved efficiency as compared with the digital domain filtering, an important property of the presented approach is that, while being nonlinear in general, the proposed filters largely behave linearly. They exhibit nonlinear behavior only intermittently, in response to noise outliers, thus avoiding the detrimental effects, such as instabilities and intermodulation distortions, often associated with nonlinear filtering.