Abstract:Being complex-valued and low in signal-to-noise ratios, magnitude-based diffusion MRI is confounded by the noise-floor that falsely elevates signal magnitude and incurs bias to the commonly used diffusion indices, such as fractional anisotropy (FA). To avoid noise-floor, most existing phase correction methods explore improving filters to estimate the noise-free background phase. In this work, after diving into the phase correction procedures, we argue that even a perfect filter is insufficient for phase correction because the correction procedures are incapable of distinguishing sign-symbols of noise, resulting in artifacts (\textit{i.e.}, arbitrary signal loss). With this insight, we generalize the definition of noise-floor to a complex polar coordinate system and propose a calibration procedure that could conveniently distinguish noise sign symbols. The calibration procedure is conceptually simple and easy to implement without relying on any external technique while keeping distinctly effective.
Abstract:The increasing success of Machine Learning (ML) and Deep Learning (DL) has recently re-sparked interest towards traffic classification. While classification of known traffic is a well investigated subject with supervised classification tools (such as ML and DL models) are known to provide satisfactory performance, detection of unknown (or zero-day) traffic is more challenging and typically handled by unsupervised techniques (such as clustering algorithms). In this paper, we share our experience on a commercial-grade DL traffic classification engine that is able to (i) identify known applications from encrypted traffic, as well as (ii) handle unknown zero-day applications. In particular, our contribution for (i) is to perform a thorough assessment of state of the art traffic classifiers in commercial-grade settings comprising few thousands of very fine grained application labels, as opposite to the few tens of classes generally targeted in academic evaluations. Additionally, we contribute to the problem of (ii) detection of zero-day applications by proposing a novel technique, tailored for DL models, that is significantly more accurate and light-weight than the state of the art. Summarizing our main findings, we gather that (i) while ML and DL models are both equally able to provide satisfactory solution for classification of known traffic, however (ii) the non-linear feature extraction process of the DL backbone provides sizeable advantages for the detection of unknown classes.