Semi-Supervised Learning with Generative Adversarial Networks on Digital Signal Modulation Classification

Deep Learning(DL)is such a powerful tool that we have seen tremendous success in areas such as Computer Vision,Speech Recognition,and Natural Language Processing.Since Automated Modulation Classification(AMC)is an important part in Cognitive Radio Networks,we try to explore its potential in solving signal modulation recognition problem.It cannot be overlooked that DL model is a complex model,thus making them prone to over-fitting.DL model requires many training data to combat with over-fitting,but adding high quality labels to training data manually is not always cheap and accessible,especially in real-time system,which may counter unprecedented data in dataset.Semi-supervised Learning is a way to exploit unlabeled data effectively to reduce over-fitting in DL.In this paper,we extend Generative Adversarial Networks(GANs)to the semi-supervised learning will show it is a method can be used to create a more dataefficient classifier.

Authentic compound-free strategy for simultaneous determination of primary coumarins in Peucedani Radix using offline high performance liquid chromatography–nuclear magnetic resonance spectroscopy–tandem mass spectrometry

Herein, a strategy is proposed for the simultaneous determination of primary coumarins in Peucedani Radix(Chinese name: Qianhu). The methodology consists of three consecutive steps: 1) Semi-preparative LC in combination with a home-made automated fraction collection module to fragment the universal metabolome standard into ten fractions(Frs. I–X); 2) LC–accurate MS/MS and quantitative1 H NMR spectroscopy conducted in parallel to acquire the qualitative and quantitative data of each fraction; 3) Robust identification and quantification of components by use of LC coupled to multiple reaction monitoring. In this final step, the most significant fractions(Frs. III–X) were pooled to serve as the pseudo-mixed standard solution. Meticulous online parameter optimization was performed to obtain the optimal parameters, including ion transitions and collision energies. Concerns were particularly paid onto pursuing the parameters being capable of monitoring regiospecific isomers, notably praeruptorin E vs. 3′-isovaleryl-4′-angeloylkhellactone. The quantitative performance of the method was validated according to diverse assays. Eleven primary coumarins(1–11) were unambiguously identified and absolutely quantified, even though no external reference compound was used. Above all, the integrated strategy not only provides a feasible pipeline for the quality assessment of Peucedani Radix, but more importantly, shows the potential for authentic compound-free quantitative evaluation of traditional Chinese medicines.