Low SNR multimirror Fabry-Perot pressure sensor optic spectrum signal analysis and demodulation via SVM-KNN regressors.

We demonstrate an ensemble learning based method to solve the problem of low SNR Fabry-Perot sensor spectrum signal demodulation. Taking the eight-layer approximate coefficients of a multilevel discrete wavelet transform as input features, an ensemble model that combines multiple SVM and KNN learners is trained. Bootstrap and booting techniques are introduced for better modeling performance and stability. It is shown that the performance of the proposed ensemble model based on SVM-KNN regressors is excellent; an accuracy of 0.46%F.S. relative mean error is achieved. This method could provide insight into the construction of a large scale fiber based Fabry-Perot sensor network.

All-fiber seawater salinity sensor based on fiber laser intracavity loss modulation with low detection limit.

An all-fiber seawater salinity sensor based on intracavity loss-modulated sensing in a fiber ring laser is proposed and experimentally demonstrated. An optical fiber multimode interferometer, which is based on single-mode-no-core-single-mode fiber structure, is cascaded with a fiber reflector and used as a reflected sensing head to enhance loss-modulated depth. It is inserted in a fiber ring laser and the intracavity loss-modulated salinity sensing is induced for the fiber laser's output intensity. The salinity sensitivity is measured to be 0.1 W/‰ with a high signal-to-noise ratio more than 49 dB and narrow full width at half maximum less than 40 pm. The temperature cross-sensitivity characteristic and stability are also analyzed. Considering the errors from cross-sensitivity, stability and resolution of the photodetector, the detection limit of the sensor system is 0.0023 ‰ (0.0002 S/m), which is comparable to the most advanced commercial electronic salinity sensor.

Preliminary Study on Rapid Serial Visualization Presentation Multi-Class Target EEG Classification.

OBJECTIVE: Brain-Computer Interface (BCI) provides a direct communication channel between the brain and external devices. After combining with the Rapid Serial Visualization Presentation (RSVP) paradigm, the RSVP-BCI system can be utilized for human vision-based fast information retrieval. Currently only binary classification of single-trial EEG can be achieved, also the research on the multi-class target RSVP is few, which limited information transfer rate and the application scenarios of the system. In this paper, we focus on the RSVP multi-class target image retrieval task that contains two classes of targets for achieving triple classification for RSVP-EEG. METHODS: Designed two experiments, each containing two tasks with different task difficulties. We recruited 30 subjects to participate in the experiments, collected EEG data, and made the data publicly available. Moreover, we conducted behavioral analysis, ERP analysis, and proposed a model, MDCNet, for EEG classification to study the feasibility of multi-class target RSVP and the impact of task difficulty. RESULTS: The experimental results indicated that (1) RSVP-EEG classification that includes non-target and 2-class target is feasibility; (2) the different targets in the same task will evoke P300 with the same latency and different amplitudes, and the hit rate of the target in EEG classification is positively correlated with its amplitude; (3) the information hidden in the time dimension play an important role in EEG classification; (4) the harder the task is, the latency of P300 is longer. CONCLUSION/SIGNIFICANCE: The experimental analysis obtained meaningful results, which provided a theoretical basis for subsequent research.