A Novel Approach to Decode Covert Spatial Attention Using SSVEP and Single-Frequency Phase-Coded Stimuli

Author(s):  
Alexandre Armengol-Urpi ◽  
Andres F. Salazar-Gomez ◽  
Sanjay E. Sarma
Author(s):  
Joana F. Reis ◽  
Catarina N. Matias ◽  
Francesco Campa ◽  
José P. Morgado ◽  
Paulo Franco ◽  
...  

Background and aim: Monitoring bioelectric phase angle (PhA) provides important information on the health and the condition of the athlete. Together with the vector length, PhA constitutes the bioimpedance vector analysis (BIVA) patterns, and their joint interpretation exceeds the limits of the evaluation of the PhA alone. The present investigation aimed to monitor changes in the BIVA patterns during a training macrocycle in swimmers, trying to ascertain if these parameters are sensitive to training load changes across a 13-week training period. Methods: Twelve national and international level swimmers (four females; eight males; 20.9 ± 1.9 years; with a competitive swimming background of 11.3 ± 1.8 years; undertaking 16–20 h of pool training and 4–5 h of dry-land training per week and 822.0 ± 59.0 International Swimming Federation (FINA) points) were evaluated for resistance (R) and reactance (Xc) using a single frequency phase sensitive bioimpedance device at the beginning of the macrocycle (M1), just before the beginning of the taper period (M2), and just before the main competition of the macrocycle (M3). At the three-time assessment points, swimmers also performed a 50 m all-out first stroke sprint with track start (T50 m) while time was recorded. Results: The results of the Hotelling T2 test showed a significant vector displacement due to simultaneous R and Xc changes (p < 0.001), where shifting from top to bottom along the major axis of the R-Xc graph from M1 to M2 was observed. From M2 to M3, a vector displacement up and left along the minor axis of the tolerance ellipses resulted in an increase in PhA (p < 0.01). The results suggest a gain in fluid with a decrease in cellular density from M1 to M2 due to decrements in R and Xc. Nevertheless, the reduced training load characterizing taper seemed to allow for an increase in PhA and, most importantly, an increase of Xc, thus demonstrating improved cellular health and physical condition, which was concomitant with a significant increase in the T50 m performance (p < 0.01). Conclusions: PhA, obtained by bioelectrical R and Xc, can be useful in monitoring the condition of swimmers preparing for competition. Monitoring BIVA patterns allows for an ecological approach to the swimmers’ health and condition assessment without resorting to equations to predict the related body composition variables.


Author(s):  
Zhipeng Chen ◽  
Yiming Cui ◽  
Wentao Ma ◽  
Shijin Wang ◽  
Guoping Hu

Machine Reading Comprehension (MRC) with multiplechoice questions requires the machine to read given passage and select the correct answer among several candidates. In this paper, we propose a novel approach called Convolutional Spatial Attention (CSA) model which can better handle the MRC with multiple-choice questions. The proposed model could fully extract the mutual information among the passage, question, and the candidates, to form the enriched representations. Furthermore, to merge various attention results, we propose to use convolutional operation to dynamically summarize the attention values within the different size of regions. Experimental results show that the proposed model could give substantial improvements over various state-of- the-art systems on both RACE and SemEval-2018 Task11 datasets.


Author(s):  
Sergio Colangeli ◽  
Walter Ciccognani ◽  
Mirko Palomba ◽  
Ernesto Limiti

In this paper a novel approach for determining the four noise parameters of FET devices over frequency is presented. Such methodology is made of two parts: the first one allows to straightforwardly extract single-frequency noise parameters from source-pull data; the second one extends this capability to multi-frequency, source-pull data to obtain a full description of device noise behavior over frequency by means of at most 10 constant parameters (depending on the required accuracy). The whole process is automated via a software routine and does not need a previous knowledge of the FET equivalent circuit's topology, or the values of its elements. This peculiarity makes the proposed method very well suited to quick characterization campaigns of active devices, avoiding the burden of a whole set of prior, different measurements and the relevant, critical extraction procedures, which are strongly dependent on the device.


2013 ◽  
Vol 31 (3) ◽  
pp. 515-521 ◽  
Author(s):  
W. Gao ◽  
S.N. Liu ◽  
Y.F. Bi ◽  
X.B. Hu

AbstractWe theoretically and experimentally investigate the conditions of obtaining a narrowband flat-top Brillouin gain spectrum (BGS) based on single-frequency and multi-frequency phase modulations. Using the unequal-amplitude spectral lines, the flat-top BGS can be realized by controlling the intensity ratio and the frequency separation between them. In experiment, we obtain the flat-top gain spectra with the bandwidths of 40 MHz and 125 MHz, and with the top fluctuation of less than 0.21 dB. Based on this, we also achieve low distortion Brillouin amplification of a probe signal pulse and spectrum.


2009 ◽  
Author(s):  
G.E. Corazza ◽  
C. Bersani ◽  
C. Palestini ◽  
A. Vanelli-Coralli

Sign in / Sign up

Export Citation Format

Share Document