Topic
Time–frequency analysis
About: Time–frequency analysis is a research topic. Over the lifetime, 5407 publications have been published within this topic receiving 104346 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: In this article, an explicit form for the reassigned Gabor spectrogram of an Hermite function of arbitrary order is given, and it is shown that the energy concentration sharply localizes outside the border of a clearance area limited by the "classical" circle where the spectrogram attains its maximum value.
Abstract: An explicit form is given for the reassigned Gabor spectrogram of an Hermite function of arbitrary order. It is shown that the energy concentration sharply localizes outside the border of a clearance area limited by the “classical” circle where the Gabor spectrogram attains its maximum value, with a perfect localization that can only be achieved in the limit of infinite order.
26 citations
TL;DR: In this paper, an online ESR estimation method of aluminum electrolytic capacitor (AEC) was proposed by using the wavelet transform (WT) based time-frequency analysis, and the relationship between ESR and the jump amount of output voltage at turn- off moments was analyzed first, and then the ESR calculation model was derived using WT with the Wavelet basis of the first derivative of Gaussian function.
Abstract: Aluminum electrolytic capacitor (AEC) is one of the most age-affected components in ac–dc conversion, and its equivalent series resistance ( ESR ) is an important index for reflecting the healthy condition of AEC. In AEC-used boost power factor correction (PFC) converters, ESR of AEC causes a small jump in the switching ripple of output voltage at switching moments, especially at turn- off moments. This small jump is hardly observed at line-frequency scale, either using time-domain analysis or frequency-domain analysis. However using time–frequency analysis this jump is very prominent due to its singularity. In this article, an online ESR estimation method of AEC is proposed by using the wavelet transform (WT) based time–frequency analysis. The relationship between ESR and the jump amount of output voltage at turn- off moments is analyzed first, and then the ESR calculation model is derived using WT with the wavelet basis of the first derivative of Gaussian function. An appropriate sampling interval for the output voltage and the inductor current is determined. Besides, the online ESR estimation scheme is implemented including the hardware and software designs. Furthermore, a prototype of boost PFC converter with 220 V ac input and 360 V dc output is built, where an average current mode control chip UC3854 is used. Four factors are discussed for estimation accuracy in the experiment, and the estimated results are consistent with the results measured by LCR meter with a relative error less than 10%.
26 citations
TL;DR: Recently developed wavelet-based solution frameworks are adopted and further generalized in this paper to account for systems with singular matrices, and a Moore–Penrose generalized matrix inverse excitation-response relationship is derived herein for determining the response EPS of linear MDOF systems.
26 citations
18 Jul 2021
TL;DR: In this article, a new time-frequency transformation layer that is based on complex frequency B-spline wavelets was proposed to improve the performance of ESC classification model, achieving accuracies of 95.20 % on the ESC-50 and 89.14 % on UrbanSound8K datasets.
Abstract: Environmental Sound Classification (ESC) is a rapidly evolving field that recently demonstrated the advantages of application of visual domain techniques to the audio-related tasks. Previous studies indicate that the domain-specific modification of cross-domain approaches show a promise in pushing the whole area of ESC forward. In this paper, we present a new time-frequency transformation layer that is based on complex frequency B-spline (fbsp) wavelets. Being used with a high-performance audio classification model, the proposed fbsp-layer provides an accuracy improvement over the previously used Short-Time Fourier Transform (STFT) on standard datasets. We also investigate the influence of different pre-training strategies, including the joint use of two large-scale datasets for weight initialization: ImageNet and AudioSet. Our proposed model out-performs other approaches by achieving accuracies of 95.20 % on the ESC-50 and 89.14 % on the UrbanSound8K datasets. Additionally, we assess the increase of model robustness against additive white Gaussian noise and reduction of an effective sample rate introduced by the proposed layer and demonstrate that the fbsp-layer improves the model's ability to withstand signal perturbations, in comparison to STFT-based training. For the sake of reproducibility, our code is made available.
26 citations
03 Dec 2020
TL;DR: In this paper, the EEG signals are decomposed into smaller segments of signal by Time Frequency Approach (T-F) like fast Fourier transform and short time Fourier Transform (STFT).
Abstract: Brain Computer Interface is a good beneficial route for severely physically challenged person who is underprivileged to communicate in conventional way or have lost their ability to speak. The cause of the work carried in the paper is to solve the problem of patients suffering from neurological disorder and its disabilities that give rise to this research. In the proposed work Electroencephalogram (EEG) based brain state signal measurement method is use to record the brain activity which is source of communication system between patient and outside world. Electroencephalogram is a non-muscular channel between the human brain and a computer system is provided by brain-computer interface (BCI) in which electrical activity are recorded for perusal of EEG signals by the brain. This signals are then decomposed into smaller segments of signal by Time frequency approaches (T-F) like fast Fourier transform & short time Fourier transform. Both these techniques acts as a feature extraction method followed by training of the data and the classification is done by using support vector machine. The performance parameters like accuracy, precision, sensitivity, specificity are calculated based on the values of evaluation metrics and overall system accuracy comes out to be 92%. The four classified signals can be used as Communication messages by the patients which will help to solve the speech impairment problem of disabled person.
26 citations