Multivariable time series prediction has been widely studied in power energy, aerology, meteorology, finance, transportation, etc. Traditional modeling methods have complex patterns and are inefficient to capture long-term multivariate dependencies of data for desired forecasting accuracy. To address such concerns, various deep learning models based on Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) methods are proposed. To improve the prediction accuracy and minimize the multivariate time series data dependence for aperiodic data, in this article, Beijing PM2.5 and ISO-NE Dataset are analyzed by a novel Multivariate Temporal Convolution Network (M-TCN) model. In this model, multi-variable time series prediction is constructed as a sequence-to-sequence scenario for non-periodic datasets. The multichannel residual blocks in parallel with asymmetric structure based on deep convolution neural network is proposed. The results are compared with rich competitive algorithms of long short term memory (LSTM), convolutional LSTM (ConvLSTM), Temporal Convolution Network (TCN) and Multivariate Attention LSTM-FCN (MALSTM-FCN), which indicate significant improvement of prediction accuracy, robust and generalization of our model.
3 December 2020 in The European Physical Journal C
The production of heavy quarks (charm and beauty) provides unique insights into the transport properties of the Quark–Gluon Plasma (QGP) in heavy-ion collisions. Experimentally, the nuclear modification factor $$R_{\mathrm{AA}}$$ and the azimuthal anisotropy coefficient $$v_{\mathrm{2}}$$ of heavy-flavor mesons are powerful observables to study the medium-related effects, such as energy loss and collectivity, on the heavy quark propagation through the QGP evolution. The latest measurements of the prompt and non-prompt open heavy-flavor hadrons allow a systematic comparison of the transport behaviors probed by charm and beauty quarks. In this work we make such an attempt utilizing our recently developed framework. By performing a quantitative investigation of $$R_{\mathrm{AA}}$$ and $$v_{\mathrm{2}}$$ , it is found that both charm and beauty quarks are efficient probes to capture the dynamical features of QGP, in particular the resulting mass hierarchy for the energy loss and azimuthal anisotropy, which are well inherited by the various D/B-meson species. Moreover, our calculations can describe simultaneously $$R_{\mathrm{AA}}$$ and $$v_{\mathrm{2}}$$ data for the prompt and non-prompt $$D^{0}$$ mesons in central ( $$0-10\%$$ ) and semi-central ( $$30-50\%$$ ) Pb–Pb collisions at $$\sqrt{s_{\mathrm{NN}}}=5.02~{\mathrm{TeV}}$$ . The predictions for B-meson observables for upcoming experimental tests are also made down to the low momentum region.
Multivariate time-series forecasting is one of the crucial and persistent challenges in time-series forecasting tasks. As a kind of data with multivariate correlation and volatility, multivariate time series impose highly nonlinear time characteristics on the forecasting model. In this paper, a new multivariate time-series forecasting model, multivariate temporal convolutional attention network (MTCAN), based on a self-attentive mechanism is proposed. MTCAN is based on the Convolution Neural Network (CNN) model, using 1D dilated convolution as the basic unit to construct asymmetric blocks, and then, the feature extraction is performed by the self-attention mechanism to finally obtain the prediction results. The input and output lengths of this network can be determined flexibly. The validation of the method is carried out with three different multivariate time-series datasets. The reliability and accuracy of the prediction results are compared with Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Long Short-Term Memory (ConvLSTM), and Temporal Convolutional Network (TCN). The prediction results show that the model proposed in this paper has significantly improved prediction accuracy and generalization.
This paper presents a digital decimation filter based on a third-order four-bit Sigma-Delta modulator. The digital decimation filter is an important part of the Sigma-Delta ADC and is designed to make the Sigma-Delta ADC (Analog-to-Digital Converter) meets the requirements of Signal-to-Noise Ratio (SNR) not less than 120 dB and Equivalent Number of Bits (ENOB) not less than 20 bits. It adopts a three-stages cascaded structure including a Cascaded Integrator Comb (CIC) decimation filter, a Finite Impulse Response (FIR) compensation filter, and a half-band (HB) filter. This structure effectively reduces about 13% multiplier cells and memory cells. The coefficient symmetry technique and CSD (Canonic Signed Digit) coding technique are used to optimize the parameters of the filter, which further reduces the computational complexity. After optimization, the circuit area is reduced by about 15%, and the logic resources are decreased by about 23%. The Verilog hardware description language is used to describe the behavior of the digital decimation filter, and the simulation is carried out based on the VCS (Verilog Compile Simulator) platform. At the same time, the prototype verification is implemented on the Xilinx Artix-7 series FPGA, and the ADC achieves 113 dB SNR and 18.5 bits ENOB. Finally, the Sigma-Delta ADC is fabricated on SMIC 0.18 m CMOS process with the layout area of 714.8 m × 628.4 m and the power consumption of 11.2 mW. The more tests for the fabricated prototypes will be performed in the future to verify that the Sigma-Delta ADC complies with the design specifications.