-
Intelligent Traffic Monitoring with Distributed Acoustic Sensing
Authors:
Dongzi Xie,
Xinming Wu,
Zhixiang Guo,
Heting Hong,
Baoshan Wang,
Yingjiao Rong
Abstract:
Distributed Acoustic Sensing (DAS) is promising for traffic monitoring, but its extensive data and sensitivity to vibrations, causing noise, pose computational challenges. To address this, we propose a two-step deep-learning workflow with high efficiency and noise immunity for DAS-based traffic monitoring, focusing on instance vehicle trajectory segmentation and velocity estimation. Our approach b…
▽ More
Distributed Acoustic Sensing (DAS) is promising for traffic monitoring, but its extensive data and sensitivity to vibrations, causing noise, pose computational challenges. To address this, we propose a two-step deep-learning workflow with high efficiency and noise immunity for DAS-based traffic monitoring, focusing on instance vehicle trajectory segmentation and velocity estimation. Our approach begins by generating a diverse synthetic DAS dataset with labeled vehicle signals, tackling the issue of missing training labels in this field. This dataset is used to train a Convolutional Neural Network (CNN) to detect linear vehicle trajectories from the noisy DAS data in the time-space domain. However, due to significant noise, these trajectories are often fragmented and incomplete. To enhance accuracy, we introduce a second step involving the Hough transform. This converts detected linear features into point-like energy clusters in the Hough domain. Another CNN is then employed to focus on these energies, identifying the most significant points. The inverse Hough transform is applied to these points to reconstruct complete, distinct, and noise-free linear vehicle trajectories in the time-space domain. The Hough transform plays a crucial role by enforcing a local linearity constraint on the trajectories, enhancing continuity and noise immunity, and facilitating the separation of individual trajectories and estimation of vehicle velocities (indicated by trajectory slopes in the Hough domain). Our method has shown effectiveness in real-world datasets, proving its value in real-time processing of DAS data and applicability in similar traffic monitoring scenarios. All related codes and data are available at https://github.com/TTMuTian/itm/.
△ Less
Submitted 5 March, 2024;
originally announced March 2024.
-
Finding emergence in data by maximizing effective information
Authors:
Mingzhe Yang,
Zhipeng Wang,
Kaiwei Liu,
Yingqi Rong,
Bing Yuan,
Jiang Zhang
Abstract:
Quantifying emergence and modeling emergent dynamics in a data-driven manner for complex dynamical systems is challenging due to the lack of direct observations at the micro-level. Thus, it's crucial to develop a framework to identify emergent phenomena and capture emergent dynamics at the macro-level using available data. Inspired by the theory of causal emergence (CE), this paper introduces a ma…
▽ More
Quantifying emergence and modeling emergent dynamics in a data-driven manner for complex dynamical systems is challenging due to the lack of direct observations at the micro-level. Thus, it's crucial to develop a framework to identify emergent phenomena and capture emergent dynamics at the macro-level using available data. Inspired by the theory of causal emergence (CE), this paper introduces a machine learning framework to learn macro-dynamics in an emergent latent space and quantify the degree of CE. The framework maximizes effective information, resulting in a macro-dynamics model with enhanced causal effects. Experimental results on simulated and real data demonstrate the effectiveness of the proposed framework. It quantifies degrees of CE effectively under various conditions and reveals distinct influences of different noise types. It can learn a one-dimensional coarse-grained macro-state from fMRI data, to represent complex neural activities during movie clip viewing. Furthermore, improved generalization to different test environments is observed across all simulation data.
△ Less
Submitted 29 November, 2023; v1 submitted 19 August, 2023;
originally announced August 2023.
-
Application of neural-network hybrid models in estimating the infection functions of nonlinear epidemic models
Authors:
Chentong Li,
Changsheng Zhou,
Junmin Liu,
Yao Rong
Abstract:
Hybrid neural network models combine the advantages of a neural network's fitting functionality with differential equation models to reflect actual physical processes and are widely used in analyzing time-series data. Most related studies have focused on linear hybrid models, but only a few have examined nonlinear problems. In this work, we use a hybrid nonlinear epidemic neural network as the ent…
▽ More
Hybrid neural network models combine the advantages of a neural network's fitting functionality with differential equation models to reflect actual physical processes and are widely used in analyzing time-series data. Most related studies have focused on linear hybrid models, but only a few have examined nonlinear problems. In this work, we use a hybrid nonlinear epidemic neural network as the entry point to study its power in predicting the correct infection function of an epidemic model. To achieve this goal, we combine the bifurcation theory of the nonlinear differential model with the mean-squared error loss and design a novel loss function to ensure model trainability. Furthermore, we find the unique existence conditions supporting ordinary differential equations to estimate the correct infection function. Using the Runge Kutta method, we perform numerical experiments on our proposed model and verify its soundness. We also apply it to real COVID-19 data to accurately discover the change law of its infectivity.
△ Less
Submitted 5 March, 2022;
originally announced March 2022.
-
El Niño Modoki thus far can be mostly predicted more than 10 years ahead of time
Authors:
X. San Liang,
Fen Xu,
Yineng Rong
Abstract:
The 2014-2015 "Monster"/"Super" El Niño failed to be predicted one year earlier due to the growing importance of a new type of El Niño, El Niño Modoki, which reportedly has much lower forecast skill with the classical models. In this study, we show that, so far as of today, this new El Niño actually can be mostly predicted at a lead time of more than 10 years. This is achieved through tracing the…
▽ More
The 2014-2015 "Monster"/"Super" El Niño failed to be predicted one year earlier due to the growing importance of a new type of El Niño, El Niño Modoki, which reportedly has much lower forecast skill with the classical models. In this study, we show that, so far as of today, this new El Niño actually can be mostly predicted at a lead time of more than 10 years. This is achieved through tracing the predictability source with an information flow-based causality analysis, which is rigorously established from first principles in the past decade. We show that the information flowing from the solar activity 45 years ago to the sea surface temperature results in a causal structure resembling the El Niño Modoki mode. Based on this, a multidimensional system is constructed out of the sunspot number series with time delays of 22-50 years. The first 25 principal components are then taken as the predictors to fulfill the prediction, which through information flow-based causal deep learning reproduces rather accurately the events 12 years in advance.
△ Less
Submitted 12 April, 2021;
originally announced April 2021.
-
A High Efficiency Aluminum-Ion Battery Using an AlCl3-Urea Ionic Liquid Analogue Electrolyte
Authors:
Michael Angell,
Chun-Jern Pan,
Youmin Rong,
Chunze Yuan,
Meng-Chang Lin,
BingJoe Hwang,
Hongjie Dai
Abstract:
In recent years, impressive advances in harvesting renewable energy have led to pressing demand for the complimentary energy storage technology. Here, a high coulombic efficiency (~ 99.7%) Al battery is developed using earth-abundant aluminum as the anode, graphite as the cathode, and a cheap ionic liquid analogue electrolyte made from a mixture of AlCl3 and urea in 1.3 : 1 molar ratio. The batter…
▽ More
In recent years, impressive advances in harvesting renewable energy have led to pressing demand for the complimentary energy storage technology. Here, a high coulombic efficiency (~ 99.7%) Al battery is developed using earth-abundant aluminum as the anode, graphite as the cathode, and a cheap ionic liquid analogue electrolyte made from a mixture of AlCl3 and urea in 1.3 : 1 molar ratio. The battery displays discharge voltage plateaus around 1.9 V and 1.5 V (average discharge = 1.73 V) and yielded a specific cathode capacity of ~73 mAh g-1 at a current density of 100 mA g-1 (~ 1.4 C). High coulombic efficiency over a range of charge-discharge rates and stability over ~150-200 cycles was easily demonstrated. In-situ Raman spectroscopy clearly showed chloroaluminate anion intercalation/deintercalation of graphite in the cathode side during charge/discharge and suggested the formation of a stage 2 graphite intercalation compound when fully charged. Raman spectroscopy and nuclear magnetic resonance suggested the existence of AlCl4-, Al2Cl7- anions, and [AlCl2. (urea)n]+ cations in the urea/AlCl3 electrolyte when an excess of AlCl3 was present. Aluminum deposition therefore proceeded through two pathways, one involving Al2Cl7- anions and the other involving [AlCl2.(urea)n]+ cations. This battery is a promising prospect for a future high performance, low cost energy storage device.
△ Less
Submitted 29 November, 2016;
originally announced November 2016.
