Article
Agriculture, Multidisciplinary
Chunling Cao, Tianli Wang, Maofang Gao, Yang Li, Dandan Li, Huijie Zhang
Summary: This study used various dimensionality reduction algorithms and inversion models to predict nitrogen content in corn leaves, with results showing that the EN-PLSR model outperformed others in accuracy and fitting effects, and dimensionality reduction of hyperspectral data significantly influenced the effectiveness of data analysis.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Mathematics, Interdisciplinary Applications
Pichid Kittisuwan, Prayoot Akkaraekthalin
Summary: In this research, the fused lasso algorithm in the wavelet domain is proposed for audio signal enhancement. The algorithm consists of the total variation method (TVM) and the penalized least squares regression (PLSR). The TVM reduces the pseudo-Gibbs phenomena, while the convex PLSR with the small bias effectively reduces noise. A novel non-convex regularization is introduced to build the convex PLSR in closed-form solution and with small bias.
FLUCTUATION AND NOISE LETTERS
(2023)
Article
Engineering, Electrical & Electronic
Debasis Kundu, Swagata Nandi, Rhythm Grover
Summary: The paper proposes the use of weighted least squares estimators for parameter estimation in chirp models. These estimators are robust to outliers and have consistent and convergent properties. Extensive simulations were performed to validate the performance of the proposed estimators.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Multidisciplinary Sciences
Hongyu Li, Xieting Xu, Yajun Lu, Xi Yu, Tong Zhao, Rufei Zhang
Summary: In this paper, a method called relaxed lad lasso is introduced to solve the problem of asymmetric heavy-tailed error distribution or outliers. By combining least absolute deviation with relaxed lasso, this method can select truly important variables in linear regression and achieve robust sparse solutions in the presence of outliers or heavy-tailed errors. The method is also immune to noise variables and maintains a better convergence rate. Through simulation and empirical results, the outstanding performance of relaxed lad lasso in terms of prediction accuracy and variable selection is further verified.
Article
Engineering, Electrical & Electronic
Debasis Kundu, Rhythm Grover
Summary: This paper investigates the use of weighted least squares estimators (WLSEs) in the presence of additive noise or outliers based on a two-dimensional sinusoidal model. The properties of the WLSEs, including consistency and asymptotic normality, are derived and compared to other robust estimators such as the least absolute deviation estimators (LADEs). Extensive simulations are performed to demonstrate the effectiveness of the proposed method, and a practical application is illustrated using a synthetic dataset.
Article
Mathematics, Applied
Lingling He, Xiaoqin Li, Yan Shen, Qiuyue Wu
Summary: In this paper, we investigate the partially linear regression model based on asymptotically almost negatively associated (AANA) random variables. Convergence results for the parametric least squares estimator and nonparametric weighted estimator are obtained under weak conditions. The results extend the corresponding ones for negatively associated (NA) errors to AANA errors and the selection of design points and weight functions is discussed. Simulation experiments are conducted to demonstrate the performance of the obtained results.
JOURNAL OF MATHEMATICAL INEQUALITIES
(2023)
Article
Automation & Control Systems
Zhaobo Liu, Chanying Li
Summary: This note investigates the convergence problem of a distributed recursive least squares estimator and finds that its convergence behavior varies significantly between scalar-parameter and high-dimensional-parameter cases.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Article
Engineering, Electrical & Electronic
Debasis Kundu
Summary: In this paper, a novel robust method for estimating the unknown parameters of a fundamental frequency and its harmonics model is proposed. The proposed weighted least squares estimators (WLSEs) are more robust and convenient to implement compared to the least squares estimators (LSEs) or periodogram estimators in the presence of outliers. The theoretical properties of the WLSEs are established and simulations show their improved performance over LSEs, periodogram estimators, least absolute deviation estimators (LADEs), and Huber's M-estimators (HMEs) in the presence of outliers.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Aerospace
Weiwei Wang, Zhangjian Lu, Ye Tian, Lang Bian, Guoyong Wang, Lixin Zhang
Summary: This paper introduces the research on the application of Doppler-aided positioning method in fused LEO navigation systems, which can improve the positioning performance and availability, and expand its application scenarios.
Review
Engineering, Mechanical
Randall J. Allemang, Rohit S. Patwardhan, Murali M. Kolluri, Allyn W. Phillips
Summary: This paper outlines various FRF estimation techniques and compares algorithms that compute FRF using different methods. It also discusses inconsistencies in some conditioned coherence metrics and provides corrected interpretations.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Automation & Control Systems
Louna Alsouki, Laurent Duval, Clement Marteau, Rami El Haddad, Francois Wahl
Summary: Relating variables X to response y is important in chemometrics. Qualitative interpretation can enhance quantitative prediction by identifying influential features. Projections (e.g. PLS) and variable selections (e.g. lasso) are used for dimension reduction in high-dimensional problems. Dual-sPLS, a variant of PLS1, provides a balance between accurate prediction and efficient interpretation through penalizations inspired by classical regression methods and the dual norm notion. It performs favorably compared to similar regression methods on simulated and real chemical data.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Multidisciplinary Sciences
Rufei Zhang, Tong Zhao, Yajun Lu, Xieting Xu
Summary: This article introduces a novel two-stage variable selection method to solve the common asymmetry problem between the response variable and its influencing factors. The proposed method achieves information symmetry and enjoys favorable asymptotic properties.
Article
Mathematics, Applied
Pingping Zhang, Qun Wang
Summary: This paper investigates the MLSSTLS problem that combines MTLS and STLS problems, providing explicit expressions of the solution under certain conditions, conducting perturbation analysis and examining the condition numbers of the solution. Numerical experiments are performed to demonstrate the validity of the results, which can be connected to previously published findings on MTLS and STLS problems.
NUMERICAL ALGORITHMS
(2022)
Article
Biochemical Research Methods
Ruoyu Tang, Xinyu He, Ruiqi Wang
Summary: The study presents a general computational method for constructing maps between different cell fates and parametric conditions by systematic perturbations. The method does not require accurate parameter measurements or bifurcations. The maps obtained can help in understanding how systematic perturbations drive cell fate decisions and transitions, providing valuable information for predicting and controlling cell states.
Article
Optics
Yong -Mei Li, Hai -Ling Liu, Shi-Jie Pan, Su-Juan Qin, Fei Gao, Dong-Xu Sun, Qiao-Yan Wen
Summary: This paper proposes a complete quantum algorithm for the k-medoids algorithm, which utilizes quantum subroutines to improve the speed of cluster assignment and center update. Compared to existing algorithms, our quantum k-medoids algorithm achieves a polynomial speedup in large data sets.