Estimation Performance Research Articles

Non-adaptive estimation for degenerate diffusion processes

We consider a degenerate system of stochastic differential equations. The first component of the system has a parameter θ 1 \theta _1 in a non-degenerate diffusion coefficient and a parameter θ 2 \theta _2 in the drift term. The second component has a drift term with a parameter θ 3 \theta _3 and no diffusion term. Parametric estimation of the degenerate diffusion system is discussed under a sampling scheme. We investigate the asymptotic behavior of the joint quasi-maximum likelihood estimator for ( θ 1 , θ 2 , θ 3 ) (\theta _1,\theta _2,\theta _3) . The estimation scheme is non-adaptive. The estimator incorporates information of the increments of both components, and under this construction, we show that the asymptotic variance of the estimator for θ 1 \theta _1 is smaller than the one for standard estimator based on the first component only, and that the convergence of the estimator for θ 3 \theta _3 is much faster than for the other parameters. By simulation studies, we compare the performance of the joint quasi-maximum likelihood estimator with the adaptive and one-step estimators investigated in Gloter and Yoshida [Electron. J. Statist 15 (2021), no. 1, 1424–1472].

MMG-Based Knee Dynamic Extension Force Estimation Using Cross-Talk and IGWO-LSTM

Mechanomyography (MMG) is an important muscle physiological activity signal that can reflect the amount of motor units recruited as well as the contraction frequency. As a result, MMG can be utilized to estimate the force produced by skeletal muscle. However, cross-talk and time-series correlation severely affect MMG signal recognition in the real world. These restrict the accuracy of dynamic muscle force estimation and their interaction ability in wearable devices. To address these issues, a hypothesis that the accuracy of knee dynamic extension force estimation can be improved by using MMG signals from a single muscle with less cross-talk is first proposed. The hypothesis is then confirmed using the estimation results from different muscle signal feature combinations. Finally, a novel model (improved grey wolf optimizer optimized long short-term memory networks, i.e., IGWO-LSTM) is proposed for further improving the performance of knee dynamic extension force estimation. The experimental results demonstrate that MMG signals from a single muscle with less cross-talk have a superior ability to estimate dynamic knee extension force. In addition, the proposed IGWO-LSTM provides the best performance metrics in comparison to other state-of-the-art models. Our research is expected to not only improve the understanding of the mechanisms of quadriceps contraction but also enhance the flexibility and interaction capabilities of future rehabilitation and assistive devices.

Method for Monitoring Wheat Growth Status and Estimating Yield Based on UAV Multispectral Remote Sensing

An efficient and accurate estimation of wheat growth and yield is important for wheat assessment and field management. To improve the accuracy and stability of wheat growth and yield estimation, an estimation method based on a genetic algorithm-improved support vector regression (GA-SVR) algorithm was proposed in this study. The correlation analysis between vegetation indices calculated from spectral data and wheat growth phenotypes and yields was performed to obtain the optimal combination of vegetation indices with high correlation and good estimation performance. At the same time, the optimal model for wheat growth monitoring was screened and constructed in experiments with 12 wheat varieties and 3 gradient nitrogen fertilizer application levels. Then, the yield estimation model was established and its applicability was verified under different nitrogen fertilizer application levels. The results showed that the constructed models for the leaf area index, plant height, and yield estimation performed well, with coefficients of determination of 0.82, 0.71, and 0.70, and root mean square errors of 0.09, 2.7, and 68.5, respectively. This study provided an effective UAV remote sensing technique for monitoring wheat growth status and estimating yield. This study provides an effective unmanned aerial remote sensing technique for monitoring wheat growth and estimating yield, and provides technical support for wheat yield assessment and field management.

Measuring rainfall using microwave links: the influence of temporal sampling

Abstract. Terrestrial microwave links are increasingly being used to estimate path-averaged precipitation by determining the attenuation caused by rainfall along the link path, mostly with commercial microwave links from cellular telecommunication networks. However, the temporal resolution of these rainfall estimates and the method to derive them are often determined by the temporal sampling strategy that is employed by the mobile network operators. Currently, the links are most often sampled at a temporal resolution of 15 min with a recording of the minimum and maximum values, while more recently, a form of instantaneous sampling with possible intervals up to 1 s has also been set up. For rainfall research purposes, often high temporal resolutions in combination with averaged values are preferred. However, it is uncertain how these various temporal sampling strategies affect the estimated rainfall intensity. Here we aim to understand how temporal sampling strategies affect the measured rainfall intensities using microwave links. To do so, we use data from three collocated microwave links, two 38 GHz and one 26 GHz, sampled at 20 Hz and covering a 2.2 km path over the city of Wageningen, the Netherlands. We aggregate the microwave link power levels to multiple time intervals (1 s to 60 min) and use a mean, instantaneous, and minimum and maximum value to characterize the signal. Based on the aggregated data, we compute rainfall intensities and compare these with 20 Hz rainfall estimates, such that we isolate errors and uncertainties caused by the sampling strategies from instrumental effects, such as different biases between instruments and representativeness errors. In general, our results show that for all sampling strategies, an increase in sampling time interval reduces the performance of the rainfall estimates, which especially holds for the instantaneous sampling strategy. Even the mean sampling strategy, which generally performs best of all strategies, is sensitive to this reduction in temporal resolution and could lead to significant underestimations. This sensitivity of the mean sampling to the temporal resolution seems to be largely affected by the non-linear relation between attenuation and rainfall. The min–max sampling strategy is mostly prone to minor underestimations or large overestimations of the path-averaged rainfall intensities. Moreover, our results, including a comparison with theoretical events, show that the attenuation due to wet antennas not only affects the comparison between the rainfall estimates obtained with a microwave link and another reference instrument but also has a significant influence on the performance of the rainfall retrieval algorithm, especially for devices with relatively long duration of the wet-antenna attenuation combined with the longer sampling time intervals. Overall, this study demonstrates the effect a selected sampling strategy can have on rainfall intensity estimates using (commercial) microwave links.

Evaluating German PISA stratification designs: a simulation study

Stratification is an important design feature of many studies using complex sampling designs and it is often used in large-scale assessment (LSA) studies, such as the Programme for International Student Assessment (PISA), for two main reasons. First, stratification variables that achieve a high between and low within strata variance can improve the efficiency of a survey design. Second, stratification allows one to, explicitly or implicitly, control for sample sizes across subpopulations. It ensures that some parts of a population are in the sample in predetermined proportions. In this study, we determine through simulation which stratification scheme is best for PISA in Germany. For this, we consider the constraints imposed by the international sampling design, the available information about schools, and specific national characteristics of the German educational system. We examine seven different stratification designs selected based on scenarios used in past LSAs in Germany and theoretical considerations for future implementations. The chosen scenarios were compared with two reference scenarios: (1) an unstratified design and (2) a synthetic optimal stratification design. The simulation study reveals that the stratification design currently applied in PISA produces satisfactory results regarding sampling precision. The present stratification design is based on Germany's federal states and school types. However, this approach leads to small strata, which has been problematic for estimating sampling variance in previous cycles. Therefore, alternative stratification scenarios were considered and, in addition to overcoming the small-strata problem, also led to smaller standard errors for estimates of student mean performance in mathematics, science, and reading. As a result of this study, we recommend considering three different stratification designs for Germany in future cycles of PISA. These recommendations aim to: (1) improve the sampling efficiency while keeping the sample size constant, (2) follow a sound methodological approach, and (3) make conservative and cautious changes while maintaining a reflection of the structure of the German federal school system with different school types. These suggestions include a reinvented stratification of grouped German federal states and designs with school types as explicit stratifiers and federal states as implicit stratifiers.

A comparison of RNA-Seq data preprocessing pipelines for transcriptomic predictions across independent studies

BackgroundRNA sequencing combined with machine learning techniques has provided a modern approach to the molecular classification of cancer. Class predictors, reflecting the disease class, can be constructed for known tissue types using the gene expression measurements extracted from cancer patients. One challenge of current cancer predictors is that they often have suboptimal performance estimates when integrating molecular datasets generated from different labs. Often, the quality of the data is variable, procured differently, and contains unwanted noise hampering the ability of a predictive model to extract useful information. Data preprocessing methods can be applied in attempts to reduce these systematic variations and harmonize the datasets before they are used to build a machine learning model for resolving tissue of origins.ResultsWe aimed to investigate the impact of data preprocessing steps—focusing on normalization, batch effect correction, and data scaling—through trial and comparison. Our goal was to improve the cross-study predictions of tissue of origin for common cancers on large-scale RNA-Seq datasets derived from thousands of patients and over a dozen tumor types. The results showed that the choice of data preprocessing operations affected the performance of the associated classifier models constructed for tissue of origin predictions in cancer.ConclusionBy using TCGA as a training set and applying data preprocessing methods, we demonstrated that batch effect correction improved performance measured by weighted F1-score in resolving tissue of origin against an independent GTEx test dataset. On the other hand, the use of data preprocessing operations worsened classification performance when the independent test dataset was aggregated from separate studies in ICGC and GEO. Therefore, based on our findings with these publicly available large-scale RNA-Seq datasets, the application of data preprocessing techniques to a machine learning pipeline is not always appropriate.

Ship Flow of the Ryuko-maru Calculated by the Reynolds Stress Model Using the Roughness Function at the Full Scale

The k-omega SST turbulence model is extensively employed in Reynolds-averaged Navier–Stokes (RANS)-based Computational Fluid Dynamics (CFD) calculations. However, the accuracy of the estimation of viscous resistance and companion flow distribution for full-sized vessels is not sufficient. This study conducted a computational analysis of the flow around the Ryuko-maru at model-scale and full-scale Reynolds numbers utilizing the Reynolds stress turbulence model (RSM). The obtained Reynolds stress distribution from the model-scale computation was compared against experimental measurements to assess the capability of the RSM. Furthermore, full-scale computations were performed, incorporating the influence of hull surface roughness, with the resulting wake distributions juxtaposed with the actual ship measurements. The full-scale calculation employed the sand-grain roughness function, and an optimal roughness length scale was determined by aligning the computed wake distribution with Ryuko-maru’s measured data. The results of this study will allow for the direct performance estimation of full-scale ships and contribute to the design technology of performance.

An evaluation of the adequacy of Lévy and extreme value tail risk estimates

This study investigates the simplicity and adequacy of tail-based risk measures—value-at-risk (VaR) and expected shortfall (ES)—when applied to tail targeting of the extreme value (EV) model. We implement Lévy–VaR and ES risk measures as full density-based alternatives to the generalized Pareto VaR and the generalized Pareto ES of the tail-targeting EV model. Using data on futures contracts of S&P500, FTSE100, DAX, Hang Seng, and Nikkei 225 during the Global Financial Crisis of 2007–2008, we find that the simplicity of tail-based risk management with a tail-targeting EV model is more attractive. However, the performance of EV risk estimates is not necessarily superior to that of full density-based relatively complex Lévy risk estimates, which may not always give us more robust VaR and ES results, making the model inadequate from a practical perspective. There is randomness in the estimation performances under both approaches for different data ranges and coverage levels. Such mixed results imply that banks, financial institutions, and policymakers should find a way to compromise or trade-off between “simplicity” and user-defined “adequacy”.

The estimation of b-value of the frequency-magnitude distribution and of its one-sigma intervals from binned magnitude data

Summary The estimation of the slope (b-value) of the frequency magnitude distribution of earthquakes is based on a formula derived by Aki decades ago, assuming a continuous exponential distribution. However, as the magnitude is usually provided with a limited resolution, its distribution is not continuous but discrete. In the literature, this problem was initially solved by an empirical correction (due to Utsu) to the minimum magnitude, and later by providing an exact formula such as that by Tinti and Mulargia, based on the geometric distribution theory. A recent paper by van der Elst showed that the b-value can be estimated also by considering the magnitude differences (which are proven to follow an exponential discrete Laplace distribution) and that in this case the estimator is more resilient to the incompleteness of the magnitude dataset. In this work we provide the complete theoretical formulation including i) the derivation of the means and standard deviations of the discrete exponential and Laplace distributions; ii) the estimators of the decay parameter of the discrete exponential and trimmed Laplace distributions; and iii) the corresponding formulas for the parameter b. We deduce iv) the standard one-sigma intervals for the estimated b. Moreover, we are able v) to quantify the error associated with the Utsu minimum-magnitude correction. Furthermore, we have discussed the formulas to produce statistically independent magnitude differences. We tested extensively the b-value estimators on simulated synthetic datasets including complete catalogues as well as catalogues affected by a strong incompleteness degree such as aftershock sequences where the incompleteness is made to vary from one event to the next. We have also analysed the real aftershock sequence of the 30/10/2016 Norcia (central Italy) to integrate the finding of the simulations. To judge the performance of the various estimators we have introduced an index p that can be seen as a non-parametric extension of the Student's t index. The main outcomes of this paper are that 1) the b-value estimators devised for continuous magnitude data are not adequate for binned magnitudes, 2) for complete datasets, estimators based on magnitudes and on magnitude differences provide substantially equivalent results, 3) for incomplete magnitude datasets, estimators based on magnitude differences provide better results, 4) for incomplete aftershock sequences there is no evidence that methods based on positive magnitude differences are superior than other methods using differences. This conclusion is further confirmed by our analysis of the above-mentioned Norcia seismic sequence. This last finding contrasts with the van der Elst's claim that the so called ${b}_ + $ method is the most adequate to treat real aftershock sequences.

3D hand pose and mesh estimation via a generic Topology-aware Transformer model

In Human-Robot Interaction (HRI), accurate 3D hand pose and mesh estimation hold critical importance. However, inferring reasonable and accurate poses in severe self-occlusion and high self-similarity remains an inherent challenge. In order to alleviate the ambiguity caused by invisible and similar joints during HRI, we propose a new Topology-aware Transformer network named HandGCNFormer with depth image as input, incorporating prior knowledge of hand kinematic topology into the network while modeling long-range contextual information. Specifically, we propose a novel Graphformer decoder with an additional Node-offset Graph Convolutional layer (NoffGConv). The Graphformer decoder optimizes the synergy between the Transformer and GCN, capturing long-range dependencies and local topological connections between joints. On top of that, we replace the standard MLP prediction head with a novel Topology-aware head to better exploit local topological constraints for more reasonable and accurate poses. Our method achieves state-of-the-art 3D hand pose estimation performance on four challenging datasets, including Hands2017, NYU, ICVL, and MSRA. To further demonstrate the effectiveness and scalability of our proposed Graphformer Decoder and Topology aware head, we extend our framework to HandGCNFormer-Mesh for the 3D hand mesh estimation task. The extended framework efficiently integrates a shape regressor with the original Graphformer Decoder and Topology aware head, producing Mano parameters. The results on the HO-3D dataset, which contains various and challenging occlusions, show that our HandGCNFormer-Mesh achieves competitive results compared to previous state-of-the-art 3D hand mesh estimation methods.

Improving accuracy in predictions about future performance

AbstractWithin the academic context, students often expect to perform significantly better on an upcoming exam when compared to their actual performance. Unfortunately, students who are least accurate in their performance predictions are also the most at risk of being underprepared compared to their peers. Thus, efforts to enhance accuracy in performance predictions would benefit the student population that is most at risk of academic shortcomings and failure. Two studies examined whether incentives improve accuracy of academic performance predictions. In Study 1, 126 students in a 200-level undergraduate course provided estimates of how well they would perform on an upcoming exam. Two weeks later, the same students were randomly assigned to receive a reward incentive, punishment incentive, or no incentive for accuracy in performance predictions and their exam performance estimates were reassessed then compared to their actual scores on an exam completed one week later. In Study 2, 144 students from multiple 300-level courses provided performance predictions for an upcoming exam that were then compared to their actual exam scores for accuracy. For a subsequent exam, the same students were randomly assigned to varying levels of reward incentives (no incentive, modest incentive, large incentive) for prediction accuracy and their performance estimates were examined and compared to actual performance on a second exam. Findings from both studies indicated performance estimates are malleable and with the appropriate incentives, accuracy in predictions may be improved and contribute to better actual performance.

Decentralized Fuzzy Fault Estimation Observer Design for Discrete-Time Nonlinear Interconnected Systems

In this paper, a fault estimation technique is proposed for discrete-time nonlinear interconnected systems with uncertain interconnections. To achieve the fault estimation, the decentralized fuzzy observer is adopted based on the Takagi–Sugeno fuzzy model. Based on the estimation error model with the subsystems of the interconnected system and its decentralized fuzzy observer, the fault estimation condition with H∞ performance is presented. The main idea of this paper is that a novel inequality condition for H∞ performance is used, and the sufficient condition is presented to guarantee the good fault estimation performance. Also, the decentralized fuzzy observer design condition for the fault estimation is converted into linear matrix inequality formats. Finally, a simulation example is provided, and the effectiveness of the proposed fault estimation technique is verified by comparison of the fault estimation performance.

Evaluating the quantum optimal biased bound in a unitary evolution process

Seeking the available precision limit of unknown parameters is a significant task in quantum parameter estimation. One often resorts to the widely utilized quantum Cramér-Rao bound (QCRB) based on unbiased estimators to finish this task. Nevertheless, most actual estimators are usually biased in the limited number of trials. For this reason, we introduce two effective error bounds for biased estimators based on a unitary evolution process in the framework of the quantum optimal biased bound. Furthermore, we show their estimation performance by two specific examples of the unitary evolution process, including the phase encoding and the SU(2) interferometer process. Our findings will provide an useful guidance for finding the precision limit of unknown parameters.

Environmental-structure-perception-based adaptive pose fusion method for LiDAR-visual-inertial odometry

Environmental-structure-perception-based adaptive pose fusion method for LiDAR-visual-inertial odometry

CiGNN: A Causality-Informed and Graph Neural Network Based Framework for Cuffless Continuous Blood Pressure Estimation.

Causalityholds profound potentials to dissipate confusion and improve accuracy in cuffless continuous blood pressure (BP) estimation, an area often neglected in current research. In this study, we propose a two-stage framework, CiGNN, that seamlessly integrates causality and graph neural network (GNN) for cuffless continuous BP estimation. The first stage concentrates on the generation of a causal graph between BP and wearable features from the the perspective of causal inference, so as to identify features that are causally related to BP variations. This stage is pivotal for the identification of novel causal features from the causal graph beyond pulse transit time (PTT). We found these causal features empower better tracking in BP changes compared to PTT. For the second stage, a spatio-temporal GNN (STGNN) is utilized to learn from the causal graph obtained from the first stage. The STGNN can exploit both the spatial information within the causal graph and temporal information from beat-by-beat cardiac signals for refined cuffless continuous BP estimation. We evaluated the proposed method with three datasets that include 305 subjects (102 hypertensive patients) with age ranging from 20-90 and BP at different levels, with the continuous Finapres BP as references. The mean absolute difference (MAD) for estimated systolic blood pressure (SBP) and diastolic blood pressure (DBP) were 3.77 mmHg and 2.52 mmHg, respectively, which outperformed comparison methods. In all cases including subjects with different age groups, while doing various maneuvers that induces BP changes at different levels and with or without hypertension, the proposed CiGNN method demonstrates superior performance for cuffless continuous BP estimation. These findings suggest that the proposed CiGNN is a promising approach in elucidating the causal mechanisms of cuffless BP estimation and can substantially enhance the precision of BP measurement.

A data-driven monitoring scheme for multivariate multimodal data

Multivariate multimodal data, obtained from complex processes with multiple correlated quality characteristics and multiple operating modes, are commonly used in many applications. Online monitoring of multivariate multimodal data has been receiving increasing attention because timely detection of process problems is important. However, existing statistical process monitoring methods are often inadequate for addressing the challenges of multivariate multimodal data monitoring. For example, they may require the process data to be unimodal or identically distributed or may assume the information regarding the multimodal processes to be known a priori. Therefore, a data-driven monitoring scheme is proposed in this paper. First, the Dirichlet process Gaussian mixture model is adopted, which can depend entirely on data characteristics and automatically cluster the data without any prior knowledge. On this basis, an exponentially weighted moving average scheme is then constructed by incorporating the negative log-likelihood statistic. Using thorough simulations and a real example, we analyse and compare the performance of the model estimation and online monitoring for detecting various out-of-control scenarios. Extensive results show that our proposed monitoring scheme is sensitive to not only model parameter shifts but also model order changes. Moreover, the proposed scheme generally performs better and more robustly than existing methods.

Support vector regression-based state of charge estimation for batteries: cloud vs non-cloud

Embracing the potential of cloud technology in the field of electric vehicle advancements, this paper explores the application of support vector regression (SVR) for accurate state of charge (SOC) estimation of lithium-ion batteries in various computational landscapes. This study aims to scrutinize and compare the performance of SOC estimation, with a specific focus on precision, computational efficiency, and execution speed. The investigation is conducted across diverse environments, including a traditional non-cloud setup and two cloud-based platforms-a standard cloud environment employing Amazon web services (AWS) EC2 servers and an enhanced configuration utilizing the MATLAB production server. The investigation not only emphasizes the effectiveness of cloud integration but also provides valuable insights into the strengths and weaknesses of the proposed methodology. The experimental results contribute to a nuanced understanding of the methodology’s performance, shedding light on its potential implications for advancing electric vehicle technologies. This study thus extends its significance beyond technical considerations, providing a broader perspective on its relevance to global electrification initiatives.

A hybrid network of NARX and DS-attention applied for the state estimation of lithium-ion batteries

Monitoring the state of the battery, including the state of charge (SOC) and state of health (SOH), is crucial for ensuring the safety and reliability of electrical equipment. The paper presents a novel hybrid network that combines nonlinear autoregressive model with exogenous inputs (NARX) and DS-attention. The proposed DS-attention method establishes a robust mapping relationship between inputs and outputs, it is a specialized method of the recurrent neural network that enhances the estimation performance by incorporating division function and self-adaptive function into the attention mechanism. The division function is designed to efficiently differentiate between exogenous inputs and state outputs, thereby minimizing the potential for cross-interference between them, and the self-adaptive function optimizes the query features within the attention mechanism. The results demonstrate that the hybrid method of NARX and DS-attention achieves higher prediction accuracy for SOH and SOC estimation of lithium-ion batteries. Specifically, compared with the best-performing similar methods, the proposed hybrid method enhanced the prediction accuracy by 22.9%, 60.7%, and 51.2% across three different SOH datasets, respectively. In terms of long sequence SOC forecasting, the improvements in prediction accuracy under two different working conditions are 82.5% and 60.5%, respectively. The proposed algorithm optimizes the attention mechanism based on the characteristics of the NARX, resulting in higher estimation accuracy for predicting the state of lithium batteries.

State-of-health estimation for lithium-ion batteries based on GWO–VMD-transformer neural network

State-of-health (SOH) estimation of lithium-ion batteries stands as a fundamental metric within the battery management system. It reflects the current level of battery aging and is important for early warning of battery failure to avoid unsafe battery behavior. Therefore, accurate SOH estimation can ensure safe and reliable battery operation. In this paper, the capacity data of the discharge phase are used as the input of the SOH estimation model, and a gray wolf optimization (GWO)–variable mode decomposition (VMD)-transformer-based SOH estimation method for lithium-ion batteries is proposed in a data-driven framework. First, the GWO algorithm is adopted to optimize VMD to decompose the original battery capacity degradation sequence into a series of intrinsic mode functions (IMFs). Then, the transformer is used to separately predict each of these IMFs. Finally, the predicted values of each IMF are integrated to obtain the final prediction of the battery capacity degradation sequence. The model undergoes testing across various datasets, and comparative evaluations are conducted against other data-driven prediction models. The experimental findings underscore the superior SOH estimation performance of the proposed method, along with its robustness when confronted with diverse types of lithium-ion batteries, spanning distinct operational conditions and different aging degrees.

DOA estimation based on a novel shifted coprime array structure

This paper proposes a novel array structure to shift the position of physical sensors through a special coprime array to achieve a high degree of freedom (DOF) with a low mutual coupling effect. The sensor position and consecutive DOF of the closed-form formula of the novel array structure are given. Specifically, when Ω is odd, the consecutive DOF is 3MN+2M+2N−1, and when Ω is even, the consecutive DOF is 3MN+2M+2N+(M−1). The numerical simulation results show that the proposed array structure has the highest continuous DOF and relatively low mutual coupling effects compared to three coprime array structures. Compared with the two nested array structures, although the continuum DOF is close, the mutual coupling effect is much lower than that of the two nested array structures. As a result, a better balance between DOF and mutual coupling effects is achieved, and better DOA estimation performance is obtained.