Pub Date : 2026-03-04DOI: 10.1109/TIM.2026.3670973
{"title":"2026 Index IEEE Transactions on Instrumentation and Measurement Vol. 74","authors":"","doi":"10.1109/TIM.2026.3670973","DOIUrl":"https://doi.org/10.1109/TIM.2026.3670973","url":null,"abstract":"","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-951"},"PeriodicalIF":5.9,"publicationDate":"2026-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11421662","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-23DOI: 10.1109/TIM.2026.3667341
Wenjingping Zhang;Hongpeng Wang;Huan Liu;Haobin Dong;Zheng Liu;Xiangyun Hu
To address the challenges of extracting weak, rapidly decaying, and noise-corrupted free induction decay (FID) signals from overhauser magnetometers, this study introduces the rank-sequential truncated tensor decomposition (RSTD) method for effective noise suppression and high signal-to-noise ratio (SNR) signal recovery in high-noise environments. The proposed approach involves acquiring multichannel FID signals using an equal-delay strategy and organizing them into a third-order tensor through data segmentation. Subsequently, the tensor undergoes a higher order singular value decomposition (SVD) combined with matrix rank decomposition, following a predefined modal decomposition sequence to enhance SNR. Comparative evaluations against conventional techniques, including SVD, principal component analysis (PCA)-assisted SVD (C-PCASVD), and multilinear SVD (MLSVD), demonstrate that RSTD achieves a signal-to-noise improvement ratio (SNIR) of 40 dB across initial SNRs ranging from −20 to 0 dB. The method matches MLSVD in denoising efficacy while outperforming SVD and C-PCASVD by more than 10 dB in SNIR. In particular, the proposed method completes processing within 100 ms, representing more than 50% improvement in computational efficiency over MLSVD. These results confirm that the proposed method substantially improves real-time processing capabilities without compromising the noise reduction performance.
{"title":"A Novel End-to-End Framework for Low-SNR FID Signal Denoising via Rank-Sequential Truncated Tensor Decomposition","authors":"Wenjingping Zhang;Hongpeng Wang;Huan Liu;Haobin Dong;Zheng Liu;Xiangyun Hu","doi":"10.1109/TIM.2026.3667341","DOIUrl":"https://doi.org/10.1109/TIM.2026.3667341","url":null,"abstract":"To address the challenges of extracting weak, rapidly decaying, and noise-corrupted free induction decay (FID) signals from overhauser magnetometers, this study introduces the rank-sequential truncated tensor decomposition (RSTD) method for effective noise suppression and high signal-to-noise ratio (SNR) signal recovery in high-noise environments. The proposed approach involves acquiring multichannel FID signals using an equal-delay strategy and organizing them into a third-order tensor through data segmentation. Subsequently, the tensor undergoes a higher order singular value decomposition (SVD) combined with matrix rank decomposition, following a predefined modal decomposition sequence to enhance SNR. Comparative evaluations against conventional techniques, including SVD, principal component analysis (PCA)-assisted SVD (C-PCASVD), and multilinear SVD (MLSVD), demonstrate that RSTD achieves a signal-to-noise improvement ratio (SNIR) of 40 dB across initial SNRs ranging from −20 to 0 dB. The method matches MLSVD in denoising efficacy while outperforming SVD and C-PCASVD by more than 10 dB in SNIR. In particular, the proposed method completes processing within 100 ms, representing more than 50% improvement in computational efficiency over MLSVD. These results confirm that the proposed method substantially improves real-time processing capabilities without compromising the noise reduction performance.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-9"},"PeriodicalIF":5.9,"publicationDate":"2026-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-20DOI: 10.1109/TIM.2026.3662422
M. Saad Shakeel;Kun Liu;Xiaochuan Liao;Wenxiong Kang
Presents corrections to the paper, (TAG: A Temporal Attentive Gait Network for Cross-View Gait Recognition).
对论文（TAG：一种用于横视步态识别的时间关注步态网络）进行了修正。
{"title":"Corrections to “TAG: A Temporal Attentive Gait Network for Cross-View Gait Recognition”","authors":"M. Saad Shakeel;Kun Liu;Xiaochuan Liao;Wenxiong Kang","doi":"10.1109/TIM.2026.3662422","DOIUrl":"https://doi.org/10.1109/TIM.2026.3662422","url":null,"abstract":"Presents corrections to the paper, (TAG: A Temporal Attentive Gait Network for Cross-View Gait Recognition).","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-1"},"PeriodicalIF":5.9,"publicationDate":"2026-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11404361","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146223652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-13DOI: 10.1109/TIM.2026.3664552
Zhen Sun;Tengfei Wang;Zhenjie Wang;Mingquan Lu
Ultrashort baseline (USBL) underwater positioning techniques are widely used to provide location-based services for unmanned underwater vehicles (UUVs). Angle misalignment is a primary source of error that reduces the accuracy of USBL positioning. The conventional least-squares (LS) alignment method assumes that the position of the seafloor transponder, which is derived from global navigation satellite system/acoustic (GNSS/A) underwater positioning, can be determined without error and that the alignment observations are independent. However, errors inevitably occur in the estimation of the transponder position by GNSS/A positioning, and the precision varies in three dimensions. Owing to geometric constraints, the observations are statistically correlated rather than independent. Ignoring these assumptions significantly reduces the precision of angle misalignment estimation. In this contribution, we propose an adaptive joint alignment (AJA) method, where both angle misalignment and the position of the seafloor transponder are treated as unknown parameters. The raw USBL data, including the acoustic range and bearing angle, are treated as observations, whereas the position of the seafloor transponder is treated as a virtual observation. An adaptive weighting factor derived from the maximum likelihood estimator is introduced to balance the contributions of the real and virtual observations to parameter estimation. Simulation results show that the proposed method improves angle misalignment estimation accuracy by 61.61% and 48.06% relative to the LS and two-step nonlinear Gauss–Markov (TSNGM) methods, respectively. Sea-trial results, evaluated in terms of USBL positioning accuracy, show that the AJA method achieves improvements of 66.02%–68.15% over the uncalibrated USBL system and 8.07%–10.08% over the LS method.
{"title":"An Adaptive Joint Alignment Method of Angle Misalignment and Seafloor Transponder for Ultrashort Baseline Underwater Positioning","authors":"Zhen Sun;Tengfei Wang;Zhenjie Wang;Mingquan Lu","doi":"10.1109/TIM.2026.3664552","DOIUrl":"https://doi.org/10.1109/TIM.2026.3664552","url":null,"abstract":"Ultrashort baseline (USBL) underwater positioning techniques are widely used to provide location-based services for unmanned underwater vehicles (UUVs). Angle misalignment is a primary source of error that reduces the accuracy of USBL positioning. The conventional least-squares (LS) alignment method assumes that the position of the seafloor transponder, which is derived from global navigation satellite system/acoustic (GNSS/A) underwater positioning, can be determined without error and that the alignment observations are independent. However, errors inevitably occur in the estimation of the transponder position by GNSS/A positioning, and the precision varies in three dimensions. Owing to geometric constraints, the observations are statistically correlated rather than independent. Ignoring these assumptions significantly reduces the precision of angle misalignment estimation. In this contribution, we propose an adaptive joint alignment (AJA) method, where both angle misalignment and the position of the seafloor transponder are treated as unknown parameters. The raw USBL data, including the acoustic range and bearing angle, are treated as observations, whereas the position of the seafloor transponder is treated as a virtual observation. An adaptive weighting factor derived from the maximum likelihood estimator is introduced to balance the contributions of the real and virtual observations to parameter estimation. Simulation results show that the proposed method improves angle misalignment estimation accuracy by 61.61% and 48.06% relative to the LS and two-step nonlinear Gauss–Markov (TSNGM) methods, respectively. Sea-trial results, evaluated in terms of USBL positioning accuracy, show that the AJA method achieves improvements of 66.02%–68.15% over the uncalibrated USBL system and 8.07%–10.08% over the LS method.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-15"},"PeriodicalIF":5.9,"publicationDate":"2026-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-13DOI: 10.1109/TIM.2026.3664370
Xuebo Zhang;Peixuan Yang;Junfeng Wang;Jiahua Zhu
The range-Doppler (RD) technique is extensively utilized by the advanced imaging sonar that features the resolution regardless of range and frequency. Unluckily, the focusing performances are mainly limited by two factors when classical RD techniques are extended to the multireceiver synthetic aperture sonar (SAS). The primary factor lies in the development of the spectrum by using the approximation of range, which brings the range error. To design the RD algorithm, the Taylor approximation of the spectrum, which must be carried out, brings extra phase error. Both errors would limit the focusing performance of a multireceiver SAS system. To strengthen the performance, a new RD technique is offered in this work. In accordance with the one-by-one mapping between the sonar bearing angle and azimuth Doppler frequency, an approximated range, including the monostatic equivalent range and bearing angle-dependent range error, is proposed. The presented model shows much higher accuracy than traditional range models. After correcting the bearing angle-dependent error via the multiplication and interpolation, the datasets are viewed as the monostatic equivalent signal. Then, an advanced RD algorithm correcting the phase error of the approximated spectrum is provided. In comparison with conventional approaches, the proposed approach is efficient and has the potential to improve the focusing performance. The computer simulations and experiments based on SAS datasets further indicate that the proposed RD technique has the ability to achieve much more outstanding images than conventional approaches.
{"title":"Focus Improvement of Multireceiver SAS Based on Range-Doppler Algorithm","authors":"Xuebo Zhang;Peixuan Yang;Junfeng Wang;Jiahua Zhu","doi":"10.1109/TIM.2026.3664370","DOIUrl":"https://doi.org/10.1109/TIM.2026.3664370","url":null,"abstract":"The range-Doppler (RD) technique is extensively utilized by the advanced imaging sonar that features the resolution regardless of range and frequency. Unluckily, the focusing performances are mainly limited by two factors when classical RD techniques are extended to the multireceiver synthetic aperture sonar (SAS). The primary factor lies in the development of the spectrum by using the approximation of range, which brings the range error. To design the RD algorithm, the Taylor approximation of the spectrum, which must be carried out, brings extra phase error. Both errors would limit the focusing performance of a multireceiver SAS system. To strengthen the performance, a new RD technique is offered in this work. In accordance with the one-by-one mapping between the sonar bearing angle and azimuth Doppler frequency, an approximated range, including the monostatic equivalent range and bearing angle-dependent range error, is proposed. The presented model shows much higher accuracy than traditional range models. After correcting the bearing angle-dependent error via the multiplication and interpolation, the datasets are viewed as the monostatic equivalent signal. Then, an advanced RD algorithm correcting the phase error of the approximated spectrum is provided. In comparison with conventional approaches, the proposed approach is efficient and has the potential to improve the focusing performance. The computer simulations and experiments based on SAS datasets further indicate that the proposed RD technique has the ability to achieve much more outstanding images than conventional approaches.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-14"},"PeriodicalIF":5.9,"publicationDate":"2026-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nondestructive testing (NDT) with ultrasonic imaging is pervasive in industry. Synthetic Aperture Focusing Technique (SAFT) can significantly improve spatial resolution and imaging contrast, but requires intensive computational resources and storage. Additionally, SAFT is typically implemented with delay-and-sum (DAS) beamforming, which suffers from limited resolution and inadequate interference rejection. This article reports $p$ th coherence factor weighted delay and sum beamforming (pCFwDAS), a nonlinear approach that integrates $p$ th root algebra with coherence factor weighting. Data was collected with an aluminum test block and a Verasonics Vantage 128 system equipped with a 32-element linear array transducer (5 MHz frequency) to evaluate imaging performance. A sparse implementation of pCFwDAS was performed with 16 and 8 elements to reduce computational overhead along with graphics processing unit (GPU) implementation. The pCFwDAS approach was compared to $p$ th root DAS and standard DAS approaches. The results shown enhanced contrast-to-noise ratio (CNR) and reduced sidelobe artifacts with pCFwDAS for both the full array and sparse array configurations. With the fully populated array, CNR enhancements of up to $14.7~pm ~1.3$ dB were observed, and the side lobes were suppressed by up to $33.3~pm ~5.7$ dB. For sparse array imaging, the CNR was enhanced up to $13.9~pm ~3.1$ dB, and side lobes were reduced by up to $18.3~pm ~3.1$ dB. These findings demonstrate the potential of pCFwDAS beamforming to enhance SAFT and enable its implementation with limited computational resources.
{"title":"Sparse Array Synthetic Aperture Focusing With pth Coherence Factor Weighted Delay and Sum Beamforming for Nondestructive Testing","authors":"Abhinav Kumar Singh;Pranaba Kumar Mishro;Shaswata Das;Ruchika Dhawan;Arjun Anand Mallya;Himanshu Shekhar","doi":"10.1109/TIM.2026.3661698","DOIUrl":"https://doi.org/10.1109/TIM.2026.3661698","url":null,"abstract":"Nondestructive testing (NDT) with ultrasonic imaging is pervasive in industry. Synthetic Aperture Focusing Technique (SAFT) can significantly improve spatial resolution and imaging contrast, but requires intensive computational resources and storage. Additionally, SAFT is typically implemented with delay-and-sum (DAS) beamforming, which suffers from limited resolution and inadequate interference rejection. This article reports <inline-formula> <tex-math>$p$ </tex-math></inline-formula>th coherence factor weighted delay and sum beamforming (pCFwDAS), a nonlinear approach that integrates <inline-formula> <tex-math>$p$ </tex-math></inline-formula>th root algebra with coherence factor weighting. Data was collected with an aluminum test block and a Verasonics Vantage 128 system equipped with a 32-element linear array transducer (5 MHz frequency) to evaluate imaging performance. A sparse implementation of pCFwDAS was performed with 16 and 8 elements to reduce computational overhead along with graphics processing unit (GPU) implementation. The pCFwDAS approach was compared to <inline-formula> <tex-math>$p$ </tex-math></inline-formula>th root DAS and standard DAS approaches. The results shown enhanced contrast-to-noise ratio (CNR) and reduced sidelobe artifacts with pCFwDAS for both the full array and sparse array configurations. With the fully populated array, CNR enhancements of up to <inline-formula> <tex-math>$14.7~pm ~1.3$ </tex-math></inline-formula> dB were observed, and the side lobes were suppressed by up to <inline-formula> <tex-math>$33.3~pm ~5.7$ </tex-math></inline-formula> dB. For sparse array imaging, the CNR was enhanced up to <inline-formula> <tex-math>$13.9~pm ~3.1$ </tex-math></inline-formula> dB, and side lobes were reduced by up to <inline-formula> <tex-math>$18.3~pm ~3.1$ </tex-math></inline-formula> dB. These findings demonstrate the potential of pCFwDAS beamforming to enhance SAFT and enable its implementation with limited computational resources.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-9"},"PeriodicalIF":5.9,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cilium-based micro-electromechanical system (MEMS) vector hydrophones are critical in underwater acoustic detection, but their performance is highly sensitive to cilium geometry. Traditional trial-and-error methods are inefficient under complex constraints. This study presents an optimization approach using the constrained optimization by linear approximation (COBYLA) algorithm, combined with multiphysics finite element analysis, to improve sensitivity while precisely controlling the first-order characteristic frequency (FOCF). The cilium is parameterized with a seven-node interpolation curve, and an acoustic-structural coupled model is built in COMSOL. Under a 1000-Hz FOCF constraint, the maximum stress at the beam (MSB) is selected as the optimization objective. COBYLA enables automatic iterative shape optimization with real-time performance feedback. Simulation results show a frequency deviation of less than 0.17% and a 21.3 dB@1-kHz improvement in sensitivity. A prototype fabricated using MEMS and 3-D printing technologies was tested in a standing wave tube, yielding a measured linear bandwidth of the sensor of 1000 Hz and a sensitivity of −175.8 dB@1 kHz re 1 V/$mu $ Pa, matching simulation predictions. Meanwhile, the fabricated sensor demonstrated stable operation after withstanding a 100-g shock and a hydrostatic pressure of 7 MPa. This work demonstrates a high-efficiency, high-accuracy framework for MEMS hydrophone design, offering enhanced performance in complex underwater acoustic environments.
{"title":"Design of a SpinCilium MEMS Vector Hydrophone Driven by the COBYLA Algorithm","authors":"Zhengyu Bai;Yujia Chai;Guojun Zhang;Zhiyuan Cheng;Shilin Liu;Yanan Geng;Jie Zhang;Li Jia;Wenshu Dai;Yuhua Yang;Renxin Wang;Wendong Zhang","doi":"10.1109/TIM.2026.3661687","DOIUrl":"https://doi.org/10.1109/TIM.2026.3661687","url":null,"abstract":"Cilium-based micro-electromechanical system (MEMS) vector hydrophones are critical in underwater acoustic detection, but their performance is highly sensitive to cilium geometry. Traditional trial-and-error methods are inefficient under complex constraints. This study presents an optimization approach using the constrained optimization by linear approximation (COBYLA) algorithm, combined with multiphysics finite element analysis, to improve sensitivity while precisely controlling the first-order characteristic frequency (FOCF). The cilium is parameterized with a seven-node interpolation curve, and an acoustic-structural coupled model is built in COMSOL. Under a 1000-Hz FOCF constraint, the maximum stress at the beam (MSB) is selected as the optimization objective. COBYLA enables automatic iterative shape optimization with real-time performance feedback. Simulation results show a frequency deviation of less than 0.17% and a 21.3 dB@1-kHz improvement in sensitivity. A prototype fabricated using MEMS and 3-D printing technologies was tested in a standing wave tube, yielding a measured linear bandwidth of the sensor of 1000 Hz and a sensitivity of −175.8 dB@1 kHz re 1 V/<inline-formula> <tex-math>$mu $ </tex-math></inline-formula>Pa, matching simulation predictions. Meanwhile, the fabricated sensor demonstrated stable operation after withstanding a 100-g shock and a hydrostatic pressure of 7 MPa. This work demonstrates a high-efficiency, high-accuracy framework for MEMS hydrophone design, offering enhanced performance in complex underwater acoustic environments.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-14"},"PeriodicalIF":5.9,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146223687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-03DOI: 10.1109/TIM.2026.3653090
{"title":"2025 List of Reviewers","authors":"","doi":"10.1109/TIM.2026.3653090","DOIUrl":"https://doi.org/10.1109/TIM.2026.3653090","url":null,"abstract":"","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-153"},"PeriodicalIF":5.9,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11370739","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-03DOI: 10.1109/TIM.2026.3659592
Hui Li;Dajun Sun;Zhongyi Cao
Doppler sonar estimates the velocity by extracting Doppler information from bottom reverberation, serving as a critical component for high-precision autonomous navigation of underwater vehicles. The configuration of inclined beams results in nonuniform reverberation energy within the illuminated footprint, leading to dynamic velocity bias related to propagation-scattering process. This study establishes a quantitative relationship between the unique transition characteristics of backscattered reverberation in narrow beam and the spatial pattern of reverberation energy. The strictly monotonic relationship with the energy ratios of transitional zones is used to invert the “reverberation energy slope,” quantifying dynamic proportional error caused by channel and achieving velocity bias compensation. The experiment demonstrates that after compensation, the temporal oscillation of velocity bias is significantly narrowed, with the range reduced by 29.94% and the variance by 73.50% within 5 h (over 60-km voyage). The method effectively mitigates the dynamic variation of velocity biases caused by channel differences, enhancing the accuracy of Doppler sonar.
{"title":"A Bias Compensation Method for Doppler Sonar in Underwater Autonomous Navigation","authors":"Hui Li;Dajun Sun;Zhongyi Cao","doi":"10.1109/TIM.2026.3659592","DOIUrl":"https://doi.org/10.1109/TIM.2026.3659592","url":null,"abstract":"Doppler sonar estimates the velocity by extracting Doppler information from bottom reverberation, serving as a critical component for high-precision autonomous navigation of underwater vehicles. The configuration of inclined beams results in nonuniform reverberation energy within the illuminated footprint, leading to dynamic velocity bias related to propagation-scattering process. This study establishes a quantitative relationship between the unique transition characteristics of backscattered reverberation in narrow beam and the spatial pattern of reverberation energy. The strictly monotonic relationship with the energy ratios of transitional zones is used to invert the “reverberation energy slope,” quantifying dynamic proportional error caused by channel and achieving velocity bias compensation. The experiment demonstrates that after compensation, the temporal oscillation of velocity bias is significantly narrowed, with the range reduced by 29.94% and the variance by 73.50% within 5 h (over 60-km voyage). The method effectively mitigates the dynamic variation of velocity biases caused by channel differences, enhancing the accuracy of Doppler sonar.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-15"},"PeriodicalIF":5.9,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146223557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-03DOI: 10.1109/TIM.2025.3636680
Weiguo Hu;Yabin Zhang;Bowen Xu;Mingyu Dong;Tao Liu;Tianxu Hao;Min Liu
Various disturbances and uncertainties existing in actual industrial production environments can degrade the measurement accuracy of soft sensors based on single-value deterministic estimation. In addition, outliers caused by operational errors or recording mistakes may affect the generalization ability of soft sensors. Inspired by this, two Bayesian twin extreme learning machines based on symmetric skewed distributions, BTELM-ALD and BTELM-STD, are proposed. Both soft sensing methods perform parameter learning in a Bayesian framework and train a pair of twin models based on combined weights to provide estimation intervals for key indicators. They use skewed heavy-tailed distributions to model the residuals, which enhances robustness to outliers. BTELM-ALD uses an asymmetric Laplace distribution (ALD) instead of Gaussian distribution and constructs a pair of twin models based on the combined weights ($p$ , $1-p$ ). The introduction of suitable surrogate functions makes the posterior distribution and marginal likelihood easy to solve. In BTELM-STD, a univariate skewed t-distribution (STD) is presented and written as a hierarchical representation. The corresponding twin models are constructed based on the combined weights ($s$ , $-s$ ), and then variational inference and the Newton method are used to optimize the parameters. Experimental results on several cases including an actual PTA oxidation process illustrate the validity and advantages of our proposed methods.
{"title":"Two Industrial Twin Soft Sensing Methods With Estimation Interval Based on Symmetric Skewed Distributions and Combined Weights","authors":"Weiguo Hu;Yabin Zhang;Bowen Xu;Mingyu Dong;Tao Liu;Tianxu Hao;Min Liu","doi":"10.1109/TIM.2025.3636680","DOIUrl":"https://doi.org/10.1109/TIM.2025.3636680","url":null,"abstract":"Various disturbances and uncertainties existing in actual industrial production environments can degrade the measurement accuracy of soft sensors based on single-value deterministic estimation. In addition, outliers caused by operational errors or recording mistakes may affect the generalization ability of soft sensors. Inspired by this, two Bayesian twin extreme learning machines based on symmetric skewed distributions, BTELM-ALD and BTELM-STD, are proposed. Both soft sensing methods perform parameter learning in a Bayesian framework and train a pair of twin models based on combined weights to provide estimation intervals for key indicators. They use skewed heavy-tailed distributions to model the residuals, which enhances robustness to outliers. BTELM-ALD uses an asymmetric Laplace distribution (ALD) instead of Gaussian distribution and constructs a pair of twin models based on the combined weights (<inline-formula> <tex-math>$p$ </tex-math></inline-formula>, <inline-formula> <tex-math>$1-p$ </tex-math></inline-formula>). The introduction of suitable surrogate functions makes the posterior distribution and marginal likelihood easy to solve. In BTELM-STD, a univariate skewed t-distribution (STD) is presented and written as a hierarchical representation. The corresponding twin models are constructed based on the combined weights (<inline-formula> <tex-math>$s$ </tex-math></inline-formula>, <inline-formula> <tex-math>$-s$ </tex-math></inline-formula>), and then variational inference and the Newton method are used to optimize the parameters. Experimental results on several cases including an actual PTA oxidation process illustrate the validity and advantages of our proposed methods.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"75 ","pages":"1-11"},"PeriodicalIF":5.9,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146169933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：