Signal-to-noise Ratio Of Signal Research Articles

Weak signal detection technique based on Durbin-Watson test and one-bit sampling.

Correlation-based detection techniques are widely used in the weak periodic signal detection field. Traditionally, they are based on extracting the correlation of a weak signal from noise. Considering the impact of a weak signal on the randomness of background noise, this article takes the opposite approach and proposes a weak signal detection technique based on the Durbin-Watson (DW) test and one-bit sampling, detecting the weak signal due to the extent to which the randomness of noise is affected. The randomness of noise is analyzed through the DW test, which is a method for detecting the randomness of data sequences through first-order autocorrelation. One-bit sampling is adopted to reduce the complexity of the sampling circuit and data processing algorithm. The effectiveness of the DW test in the situation of one-bit sampling is demonstrated through simulation and analysis. Simulation results show that the proposed technique is capable of detecting weak sinusoidal and square-wave signals with a signal-to-noise ratio (SNR) above -30 dB, and the frequency or SNR of a weak signal can be further estimated based on mutual constraints. The measured results confirm the capability. In addition, the factors of coherent sampling, noise bandwidth, and comparator threshold that influence the performance of the proposed technique are simulated and discussed in detail.

Two-stage benefits of internal and external noise to enhance early fault detection of machinery by exciting fractional SR

For early mechanical fault diagnosis, stochastic resonance (SR) breaks the previous perception that “noise is useless” and uses internal noise or external noise to enhance weak fault characteristics. Moreover, the fractional-order derivative can reinforce noise-enhanced weak fault detection. However, in the face of extremely low signal-to-noise ratio conditions, the enhancement effect of fractional SR induced by a single excitation of internal noise or external noise is unsatisfactory. To solve the above drawbacks, this paper investigates two-stage benefits of both internal and external noise to enhance early fault detection of machinery by exciting parallel array of fractional SR, and using the signal-to-noise ratio (SNR) to adjust these parameters of fractional SR. Then, two experiments including rolling element bearings and gearboxes were performed to validate it. Experimental results show that the proposed method takes on obvious detection ability for low SNR signals, where the amplitude at the fault characteristic frequency is amplified by beyond 300 times in bearing and gearbox fault experiments than one-stage those. In addition, it is also found that as the noise intensity and the number of iterations increase, the amplitude at the fault characteristic frequency tends to the peak value and then falls into a saturation value. Finally, compared with empirical mode decomposition (EMD), the proposed method can amplify the amplitude at the fault characteristic frequency beyond 1000 times in bearing and gearbox fault experiments. It was concluded that the proposed method has obvious advantages in extracting weak fault characteristics of machinery submerged by strong background noise.

A systematic investigation of detectors for low signal-to-noise ratio EMG signals.

Active participation of stroke survivors during robot-assisted movement therapy is essential for sensorimotor recovery. Robot-assisted therapy contingent on movement intention is an effective way to encourage patients' active engagement. For severely impaired stroke patients with no residual movements, a surface electromyogram (EMG) has been shown to be a viable option for detecting movement intention. Although numerous algorithms for EMG detection exist, the detector with the highest accuracy and lowest latency for low signal-to-noise ratio (SNR) remains unknown. This study, therefore, investigates the performance of 13 existing EMG detection algorithms on simulated low SNR (0dB and -3dB) EMG signals generated using three different EMG signal models: Gaussian, Laplacian, and biophysical model. The detector performance was quantified using the false positive rate (FPR), false negative rate (FNR), and detection latency. Any detector that consistently showed FPR and FNR of no more than 20%, and latency of no more than 50ms, was considered an appropriate detector for use in robot-assisted therapy. The results indicate that the Modified Hodges detector - a simplified version of the threshold-based Hodges detector introduced in the current study - was the most consistent detector across the different signal models and SNRs. It consistently performed for ~90% and ~40% of the tested trials for 0dB and -3dB SNR, respectively. The two statistical detectors (Gaussian and Laplacian Approximate Generalized Likelihood Ratio) and the Fuzzy Entropy detectors have a slightly lower performance than Modified Hodges. Overall, the Modified Hodges, Gaussian and Laplacian Approximate Generalized Likelihood Ratio, and the Fuzzy Entropy detectors were identified as the potential candidates that warrant further investigation with real surface EMG data since they had consistent detection performance on low SNR EMG data.

High-performance deep-sea long-range underwater acoustic communication: Deconvolved conventional beamforming based approach

To address the challenges posed by large propagation loss and severe multipath delay spread, while aiming to improve communication rates in long-range underwater acoustic (UWA) communication, this paper introduces a novel approach: multi-beam (MB) Orthogonal Frequency Division Multiplexing (OFDM) based on deconvolved conventional beamforming (dCv). The proposed method establishes a signal processing framework for UWA channel receivers, wherein received signals corresponding to each angle beam are isolated and concentrated using the dCv technique. Subsequently, the separated signals from each angle undergo maximal-ratio combining (MRC) channel equalization followed by demodulation. Compared to conventional single-beam (SB) processing methods, the proposed approach capitalizes on multipath diversity gain achieved by combining MB outputs originating from various arrival angles. Moreover, employing dCv processing demonstrates superior performance compared to popularly employed beamforming techniques, especially when dealing with paths arriving from closely aligned angles. Additionally, the array beamforming utilized significantly enhances the signal-to-noise ratio (SNR) of each beam output, mitigating the elevated error rates in channel estimation associated with combining low SNR signals received from individual elements. Simulation results utilizing BELLHOP and experimental data from the South China Sea showcase notably improved bit error rate (BER) performance for the proposed method compared to SB equalization, MB-MRC equalization based on conventional beamforming (CBF), minimum variance distortionless response (MVDR), and worst-case performance optimization (WCPO), as well as standard MRC equalization techniques. The proposed receiver achieves error-free decoding results at a communication distance of 80 km with a data rate of 247 bps.

CNN-based noise reduction for multi-channel speech enhancement system with discrete wavelet transform (DWT) preprocessing.

Speech enhancement algorithms are applied in multiple levels of enhancement to improve the quality of speech signals under noisy environments known as multi-channel speech enhancement (MCSE) systems. Numerous existing algorithms are used to filter noise in speech enhancement systems, which are typically employed as a pre-processor to reduce noise and improve speech quality. They may, however, be limited in performing well under low signal-to-noise ratio (SNR) situations. The speech devices are exposed to all kinds of environmental noises which may go up to a high-level frequency of noises. The objective of this research is to conduct a noise reduction experiment for a multi-channel speech enhancement (MCSE) system in stationary and non-stationary environmental noisy situations with varying speech signal SNR levels. The experiments examined the performance of the existing and the proposed MCSE systems for environmental noises in filtering low to high SNRs environmental noises (-10 dB to 20 dB). The experiments were conducted using the AURORA and LibriSpeech datasets, which consist of different types of environmental noises. The existing MCSE (BAV-MCSE) makes use of beamforming, adaptive noise reduction and voice activity detection algorithms (BAV) to filter the noises from speech signals. The proposed MCSE (DWT-CNN-MCSE) system was developed based on discrete wavelet transform (DWT) preprocessing and convolution neural network (CNN) for denoising the input noisy speech signals to improve the performance accuracy. The performance of the existing BAV-MCSE and the proposed DWT-CNN-MCSE were measured using spectrogram analysis and word recognition rate (WRR). It was identified that the existing BAV-MCSE reported the highest WRR at 93.77% for a high SNR (at 20 dB) and 5.64% on average for a low SNR (at -10 dB) for different noises. The proposed DWT-CNN-MCSE system has proven to perform well at a low SNR with WRR of 70.55% and the highest improvement (64.91% WRR) at -10 dB SNR.

Improving power cable partial discharge pattern recognition through gustafson-kessel fuzzy clustering techniques

Existing methods for recognizing partial discharge patterns in power cables do not utilize fuzzy clustering of the discharge signals, resulting in poor quality and low recall and precision of the pattern recognition. To address this, we propose a new approach for partial discharge pattern recognition in cables using Gustafson-Kessel(GK) Fuzzy Clustering. The method involves acquiring signals from a power cable partial discharge monitoring system and then processing the signals with GK fuzzy clustering. The clustered discharge signals are filtered with wavelet packet transforms before input into an improved adaptive resonance theory(ART) neural network for final pattern recognition. Experiments demonstrate the new technique achieves up to 98.7% recall and 85.6% precision for discharge pattern recognition, with discharge signal Signal Noise Ratio(SNR) between 55 dB and 62 dB and maximum recognition accuracy reaching 98%. The proposed fuzzy clustering-based pattern recognition approach significantly enhances partial discharge diagnostics for power cable monitoring.

An FSCEEMD method for downhole weak SNR signal extraction of near-bit attitude parameters

In practice, the near-bit drilling tool confronts with strong vibrations and high-speed rotation. Therein the original signal amplitude of the tool attitude measurements is relatively feeble, and the signal-to-noise ratio (SNR) is exceptionally low. To handle this issue, this paper proposes a weak SNR signal extraction method, frequency selecting complementary ensemble empirical mode decomposition, which is based on ensemble empirical mode decomposition combining with complementary noise and frequency selecting. This method firstly adds different positive and negative pairs of auxiliary white noise to the original near-bit weak SNR signal, secondly adopts empirical mode decomposition on each pair of noise-added signals, then performs ensemble averaging on the obtained multiple sets of intrinsic mode function (IMF) to output more stable IMF of each order and set suitable weights according to designed frequency threshold, and finally reconstructs the original useful signal through weighted summing IMFs. Simulation results show that the extraction accuracy of well inclination angle ranges about ± 0.51°, and the extraction accuracy of tool face angle ranges about ± 1.35°, and meanwhile experimental results are provided compared with other advanced methods, which verifies the effectiveness of our method.

Vectorial‐Optics‐Enabled Multi‐View Non‐Line‐Of‐Sight Imaging with High Signal‐To‐Noise Ratio

AbstractNon‐line‐of‐sight (NLOS) imaging enables the reconstruction of targets beyond the direct line of sight, which has extensive applications across various fields. However, it remains unclear how and to what extent the vectorial nature of light, which plays a critical role in optical imaging and object recognition, can enhance the imaging performance of active NLOS imaging. Here, this work proposes a method that utilizes vector optical fields (VOFs) with arbitrarily designed polarization distributions to realize NLOS imaging in challenging conditions with low signal‐to‐noise ratio (SNR). A generalized VOF reflection model is established at the relay surface for active NLOS imaging to determine the optimal illumination angle and polarization as well as the received polarization. Based on this model, the feasibility of multi‐view NLOS imaging is demonstrated with enhanced SNR of echo signals under proper VOF modulation, surpassing the limitations of conventional single‐view NLOS imaging. Simulations and experiments validate the superior performance of the proposed approach to realize accurate reconstruction and recognition of hidden objects.

Multiscale dilated denoising convolution with channel attention mechanism for micro-seismic signal denoising

Denoising micro-seismic signals is paramount for ensuring reliable data for localizing mining-related seismic events and analyzing the state of rock masses during mining operations. However, micro-seismic signals are commonly contaminated by various types of complex noise, which can hinder micro-seismic accurate P-wave pickup and analysis. In this study, we propose the Multiscale Dilated Convolutional Attention denoising method, referred to as MSDCAN, to eliminate complex noise interference. The MSDCAN denoising model consists of an encoder, an improved attention mechanism, and a decoder. To effectively capture the neighborhood features and multiscale features of the micro-seismic signal, we construct an initial dilated convolution block and a multiscale dilated convolution block in the encoder, and the encoder focuses on extracting the relevant feature information, thus eliminating the noise interference and improving the signal-to-noise ratio (SNR). In addition, the attention mechanism is improved and introduced between the encoder and decoder to emphasize the key features of the micro-seismic signal, thus removing the complex noise and further improving the denoising performance. The MSDCAN denoising model is trained and evaluated using micro-seismic data from Stanford University. Experimental results demonstrate an impressive increase in SNR by 11.237 dB and a reduction in root mean square error (RMSE) by 0.802. Compared to the denoising results of the DeepDenoiser, CNN-denoiser and Neighbor2Neighbor methods, the MSDCAN denoising model outperforms them by enhancing the SNR by 2.589 dB, 1.584 dB and 2dB, respectively, and reducing the RMSE by 0.219, 0.050 and 0.188, respectively. The MSDCAN denoising model presented in this study effectively improves the SNR of micro-seismic signals, offering fresh insights into micro-seismic signal denoising methodologies.

Research on U-Net seismic signal denoising combined with residual dense blocks

Aiming to address the limited noise reduction capabilities of conventional methods for reducing noise in seismic signals, the paper proposes a noise reduction model based on the RDBU-Net network. This model utilizes a residual dense block (RDB) instead of conventional convolutional layers in the U-Net network to enhance the feature extraction capacity for same-band noise, thereby elevating the signal-to-noise ratio (SNR) of seismic signals. The RDBU-Net model is trained, validated, and tested using the global seismic dataset from Stanford University. In comparison with the wavelet threshold method, the denoising RDB model, and the U-Net model, the RDBU-Net model demonstrates an improvement in SNRs by 7.82 dB, 6.13 dB, and 2.9 dB, respectively. Additionally, the root mean square errors are reduced by 0.4812, 0.3736, and 0.1938, and the correlation coefficients are enhanced by 0.3818, 0.2714, and 0.1205. The RDBU-Net model proposed in this study effectively enhances the SNR of seismic signals and offers fresh insights into eliminating noise within the same frequency band of seismic signals.

Eutectogels as a Semisolid Electrolyte for Organic Electrochemical Transistors.

Organic electrochemical transistors (OECTs) are signal transducers offering high amplification, which makes them particularly advantageous for detecting weak biological signals. While OECTs typically operate with aqueous electrolytes, those employing solid-like gels as the dielectric layer can be excellent candidates for constructing wearable electrophysiology probes. Despite their potential, the impact of the gel electrolyte type and composition on the operation of the OECT and the associated device design considerations for optimal performance with a chosen electrolyte have remained ambiguous. In this work, we investigate the influence of three types of gel electrolytes-hydrogels, eutectogels, and iongels, each with varying compositions on the performance of OECTs. Our findings highlight the superiority of the eutectogel electrolyte, which comprises poly(glycerol 1,3-diglycerolate diacrylate) as the polymer matrix and choline chloride in combination with 1,3-propanediol deep eutectic solvent as the ionic component. This eutectogel electrolyte outperforms hydrogel and iongel counterparts of equivalent dimensions, yielding the most favorable transient and steady-state performance for both p-type depletion and p-type/n-type enhancement mode transistors gated with silver/silver chloride (Ag/AgCl). Furthermore, the eutectogel-integrated enhancement mode OECTs exhibit exceptional operational stability, reflected in the absence of signal-to-noise ratio (SNR) variation in the simulated electrocardiogram (ECG) recordings conducted continuously over a period of 5 h, as well as daily measurements spanning 30 days. Eutectogel-based OECTs also exhibit higher ECG signal amplitudes and SNR than their counterparts, utilizing the commercially available hydrogel, which is the most common electrolyte for cutaneous electrodes. These findings underscore the potential of eutectogels as a semisolid electrolyte for OECTs, particularly in applications demanding robust and prolonged physiological signal monitoring.

Theoretical Demonstration of Improved Performance in Multi-Channel Photonic RF Signals Based on Optical Injection Locking of Optical Comb and Array Lasers

Multi-channel radio frequency (RF) signal generation, facilitated by photonic technology, offers significant potential for generating coherent signals with a high frequency and low phase noise, providing multifunctional capabilities across diverse platforms, including RF and photonic systems. Traditional methods for multi-channel photonic RF signal generation typically entail the integration of diverse optical components, such as filters and amplifiers. However, this integration often results in compromises related to power efficiency, cost-effectiveness, and implementation simplicity. To address these challenges, we propose a novel method for generating multi-channel photonic RF signals based on optical injection locking technology. This approach eliminates the necessity for traditional optical components, leading to a substantial enhancement in the performance of photonic RF signals. We present the design of an optical injection locking-based multi-channel photonic RF signal generation schematic and theoretically evaluate its Signal-to-Noise Ratio (SNR) and eye pattern performance for data modulation using the Lumerical INTERCONNECT simulator. Our results reveal a significant 1.3-dB and 3.6-dB enhancement in SNR for 30-GHz and 60-GHz signals, respectively. Furthermore, we observed an improved communication performance, as evidenced by enhanced eye patterns in 3-Gbps data transmission compared to passive photonic RF signal generation methods.

A systematic investigation of detectors for low signal-to-noise ratio EMG signals

Background Active participation of stroke survivors during robot-assisted movement therapy is essential for sensorimotor recovery. Robot-assisted therapy contingent on movement intention is an effective way to encourage patients’ active engagement. For severely impaired stroke patients with no residual movements, a surface electromyogram (EMG) has been shown to be a viable option for detecting movement intention. Although numerous algorithms for EMG detection exist, the detector with the highest accuracy and lowest latency for low signal-to-noise ratio (SNR) remains unknown. Methods This study, therefore, investigates the performance of 13 existing EMG detection algorithms on simulated low SNR (0dB and -3dB) EMG signals generated using three different EMG signal models: Gaussian, Laplacian, and biophysical model. The detector performance was quantified using the false positive rate (FPR), false negative rate (FNR), and detection latency. Any detector that consistently showed FPR and FNR of no more than 20%, and latency of no more than 50ms, was considered an appropriate detector for use in robot-assisted therapy. Results The results indicate that the Modified Hodges detector – a simplified version of the threshold-based Hodges detector introduced in the current study – was the most consistent detector across the different signal models and SNRs. It consistently performed for ~90% and ~40% of the tested trials for 0dB and -3dB SNR, respectively. The two statistical detectors (Gaussian and Laplacian Approximate Generalized Likelihood Ratio) and the Fuzzy Entropy detectors have a slightly lower performance than Modified Hodges. Conclusions Overall, the Modified Hodges, Gaussian and Laplacian Approximate Generalized Likelihood Ratio, and the Fuzzy Entropy detectors were identified as the potential candidates that warrant further investigation with real surface EMG data since they had consistent detection performance on low SNR EMG data.

Microseismic activity characteristics and range evaluation of hydraulic fracturing in coal seam

Accurately evaluating hydraulic fracturing (HF) activity characteristics and impact range is critical to optimizing the HF scheme design. This study proposes an adaptive threshold peak detection method and a wavelet transform-support vector machine (WT-SVM) polarity analysis method based on microseismic (MS) monitoring data from a coal mine in Guizhou, China. The results show that the adaptive threshold peak detection method effectively improves the accuracy and stability of the P-wave initial arrival pickup for low signal-to-noise ratio (SNR) signals. The main frequency band of the MS signal of HF is focused on 150–300 Hz. The improved particle swarm optimization algorithm was used by introducing a random-weight strategy to determine the high-density impact areas of the HF within approximately 20 m on both sides of the fracturing borehole. Density cloud map analysis revealed a small area of the blank zone at the junction of the fracturing section, which provided a basis for fracturing parameter design and scheme optimization. The directional characteristics of the MS waveform were quantitatively described by polarity analysis, where the dip angle distribution ranged from 39° to 49°, the strike angle ranged from 61° to 74°, the slip angle was concentrated near 44°, and the stretch angle ranged from 61° to 74°. The WT-SVM proposed in this study overcame the effects of low SNR signals and the limitations of a small-area station network; 62% of the shear ruptures and 38% of the tensile ruptures of the HF were obtained. Shear cracks were the primary expression of the HF rupture type.

MRI compatibility testing of commercial high intensity focused ultrasound transducers

PurposeThe study aimed to compare the performance of eight commercially available single-element High Intensity Focused Ultrasound (HIFU) transducers in terms of Magnetic Resonance Imaging (MRI) compatibility. MethodsImaging of an agar-based MRI phantom was performed in a 3 T MRI scanner utilizing T2-Weighted Fast Spin Echo (FSE) and Fast low angle shot (FLASH) sequences, which are typically employed for high resolution anatomical imaging and thermometry, respectively. Reference magnitude and phase images of the phantom were compared with images acquired in the presence of each transducer in terms of the signal to noise ratio (SNR), introduced artifacts, and overall image quality. ResultsThe degree of observed artifacts highly differed among the various transducers. The transducer whose backing material included magnetic impurities showed poor performance in the MRI, introducing significant susceptibility artifacts such as geometric distortions and signal void bands. Additionally, it caused the most significant SNR drop. Other transducers were shown to exhibit high level of MRI compatibility as the resulting images closely resembled the reference images with minimal to no apparent artifacts and comparable SNR values. ConclusionsThe study findings may facilitate researchers to select the most suitable transducer for their research, simultaneously avoiding unnecessary testing. The study further provides useful design considerations for MRI compatible transducers.

Selective chronic recording in small nerve fascicles of sciatic nerve with carbon nanotube yarns in rats

Objective. The primary challenge faced in the field of neural rehabilitation engineering is the limited advancement in nerve interface technology, which currently fails to match the mechanical properties of small-diameter nerve fascicles. Novel developments are necessary to enable long-term, chronic recording from a multitude of small fascicles, allowing for the recovery of motor intent and sensory signals. Approach. In this study, we analyze the chronic recording capabilities of carbon nanotube yarn electrodes in the peripheral somatic nervous system. The electrodes were surgically implanted in the sciatic nerve’s three individual fascicles in rats, enabling the recording of neural activity during gait. Signal-to-noise ratio (SNR) and information theory were employed to analyze the data, demonstrating the superior recording capabilities of the electrodes. Flat interface nerve electrode and thin-film longitudinal intrafascicular electrode electrodes were used as a references to assess the results from SNR and information theory analysis. Main results. The electrodes exhibited the ability to record chronic signals with SNRs reaching as high as 15 dB, providing 12 bits of information for the sciatic nerve, a significant improvement over previous methods. Furthermore, the study revealed that the SNR and information content of the neural signals remained consistent over a period of 12 weeks across three different fascicles, indicating the stability of the interface. The signals recorded from these electrodes were also analyzed for selectivity using information theory metrics, which showed an information sharing of approximately 1.4 bits across the fascicles. Significance. The ability to safely and reliably record from multiple fascicles of different nerves simultaneously over extended periods of time holds substantial implications for the field of neural and rehabilitation engineering. This advancement addresses the limitation of current nerve interface technologies and opens up new possibilities for enhancing neural rehabilitation and control.

Coherent DOA Estimation Algorithm with Co-Prime Arrays for Low SNR Signals

The Direction of Arrival (DOA) estimation of coherent signals in co-prime arrays has become a popular research topic. However, traditional spatial smoothing and subspace algorithms fail to perform well under low signal-to-noise ratio (SNR) and small snapshots. To address this issue, we have introduced an Enhanced Spatial Smoothing (ESS) algorithm that utilizes a space-time correlation matrix for de-noising and decoherence. Finally, an Estimating Signal Parameter via Rotational Invariance Techniques (ESPRIT) algorithm is used for DOA estimation. In comparison to other decoherence methods, when the SNR is −8 dB and the number of snapshots is 150, the mean square error (MSE) of the proposed algorithm approaches the Cramér–Rao bound (CRB), the probability of resolution (PoR) can reach over 88%, and, when the angular resolution is greater than 4°, the estimation accuracy can reach over 90%.

Optimized Two‐Color Single‐Molecule Tracking of Fast‐Diffusing Membrane Receptors

AbstractSingle particle tracking (SPT) combined with total internal reflection fluorescence microscopy (TIRFm) is an outstanding approach to decipher mechanisms on the cell membrane at the nanoscale. Multicolor configurations, needed to investigate interactions, are still hindered by several challenges. This work systematically and quantitatively analyzes the impact of necessary elements of SPT‐TIRFm setups on the signal‐to‐noise ratio (SNR), which must be optimized especially in dynamic studies needing minimally invasive dyes for biomolecule labeling. Autofluorescence originating from commonly used optical glass results in the dominant limiting factor in TIRFm, and a cover glass material is tested yielding significant SNR improvements in multichannel TIRFm. Moreover, methodologies are optimized for reducing fluorophore photobleaching in multicolor implementations requiring simultaneous stabilization of multiple dyes. The developed strategies are applied to the fast p75NTR receptors labeled by two fluorophores on the membrane of living cells, achieving reliable, simultaneous two‐color SPT, contrary to configurations using standard cover glasses. This work highlights the importance of optical materials suitable for microscopy and with reduced autofluorescence for increasing sensitivity toward ultimate spatiotemporal resolutions. In particular, the present protocols can pave the way for multicolor super‐resolved localization and tracking of single molecules by TIRFm, greatly expanding the potential of SPT.

GNSS-IR Soil Moisture Inversion Derived from Multi-GNSS and Multi-Frequency Data Accounting for Vegetation Effects

The Global Navigation Satellite System Interferometric Reflectometry (GNSS-IR) technique provides a new remote sensing method that shows great potential for soil moisture detection and vegetation growth, as well as for climate research, water cycle management, and ecological environment monitoring. Considering that the land surface is always covered by vegetation, it is essential to take into account the impacts of vegetation growth when detecting soil moisture (SM). In this paper, based on the GNSS-IR technique, the SM was retrieved from multi-GNSS and multi-frequency data using a machine learning model, accounting for the impact of the vegetation moisture content (VMC). Both the signal-to-noise ratio (SNR) data that was used to retrieve SM and the multipath data that was used to eliminate the vegetation influence were collected from a standard geodetic GNSS station located in Nanjing, China. The normalized microwave reflectance index (NMRI) calculated by multipath data was mapped to a normalized difference vegetation index (NDVI), which was derived from Sentinel-2 data on the Google Earth Engine platform to estimate and eliminate the influence of VMC. Based on the characteristic parameters of amplitude and phase extracted from detrended SNR signals and NDVI derived from multipath data, three machine learning methods, including random forest (RF), multiple linear regression (MLR), and multivariate adaptive regression spline (MARS), were employed for data fusion. The results show that the vegetation effect can be well eliminated using the NMRI method. Comparing MLR and MARS, RF is more suitable for GNSS-IR SM inversion. Furthermore, the SM reversed from amplitude and phase fusion is better than only those from either amplitude fusion or phase fusion. The results prove the feasibility of the proposed method based on a multipath approach to characterize the vegetation effect, as well as the RF model to fuse multi-GNSS and multi-frequency data to retrieve SM with vegetation error-correcting.

A data fusion control-based broad-band and high-dynamic excitation generation method for calibrating the low-frequency linear vibration sensors

The low-frequency vibration sensors are widely used for monitoring vibration in various engineering, and their sensitivity must be calibrated in order to ensure monitoring performance. Currently, the linear shaker is usually applied to provide an excitation for calibrating the sensors. However, this calibration process has the disadvantages of narrow low-frequency range, low signal-to-noise ratio (SNR), and high total harmonic distortion, which severely limits the improvement of calibration ability. In this study, a data fusion control-based broad-band and high-dynamic excitation generation method was proposed, which provides the linear excitation with broad-band high-dynamic for the sensors to achieve high-performance calibration. The simulations demonstrate that the amplitude attenuation and noises can be effectively eliminated in the range of DC (quasi-static)-150 Hz. The comparison experiments with the shaker show that the proposed method has the ability to generate high-dynamic and low-distortion excitation acceleration in 0.01–5 Hz and improve the SNR 40 dB of the excitation. Moreover, the low SNR of the sensor output signal is also well suppressed.