Weiqiao Shan, Yuang Li, Yuhao Zhang, Yingfeng Luo, Chen Xu, Xiaofeng Zhao, Long Meng, Yunfei Lu, Min Zhang, Hao Yang, Tong Xiao, JingBo Zhu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Accurate decoding of human motion intention from surface electromyography (sEMG) signals recorded non-invasively from the skin surface is critical for enabling intuitive control in assistive robotics and human–machine interactions. With the advancement of high-density sEMG (HD-sEMG), neural decoding methods based on motor unit (MU) activity have shown promise due to their potential to capture finely controlled movement information. However, the effects of data segmentation parameters on the decomposition and decoding accuracy remain underexplored. In this study, we systematically investigated how the segmentation length and data size of sEMG signals used for decomposition affect the performance of finger force decoding. Specifically, HD-sEMG signals were recorded from eight human participants during single- and multi-finger isometric force tasks. A neural decoding pipeline was developed for finger force predictions. We first evaluated the impact of four segmentation window lengths (10 s, 20 s, 40 s, and 80 s) on decoding accuracy, and found that a 20-second window was sufficient to ensure accurate decoding, with no additional benefit from using longer segments. Using this setting, we further examined the effect of training data size by comparing decoders trained with different data sizes. Our results showed that using the full training dataset significantly improved decoding performance compared to using only half of the training dataset. These findings offer practical guidelines for optimizing data usage in MU-based motion intention decoding systems.

Abstract The World Wide Web Consortium (W3C) has established standards for decentralized identities (DIDs) and verifiable credentials (VCs). A DID serves as a unique identifier for an entity, while a VC validates specific attributes associated with the DID holder. To prove ownership of credentials, users generate verifiable presentations (VPs). To enhance privacy, the W3C standards advocate for randomizable signatures in VC creation and zero-knowledge proofs for VP generation. However, these standards face a significant limitation: they cannot effectively verify cross-domain credentials while maintaining anonymity. In this paper, we present Anonymous Verifiable Presentations with Extended Usability (AVPEU), a novel framework that addresses this limitation through the introduction of a notary system. At the technical core of AVPEU lies our proposed randomizable message-hiding signature scheme. We provide both a generic construction of AVPEU and specific implementations based on Boneh–Boyen–Shacham, Camenisch–Lysyanskaya, and Pointcheval–Sanders signature. Our experimental results demonstrate the feasibility of these schemes.

Finger force monitoring has become increasingly prevalent in the field of the Internet of Medical Things (IoMT) as a key indicator of muscle strength and health status, facilitating remote rehabilitation and personalized health monitoring. However, existing methods are limited by inaccurate decoding performance or complex procedures when derived in a supervised manner. To address these challenges, we developed a novel unsupervised approach featuring a robust and lightweight neural-drive decoder for multifinger force predictions. High-density surface electromyogram (sEMG) signals were recorded from the finger extensor muscles during isometric finger extension tasks. Each MU was then assigned a probability indicating its association with the target finger, based on its mean firing rates during the activation periods of individual fingers. MUs with probabilities exceeding a predefined threshold were retained for the final force prediction. Our results demonstrate that the neural-drive decoder achieved a computation time of$68.83\pm 13.63$ms, making it suitable for real-time applications. Furthermore, our decoder outperformed the sEMG-amplitude-based approach ($R^{2}$:$0.79\pm 0.039$versus$0.64\pm 0.080$, root mean-square error (RMSE):$4.89\pm 0.73$versus$7.31\pm 1.88$% of maximum force, Pearson correlation coefficient (PCC):$0.87\pm 0.028$versus$0.76\pm 0.06$, and mean absolute error (MAE):$3.86\pm 0.62$versus$6.08\pm 1.51$% of maximum force). The developed neural decoder demonstrated advantages over the state-of-the-art neural decoders in terms of accuracy, training procedures, and practicality. Additionally, our approach exhibited robust performance across various probability thresholds, data sources, and background noise, highlighting its potential for finger force monitoring applications in diverse IoMT scenarios.

Neural or muscular injuries, such as due to amputation, spinal cord injury, and stroke, can affect hand functions, profoundly impacting independent living. This has motivated the advancement of cutting-edge assistive robotic hands. However, unintuitive myoelectric control of these devices remains challenging, which limits the clinical translation of these devices. Accordingly, we developed a robust motor-intent decoding approach to continuously predict the intended fingertip forces of single and multiple fingers in real time. We used population motor neuron discharge activities (i.e., neural drive from brain to spinal cord) decoded from a high-density surface electromyogram (HD-sEMG) signals as the control signals instead of the conventional global sEMG features. To enable real-time neural-drive prediction, we employed a convolutional neural network model to establish the mapping from global HD-sEMG features to finger-specific neural-drive signals, which were then employed for continuous and real-time control of three prosthetic fingers (index, middle, and ring). As a result, the neural-drive-based approach can decode the motor intent of single-finger and multifinger forces with significantly lower force estimation errors than that obtained using the global HD-sEMG-amplitude approach. Besides, the force prediction accuracy was consistent over time and demonstrated strong robustness to signal interference. Our network-based decoder can also achieve better finger isolation with minimal forces predicted in unintended fingers. Our work demonstrates that the accurate and robust finger force control could be achieved through this new decoding approach. The outcomes offer an efficient intent prediction approach that allows users to have intuitive control of prosthetic fingertip forces in a dexterous way.

In industrial control systems, terminal devices in fieldbus networks are vulnerable to physical intrusion attacks, where attackers can directly install external intrusion devices. Currently, the localization capabilities of existing methods for intrusion devices are limited. As the reflection signals in transmitted signals are imperceptible and difficult to extract, many methods focus on actively transmitting pulse signals to locate intrusion devices. In this article, we enhance the reflection signals in transmitted signals by parallel connection of an appropriate resistor with the gateway and propose a localization method based on the collaboration of multiple devices' reflection signals. This method can significantly improve localization precision while reducing the sampling rate and does not occupy communication bandwidth. Experimental results on a real-world controller area network testbed demonstrate that our method can achieve a localization precision of 5 cm when locating intrusion devices under a sampling rate of 50 MS/s.

Accurate control over individual fingers of robotic hands is essential for the progression of human-robot interactions. Accurate prediction of finger forces becomes imperative in this context. The state-of-the-art neural decoders can extract neural signals from surface electromyogram (sEMG) signals. However, these decoders require labeled data for decoder training, which is challenging to obtain in cases such as limb loss and limits decoder generalizability. In our study, we extracted motoneuron firing information by decomposing high-density sEMG signals from both finger flexor and extensor muscles. We assigned each neuron a probability, reflecting its association with the targeted fingers, based on its temporal firing rate distribution. We then employed a probability thresholding and weighting strategy to select and prioritize neurons for finger force predictions. Our results revealed that the unsupervised neural decoder significantly outperformed both the supervised neural decoder and sEMG-amplitude approaches (: 0.74 ± 0.028 vs. 0.70 ± 0.028 vs. 0.63 ± 0.031, root mean square error: 6.74 ± 0.60% vs. 8.41 ± 0.56% vs. 10.33 ± 0.59% of maximum force), thereby offering a promising and practical solution for accurate force controls. Our results also demonstrated high computational efficiency (96.26 ± 24.16 ms), viable for real-time implementations. The outcomes offer an unsupervised decoder with simplified data requirements for decoder training. The decoder boasts enhanced functionality and adaptability in predicting finger flexion and extension forces. In addition, our approach holds promise for broader applications in scenarios where force measurement proves challenging.

Surface electromyogram (sEMG)-based multidegree of freedom (DoF) finger force estimation for the prosthesis and armband applications has obtained increasing attention in the human–machine interface (HMI) field. However, few studies have explored the relation between force estimation performance and coverage area of sEMG electrodes. To address the needs of transradial amputees with varying stump lengths, we investigated the force estimation performance using 16 different electrode layouts covering different forearm areas. Additionally, since the position of the armband affects force estimation performance, we evaluated how model performance varies with the armband worn from the wrist to the elbow. This allows users to select their armband position based on a tradeoff between model performance and practical convenience. We acquired 256-channel forearm sEMG and multi-DoF finger force data from 20 intact subjects. Each subject participated in the experiment on two different days (3 to 25 days apart). Benchmark features were extracted and least squares-based linear finite impulse response models were constructed to estimate the multi-DoF finger force. Both intra-day and interday results were reported for comparison. As a result, the interday regression root mean square error ranged from 8.71±0.80% to 10.98±0.98% of maximum force for prosthesis application and from 9.45±0.79% to 10.82±0.90% of maximum force for armband application. In summary, this work enables users to customize their systems based on their physical conditions and requirements.

Attribute-Based Encryption (ABE) provides fine-grained access control to encrypted data and finds applications in various domains. The practicality of ABE schemes hinges on the balance between security and efficiency. The state-of-the-art adaptive secure ABE scheme, proven to be adaptively secure under standard assumptions (FAME, CCS'17), is less efficient compared to the fastest one (FABEO, CCS'22) which is only proven secure under the Generic Group Model (GGM). These traditional ABE schemes focus solely on message privacy. To address scenarios where attribute value information is also sensitive, Anonymous ABE (A2BE) ensures the privacy of both the message and attributes. However, most A2BE schemes suffer from intricate designs with low efficiency, and the security of the fastest key-policy A2BE (proposed in FEASE, USENIX'24) relies on the GGM.

Concerns about how third parties manage personal information have led to the development of decentralized identities (DIDs) and verifiable credentials (VCs). The World Wide Web Consortium (W3C) working group has been developing standards for DIDs and VCs. In the W3C standards, a DID identifies an entity (a DID holder) and a VC confirms that this DID holder has some associated attributes. A DID holder can obtain many VCs and confirm any number of these VCs to others (verifiers) in verifiable presentations (VPs). In order to keep a holder’s identity and attributes private, it is necessary to achieve anonymous VPs that allows this information to be kept confidential. The W3C working group recommends using randomizable signatures to create VCs with zero-knowledge proofs for this purpose. However, the anonymous VPs provided by the this method are limited that in the real world, credentials in cross domains cannot be universally verified. To overcome this limitation, in this paper, we propose a new scheme, called Verifiable Credentials with anonymous DIDs (VCaDID), which aims to achieve anonymous VPs in cross-domain settings. The main technique in our VCaDID scheme is a ring signature with multiple attributes by hiding a holder’s public key among a ring of holders. In our scheme, we set private keys associated with the holder’s DID and attributes, which allow the holder to anonymously present these credentials in a verifiable way. We also prove that the proposed VCaDID scheme satisfies correctness, anonymity and unforgeability under security assumptions of discrete log and random oracle model. Finally, we implement our scheme to demonstrate its feasibility.

Surface electromyography (sEMG)-based gesture recognition can achieve high intra-session performance. However, the inter-session performance of gesture recognition decreases sharply due to the shift in data distribution. Therefore, developing a robust model to minimize the data distribution difference is crucial to improving the user experience. In this work, based on the inter-session gesture recognition task, we propose a novel algorithm called locality preserving and maximum margin criterion (LPMM). The LPMM algorithm integrates three main modules, including domain alignment, pseudo-label selection, and iteration result selection. Domain alignment is designed to preserve the neighborhood structure of the feature and minimize the overlap of different classes. The pseudo-label selection and iteration result selection can avoid the decrease in accuracy caused by mislabeled samples. The proposed algorithm was evaluated on two of the most widely used EMG databases. It achieves a mean accuracy of 98.46% and 71.64%, respectively, which is superior to state-of-the-art domain adaptation methods.

Gaze estimation based on electrooculograms (EOGs) has been widely explored. However, the inter-subject variability of EOGs still leaves a significant challenge for practical applications. It contributes to performance degradation when handling inter-subject issues. In this paper, an unsupervised transfer learning approach with an adaptive reweighting and resampling (ARR) strategy to fully consider individual variability is proposed for EOG-based gaze angle estimation. It allows quantifying domain shifts by leveraging the source-target similarities, reweighting and resampling the source data to retain relevant instances and disregard irrelevant instances during adaptation. Specifically, our proposed methodology first assesses the domain shifts via decomposing transformation matrices, which are estimated between the training subjects (denoted as multi-source domains) and the test subject (denoted as target domain). Then, the multi-domain shifts are assigned as weighted indicators to resample the multi-source domains for model training. Comparative experiments with several prevailing transfer learning methods including CORrelation ALignment (CORAL), Geodesic Flow Kernel (GFK), Joint Distribution Adaptation (JDA), Transfer component analysis (TCA), and Balanced distribution adaption (BDA) using two different normalization processes were conducted on a realistic scenario across 18 subjects. Experimental results demonstrate that the ARR strategy can significantly improve performance (mean absolute error (MAE) reduction: 7.0%, root mean square error (RMSE) reduction: 6.3%), outperforming the prevailing methods. Besides, the impacts of data diversity and data size on ARR strategy are further investigated. It exhibits that data size is more important than data diversity for EOG-based gaze angle estimation, and also presents the benefits of the ARR strategy for dealing with practical scenarios.

Due to the limitations of large size, non-random movement, high cost and complex equipment, the video recognition and pressure testing platform used for gait monitoring can only be operated in laboratory environment. Some insole devices invented in recent years are basically equipped with other modules, making them inconvenient to wear. To solve the above problems, this paper proposes intelligent flexible pressure insoles with low cost and high resolution, as well as convenience and comfortness. The insole collects the plantar pressure distribution and motion data of daily walking through 12 x 4 flexible pressure array and accelerometer sensor, and then wirelessly transmits the data to the upper computer through Bluetooth. Material property tests were conducted and the results showed robustness and stability of the pressure sensor. The feasibility of the insole for gait monitoring is verified through preliminary walking experiments. The results show that of the proposed insole is characterized by low cost, high resolution, portability and comfortableness. Additionally, it can be used directly into the shoe without any external module, regardless of site and environment restrictions. The insole can meet the needs of daily gait monitoring and provide more quantitative reference data for gait rehabilitation.

With massive data collected in Internet of Things (IoT)-based smart environment, improving privacy preservation via client verification and identification is crucial. Surface electromyogram (sEMG) has emerged as a cancelable neuromuscular biometric trait, which makes up the noncancelability flaw of the traditional face and fingerprint biometrics. Current studies are in the proof-of-concept stage. In-depth studies to find the optimal solution to decode sEMG biometrics with excellent cross-day performance are very scarce. For neurophysiological biometrics, the permanence across time is a crucial factor. Our work aims to optimize the cross-day performance of the sEMG biometric decoder. We systematically evaluated the performance of 28 hand gestures to generate sEMG, 55 temporal–spectral–spatial features to represent sEMG, 9 distance measures and 9 classifiers to make decisions. Both biometric verification and identification were investigated in rigorous cross-day validations. Results show that the optimal combination of ≥ 10 temporal–spectral–spatial features achieved the best cross-day performance with city-block distance and support vector machine (SVM) applied. EMG generated by middle finger extension and hand close is preferred as biometric tokens. Using the optimized decoder, a cross-day identification accuracy of 88.75% and verification error rate of 9.85% were achieved. The verification error rate could be further reduced to 2.45% if impostors input sEMG under random gestures. Moreover, our work proved the reliability of sEMG biometrics even under muscle fatigue for the first time. This is also the first study to systematically evaluate the cross-day performance of different components in sEMG biometric decoding systems, serving as a technique-screening tool for future studies.

Sleep posture, which affects the quality of sleep and could lead to medical conditions, such as pressure ulcers, is a key metric for sleep analysis in Internet of Medical Things (IoMT). In this article, a real-time and low-cost smart mat system for sleep posture recognition based on frequency channel selection is proposed. The system can recognize postures unobtrusively with a dense flexible sensor array. In addition, to enable real-time recognition with a relatively low-cost STM32 processor system, a lightweight algorithm that includes frequency channel selection, model pretraining, and real-time classification is proposed. Through a series of short-term and overnight experiments with 21 subjects, the feasibility and reliability of the proposed system were evaluated. Experimental results show that the accuracy of the short-term experiment is up to 95.43% and of the overnight experiment is up to 86.80% for four posture categories (supine, prone, right, and left) classification. The model size is just 56 kB which is much smaller than other methods. The runtime of the complete algorithm is about 6 ms with a low-power STM32 embedded system, which shows the system’s ability to provide real-time posture recognition. As an edge device, the proposed system could lead to the development of fast, convenient, and low-cost sleep posture recognition products for IoMT.

In this paper we propose a novel feature, called directional entropy feature (DEF), to improve the performance of human detection under complicated background in images. DEF describe the regularity of region by computing the entropy value of edge pointspsila spatial distribution in specific direction, so DEF has the discriminating power for regular and random pattern. We combine histogram of oriented gradient (HOG) feature with DEF to construct a human detection classifier to test DEFpsilas performance. Experimental results show that DEF can help HOG to decreases false alarms caused by random complicated and rigid shaped background.