Sergio Guillén, Lorenzo J. Tardón, Ana M. Barbancho, Isabel Barbancho
Electroencephalography (EEG) has proven to be a powerful tool for decoding neural activity in real-life scenarios, including the perception and processing of emotions elicited by music. Among the various musical features that modulate these responses, musical mode and melodic direction (or contour) play a central role. This work proposes a novel EEG feature extraction framework based on Empirical Mode Decomposition (EMD) to decompose EEG signals recorded during active piano performance for the detection of these two musical features. Energy-based descriptors were computed from the first four Intrinsic Mode Functions (IMFs) obtained through EMD and used to train a Wide Neural Network classifier. The model performance was benchmarked against several standard classifiers commonly employed in EEG classification tasks. Two experimental scenarios were explored: intra-subject and intersubject. In the former case, the proposed framework achieved average accuracies of 84%, whereas in the inter-subject case, accuracies reached 81% on average using the first IMF, with AUC-ROC values of 0.9 on average in both cases. A frequency-domain analysis revealed a strong correspondence between the IMFs and conventional EEG sub-bands, except for certain specific frequency components that were not fully represented. Predominant activations were observed over the frontal, temporal, and parietooccipital regions, areas consistently associated with music processing. Furthermore, the proposed method demonstrated robustness to handedness, successfully classifying EEG data across participants regardless of their dominant hand. These findings highlight the potential of EMD-based feature extraction for identifying music-related neural patterns and contribute to advancing research in Music Information Retrieval (MIR), affective neuroscience, and the broader understanding of the neural encoding of musical structure.
Zihang Song, Matteo Zecchin, Bipin Rajendran, Petar Popovski, Osvaldo Simeone
Multiple-input multiple-output (MIMO) turbo receivers iteratively exchange information between a soft-input soft-output detector and a decoder. Traditional implementations rely on pilot-based channel estimates, which can be poor in the case of short pilot sequences or mismatched channel models, resulting in significant performance degradation. To address this limitation, we introduce the EXTrinsic In-Context Learning Network (EXT-ICLNet), a Transformer-based detector that takes the pilots as in-context examples and is trained to approximate the symbol extrinsic posterior distribution given the pilots, without explicit channel estimation. EXT-ICLNet is thus compatible with existing turbo receivers, as it implements a log-domain prior skip connection, ensuring that the network learns only the extrinsic information. Numerical results for a $4\times4$ MIMO system with 16-QAM and a quantized front end show that EXT-ICLNet outperforms baselines based on explicit channel state information (CSI) estimation at low-to-medium signal-to-noise ratios (SNRs), as well as expectation-propagation (EP) baselines with perfect CSI at high SNRs. The results are further validated by comparing the extrinsic information transfer (EXIT) charts of EXT-ICLNet, a neural EP baseline, and EXT-ICLNet without skip connection, demonstrating how EXT-ICLNet provides a more favorable transfer of a priori and extrinsic information.
Nisal Ranasinghe, Dzung Do-Ha, Chathura J. Gamage, Simon Maksour, Tamasha Malepathirana, Sachith Seneviratne, Lezanne Ooi, Saman Halgamuge
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by motor neuron degeneration, with changes to neuronal excitability being an early phenotype for ALS identified in patients and animal models. Recent advancements in induced pluripotent stem cell (iPSC) technology have enabled the generation of human iPSC-derived neuronal cultures, which, when combined with multi-electrode array (MEA) electrophysiology, provide rich spatial and temporal electrophysiological data. Traditionally, MEA data is analyzed using handcrafted features based on potentially imperfect domain knowledge, which while useful may not fully capture all the relevant characteristics inherent in the MEA data. Machine learning, in particular deep learning has the potential to automatically learn relevant characteristics (features) from raw data, without solely relying on handcrafted feature extraction. However, handcrafted features remain critical for encoding domain knowledge and improving model interpretability, especially in scenarios with limited or noisy data, as is often the case in most experimental studies. This study introduces FRAME-C, a knowledge-augmented machine learning pipeline that combines domain knowledge, raw spike waveform data, and deep learning techniques to classify MEA signals and identify ALS-specific phenotypes. FRAME-C leverages deep learning to extract important features from spike waveforms, while also incorporating handcrafted features such as spike amplitude, inter-spike interval, and spike duration, thus preserving key spatial and temporal information. We validated FRAME-C on both simulated and real-world MEA data from human iPSC-derived neuronal cultures, demonstrating its superior performance compared to existing methods for MEA classification.
Mahmoud Saad Abouamer, Henrik C. M. Frederiksen, Robin J. Williams, Shashi Raj Pandey, Petar Popovski
This work investigates how site-specific digital twins (DTs) can support wireless localization using simplified distributed infrastructure. We propose PRISM (Probabilistic Refinement using Inverse learning and a Site-specific Model), a DT-guided probabilistic localization scheme with single-antenna access points (APs). Localization is formulated as a probabilistic inverse problem in which a conditional proposal distribution over UE locations is learned from multi-node channel impulse responses (CIRs). The proposal generates candidate locations, with its spread regularized during training and DT consistency used to reweight candidates and refine the estimate at inference. The DT serves as an approximate forward-consistency model, avoiding assumptions of exact channel fidelity or differentiability with respect to user position. Numerical experiments demonstrate gains over data-driven and classical fingerprinting benchmarks, while achieving a median error of approximately $0.7$ m using 10 single-antenna APs. Without retraining, PRISM remains effective under degraded inference-time DT features. PRISM also yields uncertainty estimates from its proposal covariance, with empirical $χ^2$ calibration approaching ideal behavior as APs increase.
Protection levels (PLs) bound position errors under specified fault hypotheses and risk budgets, but advanced receiver autonomous integrity monitoring (ARAIM) is formulated for snapshot GNSS estimation rather than the correlated posterior of tightly coupled GNSS/IMU factor graph optimization (FGO). This paper formulates ARAIM solution separation and PL computation on the FGO posterior. Because current-epoch pseudorange factors enter only the current-state block of the fixed-lag information matrix, the fault-excluded position covariance, solution separation, and nominal-bias propagation follow exactly from the linearized current-state marginal, which re-tains correlations with velocity, IMU-bias, and clock states and with historical information. Woodbury updates with selected-output evaluation compute these quantities for fault-free, single-satellite, constellation-wide, and second-order hypotheses with-out repeated full-state inversion or graph reoptimization. Surrounding monitoring layers screen measurements, handle post-fit outliers, and determine PL validity and availability. In two road experiments, both nominal sequences achieve 100% horizontal and vertical empirical bounding over PL-valid epochs, with a maximum error-to-PL ratio of 0.96. Normalized errors show that the posterior covariance alone does not capture slowly varying ranging biases: a covariance-only bound computed from the same posteriors retains no resolvable margin, and strong temporal correlation limits the empirical evidence to consistency checks rather than validation of the integrity budget. On identical Experiment A posteriors, the proposed method reduces fault-hypothesis evaluation and PL solving time by 65.2% relative to direct full-state inversion with matching integrity decisions, and controlled fault injections demonstrate fail-safe PL invalidation while navigation continues.
Xinyue Jiang, Tong Li, Shiceng Zhang, Ke Chen, Zhu Xiao
Channel state information (CSI) prediction forecasts future channel states from historical observations, supporting transceiver design and resource allocation with reduced measurement overhead. However, existing deep learning predictors mainly encode channel knowledge in model parameters, which can limit their ability to represent diverse patterns of channel dynamics across scenes. In this paper, we propose RAG-CSI, a retrieval-enhanced Transformer framework that augments prediction with relevant historical channel knowledge. Specifically, we construct a global--local CSI knowledge store containing representative observations across scenes and historical observations from the current serving scene. We further develop joint state--dynamics retrieval (JSDR), which jointly characterizes channel states, phase dynamics, and amplitude dynamics to retrieve relevant CSI sequences. A Transformer predictor then uses cross-attention to fuse the retrieved knowledge with the current CSI history and predict future CSI with limited additional model parameters. Experiments on a multi-scene CSI dataset show overall improvements exceeding 10\% across all five evaluation metrics over the best baseline for each metric, including a 15.42\% reduction in normalized mean squared error.
Jarosław Wojtuń, Bartosz Czaja, Tomasz Grasza, Jan M. Kelner, Cezary Ziółkowski, Radek Zavorka, Josef Vychodil, Tomas Mikulasek 等 11 位作者
Root mean square (RMS) delay spread is a key parameter characterizing temporal dispersion in radio propagation channels. Its estimation from measured power delay profiles (PDPs), however, depends strongly on the adopted power threshold. This paper investigates the sensitivity of RMS delay spread estimates to relative PDP thresholding in a time-varying millimeter-wave channel at 33 GHz. The analysis uses wideband channel sounding measurements conducted in a dynamic road environment with moving vehicles. The temporal evolution and empirical cumulative distributions of RMS delay spread are evaluated for thresholds ranging from -20 to -50 dB. Reducing the threshold increases the median RMS delay spread from approximately 140 to 168 ns and the 90th percentile from 200 to 210 ns. Both statistics converge at lower thresholds, with -40 dB providing a practical convergence level. The results emphasize the importance of threshold selection for reliable channel characterization and consistent comparison of measurements.
Seungjae Han, Joshua Yedam You, Soi Kim, Minho Eom, Jinho Park, Seunghee Park, Byeongwook Lee, Young-Gyu Yoon
Individual neurons exhibit diverse molecular, morphological, and electrophysiological properties that shape their distinct roles in neural circuits. While these multimodal features are largely shared across individuals, they remain an underutilized source of information in analyzing neural population dynamics. We posit that these shared biological profiles provide a common reference for relating neural dynamics across subjects and recording sessions. Our framework leverages this structure by routing neurons into latent groups based on their biological profiles and analyzing population dynamics through group-level representations. Because these assignments are anchored in biological properties that are consistent across subjects, the learned organization supports transfer to unseen animals without retraining or neuron-wise correspondence. We evaluate our framework on whole-brain calcium imaging in larval zebrafish and large-scale electrophysiology in the mouse visual cortex. Across both species and recording modalities, the framework delivers state-of-the-art performance in both neural decoding and manifold analysis: it achieves superior decoding of behavior and stimulus labels while better preserving the temporal consistency and local geometric structure of latent trajectories on unseen data. Our experiments demonstrate that not only does providing biological features improve baseline performance, but our proposed framework achieves even stronger results by using these properties to explicitly structure the population rather than treating them as auxiliary inputs. Together, these results suggest that organizing neural populations by their multimodal biological features offers a promising framework for generalizing population dynamics across subjects and recording sessions.
Yannan Chen, Beichen Huang, Licheng Zhao, Chris Ding, Kaiming Shen
The normalized cut (NCut) problem is a fundamental one in the area of data clustering. Despite its fractional structure, the NCut problem is seldom considered from a fractional programming (FP) perspective in the existing literature. The main reason is that the conventional FP method is inadequate for this problem case: the classical Dinkelbach's transform can only handle a single ratio and hence is limited to the two-class clustering, whereas the multi-class clustering involves multiple ratio terms. This work proposes a novel so-called multidimensional quadratic transform that is capable of dealing with the multi-ratio case; it recasts the NCut problem into a bipartite matching problem---which can be immediately solved by linear search. Furthermore, we explore the connection between the proposed FP method and the minorization-maximization (MM) theory, and in turn characterize the convergence behavior of our algorithm. Finally, we conduct experiments to numerically demonstrate the advantages of the proposed FP-based algorithm in a variety of data clustering and image segmentation tasks.
Md Wahiduzzaman Suva, Esm E Moula Chowdhury Abha, Md Imtiaj Alam Sajin, Md Iftekharul Mobin, Jia Uddin
Automated polysomnography after ischemic stroke is challenging because of altered sleep-related neural activity and concurrent respiratory disturbances. Existing stroke-specific approaches mainly address sleep-stage classification, leaving joint analysis of sleep architecture and respiratory events underexplored. Moreover, adapting pretrained EEG encoders introduces a representation inconsistency when downstream models are trained on embeddings from previously seen patients but evaluated on unseen individuals. We propose MM-Net, a multimodal multi-task framework that resolves this inconsistency through cross-fitting and ridge-based embedding alignment. Sleep-adapted EEG-EOG representations and engineered neural descriptors are fused with learned cardiorespiratory features through cross-attention and shared temporal modeling for simultaneous sleep staging and respiratory-event detection. On iSLEEPS (99 patients with subacute ischemic stroke), ten-fold patient-independent evaluation achieved 76.3% staging accuracy (Cohen's $κ$ = 0.656) and 0.825 respiratory-event AUC, outperforming the evaluated baselines under matched protocols. The sleep-adapted encoder improved the downstream model when its embeddings were aligned, raising staging accuracy by 1.4 percentage points over the frozen encoder ($p$ = 0.020) versus 0.2 points without alignment. MM-Net also achieved 86.4% staging accuracy on Sleep-EDF, above every single-EEG architecture reimplemented on the same recordings, and 80.3% staging accuracy with 0.889 respiratory AUC on ISRUC-I, exceeding its retrained staging baselines. Without retraining, the iSLEEPS-trained model retained respiratory AUCs of 0.833 on ISRUC-I and 0.841 on UCD. These findings support representation-consistent neural adaptation and joint physiological modeling across stroke, healthy-sleep, and sleep-disorder cohorts.
Compact electroencephalography (EEG) decoders enable embedded brain--computer interfaces (BCIs), but compression can alter adversarial robustness without proportional hardware savings. We present ORBITAL, a robustness-constrained framework combining hardware-realizable structured compression, recovery-aware operating-point selection, multi-attack evaluation, and native edge validation. Across nine subjects and three seeds on BCI Competition IV-2a, S25-TRADES achieves the highest observed mean true-union robustness among compressed EEGNet models (23.06% vs 21.27% for standard Full), with 26.27% fewer parameters and 25.10% fewer MACs. Matched controls identify TRADES as the main source of robustness recovery, improving S25 union accuracy by 6.74 percentage points over clean recovery (p=0.0039), while per-example analysis reveals boundary redistribution rather than uniform margin expansion. On Jetson Orin Nano, FP16 compression reduces board energy by up to 16.20% and latency by 8.00%. Despite 99.3% prediction agreement, native INT8 Q/DQ incurs about twice the FP16 latency and energy, demonstrating the need for joint robustness and hardware-aware optimization.
Radio-frequency (RF) digital twins (DTs) enable site-specific wireless propagation prediction by explicitly reconstructing the physical environment and modeling electromagnetic interactions. Although physics-based DTs reduce reliance on exhaustive measurements, systematic bias relative to real radio data arises from environmental uncertainty, material inaccuracies, and geometric abstractions. Calibration using field measurements is therefore necessary but costly under limited sampling budgets. This paper studies active RF DT calibration, in which measurement locations are selected sequentially and in an online manner to estimate a calibration offset under strict measurement budget constraints. A physics-based RF DT pipeline is developed, integrating semantic geospatial data from OpenStreetMap, 3D modeling in Blender, and deterministic ray tracing in Sionna to capture dominant building and foliage effects. This RF DT is calibrated using path-loss measurements collected with software-defined radios. Multiple online calibration strategies that select radio measurement locations are compared, including random sampling, rank-structured sampling with balanced coverage, quasi-Monte Carlo ranked selection, progressive structure-aware allocation, and fully adaptive uncertainty-driven sampling with estimator stabilization. Results demonstrate that adaptive and structure-aware schemes significantly improve both mean and worst-case calibration accuracy, particularly in the low-budget regime for radio measurements.
In federated learning (FL), secure aggregation enables a server to compute the sum of local model updates of $N$ clients, without revealing any information on the individual model updates to the server, beyond their sum. Different clients may be subject to heterogeneous bandwidth constraints, and consequently may choose different numbers of quantization levels to represent the information they send. Most secure aggregation works require the clients to project the values of their update parameters to a common finite group (or field), which does not directly accommodate heterogeneous precision constraints on their transmitted masked updates. Existing approaches of secure aggregation with heterogeneous quantization levels reveal partial aggregates of model-updates, and can reveal segments of individual model updates, under certain dropout and client-server collusion patterns. To prevent this privacy breach, we propose an information-theoretic secure aggregation scheme that tolerates client drop-outs, and up to $T$ semi-honest clients that collude with each other and with the server, under heterogeneous quantization at clients. Our protocol utilizes the $(T+1, N)$ threshold black-box secret sharing construction, where secret shares of the client updates, despite belonging to distinct algebraic groups owing to different quantization, are generated from identical linear integer systems.
Wide-area 6DoF rendering from sparse Ambisonic recordings is commonly addressed by interpolating parametric spatial descriptors rather than the waveforms themselves. This preserves the fidelity of a nearby measured signal and avoids phase artifacts caused by mixing distant arrays. However, existing DirAC-based approaches rely on geometric interpolation weights that do not depend on the acoustic scene observed by the microphones. We propose a lightweight Graph Neural Network (GNN) that refines only this interpolation stage. The known Ambisonic arrays and the query position are represented as a graph whose nodes contain local DirAC features and whose edges encode relative distance and directional relations. The network predicts positive query-conditioned weights over the measured arrays, while the target energy, diffuseness, and direction of arrival are reconstructed by an analytical DirAC layer. Experiments on unseen rooms, unseen speech signals, and unseen array configurations show that the proposed weighting improves energy reconstruction and radiation-pattern similarity with respect to baseline interpolation.
Root-MUSIC is a spectral estimation algorithm that approximates the unknown signal frequencies by constructing a high-degree polynomial and finding a subset of roots which are closest to the complex unit circle. Previous works found asymptotic expectation formulas for the performance of Root-MUSIC under the implicit assumption that the aforementioned root selection criterion does not select extraneous roots -- those which are unrelated to the correct parameters. This paper removes the need for this assumption by showing all extraneous roots lie outside an annulus of a certain thickness and therefore are not selected by the algorithm. This paper also provides sharp and non-asymptotic error bounds for the correct roots in terms of fundamental model parameters. All results hold under a natural separation condition on the correct signal frequencies and are applicable in both the single- and multi-snapshot models. Numerical simulations confirm our theory. The main mathematical insight of this paper is a geometric property of the Root-MUSIC polynomial: its correct roots are highly stable to noise while its extraneous roots must lie outside of an annulus.
Hari Vishnu, Jin Xinhui, Mandar Chitre, Simen Hexeberg, Koay Teong Beng
Passive acoustic monitoring in tropical waters such as those in Southeast Asia is hampered by the persistent, high-level impulsive noise of snapping shrimp. This is a particularly daunting challenge for detecting broadband transient vocalizations of dolphins - clicks, buzzes and burst-pulses. These vocalizations, unlike frequency-modulated whistles, share the broadband, impulsive character of the shrimp snaps and cannot be easily separated from them by frequency-localized or energy-based means. We develop TRIDENT, a lightweight waveform-based detector for these vocalizations, and apply it to 54.2 recorder-months of data from ten island sites around Singapore. We show that the training-data pipeline and the architecture are both important: controlled signal-to-noise-ratio based mixing to generate training data, appropriate normalization and hard negative mining, together steer the detector toward learning transient temporal structure rather than energy shortcuts. A single joint detector handles all three signal types better than specialized models for each, and outperforms established baselines. Applied to long-term recordings, it reveals tightly clustered, predominantly daytime activity, motivating days-with-detections as a robust indicator of cetacean presence.
Experts can master fine hand-object manipulation skills, yet limited conscious perception of rapid tactile interactions hinders their ability to articulate the underlying control mechanisms. Badminton racket control requires rapid, fine hand-force coordination, providing a demanding testbed for uncovering these mechanisms. An 864-node tactile-inertial glove with synchronized 100-Hz acquisition recorded over 14,000 strokes from more than 70 participants across five skills. Inspired by sensorimotor control, our mechanics-grounded two-layer framework inferred object-driving and grip-stabilizing control from tactile history, while resolving their spatiotemporal force realization. We found that higher proficiency was associated with selective rather than extensive control: greater skill specificity, anticipatory regulation, focal distal modulation and lower geometric torque cancellation, while spatiotemporal coordination showed a non-monotonic acquire-refine pattern. We translated these mechanisms into personalized guidance targeting force control and force-motion coordination. In a randomized cohort, mechanism-guided learners gained 0.72 expert-rated skill levels versus 0.05 with level-only feedback, making control mechanisms actionable.
Kyriakos Stylianopoulos, Mattia Fabiani, Davide Dardari, George C. Alexandropoulos
The recently envisioned goal-oriented communications paradigm requires machine learning inference to be performed directly on wirelessly transferred data, by splitting neural network computations between the transmitter, the Receiver (RX), and the wireless channel. Such approaches aim to offload large portions of the computation onto the channel itself, which is often controlled by reconfigurable MetaSurfaces (MSs). Typical MS-parametrized systems, however, enable only limited computation offloading in the wave domain, usually corresponding to a single neural network layer, due to adverse channel fading effects and the almost linear operation of the channel and conventional MS models. In this paper, we demonstrate that an eXtremely Large (XL) Multiple-Input Multiple-Output (MIMO) system equipped with Stacked Intelligent MSs (SIM) at the RX side, together with a single antenna and its accompanying radio-frequency chain, may act as a two-layer neural network operating exclusively over the air. The first (uncontrollable) network layer corresponds to the transformations offered by wireless fading, while the second (controllable) layer is implemented via SIM operations. In particular, only the first, front-end MS layer needs to exhibit nonlinear response to serve as an activation function. Leveraging the Extreme Learning Machine (ELM) framework, the proposed XL MIMO system architecture is shown to be a universal approximator, and the optimal linear-SIM-based combining is derived in closed form, allowing for fast training even and partial signal observability. Numerical evaluations across diverse datasets showcase that the proposed XL-MIMO-ELM framework achieves classification accuracy comparable to idealized digital models, thereby, proving the viability of low complexity, wave-domain neural computations.
Federated learning (FL) has emerged as a key enabler of distributed model training across ubiquitous wireless edge devices while preserving data privacy in next-generation wireless systems. One communication-efficient approach to realize this is leveraging Over-the-Air Federated Learning (OTA-FL), which exploits the superposition property of wireless multiple access channels (MAC) to perform low-latency model aggregation. However, practical OTA-FL deployments are significantly affected by the joint effects of channel noise at the server, fading, and both data and system heterogeneity across participating edge devices, which can degrade the model performance and thereby convergence. In this work, we propose the novel Robust and Convergent Federated Learning (RoCo-FL), a framework that jointly handles these challenges in federated wireless networks. RoCo-FL formulates local updates through inexact optimization, capturing the effects of wireless noise, partial participation, and heterogeneous system capabilities via structured inexactness conditions. These conditions naturally manifest as a proximal constraint at the edge devices that stabilizes local updates and mitigates the impact of channel impairments, data heterogeneity, and stragglers. We establish convergence guarantees for non-convex objectives under the proposed zeroth- and first-order inexactness models, providing theoretical insight into the interplay between channel impairments, data, and system heterogeneity. Experimentally, we validate RoCo-FL on real-world datasets, including CIFAR10, CIFAR100, and FEMNIST, demonstrating its robustness in noisy and heterogeneous environments. Compared to state-of-the-art baselines, RoCo-FL achieves significantly more stable convergence and higher accuracy, particularly in the presence of stragglers.
Anti-jamming capability is of paramount importance in modern wireless communications, yet conventional receivers are severely constrained by their fixed array geometries, which limit the available spatial degrees of freedom and leave performance highly susceptible to strong interference. By contrast, movable antennas (MAs) provide additional spatial degrees of freedom through antenna-position optimization. This paper investigates MA-enabled anti-jamming by jointly optimizing the antenna-position vector (APV) and receive beamformer to maximize the output signal-to-interference-plus-noise ratio (SINR). For the canonical setting with a single jammer and a single line-of-sight (LoS) propagation path on the legitimate link, we develop a damped projected-gradient descent (DPGD) algorithm with low computational complexity, which scales linearly with the number of receive MAs. We establish a generic active strict-saddle landscape for the resulting steering-coherence minimization problem and prove that DPGD converges almost surely to a local minimizer rather than merely to a Karush-Kuhn-Tucker (KKT) point. For the general setting with multiple channel paths on the legitimate link and/or multiple jammers, we derive an equivalent ridge least-squares formulation and develop a low-complexity soft-active projected-gradient (SAPG) algorithm, which also scales linearly with the number of receive MAs. We further show that the proposed antenna position optimization framework applies to both minimum variance distortionless response (MVDR) and zero-forcing (ZF) reception. Numerical results validate the proposed algorithms, demonstrating substantial gains in both anti-jamming performance and computational efficiency.
Electrocardiogram (ECG) models need supervision that targets evolving weaknesses and remains faithful to waveform evidence. We introduce ECG-RSI, a recursive self-improvement framework combining verified waveform synthesis with feedback-guided intervention search. Signal remeasurement verifies waveform edits; construction errors guide generator repair, while learner errors motivate interventions compared on a separate selection set. Across seven benchmarks, three rounds raise Qwen3.5-35B-A3B's mean question-answering accuracy from 34.57% to 56.04% and mean classification F1 from 24.27 to 34.03. Across the retained trajectory, the preferred supervision shifts from direct reinforcement to confusion-focused contrasts. Continued feedback correction yields diminishing external gains. These results motivate separating error diagnosis from intervention selection and checking whether gains on the feedback set transfer to independent benchmarks.
Purpose. Electroencephalography (EEG) montages, channel counts and recording lengths vary widely, but most EEG foundation models are bound to the channel set and input length they were trained on. We aimed to build a generative EEG foundation model independent of the number, identity and order of electrodes and of the input length. Methods. NOMAD-EEG is a 12.0 M-parameter diffusion model that alternates state-space layers along time, without patches, with attention across electrodes identified only by their 3D scalp position. It was pre-trained on 147 h of unlabelled EEG from a single corpus and evaluated with frozen linear probes on ten datasets with 1 to 64 electrodes against seven foundation models run under one protocol, and on two unseen populations. Results. Despite its small size (12.0 M parameters against 69 M for REVE) and single pre- training corpus, NOMAD-EEG outperformed all seven foundation models on EEGMMIDB mo- tor imagery (balanced accuracy 0.442), Siena seizure detection (macro F1 0.622) and Bonn seizure detection (AUROC 0.999), had the highest accuracy of the seven compared models on TUEP epilepsy diagnosis (0.640) and, after fine-tuning, the highest \k{appa} (0.635) and weighted F1 (0.814) on TUEV event classification. It was the best of five models on P300 detection in autism with an unseen eight-electrode montage (AUROC 0.718). With only 5 of 19 electrodes, its mean AUROC fell only from 0.942 to 0.891. Conclusion. Identifying electrodes by position and modelling time without patches let one small model, pre-trained on one corpus, outperform much larger EEG foundation models on seizure detection, epileptiform event classification, epilepsy diagnosis and motor imagery.
Po-Heng Chou, Jia-Qing Lin, Wei-Chen Lin, Wan-Jen Huang, Ronald Y. Chang
In this paper, we propose a feedback-efficient hybrid precoding framework for wideband millimeter-wave (mmWave) multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. To reduce the hardware cost of radio-frequency (RF) chains and the feedback burden associated with frequency-selective digital precoding, frequency-flat analog precoders and combiners are first constructed from the dominant angle-of-arrival (AoA) and angle-of-departure (AoD) directions associated with the strongest propagation paths. For digital precoding, a shared quantized codebook is designed using the Lloyd algorithm, and a binary-search-based hierarchical interpolation method is developed to identify codeword switching locations between adjacent pilot subcarriers. By avoiding exhaustive codeword comparisons over all intermediate subcarriers, the proposed method reduces online search complexity and enables compact digital-precoder feedback. For the considered system configuration, the proposed method achieves a worst-case feedback reduction of approximately 98.34% relative to full-grid per-subcarrier feedback. Simulation results further demonstrate competitive spectral efficiency and bit error rate (BER) performance, low computational cost, and robustness under imperfect effective channel state information (CSI).
In over-the-air split learning over an aging single-antenna Rayleigh channel, the mean returned gradient equals the antenna-domain gradient scaled by the temporal correlation. Uncorrected training collapses in negative-correlation lobes. Near a correlation zero, no nonzero scalar calibration improves the ratio of mean power to innovation power for an individual transmission, and, on the tested grid, using sixteen times as many pilots recovers only 7 percent of the training gap. Returning at the first correlation extremum with phase correction, 1.3 ms later at 3.5 GHz and 15 m/s, achieves 90.2 percent accuracy, compared with 91.8 percent with CSI. In the tested line-of-sight setting, the near-zero gap becomes negligible.
Over-the-air split learning returns the receiver's loss gradient through the reciprocal channel, so terminal motion between the forward and backward transmissions corrupts the backward pass. We repair the aged return at a fixed receive array using only second-order statistics. Two numbers taken from the forward-backward cross-covariance, the alignment reachable by the best common phase and by the best unitary correction, bound the main term of every repair. In the 3GPP tapped-delay-line channels TDL-A/B/C, where every antenna ages by the same scalar, the two numbers coincide, and at a correlation zero all tested fixed corrections stay between 58% and 62% accuracy, whereas a return delayed to the next correlation lobe reaches 98.4%. In 3GPP clustered-delay-line channels a statistical unitary correction stays within 0.02 in mean cross-entropy of ideal channel knowledge in all six tested profile and speed combinations, while common-phase correction falls to 69% for CDL-A at 3 km/h and the tested delay loses up to 10 points. In a two-cluster model spatial correction is decisive only where the common-phase alignment nearly vanishes (64% to 95%). With a ResNet-18 on CIFAR-10, only the delayed return trains at a separable correlation zero (86% against at most 16%), and at 30 km/h in a clustered-delay-line channel the unitary correction reaches 80% where the common phase reaches 60%. Given geometry, displacement and angular statistics, a port permutation chosen by linear assignment reaches 99.5-99.7% accuracy for hexagonal and square arrays in oblique motion without pilots or digital feedback during training, where nearest-position matching drops to 89%.
Fengchong Yao, Jianbing Li, Qing Liu, Qikun Liu, Kefeng Song, Haitao Li, Song Wang
Radio frequency fingerprint identification (RFFI) exploits transmitter-specific hardware imperfections as physicallayer identity cues for Internet of Things (IoT) devices, but deep models often degrade across acquisition environments. In multi-antenna reception, antenna topology and frequencyoffset dynamics structure receiver observations, while capturedependent variation distorts target embeddings and misaligns source-trained decision boundaries. This article proposes physicsinformed structure anchoring with capture-aware prototype calibration (PISA-CAPC) to address both representation and decision mismatches. The two stages separate source representation construction from target decision correction. During source training, PISA organizes antenna tokens through a topology-guided graph, conditions propagation on CFO-derived acquisition dynamics, and applies bounded contextual residual suppression to preserve identity evidence. At deployment, unlabeled capture-aware prototype calibration (U-CAPC) estimates capture-local prototypes and recalibrates target decision scores while keeping the representation and source classifier fixed. Thus, calibration uses neither target labels nor target-domain backbone updates. On a measured WiFi benchmark with four receive antennas and ten transmitters, PISA-CAPC achieves a mean target-domain Macro-F1 of 0.9257 under a balanced transductive setting. Component ablations support complementary roles for topology-guided anchoring, CFO-conditioned modulation, reliability-aware token aggregation, contextual suppression, and capture-aware calibration. These results indicate that physically motivated representation learning can be combined with labelfree decision calibration to improve cross-environment RFFI under the evaluated protocol without changing the deployed backbone.
Most in-field crop-monitoring devices sense a single modality -- either a camera looking at the canopy or a probe reading the soil -- and so cannot detect stress that begins below ground before it becomes visible above it. We present AgriDefend, a self-contained consumer-grade edge device that fuses both. A MobileNetV2 convolutional neural network (CNN) analyses canopy imagery, and a Random Forest classifier integrates trend features from a single soil probe: moisture, the rate of change in pH and in N, P, K, and belowground CO2 relative to ambient (a proxy for soil respiration). A logistic-regression meta-learner fuses the two into a five-class crop-health prediction, all on a Raspberry Pi 5 with no cloud or cellular link. On a controlled greenhouse drought trial with wild radish, the fused model reached 87% accuracy (R^2=0.88, MAE=0.05) at 1.4 s INT8 TensorFlow Lite inference, and its soil signals preceded visible wilting by about four days. The complete bill of materials is $205, roughly a fifth of comparable commercial nodes, positioning multimodal, sustainable crop sensing as a practical consumer-electronics product for connectivity-limited settings.
Cross-environment generalization remains a fundamental challenge for Wi-Fi-based human pose estimation (HPE), as environmental multipath can drastically alter Channel State Information (CSI) for the same human motion. Existing approaches often discard CSI phase or collapse complex signals into real-valued features early, losing motion-sensitive phase dynamics. We present C-MambaPose, a phase-preserving complex-valued state-space framework for robust cross-environment Wi-Fi 3D HPE. Our key insight is to preserve complex phase relationships throughout spatiotemporal modeling, rather than using phase only as an input feature. C-MambaPose sanitizes synchronization-induced phase errors and employs a Spatiotemporal Complex Mamba encoder with dynamic multi-scale receptive fields to capture long-range phase-amplitude dynamics. Cross-attention then maps the learned representations to human joints, followed by a skeleton-aware GraFormer decoder for anatomically coherent pose estimation. Experiments on MM-Fi show that C-MambaPose achieves strong performance across all settings, with the largest gains in unseen environments, reaching $298.5$ mm and $317.6$ mm MPJPE on the two cross-environment protocols using only $3.78$ M parameters and no pretraining. Ablations further show that prematurely discarding complex phase relationships substantially degrades performance, demonstrating their importance for cross-environment Wi-Fi HPE.
In this work, we propose a movable-element (ME) multi-functional reconfigurable intelligent surface (MF-RIS) architecture, termed MM-RIS, which integrates element mobility with adaptive signal reflection, transmission, amplification, and energy harvesting (EH) to enhance communication flexibility and energy self-sustainability. A hybrid rate-splitting multiple access (RSMA) and non-orthogonal multiple access (NOMA) transmission scheme is further designed to improve spectral efficiency. We formulate an energy efficiency (EE) maximization problem by jointly optimizing transmit beamforming, power and rate allocation, MM-RIS configurations, panel deployment, and ME positions. To address the resulting non-convex mixed continuous-discrete optimization problem, we propose constrained hybrid deep reinforcement learning with federated transformer and parametrized sharing (CRAFTS). CRAFTS integrates multi-agent proximal policy optimization (MAPPO) and multi-agent deep Q-networks (MADQN) through bi-directional parametrized sharing to handle continuous and discrete actions, respectively. Moreover, a transformer-based architecture and federated learning (FL) are incorporated to enhance state representation and distributed knowledge sharing, whilst a Lagrangian-based adaptive penalty mechanism facilitates the constraint satisfaction. Simulation results have demonstrated that the proposed MM-RIS architecture and CRAFTS framework substantially improve EE and constraint satisfaction compared to the benchmarks in the open literature. Moreover, the results validate the benefits of spatially reconfigurable MM-RIS, hybrid RSMA-NOMA transmissions, and intelligent collaborative learning, highlighting the potential of the proposed framework for energy-efficient and adaptive next-generation wireless networks.
Ambient intelligence seeks to continuously understand human presence, activity, and physiology in physical spaces, enabling smart environments, health monitoring, and human-computer interaction. WiFi infrastructure offers a ubiquitous, always-on, and privacy-preserving sensing substrate across billions of IoT devices. Yet ambient sensing with WiFi remains largely fragmented, with most systems relying on task-specific models, labeled data collection, and customized training pipelines. We present AmbientFM, a foundation model for ambient sensing through ubiquitous WiFi signals. AmbientFM is pre-trained on 9.2 million unlabeled Channel State Information (CSI) samples collected over 439 days from 20 commercial device types deployed in real homes, the largest real-world WiFi CSI pre-training corpus reported to date. It learns transferable wireless representations through contrastive learning, masked reconstruction, and physics-informed objectives tailored to wireless signals. With lightweight adaptation, a single backbone transfers to 9 downstream tasks, achieving above 0.90 AUROC on all classification benchmarks and supporting dense spatial reconstruction. AmbientFM reduces the need for labeled data and task-specific design, enabling scalable ambient intelligence on existing wireless infrastructure.