New Algorithm Improves Channel Estimation in NB-IoT Links
A new KANN-DBSCAN algorithm improves blind channel estimation in NB-IoT links by adapting clustering parameters and bandwidths to complex signal ...
Tag: Signal Processing
Dual-Branch Deep Learning Network Advances Motor Imagery Brain-Computer Interface Decoding
A novel dual-branch neural network called EEG-DBNet simultaneously decodes temporal dynamics and spectral representations to significantly enhance motor imagery classification ...
New Algorithm Accelerates Polarization and Direction-of-Arrival Estimation for Curved Sensor Arrays
A new signal processing algorithm combining phase-mode transformation with Hough transform parameter extraction resolves the computational latency of joint direction ...
Hybrid ChaCha20 Framework Encrypts High Bit-Depth Audio with Near-Lossless Decryption
Researchers have demonstrated a hybrid audio encryption framework combining ChaCha20 and non-linear transformations that achieves near-lossless reconstruction and mitigates leakage ...
Cyclostationary Signal Modeling Enhances Multilingual BERT Sentiment Analysis Accuracy
Modeling text embeddings as cyclostationary sequences improves BERT-based multilingual sentiment analysis accuracy by up to 28% across five diverse datasets.
Dual-branch AI network cleans wearable ECG signals while protecting diagnostic heart waves
Researchers report a dual-branch gated U-Net that removes motion artifacts from wearable ECG recordings while measurably preserving P and T ...
From Fourier to Deep Learning: How Bridge Health Monitoring Signals Are Evolving
A new review of 201 studies traces how bridge structural health monitoring evolved from classical Fourier and wavelet analysis to ...
Modified Fire Hawk Optimizer Sharpens Direction Finding for Humanoid Robots
A simulation study reports that a modified Fire Hawk Optimizer with time-varying control parameters improves direction-of-arrival estimation accuracy for a ...
Deep Learning and IoT Framework Records, Preserves and Recreates Traditional Music
Researchers report an IoT-and-deep-learning framework that captured nearly 3,000 music recordings and achieved 94% accuracy in recognising traditional music patterns ...
Drones That Listen for Wi-Fi Could Map Survivors When Cameras Fail
A simulation study presents a multi-UAV system that denoises passive Wi-Fi signals from survivors' devices to generate zone-level rescue priorities.
Simple Muscle Signal Pipeline Drives Robotic Arm, But Speed Remains a Hurdle
A Cleveland State University study shows that a simplified surface EMG onset-detection pipeline can drive a Baxter robotic arm to ...
Algorithm Tracks Shifting Electrodes to Make Muscle Sensor Readings Reliable Across Sessions
A new high-density sEMG algorithm estimates how far electrodes shift between recording sessions, cutting feature errors and bringing reliable muscle-based ...
Kalman Filter Approach Offers Interpretable Real-Time Sleep Spindle Detection
A new adaptive Kalman filtering method frames sleep spindle detection as state-space inference, producing interpretable, low-latency sigma-event catalogues from full-night ...
Diffusion Model Reclaims Lost Bits in High-Speed Analog-to-Digital Converters
A conditional diffusion model with physics-based constraints boosted effective resolution, SNDR, and spurious-free dynamic range in high-speed ADCs tested from ...
DCT and high-boost filtering improve grayscale multi-focus fusion in four-dataset study
A two-stage framework combining high-boost filtering with discrete cosine transformation achieved the strongest reported performance on four grayscale multi-focus image ...
Analogue Memory Chips Run Fast Fourier Transforms for Edge Signal Processing
Researchers have experimentally demonstrated analogue fast Fourier transforms on charge-trapping memory arrays, computing transforms of up to 65,536 points across ...
Quantum Hybrid Model Tested for ECG Arrhythmia Detection
A new study evaluates a hybrid quantum algorithm for ECG arrhythmia classification, finding it competitive at moderate resolutions but lacking ...
Sliding Windows and Fourier Decomposition Give Brain-Computer Interfaces a Lightweight Edge
Researchers in India have built a lightweight brain-computer interface framework that combines sliding-window segmentation, Fourier decomposition, and a compact neural ...
Wearable Sensors Turn Cattle Estrus Into Scalable Fertility Data
A New Zealand proof-of-concept study shows that leg-mounted accelerometers can detect first estrus in grazing heifers and first-lactation cows, opening ...
Wavelet-Fourier hybrid sharpens I/O phase predictions on HPC clusters
A hybrid wavelet-Fourier analysis in the FTIO tool pinpoints when periodic I/O phases are actually valid, cutting prediction error and ...
