arXiv Machine Learning
Comparative Evaluation of Deep-Learning based DoA Estimators for Real-Time ISAC Beamforming
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Jul 15
Self-Evolving In-Context Learning for Direct Pilot-to-Beamformer Design in MU-MISO Systems
arXiv:2607. 11970v1 Announce Type: cross Abstract: We develop an enhanced in-context learning (ICL) framework to improve the performance of pilot-based beamforming in multi-user multiple-input single-output (MU-MISO) systems.
arXiv Machine Learning
Aug 5
Task-Oriented Candidate-Latent Feedback for Coarse-to-Fine Sensing in Distributed OFDM-ISAC Networks
arXiv:2608. 03319v1 Announce Type: cross Abstract: Future integrated sensing and communication (ISAC) architectures separate the sensing entity (SE) that acquires measurements from the sensing function (SF) that performs inference, creating a need for compact, task-oriented feedback on the SE-SF interface.
arXiv AI
Jul 2
Meta-Transfer Learning for mmWave Beam Alignment
arXiv:2607. 00860v1 Announce Type: cross Abstract: Millimeter-wave (mmWave) beam alignment plays a critical role in next-generation wireless systems, yet its efficient implementation remains challenging.
arXiv AI
1d ago
Spectral-Convolutional Factorized Transformer for Joint Multipath Parameter Estimation: Applications to mmWave Positioning and Radio SLAM
arXiv:2611.01491v2 Announce Type: replace-cross Abstract: Future 6G wireless networks are expected to evolve toward AI-native systems that support intelligent, environment-aware services, including p...
arXiv AI
Oct 12
Energy-Efficient Spiking Neural Networks for Sensing-Aided Beam Prediction
The paper presents energy‑efficient spiking neural network (SNN) architectures for sensing‑aided beam prediction in millimeter‑wave systems, comparing them with conventional artificial neural networks (ANNs) across position‑only, vision‑only, and multimodal inputs. While compact ANNs outperform SNNs for position‑only data, SNNs achieve higher accuracy than ANNs for vision and multimodal sensing at comparable or lower energy costs, with the multimodal SNN reaching 58.08% top‑1 accuracy—an 18.11‑percentage‑point improvement over the nearest‑energy ANN. The study also finds that SNN accuracy is constrained by binary inter‑layer communication rather than model size.
arXiv AI
Oct 12
CNN-Transformer with Sigmoid Label for High-Resolution Sub-Degree DOA Estimation
The paper introduces a sigmoid label for deep‑learning based direction‑of‑arrival (DOA) estimation, encoding DOAs as sums of angle‑shifted monotonic sigmoid functions. Using a CNN‑Transformer network trained with this label, the authors achieve sub‑degree accuracy and 1‑degree resolution with a 0.25° grid step, surpassing traditional Gaussian, hard, and BCE labels. Numerical results confirm the sigmoid label’s superior accuracy and resolution over other labeling approaches.