Wireless Artificial Intelligent Computing Systems and Applications

The 3-volume set LNCS 15686 - 15688 constitutes the proceedings of the 19th International Conference on Wireless Artificial Intelligent Computing Systems and Applications, WASA 2025, which took place in Tokyo, Japan, during June 24-26, 2025.

The 70 full papers and 34 short papers included in the proceedings were carefully reviewed and selected from 282 submissions. The proceedings also contain 10 papers from the AICom2 symposium.

WASA is a prestigious annual gathering that serves as a global platform for researchers, academics, and industry professionals to explore and exchange cuttingedge ideas, research findings, and innovative solutions at the dynamic intersection of wireless technologies and artificial intelligence (AI) computing systems.

Inhalt

.- GAT-Enhanced DRL Decision Framework for Joint Routing and Scheduling under Vehicular Time-Sensitive Networking.

.- CSI-Cro A Cross-Domain Sign Language Recognition System Based on Dual-Attention Feature Decoupled Network.

.- Latency-Optimal and Memory-Aware Model Partitioning for Cooperative Inference at the Edge.

.- A Sub-Domain Index System for Network Security Situation Assessment.

.- Integrating Pretrained Models with Graph Neural Networks for Smart Contract Interpretability.

.- A Simple Heuristic Finding Connectivity Bottleneck in Networks with Shared Risk Resource Groups.

.- Improving the Freshness of Digital Twins in Edge Computing.

.- Joint Offloading and Resource Allocation Optimization based on Multi-coupled Directed Acyclic Graphs for IIoT.

.- SeRed: A Selective Reduction Method for Efficient Network Flow Storage.

.- CAG: A Constraint-Driven Adversarial Traffic Generation Scheme Based on Feature Correlations.

.- A Supervisor-Oriented Privacy-Preserving Fair Exchange Scheme for V2G.

.- MACROSS: End-to-End Network Attack Detection Model for EV Charging Station Based on Temporal-Frequency Feature Fusion.

.- Game Theory-Based Secure Deduplication Method.

.- A Cuckoo Filter-based Anomaly Detection and Localization Mechanism for SDN-based Multidomain IoT.

.- Enhancing Federated Learning in IoT Through Dynamic Spectral Clustering.

.- Differentially Private Vertical Federated Learning with Dual-Sparsification.

.- Bandwidth Prediction within Playback-time of Buffer Duration based on Frame-level Stable Segment for Live Streaming.

.- MissingClip: An Industrial Anomaly Detection Method under Modality Missing.

.- Multi-Preference Sequence Recommendation Transformer.

.- Exploring Heterogeneity in Federated Learning.

.- Diverse Delay-Sensitive Task Offloading and Resource Allocation for IIoT: An SoI Enhanced Approach.

.- DEFA: A Prairie Rat-hole Target Detection Algorithm Integrating Depth Estimation Based on MiDas.

.- PIS-EnsemNet: A novel method for predicting the number of devices in hotspot sharing networks based on passive measurement.

.- Joint Hierarchical Feature Fusion and Progressive Learning for Topology Robustness Prediction.

.- A SCA-Based Method for RIS Assisted Over-the-Air Computation.

.- MPAM: Dual-Transformer for Millimeter-Wave Sensing Based Multi-person Activity Monitoring System.

.- Dynamic Graph-based Semi-Supervised Anomaly Detection for In-Vehicle CAN Bus Network.

.- A Model of Multi-order Sampling Neighbor Aggregation for Traffic Flow Prediction.

.- Physics-Informed Transformer for Efficient Fluid Dynamics Predictions.

.- ICSPFuzzer: An Efficient Fuzzing Technique for ICS Protocols.

.- MRT-DETR: A New Real-Time Object Detection Method.

.- MASS: Empowering Wi-Fi Human Sensing with Metasurface-Assisted Sample Synthesis.

.- DDC Sketch: A Dynamically Adaptive Framework for Network Traffic Measurement.

.- Timely Watchers: Cost-Effective Schedule for Urban Sensor Patrols.

.- vGPU Performance Testing Framework for Large Model Inference.