The GPC 2025 proceedings deal with green, pervasive, and cloud computing focusing on pervasive and environmentally sustainable computing.

The 33 full papers and 2 short papaers were carefully reviewed and selected from 69 submissions. They were organized in topical sections as follows: Security, Trust, and Resilience,Wireless, Vehicular, and Underwater Networking,Federated, Edge, and Green Learning Systems,Mobile Crowdsensing, Scheduling, and Collaborative Optimization,Intelligent Sensing, Perception, and Applied Analytics and Graphs, Simulation, and Robust AI Models.

Inhalt

.- Security, Trust, and Resilience

.- CrossDID: A Secure and Accountable Cross-Chain Decentralized Identity System.
.- Lamina: An Efficient Asynchronous BFT through Lightweight Broadcast and Optimistic Paths.
.- Reliable and efficient data sharing framework based on blockchain in VANETs.
.- Cross-Network Malicious Traffic Detection Via Graph Neural Networks and Dual Distribution Alignment.
.- ACCG: A Collaborative-Adversarial Framework for Robust Code Generation.
.- A Game Theory Guided Attack-Defense Framework for Character-Level Modality Perturbations.
.- A Poisoning Attack System Targeting Critical Layers of Backdoors in Federated Learning Models.

.- Wireless, Vehicular, and Underwater Networking

.- Joint Semantic-Fidelity and Energy Optimization for AUV Collaborative Sensing via Multi-Agent Reinforcement Learning.
.- LoRaSim: A LoRa Network simulator via Hybrid Physics-Data Synergy.
.- Research on a Message Transmission Strategy in VANETs Based on Implicit Acknowledgment with Adaptive Beacon Intervals.
.- CICC: Concurrent Inter-channel Collision Cancellation in LoRa Networks.
.- Asymmetric IoT communication and sensing technology based on RF-Computing.
.- P-VBF: A Vertical Projection Vector-Based Forwarding Routing Protocol for Multi-UUV Cooperative Perception.

.- Federated, Edge, and Green Learning Systems

.- FedMIS: Federated Medical Image Segmentation with Knowledge Distillation.
.- DCML: An Efficient Cluster-Based Meta-Learning Algorithm with Knowledge Selective Reuse.
.- Lightweight Gated Spiking Neural Network Based on Temporal Residual Connection.
.- What Data Changed: Empirical Analysis of the Impact Mechanism of Streaming Data on Model Training in AIoT Environment.
.- FUPEC: A Federated Unlearning-Empowered Personalized Edge Caching Paradigm.
.- Programmable Context-Aware Orchestration of Collaborative Inference Using Software-Defined AIoT in Personal Healthcare Wearables.
.- Energy-Efficient Federated Fine-Tuning of Large Language Models in Heterogeneous Edge Environments.

.- Mobile Crowdsensing, Scheduling, and Collaborative Optimization

.- UWCS: A Unified Online Capability Assessment Framework for Hybrid Workers in Mobile Crowdsensing.
.- NSP-MCS: Reinforcement Learning-Based Navigation-Aware Task Scheduling for Mobile Crowdsourcing.
.- Low-Latency Onboard Compression Task Scheduling and Routing Co-Optimization Algorithm for LEO Satellites.
.- Collaborative Optimization of Heterogeneous Multi-Agent Integrated Sensing and Computing Based on QMIX Algorithm.
.- Optimizing Exploration-Coverage Trade-offs for Multi-UAV Systems Under Energy-Constrained Path Planning.
.- Multi-Agent Deep Reinforcement Learning for Cloud-UAV-RSU Collaborative Vehicular Computation Offloading.

.- Intelligent Sensing, Perception, and Applied Analytics

.- Scalable Data-Driven Modeling of Postoperative Pain Risk Factors Using Distributed Machine Learning Algorithms.
.- A Lightweight Traffic Flow Prediction Model Based on Multi-Scale Spatiotemporal Characteristics.
.- Inductive adaptive spatio-temporal Kriging for urban air quality inference.
.- ParaDepth: Distributed Multi-View 3D Reconstruction for Fire Scene.
.- CogPlanner: A Cognitive Navigation Method for Embodied Intelligence Guided by Indoor Floor Plans.
.- MultiMo-Concen: Multimodal Learning Concentration Recognition Based on Vision and Millimeter Wave.
.- MoiréTemp: High-precision Temperature Measurement Based on Moiré Pattern.

.- Graphs, Simulation, and Robust AI Models

.- BadFF: Inspect the weakness and robustness of forward-forward neural networks.
.- Accelerating Graph Similarity Search with Maximum Common Subgraphs.