International Conference on Edge Computing [Services Society]

Edge computing is undergoing a fundamental transformation from monolithic deployments to dynamic microservices architectures with distributed AI-driven inference. This evolution exposes critical weaknesses in traditional Layer 3/4 networking, which relies on static IP addressing, perimeter-based security and location-centric routing policies. As applications fragment into ephemeral components and AI agents are coordinated across trust boundaries, existing connectivity models become operationally complex and architecturally brittle. In parallel, as complexity increases organizations pay an increasing "networking tax" in slower development, higher maintenance and more frequent disruptions caused by the tight coupling between edge devices, applications and network configuration. This paper argues that edge computing requires a paradigm shift toward application-centric connectivity built on Layer 7 identity-anchored service networking and zero trust principles. We identify key limitations of location-based networking in edge environments, examine the challenges posed by distributed AI systems, and propose design principles and an initial technical implementation for application-centered networking that can accommodate the dynamic, distributed nature of modern edge applications.

To address the insufficient processing capability and low energy efficiency of existing smart electricity meters using traditional embedded software protocol stacks, this paper proposes a hardware acceleration and energy efficiency optimization design. The functional requirements of smart electricity meters are analyzed, and a design method based on the MQTT protocol is developed. Experimental tests verify the effectiveness and feasibility of the proposed method, which improves communication efficiency and reduces power consumption, providing a viable technical path for the development of new-generation smart electricity meters.

This paper presents a Fine-grained Feature Extraction U-Net (FgFEU-Net) model to realize automatic localization and segmentation of breast cancer lesions with breast ultrasound images. In the FgFEU-Net model, we used the transfer learning method to replace the coding part of the backbone network with the VGG16 model to achieve deep and fine-grained feature extraction. The extracted low-level information is combined with the high-level knowledge of the U-Net decoding layer at the same layer and converted into high-resolution information to pinpoint lesions precisely. Finally, the high-resolution information is transformed into a high-resolution image through the final convolutional layer. In addition, the combo loss is adopted to deal with the imbalance between organ input images and their output images. The proposed model was tested and predicted in the BUS2018 (Breast Ultrasound Images 2018) dataset, and the proposed evaluation metric of accuracy, precision, and sensitivity reached 99.03%, 96.83%, and 99.35%, respectively.

Mobile Edge Computing (MEC) enables latency-critical applications by offloading tasks from resource-limited wireless devices (WDs) to edge servers. However, optimal task offloading and resource allocation remain challenging under dynamic wireless channels and diverse task demands. This paper presents a two-step adaptive solution that combines deep neural networks (DNN) and projected gradient descent. A DNN first predicts binary offloading decisions from real-time channel conditions. Conditioned on these decisions, we solve a constrained optimization problem to determine task partitioning and CPU allocation that minimize the weighted sum delay. To enable rapid adaptation to new task scenarios, we embed meta-learning into DNN training. Simulation results show that our method achieves low delay and strong generalization in multi-user, multi-server MEC environments.

In today’s digital landscape, the management and trading of personal data have emerged as critical issues, particularly concerning privacy, security, and economic value. The traditional centralized model of personal data trading has significant drawbacks, including a lack of transparency, user control, and equitable compensation, leading to privacy violations and data breaches. This paper introduces a blockchain-enabled platform for personal data trading that leverages smart contracts, RSA-encrypted NFTs, and IPFS to enhance transparency and user control. The system was evaluated on four EVM-compatible blockchain platforms–Binance Smart Chain, Polygon, Fantom, and Celo–focusing on key functionalities such as data transaction recording, RSA-encrypted NFT creation, and secure NFT transfer. The evaluation provided insights into transactional efficiency, resource utilization, and economic aspects. Our findings suggest that the proposed platform can significantly improve the management and trading of personal data by ensuring data integrity and equitable value distribution.