Edgesl: Edge-computing architecture on smart lighting control with distilled knn for optimum processing time

AG Putrada, M Abdurohman, D Perdana… - IEEE Access, 2023 - ieeexplore.ieee.org
Our previous research applied a novel classification-integrated moving average (CIMA)
method, an intelligence method that improves the performance of passive infrared (PIR) …

Subsample, Generate, and Stack Using the Spiral Discovery Method: A Framework for Autoregressive Data Compression and Augmentation

ÁB Csapó - IEEE Transactions on Systems, Man, and …, 2024 - ieeexplore.ieee.org
This article addresses the challenge of efficiently managing datasets of various sizes
through two key strategies: 1) dataset compression and 2) synthetic augmentation. This …

[HTML][HTML] Low-Power Preprocessing System at MCU-Based Application Nodes for Reducing Data Transmission

D Kim, C Roh, D Baek, S Choi - Electronics, 2024 - mdpi.com
Edge computing enables prompt responses in IoT environments, such as the operation of
autonomous vehicles and unmanned aerial vehicles. However, with the increase in sensor …

Federated Learning in Mesh Networks

X Wang, Y Chen, OA Dobre - Artificial Intelligence for Future …, 2025 - Wiley Online Library
Federated learning, emphasizing decentralized machine learning without central data
aggregation, has expanded notably, attributed to its data privacy benefits and reduced …

Preprocessing at Application Nodes for Reduction of Data Transmission in Edge Computing

D Kim, C Roh, D Baek - 2023 IEEE International Conference on …, 2023 - ieeexplore.ieee.org
Existing edge computing collects data from applications and transmits it to edge nodes.
However, as the amount of data transmission increases, the network burden also increases …