Enabling Efficient Data Integration of Industry 5.0 Nodes Through Highly Accurate Neural CSI Feedback | IEEE Journals & Magazine | IEEE Xplore

Enabling Efficient Data Integration of Industry 5.0 Nodes Through Highly Accurate Neural CSI Feedback


Abstract:

Industry 5.0 refers to the fifth industrial revolution that leverages advanced technologies such as Internet of things (IoT) and artificial intelligence (AI) to increase ...Show More

Abstract:

Industry 5.0 refers to the fifth industrial revolution that leverages advanced technologies such as Internet of things (IoT) and artificial intelligence (AI) to increase efficiency, productivity, and flexibility in manufacturing and other industries. Wireless IoT devices help collect and transmit real-time data to support intelligent decision-making, while AI algorithms process and analyze the data to optimize production processes, predict equipment failure, and enhance supply chain management. To achieve efficient integration for the data fused by various sensors, the data should be perfectly synchronized and out-of-errors. In cellular-based sensors, this requires the base station (BS) to know the state of the channel at each node. In this work, we propose a novel method for CSI compression by learning an approximation for a sufficient statistics function. Our method establishes a new category of compression techniques based on the theory of sufficient statistics. Moreover, we present a detailed analysis of the upper bound of the prediction error in our specific scenario. We develop a Bayesian optimization framework to optimally select the adopted neural network architecture. The experimental results confirm that our solution outperforms both conventional and learning-based solutions in terms of reconstruction error, model size, and scalability.
Published in: IEEE Transactions on Consumer Electronics ( Volume: 69, Issue: 4, November 2023)
Page(s): 813 - 824
Date of Publication: 12 July 2023

ISSN Information:

Funding Agency:


References

References is not available for this document.