Noncoherent OOK Symbol Detection with Supervised-Learning Approach for BCC

TL;DR

This paper introduces a supervised-learning based noncoherent detection method for OOK symbols in body channel communication, enhancing accuracy in WBANs for medical device applications.

Contribution

It proposes novel detection techniques that leverage distributed reception and supervised learning to improve noncoherent OOK symbol detection in BCC systems.

Findings

Validated the proposed detection techniques through numerical simulations.

Demonstrated improved performance over traditional methods.

Showed applicability for real-world WBAN medical devices.

Abstract

There has been a continuing demand for improving the accuracy and ease of use of medical devices used on or around the human body. Communication is critical to medical applications, and wireless body area networks (WBANs) have the potential to revolutionize diagnosis. Despite its importance, WBAN technology is still in its infancy and requires much research. We consider body channel communication (BCC), which uses the whole body as well as the skin as a medium for communication. BCC is sensitive to the body's natural circulation and movement, which requires a noncoherent model for wireless communication. To accurately handle practical applications for electronic devices working on or inside a human body, we configure a realistic system model for BCC with on-off keying (OOK) modulation. We propose novel detection techniques for OOK symbols and improve the performance by exploiting distributed reception and supervised-learning approaches. Numerical results show that the proposed techniques are valid for noncoherent OOK transmissions for BCC.