Abstract
Optical communication technologies are increasingly
being used in implantable and wearable biomedical systems,
particularly for applications such as phototherapy, photometry, and
biosensing. These optical biomedical implants show strong
potential in the study and treatment of neurological conditions like
Parkinson’s and Alzheimer’s disease, as well as cardiovascular
disorders. They are also promising for managing depression-like
behaviours, restoring vision, and supporting cancer therapies.
However, optical channels within the body or on the skin are
complex. They are highly nonlinear, time-varying, and subject to
scattering and multipath fading. In addition, noise can distort the
transmitted signals, and simple amplification alone cannot fully
restore signal clarity or shape. To address these challenges, this
study proposes an AI-enabled optical wireless communication
system tailored for implantable and wearable biomedical
applications. A network equalisation technique was integrated to
compensate for distortions caused by channel impairments, thereby
improving the fidelity of transmitted signals. Simulation results
indicate that while amplifier gain flattening enhanced signal
amplification, the neural network-based equalisation achieved a
notable reduction in bit error rate (BER) compared with preequalisation
conditions, confirming its effectiveness in improving
communication performance in biomedical systems.