Modeling and Modulation Optimization for OWC Limited by Electronic and Photonic Bandwidth

TL;DR

This paper models bandwidth limitations in optical wireless communication systems and optimizes the signal spectrum to enhance throughput beyond traditional bandwidth constraints.

Contribution

It introduces a pole-zero transfer function model for component bandwidth limitations and proposes a novel optimization algorithm for spectral enhancement.

Findings

Optimization improves throughput in bandwidth-limited OWC links.

The Newton-based algorithm outperforms Hughes-Hartogs in finding near-optimal spectra.

Employing the multi-stage response model enhances performance under bandwidth constraints.

Abstract

In contrast to radio frequency (RF), where the modulation bandwidth is restricted by regulations to avoid interference, the available bandwidth in optical wireless communication (OWC) is primarily constrained by system components. To investigate their frequency characteristics, we review the bandwidth limitations of components in the PHY layer of OWC links. Such limitations typically contribute to a decay in the frequency profile of the gain-to-noise ratio (GNR), which can be modeled by a pole-zero transfer function that is generally low-pass. To boost performance, we optimize the signal power spectral density (PSD) of DC-biased optical orthogonal frequency-division multiplexing (DCO-OFDM) which allows for modulation beyond the 3-dB end-to-end bandwidth. We express the Lagrangian-optimized throughput versus the maximum modulation frequency, for an M-zero N-pole low-pass GNR optical link. For optimization implementation, we compare a novel Newton-based algorithm with a newly accelerated version of the Hughes-Hartogs (HH) algorithm, to find the (near-) optimal signal spectrum for theoretical, measured and simulated GNR responses. As demonstrated numerically, employing the proposed multi-stage response model for optimization improves performance in dealing with the successive bandwidth limitations.