Clipping and Quantization Impact in Power-Imbalanced Upstream DSCM Coherent PONs

Future passive optical networks (PONs) are target- ing data rates of 200 Gbps/λ and beyond, while moving toward end-to-end coherent transmission from metro to access networks. In this context, deploying digital subcarrier multiplexing (DSCM) is attractive for enabling low-latency and high-flexibility services, such as 5G/6G fronthauling. DSCM offers finer granularity in bit-rate allocation compared to single-carrier coherent solution. However, adapting coherent transmission in general, and DSCM in particular, to the shared PON infrastructure is challenging due to several constraints, including the wide dynamic range in upstream transmission. Additionally, the multi-carrier nature of DSCM increases the peak values relative to the signal average in the Tx and Rx electrical signals, and thus higher penalties are exhibited in systems with low-resolution analog-to- digital converters (ADCs). This penalty further increases when considering the inherent power imbalance in PON architectures, making it a critical limitation in DSCM-PON performance that must be carefully considered in physical-layer design and network planning. In this context, clipping can be employed to mitigate high signal peaks, but it must be carefully optimized to balance clipping and quantization penalties. Therefore, this paper investigates the impact of subcarrier power imbalance on the performance of upstream DSCM coherent PON, with a particular focus on clipping and quantization effects introduced by finite-resolution ADCs.