Analytical Model for Blocking Probabilities Considering Security-Aware Reconfiguration in Spectrally-Spatially Elastic Optical Networks

Spectrally-spatially elastic optical networks (SSEONs) are designed to support high data rates, making physicallayer security a key concern to safeguard against eavesdropping and unauthorized access. A promising approach to enhancing physical-layer security lies in resource allocation pro-cess-specifically, the reconfiguration of lightpath requests to reduce exposure. However, securing SS-EONs during this process is challenging due to the complex interplay of spectral and spatial constraints. In this paper, we propose a reconfigurable service model (RSM) based on a continuous-time Markov chain framework to address this challenge. RSM analytically evaluates the impact of reconfiguration on both security and blocking performance while mitigating inter-core and inter-mode crosstalk effects. The model accommodates both single-class and multiclass requests, where the former involves uniform slot demands and the latter includes variable bandwidth requirements. To address scalability issues in single-hop scenarios, we introduce an iterative approximation model. Analytical results demonstrate that random reconfiguration can significantly enhance security with only a slight increase in blocking probability, depending on reconfiguration rate, link load, and link capacity. Our findings confirm that RSM effectively balances security and performance, offering a feasible solution for secure resource allocation in SSEONs.