Skip to content

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

2026

Rearrangeable Route Assignment for Flexible-Grid OXC-Clos Networks

With the surge in bandwidth demand, optical cross-connects (OXCs) face scalability issues. The flexible-grid OXC-Clos network composed of small-scale OXC modules offers a scalable solution. However, strictly nonblocking (SNB) and wide-sense nonblocking (WSNB) designs suffer from high costs. To remarkably reduce the network cost, this paper explores rearrangeable route assignment for flexible-grid OXC-Clos networks without wavelength converters (WCs). We generalize the bipartite graph model for classical Clos networks and propose an extended bipartite graph model, which constructs a distinct bipartite graph for each frequency slot (fSlot), to capture the feature that a lightpath (LP) may encounter distinct conflicts in successive fSlots it uses. The extended model maps the LPs occupying multiple fSlots to different bipartite graphs, rendering the coloring of different graphs mutually coupled. We propose a recursive coloring process to resolve this coupling and properly color the graphs, thereby realizing routing assignment and network reconfiguration. Based on this process, we derive the rearrangeably nonblocking (RNB) condition, which is independent of the number of LP granularity types, delivering much lower costs than that of WSNB networks. For further cost reduction, we explore blocking flexible-grid OXC-Clos networks where a very low blocking probability is permitted. We first reveal a blocking property via simulation and then devise a recursive first-fit (FF) routing strategy by analyzing blocking scenarios. Leveraging this property and the recursive FF strategy, we demonstrate that blocking networks achieve a substantial cost reduction (up to 38.2%) relative to RNB networks.

Qi-Xiang Lai, Tong Ye, Yi-Bei Yao et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.