2017/09/14 by Sadananda Behera, Behera, Sadananda, Jithin S. George +4
Computer Science · Engineering · #Advanced Optical Network Technologies #Advanced Photonic Communication Systems #Bandwidth (computing) #Computer network #Computer science #Electronic engineering #Engineering #FOS: Computer and information sciences #Frequency allocation #Multiplexing #Networking and Internet Architecture (cs.NI) #Optical Network Technologies #Orthogonal frequency-division multiplexing #Physical layer #Spectral efficiency #Statistical time division multiplexing #Telecommunications #cs.NI
paper · pdf · doi:10.48550/arxiv.1709.04616
published in arXiv (Cornell University) (Cornell University)
arxiv created 2017/09/14 · openalex publication_date 2017/09/14 · arxiv updated 2017/09/15 · openalex created_date 2022/10/03 · openalex updated_date 2026/08/05
Coherent Optical Orthogonal Frequency Division Multiplexing (CO-OFDM) based\nElastic Optical Network (EON) is one of the emerging technologies being\nconsidered for next generation high data rate optical network systems. Routing\nand Spectrum Allocation (RSA) is an important aspect of EON. Apart from\nspectral fragmentation created due to spectrum continuity and contiguity\nconstraints of RSA, transmission impairments such as shot noise, amplified\nspontaneous emission (ASE) beat noise due to coherent detection, crosstalk in\ncross-connect (XC), nonlinear interference, and filter narrowing, limit the\ntransmission reach of optical signals in EON. This paper focuses on the\ncross-layer joint optimization of delay-bandwidth product, fragmentation and\nlink congestion for RSA in CO-OFDM EON while considering the effect of physical\nlayer impairments. First, we formulate an optimal Integer Linear Programming\n(ILP) that achieves load-balancing in presence of transmission impairments and\nminimizes delay-bandwidth product along with fragmentation. We next propose a\nheuristic algorithm for large networks with two different demand ordering\ntechniques. We show the benefits of our algorithm compared to the existing load\nbalancing algorithm.\n