[Paper Review] A survey on Crosstalk and Routing, Modulation Selection, Core and Spectrum Allocation in Elastic Optical Networks

Ítalo Brasileiro, Lucas Rodrigues Costa+1 more|arXiv (Cornell University)|Jul 19, 2019

Advanced Optical Network Technologies29 references11 citations

TL;DR

This survey provides a comprehensive analysis of Routing, Modulation, Spectrum, and Core Allocation (RMSCA) in Elastic Optical Networks (EONs) enhanced with Multi-Core Fibers (MCF) and Space-Division Multiplexing (SDM). It identifies crosstalk as the dominant physical-layer impairment, evaluates 32 key papers on RMSCA, and highlights the lack of cross-comparison among advanced algorithms and limited consideration of other physical layer effects beyond crosstalk.

ABSTRACT

Elastic Optical Networks (EON) emerge as a viable solution to supply the current growing demand for bandwidth. With the application of multi-core fibers (MCF) in EON links, it is possible to increase the availability of spectral resources. An EON network with MCF enables Space-Division Multiplexing (SDM), allowing the use of more resources in the fibers and increasing the capacity of attending circuit requests. However, the use of SDM brings some problems of interference between the circuits of a fiber, with greater emphasis on crosstalk interference. In this paper, some important concepts around EON are presented, along with the characterization of SDM technology. The Routing, Modulation, Spectrum and Core Allocation (RMSCA) problem is also characterized, and some solutions currently found in the literature are cited. After, the impact of crosstalk interference is discussed, and which elements are responsible for its occurrence. The paper is concludes with an evaluation of the state of the art, and the discrimination of the main points found from the study of papers related to the SDM-EON scenario.

Motivation & Objective

• To analyze the state of the art in Routing, Modulation, Spectrum, and Core Allocation (RMSCA) in Space-Division Multiplexing Elastic Optical Networks (SDM-EON).
• To evaluate the impact of crosstalk interference on network performance and resource allocation in multi-core fiber-based EONs.
• To identify research gaps in current RMSCA solutions, including lack of cross-comparison among proposed algorithms and limited modeling of physical layer impairments beyond crosstalk.
• To assess the performance of existing RMSCA algorithms against classical EON approaches such as Dijkstra and First Fit, particularly in dynamic traffic scenarios.
• To highlight challenges in SDM-EON deployment, including cost-effective switching equipment, crosstalk mitigation, and achieving performance parity with n single-core fibers.

Proposed method

• Systematic literature review of 32 papers on SDM-EON and RMSCA, focusing on routing, modulation, spectrum, and core allocation strategies.
• Categorization of proposed RMSCA algorithms based on core allocation, spectrum assignment (e.g., First Fit, Random Fit), and crosstalk awareness.
• Analysis of network configurations: 17 papers use 7-core fibers, 16 use dynamic traffic, and 11 report bandwidth demands between 1–10 slots.
• Evaluation of crosstalk impact using a threshold of -25 dB, where high crosstalk renders spectral slots unusable despite availability.
• Comparison of proposed RMSCA solutions against classical algorithms (e.g., Dijkstra, First Fit), with emphasis on blocking rate differences up to 74.13%.
• Identification of architectural challenges, including the need for specialized switches, multiplexers (e.g., photonic-lantern multiplexers), and amplifiers for multi-core fiber switching.

Experimental results

Research questions

• RQ1 How do different RMSCA algorithms perform in dynamic traffic scenarios with multi-core fiber-based EONs?
• RQ2 To what extent does crosstalk degradation affect network blocking rates and spectral resource availability in SDM-EONs?
• RQ3 Why are most RMSCA solutions compared only to classical EON algorithms (e.g., Dijkstra, First Fit), and what is the performance gap between modern RMSCA proposals?
• RQ4 What are the dominant physical layer impairments in SDM-EONs beyond crosstalk, and why are they underrepresented in current literature?
• RQ5 How can future RMSCA algorithms be designed to be impairment-aware, incorporating multiple physical layer effects for improved real-world accuracy?

Key findings

• Crosstalk is the most studied physical-layer impairment in SDM-EON literature, with 17 out of 32 papers explicitly modeling its impact on spectrum allocation.
• The blocking rate in SDM-EONs can vary by up to 74.13% depending on whether crosstalk is modeled dynamically or statically, highlighting its critical impact on network performance.
• Only 2 out of 32 papers consider physical layer effects beyond crosstalk, indicating a significant research gap in modeling comprehensive impairments such as nonlinearities or polarization crosstalk.
• Four papers address network survivability and protection, while only one explores optical traffic aggregation, underscoring limited work in resilience and traffic engineering.
• Despite 15 papers proposing RMSCA solutions, no direct comparison exists between these advanced algorithms, limiting the ability to identify optimal strategies.
• The performance of a multi-core fiber (MCF) with n cores must match that of n single-core fibers (SCF) to justify cost, making crosstalk tolerance a key design requirement for practical deployment.