Project topics and materials
Get Complete Project Material File(s) Now! »
Operator’s Network Structuring
An operator’s network can be segmented into three main parts; these divisions are made to highlight differences in structure, hierarchy and functionality between them. This section briefly explains each of these parts: The access network is the stage of the telecommunication network nearest to the end user. The Passive Optical network (PON) is among the typically deployed optical telecommunication technology over that stage. The metropolitan network is the network part that interconnects the access network with other parts in a geographic region of variable size in order to perform a traffic aggregation role. This part of operator network has the most important traffic growth rate [3]. Ring or mesh topology is typically used over this network. All end-user companies/services are interconnecting through the core optical network which ties metropolitan areas all together, with connectivity to datacenters and other network operators, including the Internet. As a consequence, the core network carries the highest traffic rate [3], usually over a meshed topology. Core and metropolitan optical networks constitute the so-called “optical transport network”, which is concerned by transmission and multiplexing functionalities 2.1.
WDM Spectral Bands
Fiber-optic communication is mainly conducted in the wavelength region where optical fibers have small transmission loss. This low-loss wavelength region ranges from 1260 nm to 1625 nm, and is divided into five wavelength bands referred to as the O-, E-, S-, C- and L-bands, as shown in table 2.1. Today optical fibers show their lowest loss in the C-band (conventional band: 1530-1565 nm), and thus it is commonly used in WDM systems for optical transport networks [14].
The L-band (long-wavelength band: 1565-1625 nm) is the second lowest-loss wavelength band. It is a usual option when the use of the C-band is not sufficient to meet the bandwidth demand. Some WDM systems can use the L-band if needed. To organize the spectrum allocation over the C- and L-bands, most of installed WDM systems use the fixed grid technology of 50 GHz channel spacing to organize the spectrum allocation. The 50 GHz of channel spacing prevents the use of high data-rate transponders (ex: 400 Gb/s and beyond), which require higher baudrates (wider channels than 50 GHz). On the contrary, the flex-grid technology permits flexible channel spacing. The ITU recommendation G.694.1 [15] supports a variety of channel spacing’s ranging from 12.5 GHz to 100 GHz and wider. Chapter 5 illustrates the benefits behind deploying the flex-grid technology over a WDM system.
Optical Transport Network Architecture
To ensure a trustworthy delivery of the transmitted information over the network, some protocols are organizing the network’s equipment workflow. Each protocol is created to serve a specific part and/or function of the network. A protocol stack refers to a group of protocols that are running concurrently to harmonize the workflow of the network equipment. The network industry abstracts and classifies networks’ protocols into groups and layers associated to functions and information to be exchanged. Optical transport networks’ protocols, like any type of network, can be divided into three main planes.
Management plane: refers to an application provided by the vendor and utilized by the network operator to handle and configure WDM services, such as provisioning, monitoring and fault location. Some of these functions can be delegated to other software such as the control plane.
Control plane: refers to a set of software that equipment uses to exchange configuration and service-related data, using messages logically separated from the main traffic . The Generalized MultiProtocol Label Switching (GMPLS) protocol suite is one of the main set of standards [16] to support communication within a control plane of optical networks;
Data plane: is responsible for functionalities like data forwarding, segmentation and assembly.
Optical transport networks can be operated using either a distributed manner, such as the case of the GMPLS protocols, using a centralized entity, such as promoted by the Software-Defined Networking concept (SDN) [17], or a mix of both, which is the typical case (management traffic over an in-band IS-IS/OSPFrouted network). The authority of both Management and Control planes might overlap from one concept to another. In this dissertation, we use the term « control and management level » to refer to all functions of both control and management planes without making assumptions on how operations are balanced between them.
Table of contents :
Acknowledgements
Abstract
Résumé
List of Figures
List of Tables
List of Acronyms
1 General Introduction
2 Research Context
2.1 Operator’s Network Structuring
2.2 Transport Evolution
2.2.1 Opaque Networks
2.2.2 Transparent Networks
2.2.3 Translucent Networks
2.3 WDM Spectral Bands
2.4 Optical Transport Network Architecture
2.5 Network Operators’ Challenges
2.5.1 First Lever: Lighting Up More Fibers
2.5.2 Second Lever: Increasing Spectral Efficiency
2.5.3 Third Lever: Applying Equipment Interoperability
3 Towards Fully Inter-operable Optical Networks
3.1 Introduction
3.2 Single-vendor Scenario: WDM System Components
3.2.1 WDM Transponders
3.2.2 Re-configurable Optical Add-Drop Multiplexers (ROADMs)
3.2.3 Optical amplifiers
3.2.4 Control and Management (C/M) level
3.3 Single Vendor Lightpath: Setup Steps
3.3.1 Before System Installation
3.3.2 Before Lightpath Setup
3.3.3 Path computation:
3.3.4 Path Configuration
3.3.5 Monitoring
3.4 Alien Wavelengths: Challenges and Classifications
3.4.1 Alien Wavelength Challenges
3.4.2 Alien Classifications
3.5 Interoperability-related initiatives
3.5.1 IETF’s work
3.5.2 OpenConfig
3.5.3 OpenROADM
3.5.4 Telecom Infra project (TIP)
3.5.5 Comparison
3.6 Alien approaches: Physical Level
3.6.1 40G/100G Coherent Alien on 10G Compensated Systems
3.6.2 100G and Beyond Bransponders on Coherent Uncompensated Systems
3.7 Alien approaches: Control/Management Level
3.7.1 Translation Approach
3.7.2 SDN Controllers’ Cooperation
3.7.3 Open Line System (OLS)
3.7.4 OpenROADM-based Approach
3.7.5 RSVP-TE-based Approach
3.8 Approaches Comparison
3.8.1 Translation Approach
3.8.2 SDN Controllers Cooperation
3.8.3 Open Line System
3.8.4 OpenROADM-based Approach
3.8.5 RSVP-TE-based Approach
3.8.6 Combinations
3.9 Conclusion
4 Automatic Alien Transponder Configuration using RSVP-TE Signaling
4.1 Introduction
4.2 RSVP-TE-Based Approach
4.3 RSVP-TE-Based Approach Combined with GNPy
4.4 RSVP-TE-Based Approach including BVTs
4.4.1 Proposal
4.5 Implementation over The Testbed
4.6 Demonstration
4.7 Conclusion
5 Path Computation Module (PCM) 75
5.1 From Fixed- to Flex-grid
5.2 PCM Execution Scenarios
5.3 PCM main structure
5.3.1 Shortest Path Selection
5.3.2 Resource availability check
5.3.3 Feasibility Check
5.4 Inputs and Outputs Files
5.4.1 Network topology.json
5.4.2 Catalog of transponders.json
5.4.3 Selected-path.json
5.5 Conclusion
6 Conclusion and Perspectives
6.1 Assumptions and Future Perspectives
7 Contributions and Distinctions
