Technische Universität München
Lehrstuhl für Kommunikationsnetze
Strategic Network Planning for
Converged Optical Networks
Elena Grigoreva M. Sc.
Vollständiger Abdruck der von der Fakultät Elektrotechnik und Informationstechnik
der Technischen Universität München zur Erlangung des akademischen Grades eines
Doktor-Ingenieurs (Dr.-Ing.)
genehmigten Dissertation.
Vorsitzender:
Prof. Dr.-Ing. Norbert Hanik
Prüfer der Dissertation:
1.
Priv.-Doz. Dr.-Ing. habil. Carmen Mas Machuca
2.
Prof. Lena Wosinska, Ph. D.
Die Dissertation wurde am 13.11.2019 bei der Technischen Universität München ein-
gereicht und durch die Fakultät für Elektrotechnik und Informationstechnik am 25.03.2020
angenommen.
Strategic Network Planning for
Converged Optical Networks
Elena Grigoreva M. Sc.
July 18, 2020
Abstract
Declining revenues, increasing network costs, new stringent and heterogeneous user
requirements place a set of challenges for the Network Providers (NPs). An example
of new requirements could be ultra-reliable communications with 0.99999 availabil-
ity requirement for mission critical services, 1ms delay for tactile Internet or 10 Gbps
capacity per sector for the centralized 5G cloud Radio Access Network (RAN). The
traditional approach of building an individual optical networks for every application,
i.e., residential access and mobile backhaul in 4G, would be cost prohibitive in this
case. This means that the NPs have to find the ways to reduce the network costs and
maximize revenues. The strategic business decisions on the network implementation
and upgrade are made based on the results of strategic network planning.
In this dissertation, we take a holistic approach on strategic network planning for
converged optical networks. By converged networks we refer to sharing of the phys-
ical infrastructure (ducts, sites, etc.) and technologies for different network types
that traditionally would be independent (optical access and backhaul networks). The
strategic network planning work-flow was implemented in a Automated Map-Based
Strategic Fixed Network Planning Tool (AMS) to allow reproducibility and compara-
bility of the results, as well as future extensions.
We start from investigating the possible generalizations of the strategic network plan-
ning results by using graph-based road topology models instead of specific area maps.
Then, we look into implications of providing for the reliability requirements of the
ultra-reliable communication heterogeneous wireless protection for the mission criti-
cal communication. The studies are done for the converged Optical Distribution Net-
work (ODN) and for a wireless access network. We continue with analyzing the en-
ergy consumption of converged ODNs, its influence on reliability and delay. To enable
energy-saving methods, we propose a network-initiated equipment wake-up method
(to switch the equipment back to the active state from the sleep or dooze state) for
packet switched networks. We compare the delay and energy performance of our
proposed method to the measure the state-of-the-art circuit switched method.
Finally, we investigate network migration planning, i.e., multi-period strategic net-
work planning problem. To guarantee the network profitability, we maximize the Net
Present Value (NPV) under the condition that user’s requirements are satisfied. We
further deal with user uncertainty, i.e., users leaving the NP or user churn, with the
rational agent based algorithm.
ii
|
