Quantum Key Distribution (QKD) is a foundational technology for future secure communications, and several QKD networks have been already deployed and tested around the world using optical fibers. However, these networks cannot scale in size due to the inefficiency of fiber QKD networks with increasing distances, making satellite networks a major candidate for long-distance QKD networks. In satellite QKD networks, satellites and ground stations can act as trusted relays, distributing keys between satellite-ground station pairs to serve requests among ground stations. Satellite QKD networks face fundamental challenges due the time-varying nature of the connection between ground stations and satellites, caused by both the satellite’s orbital movement and fluctuating atmospheric attenuation. Thus, it is necessary to design novel schemes to dynamically allocate resources for satellite QKD networks that adapt to evolving network conditions in different time intervals. In this work, we investigate the problem of resource allocation in satellite QKD networks taking into account the changing key generation rates, calculated according to evolving weather conditions and satellite visibility. We first model the achievable key rate of connections between satellite and ground stations under different weather conditions, which is used as an input for optimization. We formulate a Mixed-Integer Linear Programming (MILP) model to allocate resources in satellite QKD networks, which decides both link assignments (i.e., deciding which ground to connect to for satellites) and the appropriate routing path for the trusted relay. In addition, the MILP models multiple timeslots and considers keys stored in the quantum key pool (QKP), allowing keys generated during low-load periods to be used later during high-load periods. Moreover, we propose to decide the link configuration with heuristic algorithms and then utilize ILP to decide the appropriate routing path for the trusted relay, which significantly reduces the execution time. The numerical results show that incorporating link configuration within the ILP achieves up to 20% more total served keys compared to heuristicbased baseline approaches, but with an execution time up to 700x longer.

Network Function Virtualization (NFV), as a promising paradigm, speeds up the service deployment by separating network functions from proprietary devices and deploying them on common servers in the form of software. Any service in NFV-enabled networks is achieved as a Service Function Chain (SFC) which consists of a series of ordered Virtual Network Functions (VNFs). However, migration of VNFs for more flexible services within a dynamic NFV-enabled network is a key challenge to be addressed. Current VNF migration studies mainly focus on single VNF migration decisions without considering the sharing and concurrent migration of VNF instances. In this paper, we assume that each deployed VNF is used by multiple SFCs and deal with the optimal placement for the contemporaneous migration of VNFs based on the actual network situation. We formalize the VNF migration and SFC reconfiguration problem as a mathematical model, which aims to minimize the VNF migration between nodes or the total number of core changes per node. The approach is a two-stage MILP based on optimal order to solve the reconfiguration. Extensive evaluation shows that the proposed approach can reduce the change in terms of location or number of cores per node in a 6-node and 14-node networks while ensuring network latency compared with the model without reconfiguration.

In recent years, machine learning algorithms have been widely used in optical networks to solve complex problems such as routing, resource allocation, among others. In routing, modulation and spectrum allocation (RMSA) problems, machine learning algorithms can be used to learn patterns in historical data and find good solutions without having to explore all existing solutions. In this paper, we propose an algorithm based on a classification model to solve the routing problem in elastic optical networks. This algorithm predicts the route according to the call request information and the state of the network links. The dataset used to train the proposal is obtained through a dynamic routing algorithm. With this dataset, two versions of the proposal are evaluated with different sets of routes according to the frequency distribution of these routes. Three network topologies are used to evaluate the routing algorithms: six-node, NSFNET and European optical network. The results are compared with two other routing algorithms: Yen’s algorithm (k shortest routes) and the spectrum continuity based shortest path (SCSP) algorithm. This last algorithm is used to train our proposal. Our proposal outperformed the Yen’s algorithm in the three network topologies in terms of blocking probability. When compared to the SCSP algorithm, our proposal obtained an average performance gain of 15% and 25% in the six-node and NSFNET network topologies, respectively. In the European network topology, our proposal achieved an average performance gain at the lowest network loads of 23.19%. In all network topologies considered, our proposal reduced the time spent to find the RMSA solution compared to the SCSP algorithm.

Network protection is a key solution and an important component in the requirements of virtualization over elastic optical networks (EONs). In this paper, we examine the significance of network virtualization with survivability design against single-link failures and shared risk link groups (SRLG) under dedicated protection and bandwidth squeezing schemes. We study an optimization version of the routing, modulation and spectrum allocation (RMSA) problem with the goal of maximizing the revenue from the accommodated requests of the virtual optical networks (VONs) or slices. We present a Mixed Integer Linear Programming (MILP) formulation for the problem and evaluate some numerical results. We suggest a heuristic that utilizes techniques of decomposition of the problem, which can be employed to obtain a near optimal solution that has a per instance guarantee on the closeness to the optimal solution.

Survivability is an important component in the requirements in elastic optical networks (EONs) with virtualization. In this paper, we examine the significance of network survivability design against single-link failure under dedicated protection and bandwidth squeezing schemes under multiple virtual topologies. We proposed an integer linear programming (ILP) formulation and a genetic algorithm (GA) to derive some different types of protection for each virtual topology considering routing and a channel-based spectrum approach. The proposed ILP and GA provide efficient survivability results and resource savings (in terms of spectrum) for a full design of modern virtualized EONs with different kinds of mechanisms for protection.

Elastic optical network (EON) is fast becoming a key solution for designing optical network with better usage of spectrum resources or other objectives of interest to tenant and/or operators. A primary concern of EONs is to protect the network against failures of its elements, because this kind of event can provoke the loss of substantial amount of traffic. In this paper, we propose a new mixed integer linear programming (MILP) formulation for protecting the network traffic against multiple link failures. The key idea is to use bandwidth squeezing together with grooming capability to provide few extra traffic for protection and guarantee a minimum bandwidth for each source-destination node pair under multiple failure events. The proposed formulation solves the virtual topology design problem jointly with the grooming, routing, modulation and spectrum allocation (RMSA) tasks. Due to the non-deterministic polynomial time (NP-hard) nature of the proposed MILP formulation, a heuristic strategy (referred to as two-step MILP) for large networks is also proposed. The solutions and performance of the proposed MILP formulation and two-step MILP analyzed through case studies in a small network. In addition, the performance of three large networks is assessed for cases scenarios where connections are under different service-level agreement (SLA). In view of proposed formulation and two-step MILP, it is possible to identify the configurations that ensure better usage of spectrum resources with different kinds of protection against single or multiple link failures.

In this paper, we propose an evolutionary multiobjective regenerator placement strategy for elastic optical networks (eMORP). The proposed optimization strategy uses the genetic algorithm NSGA-II to determine non-dominated solutions when the call request blocking probability and the total amount of regenerators used in the network are taken into account. In our simulations, we considered the amplified spontaneous emission noise generated by optical amplifiers (in-line, booster, and pre-amplifier) as physical impairment. Two recently proposed heuristics for regenerator assignment have been used, together with the regenerator placement strategies proposed and analyzed in this paper, for comparison purpose. The results obtained for two different network physical topologies state the efficiency of eMORP. Our regenerator placement strategy reduced considerably the call request blocking probability for the same number of regenerators in the network, as well as it acquired efficient solutions with just a fraction of the nodes with regeneration capability in comparison to other heuristics presented in the literature.

Spectrum-sliced elastic optical path networks (SLICE) enable flexible bandwidth provisioning, which allows efficient resource utilization and support to heterogeneous bandwidth demands. This makes SLICE a very promising networking architecture. In SLICE, finding a route and a slice of the spectrum is an important design problem, which is known as the Routing and Spectrum Assignment (RSA) problem. In this paper, we present a new heuristic, based on Yen's k-shortest path algorithm and capable of dealing with the traffic heterogeneity to find appropriate sets of paths for each source-destination node pairs to mitigate the bottleneck in the network. The results show that the proposed algorithm reduces the lightpath blocking probability and achieves significantly improved spectrum efficiency. We also analyse the maximum number of alternate shortest paths for minimising the blocking probability.

This paper proposes a linear formulation and an iterative heuristic, both with traffic grooming capability, which can maximize the number of remaining available routes and minimize the number of transceivers in Elastic Optical Networks (EON). The aim of the proposal is to preserve the open capacity for the accommodation of future unknown demands. Case studies are carried out in order to analyze the basic properties of the formulation in a small network, and the heuristic is used for moderate larger networks. The results suggest that it is feasible to preserve enough open capacity to avoid blocking of future requests in EON with scarce resources.

Transparent optical networks offer a common transport platform for services with different bit-rates, transmission formats and technologies. In such a heterogeneous scenario, it is an inefficient use of network resource to provision end-to-end lightpaths without considering the different optical performance requirements of the different service interfaces. This paper proposes an efficient routing and wavelength assignment algorithm, based on service interface tolerance to optical noise and chromatic dispersion that allocates a lightpath with just enough end-to-end optical signal integrity for the connecting service. We investigate its performance in transparent optical networks with heterogeneous services at 10 and 40-Gb/s. For 40-Gb/s services, optical duobinary (ODB) and RZ-DQPSK modulation formats are assumed. We demonstrate that network blocking can significantly be reduced by the proposed algorithm.

Physical layer impairments may require highly complex implementations of the control plane in dynamic networks with quality of service (QoS) constraints. In order to alleviate this problem, we propose that dynamic wavelength assignment should be based on the elementary first-fit (FF) algorithm using off-line optimized priority lists. It is shown that this may significantly reduce the effects of physical impairments with no additional computational cost for the control of dynamic networks. A simple metric is proposed to evaluate optical QoS taking the performance of the network as a whole. Different heuristics are presented and evaluated. Results show that there are particular lists capable of reducing in four orders of magnitude the probability of FWM impacting network performance.

In this paper we present an exact analytical model for the FIFO delay-line buffer in an asynchronous network scenario considering that packets arrive to the buffer following any inter-arrival time process. The model enables us to study the impact that different inter-arrival distributions have on the buffer performance and design process. The accuracy of the model is compared with simulations for deterministic, uniform, exponential and Weibull inter-arrival time distributions. Then, the effects of these distributions, input load and delay line spacings on the buffer performance are investigated.

In this paper we address the problem of providing delay-line buffers with the capability to differentiate services in optical packet switching networks. A simple and efficient policy for acceptance and discarding of optical packets is proposed and its analytical model is presented. The model enables the exact calculation of packet loss probability and average delay for Poisson arrivals and any packet length distribution. The buffer performance as predicted by the model is compared with simulations for Pareto packet length distribution.

The paper introduces a new analytical approach for estimating blocking probabilities in linear all-optical networks. The assumptions of the classical Lee approximation are discussed and their effects are evaluated. The paper proposes a better substitute for the independent link assumption. The new assumption takes all active paths, as well as all free links, as independent objects on the network topology. The new model is shown to generate estimates that exactly fit the blocking probabilities obtained through simulations on linear topologies using Poissonian, single-wavelength spatially homogeneous traffic. Finally, the independent wavelength assumption is also evaluated.