Spectrally/spatially flexible optical networks (SS-FONs) are a promising solution to future traffic requirements in optical backbone networks. SS-FONs exploit the spatial dimension to increase the network capacity, while preserving the resource management flexibility, as they still operate within a flex-grid composed of small frequency slots (slices). A flex-grid makes it possible to realize transmission using super-channels (SChs) that comprise a set of contiguous slots. In this paper, we focus on spectral SChs, i.e., a SCh realized only on one spatial mode on each fiber. In SS-FONs, setting up and tearing down multiple lightpath requests within a flex-grid may result in spectrum fragmentation and, in turn, blocking of requests. In this work, we investigate several fragmentation metrics in a SS-FON. The problem of identifying appropriate metrics to measure fragmentation has been investigated in single-core elastic optical networks, but to the best of our knowledge, there is no such research available for SS-FONs. Therefore, we propose several fragmentation metrics for SS-FONs. We introduce the concept of bordering super-channels (B-SChs), i.e., SChs whose spectrum is allocated at the border of already allocated spectrum slots, which show promising results with regard to minimizing fragmentation. The investigation of all candidate B-SChs allows us to find the one that minimizes the network fragmentation and, in turn, spectrum waste. Hence, we propose a fragmentation-aware (FA) algorithm with bordering SChs that assigns optical resources to dynamic requests by utilizing information from the proposed fragmentation metrics and the set of candidate B-SChs. Experiments on a representative network topology show that the investigation of multiple B-SChs in the FA algorithm reduces the blocking probability when compared to the reference FA algorithms. Finally, we analyze the impact of a spatial continuity constraint on the network fragmentation.

Rapid growth of network traffic causes the need for the development of new network technologies. Artificial intelligence provides suitable tools to improve currently used network optimization methods. In this paper, we propose a procedure for network traffic prediction. Based on optical networks’ (and other network technologies) characteristics, we focus on the prediction of fixed bitrate levels called traffic levels. We develop and evaluate two approaches based on different supervised machine learning (ML) methods—classification and regression. We examine four different ML models with various selected features. The tested datasets are based on real traffic patterns provided by the Seattle Internet Exchange Point (SIX). Obtained results are analyzed using a new quality metric, which allows researchers to find the best forecasting algorithm in terms of network resources usage and operational costs. Our research shows that regression provides better results than classification in case of all analyzed datasets. Additionally, the final choice of the most appropriate ML algorithm and model should depend on the network operator expectations.

Efficient resource allocation is a key task for network operators to provide services for users. Its optimization, in short-term perspective, allows to reduce operating costs, provide required quality of service and detect dataflow anomalies. It is crucial to develop tools to better understand and forecast traffic behaviour of optical networks. In this work, we propose a Machine Learning (ML) procedure for short-term traffic volumes forecasting, using the Linear Discriminant Analysis (LDA) classifier. The main novelty, comparing to the other works in this field, is that in our approach, we predict fixed bitrates levels of the traffic, instead of exact traffic volume. In consequence, the traffic prediction problem is formulated as a classification task, while most of the previous works in this field model the traffic using time series. We also present numerical results as a proof of effectiveness of the described approach. We examine real traffic collected by Seattle Exchange Point, together with traffic generated based on real data (which consist of real traffic dependencies). We obtained up to 93% of correct predictions for real data traffic.

The paper studies efficient modeling and prediction of daily traffic patterns in transport telecommunication networks. The investigation is carried out using two historical datasets, namely WASK and SIX, which collect flows from edge nodes of two networks of different size. WASK is a novel dataset introduced and analyzed for the first time in this paper, while SIX is a well-known source of network flows. For the considered datasets, the paper proposes traffic modeling and prediction methods. For traffic modeling, the Fourier Transform is applied. For traffic prediction, two approaches are proposed—modeling-based (the forecasting model is generated based on historical traffic models) and machine learning-based (network traffic is handled as a data stream where chunk-based regression methods are applied for forecasting). Then, extensive simulations are performed to verify efficiency of the approaches and their comparison. The proposed modeling method revealed high efficiency especially for the SIX dataset, where the average error was lower than 0.1%. The efficiency of two forecasting approaches differs with datasets–modeling-based methods achieved lower errors for SIX while machine learning-based for WASK. The average prediction error for SIX reached 3.36% while forecasting for WASK turned out extremely challenging.

In this paper, we focus on the problem of resource allocation for time-varying traffic in translucent space-division multiplexing (SDM) elastic optical networks (EON) with back-to-back (B2B) signal regeneration. The performance is measured in terms of bandwidth blocking probability, usage of spectral resource (i.e. minimizing allocated bandwidth overhead compared to required traffic) and number of lightpaths' reallocations. We formulate and evaluate an effective algorithm for time-varying traffic allocation. Simulations are run on a representative network topology with time-varying traffic based on several types of network services with different underlying stochastic processes in order to properly model the real traffic.

We consider a dynamic lightpath provisioning problem in translucent spectrally–spatially flexible optical networks (SS-FONs) in which flexible signal regeneration is achieved with transponders operating in back-to-back (B2B) configurations. In the analyzed scenario, an important aspect that has a significant impact on the network performance is the decision on placement of transponders that can be used for two purposes: transmitting/receiving (add/drop) of optical signals at the source/destination nodes and regeneration of the signals at some intermediate nodes. We propose a new algorithm called scaled average used regenerators (SAUR). The key idea of the SAUR method is based on a data analytics approach, i.e., the algorithm exploits information on network traffic characteristics and the applied dynamic routing algorithm to obtain additional knowledge for the decision on transponder placement. The numerical results obtained for two representative topologies highlight that the proposed SAUR method outperforms reference algorithms in terms of the amount of traffic that can be accepted in the network. In other words, placement of transponders yielded by the SAUR method allows to increase the SS-FON throughput using only the existing resources, i.e., the network operator does not have to invest in new devices or fibers.