As the communication infrastructure that sustains critical societal services, optical networks need to function in a secure and agile way. Thus, cognitive and automated security management functionalities are needed, fueled by the proliferating machine learning (ML) techniques and compatible with common network control entities and procedures. Automated management of optical network security requires advancements both in terms of the performance and efficiency of ML approaches for security diagnostics, as well as novel management architectures and functionalities. This paper tackles these challenges by proposing what we believe to be a novel functional block called the security operation center, describing its architecture, specifying key requirements on the supported functionalities, and providing guidelines on its integration with optical-layer controller. Moreover, to boost efficiency of ML-based security diagnostic techniques when processing high-dimensional optical performance monitoring data in the presence of previously unseen physical-layer attacks, we combine unsupervised and semi-supervised learning techniques with three different dimensionality reduction methods and analyze the resulting performance and trade-offs between the ML accuracy and run-time complexity.
Optical networks are critical infrastructure supporting vital services and are vulnerable to different types of malicious attacks targeting service disruption at the optical layer. Due to the various attack techniques causing diverse physical- layer effects, as well as the limitations and sparse placement of optical performance monitoring devices, such attacks are difficult to detect, and their signatures are unknown. This paper presents a Machine Learning (ML) framework for detection and identification of physical-layer attacks, based on experimental attack traces from an operator field-deployed testbed with coherent receivers. We perform in-band and out-of-band jamming signal insertion attacks, as well as polarization modulation attacks, each with varying intensities. We then evaluate 8 different ML classifiers in terms of their accuracy, and scalability in processing experimental data. The optical parameters critical for accurate attack identification are identified and the generalization of the models is validated. Results indicate that Artificial Neural Networks (ANNs) achieve 99.9% accuracy in attack type and intensity classification, and are capable of processing 1 million samples in less than 10 seconds.
In order to accomplish cost-efficient management of complex optical communication networks, operators are seeking automation of network diagnosis and management by means of Machine Learning (ML). To support these objectives, new functions are needed to enable cognitive, autonomous management of optical network security. This paper focuses on the challenges related to the performance of ML-based approaches for detection and localization of optical-layer attacks, and to their integration with standard Network Management Systems (NMSs). We propose a framework for cognitive security diagnostics that comprises an attack detection module with Supervised Learning (SL), Semi-Supervised Learning (SSL) and Unsupervised Learning (UL) approaches, and an attack localization module that deduces the location of a harmful connection and/or a breached link. The influence of false positives and false negatives is addressed by a newly proposed Window-based Attack Detection (WAD) approach. We provide practical implementation guidelines for the integration of the framework into the NMS and evaluate its performance in an experimental network testbed subjected to attacks, resulting with the largest optical-layer security experimental dataset reported to date.
The ongoing transition towards a networked society requires reliable and secure network infrastructure and services. As networks evolve from simple point to point systems towards complex, software-defined, ultra-high capacity and reach, and distributed cloud environments, new security challenges emerge. The EU-funded RECODIS project aims at coordinating and fostering research collaboration in Europe on disaster resiliency in communication networks. One of the disaster types, considered by RECODIS Working Group (WG) 4, are deliberate human-made attacks aimed at gaining unauthorized access to the network or disrupting the service. In order to develop methods for increasing network security in the presence of attacks, it is crucial to first identify the security vulnerabilities and attack methods that exploit them, as well as the capabilities and shortcomings of existing security schemes. To this end, the members of RECODIS WG4 performed a comprehensive overview of attack methods and security approaches from the literature. This overview covers the security vulnerabilities inherent to the underlying physical layer, the implications of software-defined networking to security, and security challenges in cloud networks spanning geographically distributed data centers.
The fifth generation (5G) of mobile technology, 5G is anticipated to be a significant leap in the evolution of mobile communication. 5G will be designed to attain 1000 times higher data volumes, 10 times lower latency, and 100 times more connected devices than its predecessor, 4G. Due to 5Gs ability to sustain high bandwidth per unit area, 5G is considered to be a cost-efficient solution to provide fixed wireless access (FWA) to households on a large scale. FWA is seen as an attractive alternative for fixed broadband access in scenarios where last mile access based on wired technologies is not economically viable. While approaches for enhancing user experience in a 5G FWA environment are investigated in the research community, the problem of providing cost-effective high capacity transport for FWA deployments still remains a major challenge. This is particularly challenging due to diverse transport network architectures and requirements imposed by different 5G deployment models. This paper addresses this problem by formulating a generalized joint-optimization framework to simultaneously plan wireless access and optical transport for 5G FWA networks in order to minimize the deployment cost while meeting various network requirements. We demonstrate the applicability of the proposed framework by applying it to a real scenario with a range of deployment options and where different types of optical x-haul solutions are considered. The results provide a cornerstone for deployment strategies that will be imperative for realizing a future-proof and cost-effective broadband access network.
17 December 2015 in Journal of Lightwave Technology
Due to the high data rates in optical networks, physical-layer attacks targeting service degradation, such as power jamming, can potentially lead to large data and revenue losses. Conventional network survivability approaches which establish link-disjoint working and backup paths to protect from component faults may not provide adequate protection for such attacks. Namely, the working and the backup paths, although link-disjoint, might both be affected by a single attack scenario due to specific attack propagation characteristics. To enhance the existing survivability approaches, we utilize the concept of an attack group (AG) which incorporates these characteristics to identify connections which can simultaneously be affected by a single attack. We apply this concept to dedicated path protection (DPP) and develop attack-aware DPP (AA-DPP) approaches which aim to establish AG-disjoint primary and backup paths in a cost-effective manner. We provide a two-step ILP formulation for the routing and wavelength assignment of the working and backup paths, as well as a heuristic for larger problem instances. Numerical results indicate that the proposed approaches provide dedicated path protection schemes with enhanced attack protection without using more resources (i.e., wavelengths, average path lengths) than standard DPP methods.