We have recently proposed and demonstrated, by means of simulation, the benefits of a simple yet effective cognitive technique to enhance stateless Path Computation Element (PCE) algorithms with the aim of reducing the connection blocking probability when relying on a potentially non up-to-date Traffic Engineering Database (TED). In this paper, we employ that technique, called Elapsed Time Matrix (ETM), in the framework of the CHRON (Cognitive Heterogeneous Reconfigurable Optical Network) architecture and, more importantly, validate and analyse its performance in an emulation environment (rather than in a simulation environment) supporting impairment-aware lightpath establishment. Not only dynamic lightpath establishment on demand has been studied, but also restoration processes when facing optical link failures. Emulation results demonstrate that ETM reduces the blocking probability when establishing lightpaths on demand, and increases the percentage of successful restorations in case of optical link failure. Moreover, the use of that technique has little impact on lightpath setup time and lightpath restoration time, respectively.
Keywords-cognition, dynamic lightpath establishment, emulator, outdated database traffic engineering database 1 IntroductionOptical transport networks are increasing their complexity due to the use of heterogeneous transmission technologies and the support of services with different requirements [1]. One solution to efficiently control such heterogeneous optical networks comes from the use of cognitive techniques [2]. A cognitive network is defined as "a network with a process that can perceive current network conditions, and then plan, decide, and act on those conditions. The network can learn from these adaptations and use them to make future decisions, all while taking into account end-to-end goals" [2]. Cognitive networks involve the utilization of three types of elements: monitoring elements, which enable the network to be aware of current conditions; software adaptable elements, which enable the network to adapt to changing conditions; and cognitive processes, which learn or make use of past history to improve performance.Cognitive networks have proved to be an excellent solution for wireless networks [3], and have also been recently proposed for optical networks [4], like the CHRON (Cognitive Heterogeneous Reconfigurable Optical Network) approach [5,6]. The CHRON architecture, which will be briefly presented in Section II, is based on a central computation entity, the Cognitive Decision System (CDS) [7], which is responsible of making decisions on how to configure devices and deal with traffic demands, taking into account the current network status. As its name suggests, the CDS leverages on a learning process to improve its performance with acquired experience.When dealing with a lightpath establishment procedure, the CDS adopts a role similar to that of the Path Computation Element (PCE) [8], which has lately received increasing attention in the optical networking community. T...