A Novel Offline Physical Layer Impairments Aware RWA Algorithm With Dedicated Path Protection Consideration

Abstract

Abstract-Physical layer impairments accumulate as light propagates through a lightpath in the transparent optical networks. Therefore, it is possible to provision a lightpath, while its quality of transmission (QoT) does not meet the required threshold. Considering the physical layer impairments in the network planning phase gives rise to a set of offline Impairments Aware Routing and Wavelength Assignment (IA-RWA) algorithms. There are very few offline IA-RWA algorithms that consider dedicated path protection demands. In this work we propose a novel offline IA-RWA algorithm, called Rahyab and perform a comparative performance evaluation study, which considers two enhanced algorithms from the literature. Simulation results indicates that demand pre-processing, diverse routing, and adaptive wavelength assignment are the main reasons of lower blocking rate of Rahyab algorithm compared to the selected algorithms.Index Terms-Dedicated path protection, impairments aware RWA, physical layer impairments, routing and wavelength assignment.

Optimal regenerator placement for path protection in impairment-aware WDM networks

“…In such a case, constraint (14) forces the value of v k j to be 0. The discussions for constraints (13) and (15) are similar.…”

A novel adaptive restoration algorithm for cost minimization of translucent optical networks

“…Many heuristic algorithms proposed in technical literature (e.g. [18] and [19]) have used this technique by considering demands sequentially, converting an oine RWA into an online RWA.…”

Physical Layer Impairment Model for CO-OFDM System Considering Amplified Spontaneous Emission and Four-wave Mixing

“…Results showed that for low signal powers the blocking probability was mainly due to the amplifiers noise, whereas for high signal powers the main contribution to the blocking probability came from the FWM effect. Azodolmolky et al thought Q factor was related to the received samples' statistical distribution, in which ASE noise, FWM effect, and XPM effect were modeled respectively according to the means and standard deviations of distributions of received samples [10]. There were also other models such as Eb/No model where Eb was the average energy per bit, the ratio of signal power and duration of each bit, and No was the noise power spectral density, the ratio of the noise power and the noise bandwidth [2].…”