Planning Optical Networks for Unexpected Traffic Growth

Transmission-Aware Bandwidth Variable Transceiver Allocation in DWDM Optical Networks

“…It can be observed that the solution with RPs is 15% closer to the requested overall traffic in the early planning years, as compared to the solution without RPs. Both are closer to the requested traffic than [6] in the early years, as shown in Fig. 4(a).…”

“…A value of 13 dB and 12 dB for OSNR and GSNR, respectively, is obtained, which is in good agreement with [16] and proves the validity of the implementation. For the comparison of system throughput, required TRX number and occupied frequency range, we consider a year as a planning period and undertake a ten-year planning (2021-2030) by creating a TD matrix for a 17-node German network based on the traffic growth model discussed in our previous work [6]. Fig.…”

“…Fig. 4(a) shows the system throughput of our proposed heuristic (with and without RPs) and results from [6] with respect to the requested traffic. It can be observed that the solution with RPs is 15% closer to the requested overall traffic in the early planning years, as compared to the solution without RPs.…”

“…Therefore, we exploit in this work the advantages of using RPs along with BV-TRXs in a flex-grid C+L-Band DWDM optical network, by proposing a novel BV-TRX allocation heuristic for the RCSA problem in such networks. Our results, applying a realistic traffic model [6] on a 17-node German network, show that using RPs not only allows operators to meet the yearly requested traffic, but also leads to a reduction of up to 25% of the total number of BV-TRXs, and savings of up to 50% in spectrum utilization, as compared to a solution without regeneration.…”

“…Several RCSA algorithms have been studied by the authors in [5,6,10] , each with different objective functions, based on certain assumptions. For example, authors in [5] assume that traffic of only 37.5 GHz is requested every time, which reduces the complexity of their configuration selection.…”